Introduction

Linguistic coarseness: vulgarity & vulgarism

Vulgarity: intrinsically vulgar features

1The term “vulgarity” fundamentally appeals to judgment. Whether it be in relation to a lack of cultivation, morals, or knowledge of social etiquette, what is deemed vulgar necessarily involves perception. The online Oxford English Dictionary (2018) defines the term “vulgar” as:

1. Lacking sophistication or good taste.
2. Making explicit and offensive reference to sex or bodily functions […].

2The first definition suggests that what constitutes vulgarity may be indexed to norms that were taught to and fully integrated by individuals, though the scope of both “sophistication” and “good taste” may vary between social groups.

3Linguistically, “vulgarity” may refer to features that are perceived as vulgar because they deal with unrefined or taboo topics. This may be expressed through the use of specific lexical items as well as semantics. On the lexical level, swear words comprise a rich variety of terms which include nouns, verbs, adjectives or interjections (Montagu: 1967). They often refer to profanity and socially taboo subjects (Ljung: 2011). Virtually every speaker of the speech community is aware of the intrinsic vulgarity of such words. The semantics of an utterance may be regarded as intrinsically vulgar as well. What is considered vulgar in this case is the meaning of the utterance, but not necessarily the lexical items used. Though lexemes expressing vulgarity generally imply vulgar semantics (“fuck” for example may refer to sexual intercourse), it is possible to refer to vulgar topics without vulgar lexical items (for example discussing bodily functions without cursing).

Vulgarism: social perception of linguistic variations

4The word “vulgar” also has folk-linguistic implications. The term itself can be traced back to “Vulgar Latin,” a non-standard form of Classical Latin. Interestingly, “Vulgar Latin” refers to both a less complex form of Latin, and one spoken by the common people. Herman (2010, 7) defines it as:

[…] the set of all those innovations and trends that turned up in the usage, particularly but not exclusively spoken, of the Latin-speaking population who were little or not at all influenced by school education and by literary models.

5What was then perceived as an unrefined form of Latin was therefore linked to the fact that it was spoken by a certain category of speakers that could threaten established linguistic norms. Posner (1996, 98) states that it was believed that Vulgar Latin would eventually imply “[…] decay from God-given complexity and uniqueness towards animal-like communicative simplicity.” This implies that the term ‘vulgar’ not only referred to what was believed to be inferior linguistic practices, but also to the fact that those inferior practices were perceived as such precisely because they were associated to specific language users: the common people.

6Discussing vulgarity and linguistics indeed entails that linguistic features as well as social factors be analyzed simultaneously. It is therefore useful to turn to what lexicographer Henry Wyld called “vulgarism” (1914, 139). This notion refers to linguistic forms that are deemed vulgar in a social context, precisely because they deviate from the norm. He defines vulgarism as (quoted in Crowley: 2003, 150):

[…] a peculiarity which intrudes itself into Standard English, and is of such a nature as to be associated with the speech of vulgar or uneducated speakers.

7Wyld seems mainly concerned with phonological variations. He establishes a clear hierarchy between Standard English, “the ‘best’ type of Spoken English” to what he calls the Modified Standard, “the various vulgar forms […] heard among the inferior ranks of the population” (ibid., 236). He provides an example of a phonemic variation with the non-standard pronunciation of the word “soot” as [sat] (ibid., 139). Talking about Standard English, or any form of Standard Language, is somewhat of a fallacy, though. What Wylde fails to mention is that Standard English is only a variety of English, a dialect (Trudgill: 1999) which is considered as the standard because it happens to be used by speakers who are associated to education, central government, and other institutions of national or global power (Eckert: 2011, 57), what Lippi-Green calls the “dominant bloc” (1994, 167). The fact that Wyld makes a subjective appreciation of pronunciation is rather unremarkable, as the existence of a link between linguistic variation and social perception has been suggested by both sociologists (Bourdieu & Boltanski: 1975, 15; Fairclough: 1989) and linguists (D’Onofrio: 2016; Eckert: 2008; Labov: 1966; Podesva: 2013; Preston: 1996).

8Though not explicitly mentioned by Wyld, it is argued here that the concept of vulgarism may also apply to prosodic variations, and that some may be deemed vulgar because they are associated to female speakers.

Female speech as vulgar speech

Stigmatizing female speech

9Linguistic norms are generally shared by speakers of a given speech community. They are taught which forms are “proper” and which are “incorrect,” and these norms are interiorized. It has been shown that speakers have no difficulty in stating what local varieties of English feature ‘good’ or ‘bad’ pronunciation (Preston: 1996). Though speakers can readily judge a variety as non-standard, they might actually use it themselves. Labov interviewed New-Yorkers who believed their pronunciation was non-standard, and referred to the negative evaluations of one’s own local pronunciation as “linguistic self-hatred” (1966, 329).

10Stigmatized ‘accents’ are not only regional though. Varieties that are associated to either racial minorities or (supposedly) less educated speakers are also more likely to be stigmatized (Siegel: 1999). The same can be said for linguistic features which are associated, rightly or wrongly, to female speakers.

11Robin T. Lakoff first suggested that female speakers shared what she called “women’s language” (1972, 81). This language is said to feature a greater number of some lexical items (such as the adjective ‘divine’) or syntactic constructions (like hedges). It has also been shown that female speakers’ voices may feature acoustic-phonetic specificities, such as higher fundamental frequencies (correlated to the perception of pitch) than male voices when pronouncing the same vowel (Hillenbrand et al.: 1995). Studies show that listeners therefore tend to categorize people’s voices as female or feminine because of fundamental frequency (Munson: 2007) or vowel formants (Johnson: 2006).

