Playin’ Native and Other Iterations of Sonic Brownface in Hollywood Representations of Dolores Del Río

Mexican actress Dolores Del Río is admired for her ability to break ground; her dance skills allowed her to portray roles not offered to many women of color early in the 20th Century. One of her most popular roles was in the movie Ramona (1928), directed by Edwin Carewe. Her presence in the movie made me think about sonic representations of mestizaje and indigeneity through the characters portrayed by Del Río. According to Priscilla Ovalle, in Dance and the Hollywood Latina: Race, Sex and Stardom (2010), Dolores del Río was representative of Mexican nationhood while she was a rising star in Hollywood. Did her portrayals reinforce ways of hearing and viewing mestizos and First Nation people in the American imaginary? Is it sacrilegious to examine the Mexican starlet through sonic brownface? I explore these questions through two films, Ramona and Bird of Paradise (1932, directed by King Vidor), where Dolores del Río plays a mestiza and a Polynesian princess, respectively, to understand the deeper impact she had in Hollywood through expressions of sonic brownface.
Before I delve into analysis of these films and the importance of Ramona the novel and film adaptions, I wish to revisit the concept of sonic brownface I introduced here in SO! in 2013. Back then, I argued that the movie Nacho Libre and Jack Black’s characterization of “Nacho” is sonic brownface: an aural performativity of Mexicanness as imagined by non-Mexicans. Jennifer Stoever, in The Sonic Color Line (2017), postulates that how we listen to particular body(ies) are influenced by how we see them. The notion of sonic brownface facilitates a deeper examination of how ethnic and racialized bodies are not just seen but heard.
Through my class lectures this past year, I realized there is more happening in the case of Dolores del Río, in that sonic brownface can also be heard in the impersonation of ethnic roles she portrayed. In the case of Dolores del Río, though Mexicana, her whiteness helps Hollywood directors to continue portraying mestizos and native people in ways that they already hear them while asserting that her portrayal helps lend authenticity due to her nationality. In the two films I discuss here, Dolores del Río helps facilitate these sonic imaginings by non-Mexicans, in this case the directors and agent who encouraged her to take on such roles. Although a case can be made that she had no choice, I imagine that she was quite astute and savvy to promote her Spanish heritage, which she credits for her alabaster skin. This also opens up other discussions about colorism prevalent within Latin America.
First, let us focus on the appeal of Helen Hunt Jackson’s Ramona (1884), as it is here that the cohabitation of Scots, Spanish, Mestizos, and Native people are first introduced to the American public in the late 19th century. According to Evelyn I. Banning, in 1973’s Helen Hunt Jackson, the author wanted to write a novel that brought attention to the plight of Native Americans. The novel highlights the new frontier of California shortly after the Mexican American War. Though the novel was critically acclaimed, many folks were more intrigued with “… the charm of the southern California setting and the romance between a half-breed girl raised by an aristocratic Spanish family and an Indian forced off his tribal lands by white encroachers.” A year after Ramona was published, Jackson died and a variety of Ramona inspired projects that further romanticized Southern California history and its “Spanish” past surged. For example, currently in Hemet, California there is the longest running outdoor “Ramona” play performed since 1923. Hollywood was not far behind as it produced two silent-era films, starring Mary Pickford and Dolores del Río.
The charm of the novel Ramona is that it reinforces a familiar narrative of conquest with the possibility of all people co-existing together. As Philip Deloria reminds us in Playing Indian (1999), “The nineteenth-century quest for a self-identifying national literature … [spoke] the simultaneous languages of cultural fusion and violent appropriation” (5). The nation’s westward expansion and Jackson’s own life reflected the mobility and encounters settlers experienced in these territories. Though Helen Hunt Jackson had intended to bring more attention to the mistreatment of native people in the West, particularly the abuse of indigenous people by the California Missions, the fascination of the Spanish speaking people also predominated the American imaginary. To this day we still see in Southern California the preference of celebrating the regions Spanish past and subduing the native presence of the Chumash and Tongva. Underlying Jackson’s novel and other works like Maria Amparo Ruiz De Burton The Squatter and the Don (1885) is their critique of the U.S. and their involvement with the Mexican American War. An outcome of that war is that people of Mexican descent were classified as white due to the signing of Treaty of Guadalupe Hidalgo and were treated as a class apart. (See Michael Olivas’ anthology, Colored Men and Hombres Aqui from 2006 and Ignacio M. García, White but not Equal from 2008). These novels and films reflect the larger dominant narrative of whiteness and its relationship to nation building. Through Pickford’s and del Río’s portrayals of Ramona they reinforce the whiteness of mestizaje.