12‘Female’ speech also tends to elicit negative folk-linguistic interpretations (McConnell-Ginet: 1975; Romaine, 1999). It has been suggested that this is because linguistic standards were designed by men (Spender: 2001) which implies that women’s way of speaking are assessed in relation to “androcentric” norms (Coates: 2016, 83):

The Androcentric Rule […] predicts that commentators will describe the linguistic behaviour of men as “normal” and the linguistic behaviour of women as deviating from that norm.

13In other words, some features can be negatively perceived because they are associated to female speakers. Labov (2001) has shown that young women are among the most innovative speakers; it is them who contribute to language change the most, but at the same time, language change generally induces stigma (D’Arcy: 2007):

[…] ongoing language change is often met with derision. […] it typically results in the characterization of new forms as sloppy, lazy, ignorant, or vulgar.

14Young women are therefore both more likely to use innovative linguistic features and be stigmatized by the rest of the speech community. The following section explores the case of two stigmatized prosodic features associated to female speakers.

‘Female’ prosodic vulgarisms

15Two suprasegmental features are accounted for: the High Rising Terminal contour, also called ‘uptalk,’ and creaky voice quality, also sometimes referred to as ‘vocal fry.’

16The High Rising Terminal contour (henceforth: HRT) has been extensively studied. It is generally construed as a rise in fundamental frequency at the end of a tone unit boundary or propositional content, or as “a movement in pitch from relatively low to relatively high” (Crystal: 1991). Descriptions may vary among researchers though, which makes it difficult to assign a stable definition (Di Gioacchino & Crook Jessop: 2010). Though it is not entirely clear when and where uptalk originated (Warren: 2016, 103) the main theories point to Australia (Benton: 1966; Guy et al.: 1986) and New Zealand (Holmes & Bell: 1996), though some press articles also mention California (Gorman: 1993). HRT users are said to be relatively young (Horvath: 1985; Bradford: 1997, 29; Fletcher & Harrigton: 2001). The first notable discussion of HRT regarding gender was probably Lakoff’s, who argued the contour was part of ‘woman’s language’ (1972, 49-50):

There is a peculiar intonation pattern, found in English as far as I know only among women, which has the form of a declarative answer to a question, and is used as such, but has the rising inflection typical of a yes-no question […].

17This ‘woman’s language’ theory is said to have launched language & gender studies in the United States in the 1970s (Eckert & McConnell-Ginet: 2003, 158), but some of Lakoff’s claims, like the one concerning HRT being exclusively female and powerless, have been challenged. In an early review of Lakoff’s word, McConnell-Ginet (1975) noted that HRT may have multiple social meanings, including a willingness to continue to speak. McLemore (1991), who studied sorority sisters in Texas, found that HRT was used by leaders of the group in the same way, suggesting that the contour does not necessarily signal powerlessness in all contexts. As far as speaker gender distribution is concerned, studies suggest that female speakers tend to use the contour more (Barry: 2007; Bradford: 1997). The press has also stated that it is a female linguistic feature (Gorman: 1993). HRT is also extremely prone to being stigmatized (Cameron: 2001, 112), especially in mainstream media (Warren: 2016, 129-149) where it may be referred to as an “epidemic” (Davis: 2010). Besides, when it is stigmatized, attention may be drawn to the fact that it is supposedly a female feature (Eckert: 2003, 393-394; McConnell-Ginet: 1978, 557). The ‘Valley Girl’ fad, which originated in California, is a good example of how young female speakers catalyzed negative evaluations for using this prosodic feature. Valley Girls were supposedly vain, unintelligent, materialist young women, first described by mass media as “clothes-crazy upper-middle-class girls” (Alexander: 1982), and were partly ridiculed and stigmatized for using HRT (Demarest: 1982; Eckert & McConnell-Ginet: 2003, 175-176).

19The extent to which male and female speakers use creaky voice has been studied, though no universal linguistic principle has emerged. It was first predominantly found in the speech of males. In a study of Edinburgh speakers, Esling (1978, 80-81) reported a higher incidence of creaky voice in males belonging to the higher class. Stuart-Smith (1999) as well as Henton & Bladon (1988) also argued that male speakers used creaky voice more than females in their samples of British speakers. More recently, some studies have found no significant difference in the use of creaky voice between genders (Hildebrand-Edgar: 2014, 51; Nicastri et al.: 2014). Others have claimed that creaky voice may primarily be used by specific groups of female speakers, including young educated American and Japanese speakers (Yuasa: 2010) and African American & White Washingtonians (Podesva & Sinae: 2010). Despite the on-going research within the linguistic community, the current perception of creaky voice use seems to be that it is a predominantly female prosodic contour. This belief has been spread by a vast number of press articles (Doctorow: 2011; Flanagan: 2011; Chen: 2011; Macrae: 2016; Wolf: 2015) to the point that other articles in mainstream media have attempted to show than men can indeed produce creaky voice when speaking as well (Saxena : 2015; Dries : 2015). Interestingly, like HRT, creaky voice has also been explicitly associated to Valley Girls (Warner: 2015).

20Both HRT and creaky voice stigmatization is overt and has ‘real-life’ consequences on female individuals. Some employment websites specifically advise speakers not to use them in job interviews (Diresta: 2018). Studies have also shown that using such features might undermine the success of women in finding employment (Gill: 1993; Anderson et al.: 2014). The arguments that lead to the stigmatization of both HRT and creaky voice are based on intrinsically subjective appreciations, which rely on “evaluative norms” (Moreau: 1997). These arbitrary norms connect esthetic, affective or moral values to linguistic forms so that standard forms are perceived as beautiful and elegant, and stigmatized forms as dysphonic, lazy, and vulgar. For example, HRT may be construed as a lack of assertiveness or self-confidence, or request for approval (Davis: 2010; Wolf: 2015). Creaky voice may be perceived as a laughable speech impediment [sic] (Macrae: 2016), or as ducks quacking (Wolf: ibid.) The fact that these stigmatizing appreciations are socially constructed within a specific linguistic context is made apparent when considering tonal languages. In such languages, HRT and creaky voice are essential linguistic building blocks insofar as they help differentiate between the same words (i.e. between the same phonemic realizations). The same word pronounced with or without HRT in Mandarin Chinese, or with or without creaky voice in Vietnamese, Burmese, or Danish would change its meaning. This suggests that the perception of linguistic markers is culturally construed. Though, as has been mentioned, it cannot be asserted that HRT and creaky voice are linguistic markers that are either exclusively or predominantly used by female English speakers across the board, these markers may be folk-linguistically perceived to be female-specific. Since both markers also may be heavily stigmatized, it could be possible that the stigma attached to them might be partially caused by the perceived femaleness of these features, which might be regarded as a form of linguistic misogyny.

21The present study explores the idea that prosodic features like HRT and creaky voice, because both may be regarded as female-specific vulgarisms, may be used to create a persona of a stereotypically unintelligent vulgar female character in a fiction television series. The use of these markers to construct such a character is not specific to the show under study: Parks & Recreation. A similar analysis could be done on other corpora as well. This particular episode of this particular series is meant to be taken as an example of a wider trope consisting of using linguistic markers that are both stigmatized and perceived as feminine in order to create a female character precisely meant to be stigmatized.

Material & method

Corpus & context

22The corpus is composed of scenes taken from an episode of NBC’s hit comedy series Parks and Recreation (2009-2015), created by G. Daniels & M. Schur and starring Amy Poehler, among others. The series centers on city hall employees working for the fictional town of Pawnee, Indiana. Leslie Knope (Poehler, whose performance won her a Golden Globe in 2014) is the hugely optimistic and idealistic deputy director of the Pawnee Parks and Recreation Department, and is dedicated to making her hometown a better place, though the over-the-top devotion she shows is often directed at small-scale goals. For example, the character is seen fighting bureaucracy and red-tape for two entire seasons to turn a construction site into a park. Leslie is also a feminist, which is repeatedly stated during the course of the series. She is best friends with Ann Perkins (Rashida Jones), and the first two seasons focus on the development of their friendship. Part of what makes Parks & Recreation’s situation comedy work is that Leslie’s optimism is counterbalanced by other characters, such as Ron Swanson (Nick Offerman), a libertarian who, despite working for it, despises the government as well as taxpayers. Though most of them have evolved over the course of the series, their core traits have remained the same.

23The comedic power of the show also partly relies on the fact that it is filmed as if it were a documentary à la The Office. Shots regularly include optical zooms, pans, and may also be slightly unsteady as to remind the audience that the cameras are hand-held, much like an actual documentary. The fictional camera crew members, though never seen, are fully integrated in the diegesis of the show. The characters are all aware that they are being filmed and occasionally give interviews to the crew, apart from other characters. These interviews are usually moments when they comment on an event that just occurred and discuss how they feel about it.

24The episode under study is entitled Doppelgängers [S06xE04] (Karas: 2013). The runtime is, like other episodes of the show, 22 minutes. The premise of the episode is that the town of Pawnee is merging with another: Eagleton. Pawnee employees, the main characters of the show, meet their Eagletonian counterparts (seen for the first time). Only one of each pair may keep their job due to budget cuts. Analysis focuses on interactions between a pair of female characters: April & Tynnyfer. April (series regular Aubrey Plaza) is an extremely stern and bored Pawnee employee. The presentation of the character on the website of the show (NBC website: 2018) mentions that she is a “sullen assistant.” Tynnyfer, a newcomer Eagletonian (guest star June Diane Raphael) is described as an “over-the-top parody of a vapid, entitled idiot” (Wilkins: 2013). Though April dislikes Tynnyfer’s attitude and personality, she pretends to be friends with her to trick her into leaving Pawnee so she can keep her job. The interactions between the two drastically different characters are meant to highlight Tynnyfer’s cluelessness. What is argued in this paper is that this is primarily done with linguistic features, and that prosody in particular is heavily used to stigmatize this female character. As previously stated, using ‘female’ linguistic markers to frame a female character as unrelatable is a phenomenon that is neither specific nor limited to this particular show. The episode discussed here is meant to demonstrate how such a phenomenon may occur. The narrative arc under study serves as a B story in the episode. April has a minute and a half worth of screen time, and Tynnyfer under a minute, yet this specific episode was chosen because on top of the traditional use of lexical or syntactic markers, Plaza’s performance also heavily relies on prosodic features which are immediately recognizable by the audience and frame Tynnyfer as a vulgar character.

Method

Why study media representations?

25It has been suggested that mass media do not determine “individual views” (Agha: 2003) and that it should not be assumed that “the dominant ideological meanings presented through television programs have immediate and necessary effects on the audience” (Morley & Brunsdon: 1999, 292). Yet, television programs and other forms of media representations may influence and shape viewers’ attitude. As far as gender is concerned, Tuchman (2000) suggested that the mass media both reflect dominant values, such as traditional views of sex roles (boys should work, girls should not), and teach those values to youngsters. A similar point was made by Ward & Harrison (2005) in their meta-analysis of 129 papers dealing with exposure to television and magazines. They found that greater and more frequent media exposure to stereotypical gender attitudes leads girls to believe in more traditional gender roles. Exposure to media also contributes to shaping how they view themselves, the satisfaction of their bodies, as well as sexual behaviors.

26What about the influence of the media on language? The debate concerning whether media representations affect linguistic use is still on-going and “the role of media in processes of linguistic change is not yet fully understood” (Androutsopoulos: 2014, 3). The main stance in sociolinguistics seems to be that the media do not significantly influence how speakers use language (Labov: 2001, 228), or if they do, it is only marginally with the spread of lexical innovations or idioms (Trudgill: 2006). It has been suggested that since people do not generally talk back to their television screen, the way the use language is not affected by the other dialects they hear in the programs they watch (ibid.) Similarly, it has been pointed out that television programs cannot teach a child who has deaf parents how to speak (Chambers: 1998, 127). However, some dissenting opinions have emerged. In a study of Glaswegian vernacular English, Stuart-Smith et al. (2013) have shown that a television program can play a role in a phonetic sound change (th-fronting and l-vocalization), especially if viewers experience a psychological and emotional engagement with the show, though the role of television is “neither necessary nor sufficient for ‘causing’ these changes (ibid., 531).” Rice & Woodsmall (1988) also argued that novel words and phrases may easily be transmitted to pre-schoolers after having been exposed to television. Coupland (2014, 79) also stated that “Mass media have strong involvement in [vernacularization and sociolinguistic change].” In any case, the extent to which mass media influence how individuals speak (for a thorough discussion, see Stuart-Smith & Ota: 2014) is a question that should not obscure another important topic: how hearers perceive scripted language broadcast on mass media.

27Linguistic studies focusing on television dialogues are scarce because such talk is said to lack authenticity (Richardson: 2010, 14). It has nevertheless been argued that speech broadcast on mass media may to some extent reflect (but not mimic) real-life speech (Lakoff: 1972, 40). In any case, no matter how authentic dialogues are, they provide “metadiscursive messages about speech and accent” (Agha: 2003) and therefore reflect “dominant values and attitudes in the society” (Tuchman: 2000). Such language is therefore worth studying in the sense that it may tell about how linguistic features are perceived by the creators of a program (Rey: 2001, 138), and possibly the audience. Finally, it has also been suggested that actors rely on linguistic features such as lexicon or prosody in order to portray fictional characters (Richardson: 2010, 132; Buchstaller: 2014, 211).

Prosodic & discourse analysis

29Since Praat, like other voice analysis softwares, may not always accurately detect creaky voice (Toshinori Ishi et al.: 2008), creaked segments were first identified by ear, then observed with spectrogram representations and with pulse reports. This voice quality may indeed be seen by inspecting glottal pulses (Anderson et al.: 2014, 1) as well spectrogram representations “anywhere there is voicing” (Shaw & Crocker: 2015). Since creak is most pronounced on the middle of vowel sounds though (Gordon & Ladefoged: 2001, 6), a phonemic transcription of words was used in figures. Jitter and shimmer values were automatically calculated by Praat, and values were rounded to 2 decimal places. Figures compare speech samples of similar length (all between 0.5 to 0.7 second in length).

30Similarly to creaky voice, HRTs were first identified by ear, then observed with the ‘Show Pitch’ function of Praat.

31Discourse analysis focuses on the semantics, lexical and syntactic structures of utterances as well as how scenes are designed. In order to provide context, interactions between April and Tynnyfer were transcribed (appendix). In the transcription, creaky voice and HRT use are shown in bold and italics respectively. The scenes making up the corpus are freely available online (Karas: 2013).

Analysis

Presenting Tynnyfer as an unrelatable character

Syntax

32Since Parks and Recreation is a comedy series, it is meant to make the audience laugh, either with, or at characters. In the case of Tynnyfer, the latter is achieved with various processes. As will be shown, various linguistic devices are used to frame this character as unrelatable, and in some cases vulgar. Though Tynnyfer will at times be referred to as ‘being’ vulgar in the rest of the present paper, it should be stressed here that this persona is constructed, in part thanks to linguistics.

33Syntax is for example used to frame her as unrelatable. This is conveyed to viewers with the name of the character, which is in itself meant to be a joke. The very first exchange between April and Tynnyfer reads as follows:

April: I’m sorry, was your name Jennifer?
Tynnyfer: No it’s Tynnyfer with two Y’s. I used to be Jennifer, but then I decided to rebrand myself.

34This first encounter sets the tone of the relationship between the two women, as well as how the audience is meant to perceive Tynnyfer. April is supposed to represent the more relatable, sane character of the pair, whereas Tynnyfer is immediately parodied and stigmatized. Indeed, attention is drawn to her self-centeredness as she makes no less than six references to herself in this first line using two proper names (“Jennifer” and “Tynnyfer”), two self-referential pronouns (“I”), one reflexive pronoun (“myself”), and an anaphoric pronoun (“it”) referring to the noun phrase “your name”. Syntactically, Tynnyfer is both the subject and the object of the propositions. The verb “rebrand” also points to her consumerism as she refers to herself as a product. In other words, the syntax and semantics of the exchange immediately convey to viewers Tynnyfer’s key personality traits. She is to be construed as a valley-girl-like, clueless, materialistic, self-obsessed character. The audience is therefore unlikely to relate to her, even though she only pronounced a single line.

35Syntax is also used by the writers of the show in order to portray Tynnyfer as a character who over-reacts to what she hears. For example, after learning that April (allegedly) goes to a better spinning-class than her, Tynnyfer replies: 

Seriously you need to get me in [this spinning class], like, that’s a must must must.

36There is a single prediction in this sentence: [you/get me in this spinning class]. In the first clause, root modality is expressed with the modal phrase “need to,” which implies necessity, but the same predication is simply repeated using the anaphoric proform “that” as subject of the second clause and “must” as a nominalized modal in subject complement position. Modality is therefore expressed no less than four times in the utterance. The overall excitable nature of Tynnyfer’s personality is also reinforced by the use of the sentence adverb “seriously” which conveys an inflated sense of importance to the rather mundane propositional content.

Semantics

37Right after meeting her, both April and the audience are presented with facets of Tynnyfer’s vulgarity (which is, as previously mentioned, constructed by the creators of the show), making her even more unrelatable. This is primarily done thanks to the semantics of the utterances. The character indeed discusses taboo topics right after having met April. She for instance refers to prescription drugs for anxiety disorder: “Hang on. It’s Xanax o’clock.” Though this conversation topic may not necessarily be deemed ‘vulgar’ per se by all viewers, in the sense that only part of the audience may find it lacks sophistication or good taste, this subject matter nevertheless is socially unacceptable (Rosewarne: 2013, 131), particularly in a work environment. Tynnyfer also features more prototypical vulgarity when she makes explicit reference to bodily functions. When her supervisor Leslie asks whether she has kids, she states that she does not because “[she has] had so much rejuvenation that [she doesn’t] think a baby could get out of there if it tried.” Vulgarity is here overtly expressed because of the semantics of the utterance, and is also reinforced thanks to editing. After having pronounced this line, the camera shifts from Tynnyfer’s face to Leslie’s, whose baffled expression is the last thing the audience sees in this scene. Viewers are once again supposed to sympathize with ‘sane’ characters such as Leslie and meant to laugh at Tynnyfer’s vulgarity.

38During the same exchange, taboo words are used by both April and Tynnyfer. Because Parks and Recreation was broadcast in primetime on a main US network, these are relatively scarce and mild, but some are still featured, such as:

We also came up with these nicknames for each other: ‘slut’ and ‘skank’.

39Both words refer to a taboo topic: sex, and link vulgarity to womanhood, as both these words are female-specific. These are pronounced at the same time by the two women, reinforcing the impression that April completely accommodates to Tynnyfer’s linguistic practices, both in terms of word choice and, as will be shown prosody.

Enregistered lexical items

40Tynnyfer’s lack of sophistication also transpires in her use of enregistered lexical items. Enregisterment may be defined as (Agha: 2003):

[…] processes through which a linguistic repertoire becomes differentiable within a language as a socially recognized register of forms.

41It is argued that Tynnyfer uses lexical items that are part of a stigmatized enregistered lexicon yet again reminiscent of Valley Girls.

42Tynnyfer uses the word like as a quotative. When used as such, it is preceded by the verb be and introduces reported speech. This is the case in:

I was like, ‘Shut up. Where do I get that?’

43It has indeed been suggested that the be like quotative is typically associated to Valley Girls (Bucholtz et al.: 2007, 326; D’Arcy: 2017, 145). The same can be said for the maximizer totally, which has also been found to be perceived as such (Suh: 2011; Beltrama: 2016, 10). The marker can be found in Tynnyfer’s speech:

I totally think that you’re right.

44Interestingly, April is portrayed as being very aware of the enregisterment of these markers. She uses them as well, to covertly mock Tynnyfer as in:

We’ve just been like ‘blah blah blah blah blah blah blah!’
If you worked here, you’d be like, ‘Ugh!’ and Leslie would be like, ‘Blah, blah, blah’ and you’d be like, ‘Uh.’
Totally. Tynny and I have been, like, totally bonding.
Totes!

45April uses the be like construction to report speech that is literally semantically empty (“blah blah blah”). Totally is over-used (twice in the same sentence), then it is repeated shortly after and truncated, giving the impression that her speech is saturated with the marker. This has two consequences. It frames Tynnyfer as a valley-girl-like character, by playing on the enregistered value of the markers. At the same time, it presents April as witty as only she (and presumably the audience) is aware that Tynnyfer is being mocked.

Mise-en-scène

46The fact that Tynnyfer is not meant to be liked by viewers is expressed linguistically on the semantic, syntactic and lexical levels, but also because of the very design of the scenes in which she is featured. As has been said, characters sometimes give ‘interviews’ to an unseen camera crew, as the premise of the show is that the characters are aware they are being filmed. Only April is shown in such a situation though. The audience has therefore access to only her impressions and feelings, which are extremely negative towards Tynnyfer. When alone, April states to the camera, and therefore to the audience:

She’s the worst person I’ve ever met.

47The woman is presented as stupid as she is unaware of April’s true feelings and, unlike April, is not given a chance to speak directly to the audience, once again framing her as an unrelatable character meant to be laughed at. Making characters failing to “exhibit contextually sensitive behavior” is indeed a way to create exaggerated prototypes (Culpeper: 2001, 88-89), in this case a clueless Valley Girl. The same process is used when April invites Tynnyfer to a house, which actually does not belong to her but to basketball star Dwayne Wade. After tricking her into coming over, April states, apart:

It’s Dwayne Wade’s house. I got his address off the Internet. I really hope he’s there when she walks in and he throws a basketball at her head.

48Finally, Tynnyfer is also vulgar as she both refers to taboo topics and lacks good taste, which may also be suggested by the costume chosen to portray her: a snake-leather dress. Though April ironically mentions it is “so cute,” it seems rather out of place in a work environment, which was probably what the creators of the show wanted to put forward.

49The next section demonstrates how prosodic contours, which may in this context be construed as vulgarisms, are used to further frame the character as unrelatable. Prosody plays an important part in analysis since April accommodating to Tynnyfer serves as a running gag in the episode.

April’s (feigned) accommodation to prosodic vulgarisms

High Rising Terminal

50“Doppelgängers” is the first and only episode in which the character of Tynnyfer is featured. Neither April nor the audience is aware of who she is, and how she speaks. After hearing her speak for the first time though, April immediately picks up on her prosodic patterns and accommodates to them in order to ape the way she speaks. The process of accommodation may be defined as (Richards & Schmidt: 2010):

a theory that seeks to explain shifts in the style of speaking people make such as when a person changes their way of speaking to make it sound more like or less like the speech of the person they are talking to.

51April adopts a strategy called ‘convergence,’ meaning she adapts to Tynnyfer’s communicative style. An individual may converge “in terms of a wide range of linguistic-prosodic-nonverbal features including speech rate, pausal phenomena and utterance length, phonological variants, smiling, gaze, and so on […]” (Giles et al.: 1991, 7). According to accommodation theory, converging to a speaker’s linguistic practices reflects “in the unmarked case, a speakers’ or a group’s need (often unconscious) for social integration or identification with another.” (ibid., 18). It should be stated that in the context of the episode under study though, the character of April is seen converging, but with a parodic intent, meant to be picked up by viewers. This is why the process of converging accommodation at play is in this case somewhat perverted.

52The audience is used to April’s voice being rather dull and monotone. This voice quality will henceforth be referred to as April’s ‘regular voice’. This is the type of voice quality she typically uses in the series, the one she uses when speaking to Tynnyfer for the first time (before she hears her speak), as well as when she is alone during interviews. When imitating Tynnyfer, April also uses what will be called her ‘feigned accommodation voice,’ which heavily features aforementioned prosodic markers.

53April’s regular voice typically does not feature HRT, except in environments where it is expected, such as yes/no questions (Lakoff: 1972, 49-50). Consider for example:

She’s the worst person I’ve ever met. I want to travel the world with her.

Les paramètres nécessaires sont manquants ou erronés.

54These two sentences are pronounced during an interview. Tynnyfer therefore cannot hear April and no accommodation takes place. The two utterances pronounced are with her regular voice as they feature a continuous decrease in pitch (figure 1). “She’s the worst person I’ve ever met” starts with a frequency of about 250 Hz. and drops to 150 Hz. by the end of the utterance. The first half of the utterance “She’s the worst person—“ is pronounced with a frequency above the average of the utterance whereas the second half (“—I’ve ever met”) features a frequency consistently below average. Similarly, “I want to travel the world with her” goes from 190 Hz. to 130 Hz.

55Tynnyfer on the other hand immediately features HRT. It is present in one of the first lines she utters: “It used to be Jennifer —”, which features a 40 % increase in frequency (figure 2).

56This use of HRT is typical of what might be considered a token of the character’s vulgarism, since it is featured in sentence-medial position. HRT is here attitudinal, as there is indeed no requirement for its use in this context, as the sentence is not a question, neither syntactically (no subject/auxiliary inversion) nor semantically (the character does know her own name, which is precisely the topic at stake). By using HRT in this context, the actress portraying Tynnyfer therefore relies on the enregisterment of HRT as a Valley Girl feature in order to convey to viewers that the character is a vain, materialistic, stupid character. She uses HRT shortly after in “I saw my spinning instructor wearing it” which has a 27 % increase in fundamental frequency.

57April first accommodates to Tynnyfer’s way of speaking after a 4.3-second pause, which is an extremely long time span. For reference, the amount of time between April’s lines and Tynnyfer’s (or vice versa) is on average 0.32 second (standard deviation: 0.90) in the rest of the episode. This abnormally long silence is meant to provide a cue to viewers that Tynnyfer’s persona is going to be lampooned by April. She indeed first smiles after hearing about Tynnyfer’s “rebranding” and HRT use. She sits on a table while touching her hair, which might signal to the viewer that she picked up on Tynnyfer’s egotism and vulgarisms. She then starts using her feigned accommodation voice, and features HRT (shown in italics) in sentence-medial position, like in the first sentence she says with her feigned accommodation voice:

Um, well, nice to meet you. My name’s April, and I just wanted to say that your dress is so cute it’s bonks.

Les paramètres nécessaires sont manquants ou erronés.

58As can be seen on a spectrogram (figure 3), HRT is featured no less than five times in this short utterance. Fundamental frequency increase is observed in the words “you” (+126 %), “April” (+222 %), “say” (+34 %), and “cute (+28 %). What suggests that this is a case of accommodation, and not simply a fluke, is that April never uses HRT with her regular voice in this episode (excluding yes/no questions) but does so 7 times with her feigned accommodation voice, which is much closer to Tynnyfer’s own use: 5 times (table 1, below).

59HRT is therefore used as linguistic tool with a dual purpose. It first signals to the audience that Tynnyfer is a stereotypical clueless airhead thanks to the fact that it is enregistered as a Valley Girl feature. Besides, it is also used to stigmatize Tynnyfer by having April use it in her own speech as well. The fact that Tynnyfer does not pick up on this linguistic change, as opposed to the audience, who knows April is actually imitating Tynnyfer, reinforces the impression that the audience is meant to laugh with April at Tynnyfer for using HRT.

Creaky voice

60April’s regular voice usually features the modal register, in which the vibrations of her vocal folds are regular. This can be seen on the words “I ever met” (figure 4) in which glottal pulses are regular on the waveform (above) and on the spectrogram (below).

Les paramètres nécessaires sont manquants ou erronés.

61Tynnyfer on the other hand heavily uses creaky voice. She does so virtually every time she speaks. For example, it is used when Tynnyfer first pronounces her name (figure 5).

Les paramètres nécessaires sont manquants ou erronés.

62In “Tynnyfer,” the first two syllables show regular pulses (shown by blue bars), but the last syllable, [fɚ:], is creaked and therefore features more erratic ones. The spectrogram also shows distinct vocal vibration (shown by black bars) on the syllable, which is also the longest of the word. It is indeed drawled and represents 55 % (0.3787 seconds) on the total length of the word, which is expected in creaky voice (Gordon & Ladefoged: 2001, 6). The same creaky voice is used when the woman says her name is spelled with two Y’s (note the irregular pulses in [aɪz]).

63Just like HRT, April accommodates to Tynnyfer’s use of creaky voice. After hearing her speak, April’s feigned accommodation voice heavily features creak. She uses this voice quality when saying her own name “April” or on the word “dress” (figure 6).

Les paramètres nécessaires sont manquants ou erronés.

64Both the waveform and the spectrogram suggest creak is used as pulses are irregular on the waveform and can be clearly seen individually on the spectrogram. Interestingly, the word “April” is uttered with both creaky voice and HRT. This is not prototypical, but creak has been shown to occur in environment where fundamental frequency is relatively high (Laver: 1980, 126). The fact that these two prosodic features are combined is further proof that April’s feigned accommodation voice is actually used to lampoon Tynnyfer’s. If all scenes featuring the two women are taken into account, it appears that April’s use of creak is indeed part of a converging accommodation strategy, like her use of HRT. April’s feigned accommodation voice is more similar to Tynnyfer’s than to April’s regular voice in terms of jitter and shimmer (table 1). This suggests that the creators of the show as well as both actresses, rely on the fact that creaky voice may be construed as a stigmatizing enregistered prosodic feature.

Table 1: Prosodic contours report

HRT
	April’s regular voice	April’s feigned accommodation voice	Tynnyfer
Number of HRTs in the episode (excluding questions)	0	7	5
Creaky voice
Jitter (local, in percent)	1.48	2.12	2.05
Shimmer (local, in percent)	8.72	11.49	10.91

Other suprasegmental features: fundamental frequency & intensity

65Besides HRT and creaky voice, April also accommodates to Tynnyfer by using other linguistic features. She indeed slightly changes her use of fundamental frequency (ƒ0) or the intensity with which she pronounces her lines.

67The same accommodation process occurs with intensity. April, when using her regular voice has an average intensity of 57 dB. (table 2). Tynnyfer’s voice has a lower intensity on average, with 49 dB. When using her feigned accommodation voice, April’s intensity also lowers, and drops to 54 dB. Similar to ƒ0 use, the results of the T test unsurprisingly suggest that both versions of April’s voice are significantly different from Tynnyfer’s (p<0.00001), and that the difference between April’s regular voice and Tynnyfer’s is stronger than the difference between April’s feigned accommodation voice and Tynnyfer’s in terms of intensity.

68The general decrease in ƒ0 and intensity featured in April’s feigned accommodation voice is probably due to the fact that April uses creaky voice ubiquitously when lampooning Tynnyfer, which is a voice quality that, as stated previously, prototypically occurs in the lower frequencies (Keating et al.: 2015), and which can therefore also be expected to be quieter.

Table 2: Fundamental frequency & intensity report

Fundamental frequency (ƒ0)
	April’s regular voice	April’s feigned accommodation voice	Tynnyfer
Mean ƒ0 (Hz.)	220.83	211.60	191.77
Standard deviation	55.74	55.07	49.01
Differences between voices in terms of ƒ0
	April’s regular voice & Tynnyfer	April’s feigned accommodation voice & Tynnyfer
Two-tailed student’s T test (p value)	<0.00001	<0.00001
Effect size (Cohen’s d)	0.55	0.38
Intensity (dB)
	April’s regular voice	April’s feigned accommodation voice	Tynnyfer
Mean intensity (dB)	57.04	53.74	49.02
Standard deviation	9.84	10.66	14.82
Differences between voices in terms of intensity
	April’s regular voice & Tynnyfer	April’s feigned accommodation voice & Tynnyfer
Two-tailed student’s T test (p value)	<0.00001	<0.00001
Effect size (cohen’s d)	0.64	0.37

Discussion

69As has been shown, framing Tynnyfer as a clueless, unrelatable character is achieved thanks to a variety of linguistic markers, including syntax, semantics, enregistered lexicon and enregistered prosodic features. Prosody is nevertheless an extremely important linguistic tool, which is heavily used to imitate or parody female voices. It has for example been shown that male imitations of female voices rely more heavily on pitch variations than on syntax or vocabulary (McConnell-Ginet: 1975, 47-48). HRT and creaky voice are extremely important when taking vulgarity into account as these markers are the very first to signal Tynnyfer’s vulgar persona. Indeed, as soon as April starts accommodating to Tynnyfer’s prosody she says: 

Well, nice to meet you. My name’s April and I just wanted to say […]

70These utterances do not rely on either syntax, lexicon, nor semantics in order to make the joke understood to the audience. If one were only to read this stretch of discourse, it would be difficult to grasp the underlying implications at stake. Yet the viewer is aware that April is making fun of Tynnyfer because these apparently neutral utterances both feature HRT and creak. This proves that prosody is as important as vocabulary and semantics in creating a stereotypical female character.

71It has also been said that April does not generally use HRT or creak in the rest of the series, and a similar remark could be made about actress June Diane Raphael, who portrays Tynnyfer. The guest star indeed does not use either HRT nor creak systematically in sentence-final position (as she does in the scenes under study) in talk show interviews such as the one she gave on The Late Show with Stephen Colbert (Colbert: 2018). It is therefore safe to say that these inflections were used specifically to create a persona in situ thanks to these linguistic markers.

72The fact that Tynnyfer is a female character is also probably not a coincidence. Her vulgarity and cluelessness are not only reminiscent of the Valley Girl stereotype (Ploschnitzki: n.d.) but also of a broader attack on women, as exemplified with “blond jokes” (Shifman & Lemish: 2012, 89):

First, blonde jokes are about blonde females. […] Second, scholars agree that stupidity and promiscuity are the two main features constructing the archetypal blonde joke.

73Besides being clueless, Tynnyfer is a promiscuous character indeed, as the words “slut” implies. Having her use HRT and creak implies that the actress relies on a stereotype directed specifically at women, which is why it is argued this might be a form of linguistic misogyny. Though the linguistic male equivalent to Valley Girls exists, the “Surfer Dude” (Fought: 2005), Valley Girls have been more stigmatized than the ‘dudes’ as their linguistic practices were immediately both criticized and deemed feminine (Eckert: 2003, 393-394). The same double standard still holds true today, as creak is perceived more negatively in female voices than in males’ (Anderson et al.: 2014).

74It should be stressed that it is by no means argued that Parks and Recreation is a misogynistic show, though. It has conveyed over the seven seasons it was broadcast what mainstream media described as feminist ideals (Ryan: 2015; Galo: 2015) as it depicted strong, sensitive, relatable female leads. Instead, it is suggested that, as a stock character (Baldick: 2008, 317), Tynnyfer’s portrayal relies on social conventions. Since the portrayal of stock characters, particularly when they are minor characters, tends to be perceived as both natural and inevitable (Fuller & Loukides: 1990, 4), Tynnyfer’s persona may be constructed linguistically with enregistered features that simultaneously convey both the character’s vulgarity and femaleness. Portraying women negatively by stressing their lack of intelligence is not uncommon. Television shows and movies are indeed no short of stereotypical intellectually-challenged female characters to the point that these representations may for example be listed under the ‘Valley Girl’ and ‘Dumb Blonde’ tropes (TV Tropes: 2019) both of which include dozens of examples. Though it does partly rely on a tongue-in-cheek use of such tropes in this particular episode, the show makes use of them sparingly and, generally, never does the show feature entire narrative arcs centered on making fun of female characters because of their gender. The portrayal of this stereotypical supporting female character (only seen in this single episode of the series) is therefore not part of a broader problematic representation of female characters on this show. It should nevertheless be noted that though Parks & Recreation does feature a main character who is presented as a feminist (Leslie Knope), which is rather unusual in mainstream American TV shows, the fact that female-specific tropes are used in the series might also shed light on the ambivalence of the show’s message. Despite the feminist ideals the series seems to convey, the fact that it also relies on a stereotype of an intellectually-challenged female character might have some viewers wonder whether it actually is as progressive as it first seemed to be.

Conclusion

75Though what is deemed vulgar depends on who is asked, it might be relatively easy to detect in comedy television shows, meant to entertain viewers with flawed characters. As a mass medium, television may indeed rely on the lowest common denominator for comedic purposes: stereotypes—including linguistic stereotypes. By relying on them, it contributes to their spread and continual enregisterment. Portraying Tynnyfer as vulgar and unrelatable is done with various non-linguistic elements, such as garments or mise-en-scène, but this is also achieved thanks to linguistic features. Specific lexical items as well as the semantics of utterances are used to depict the character’s vulgarity, by having her discuss socially frowned-upon topics with taboo words. Enregistered lexical and prosodic features are also subtly used to appeal to shared cultural stereotypes. Live action television is indeed a medium that requires that language be pronounced and heard, not read. As has been shown, prosodic vulgarisms can convey the character’s vulgarity, similarly to lexicon or syntax, with supposedly ‘female’ markers. HRT and creaky voice are therefore meant to let viewers know that Tynnyfer is vulgar and female, and also more specifically that she is a vulgar woman, since what is considered vulgar is gender-specific. The vulgarity of the character is indeed intrinsically linked to her femininity. This is made explicitly clear when she mentions her vagina rejuvenation to her boss-to-be, Leslie. It is argued that the same phenomenon occurs more subtly thanks to the linguistic portrayal of the character. The semantics of the utterances pronounced by Tynnyfer, which make the character unrelatable, cannot be dissociated from the vulgarisms that are consistently used in her portrayal: creaky voice and HRT. Because these enregistered linguistic markers can be both stigmatized and perceived as intrinsically feminine, the ideology referred to in this paper as ‘linguistic misogyny’ is used to frame this female character as vulgar. An ideology, be it linguistic or otherwise, is defined as “a systematic body of ideas, organized from a particular point of view” (Hodge: 1979, 6). In the case of linguistic misogyny, the point of view is definitely androcentric as the linguistic behavior perceived to be female-specific is othered and ridiculed. It is not argued here that the systematic use of HRT and creaky voice is necessarily common in media representations of women, and, to the author’s knowledge, no extensive media representation analysis (of an entire TV series for example) concerning the use of such markers exists. The analysis provided here is solely to be taken as proof of a phenomenon, and no argument is made concerning the magnitude of said phenomenon. The goal of this paper is also not to discuss whether this form of linguistic misogyny is an effective humor device or not (i.e.: is it actually funny or not). What should be stressed instead is that the way viewers feel about certain female characters (and possibly certain actual female speakers) may be rooted in misogynistic linguistic norms. Indeed, humor may be used in mainstream media in order to subvert and reinforce sexist ideologies according to Coulomb-Gully (2012), who mentions a ‘willing suspension of ethics’.

76In any case, what linguistic markers a hearer considers masculine or feminine, refined or lazy, elegant or vulgar, is indeed the result of a linguistic ideology so well-entrenched that it seems common sense. It is not. Commonly-held beliefs about linguistic phenomena are by no means neutral. To quote Giles & Niedzielski (1998, 91):

[…] sounds are in the ear of ear beholder, to be variably interpreted and socially constructed, rather than ‘out there’ as some fact to be objectively measured.

77Speakers, when they are irritated by someone’s voice, regional or social ‘accent,’ or any linguistic feature, should question their own beliefs and how they have come to perceive a stretch of discourse this way.