Screenshot from RAMONA (1910). Here, Ramona (played by Mary Pickford) finds out that she has “Indian blood.”
In 1910, Mary Pickford starred in D.W. Griffith’s short film adaptation of Ramona as the titular orphan of Spanish heritage. Henry B. Walthall played Alessandro, the Native American, in brownface to mimic physical attributes of the native Chumash of Southern California. Griffith also wanted to add authenticity by filming in Camulas, Ventura County, the land of the Chumash and where Jackson based her novel. The short is a sixteen-minute film that takes viewers through the encounter between Alessandro and Ramona and their forbidden love, as she was to be wedded to Felipe, a Californio. There is a moment in the movie when Ramona is told she has “Indian blood”. Exalted, Mary Pickford says “I’m so happy” (6:44-6:58). Their short union celebrates love of self and indigeneity that reiterates Jackson’s compassionate plea of the plight of native people. In the end, Alessandro dies fighting for his homeland, and Ramona ends up with Felipe. Alessandro, Ramona, and Felipe represent the archetypes in the Westward expansion. None had truly the power but if mestizaje is to survive best it serve whiteness.
Edwin Carewe’s rendition of Ramona in 1928 was United Artists’ first film release with synchronized sound and score. Similar to the Jazz Singer in 1927, are pivotal in our understanding of sonic brownface. Through the use of synchronized scored music, Hollywood’s foray into sound allowed itself creative license to people of color and sonically match it to their imaginary of whiteness. As I mentioned in my 2013 post, the era of silent cinema allowed Mexicans in particular, to be “desirable and allowed audiences to fantasize about the man or woman on the screen because they could not hear them speak.” Ramona was also the first feature film for a 23-year-old Dolores del Río, whose beauty captivated audiences. There was much as stake for her and Carewe. Curiously, both are mestizos, and yet the Press Releases do not make mention of this, inadvertently reinforcing the whiteness of mestizaje: Del Río was lauded for her Spanish heritage (not for being Mexican), and Carewe’s Chickasaw ancestry was not highlighted. Nevertheless, the film was critically acclaimed with favorable reviews such as Mordaunt Hall’s piece in the New York Times published May 15, 1928. He writes, “This current offering is an extraordinarily beautiful production, intelligently directed and, with the exception of a few instances, splendidly acted.” The film had been lost and was found in Prague in 2005. The Library of Congress has restored it and is now celebrated as a historic film, which is celebrating its 90th anniversary on May 20th.
In Ramona, Dolores del Río shows her versatility as an actor, which garners her critical acclaim as the first Mexicana to play a starring role in Hollywood. Though Ramona is not considered a talkie, the synchronized sound comes through in the music. Carewe commissioned a song written by L. Wolfe Gillbert with music by Mabel Wayne, that was also produced as an album. The song itself was recorded as an instrumental ballad by two other musicians, topping the charts in 1928.
In the movie, Ramona sings to Allessandro. As you will read in the lyrics, it is odd that it is not her co-star Warner Baxter who sings, as the song calls out to Ramona. Del Río sounds angelic as the music creates high falsettos. The lyrics emphasize English vernacular with the use of “o’er” and “yonder” in the opening verse. There are moments in the song where it is audible that Del Río is not yet fluent speaking, let alone singing in English. This is due to the high notes, particularly in the third verse, which is repeated again after the instrumental interlude.
I dread the dawn
When I awake to find you gone
Ramona, I need you, my own
Each time she sings “Ramona” and other areas of the song where there is an “r”, she adds emphasis with a rolling “r,” as would be the case when speaking Spanish. Through the song it reinforces aspects of sonic brownface with the inaudible English words, and emphasis on the rolling “rrrr.” In some ways, the song attempts to highlight the mestiza aspects of the character. The new language of English spoken now in the region that was once a Spanish territory lends authenticity through Dolores del Río’s portrayal. Though Ramona is to be Scottish and Native, she was raised by a Spanish family named Moreno, which translates to brown or darker skin. Yes, you see the irony too.

United Artists Studio Film Still, under Fair Use
Following Del Río’s career, she plays characters from different parts of the world, usually native or Latin American. As Dolores del Río gained more popularity in Hollywood she co-starred in several movies such as Bird of Paradise (1932), directed by King Vidor, in which she plays a Polynesian Princess. Bird of Paradise focuses on the love affair between the sailor, played by Joel McCrea, and herself. The movie was controversial as it was the first time we see a kiss between a white male protagonist with a non-Anglo female, and some more skin, which caught the eye of The Motion Picture Code commission. Though I believe the writers attempted to write in Samoan, or some other language of the Polynesian islands, I find that her speech may be another articulation of sonic brownface. Beginning with the “talkies,” Hollywood continued to reiterate stereotypical representations though inaccurate music or spoken languages.
In the first encounter between the protagonists, Johnny and Luana, she greets him as if inviting him to dance. He understands her. He is so taken by her beauty that he “rescues” her to live a life together on a remote island. Though they do not speak the same language at first it does not matter because their love is enough. This begins with the first kiss. When she points to her lips emphasizing kisses, which he gives her more of. The movie follows their journey to create a life together but cannot be fully realized as she knows the sacrifice she must pay to Pele.
I do not negate the ground breaking work that Dolores del Río accomplished while in Hollywood. It led her to be an even greater star when she returned to Mexico. However, even her star role as Maria Candelaria bares some examination through sonic brownface. It is vital to examine how the media reinforces the imaginary of native people as not well spoken or inarticulate, and call out the whiteness of mestizaje as it inadvertently eliminates the presence of indigeneity and leaves us listening to sonic brownface.
—
reina alejandra prado saldivar is an art historian, curator, and adjunct lecturer in the Women, Gender and Sexuality Studies Program and Liberal Studies Department at CSULA and in the Critical Studies Program at CALArts. As a cultural activist, she focused her earlier research on Chicano cultural production and the visual arts. Prado is also a poet and performance artist known for her interactive durational work Take a Piece of my Heart as the character Santa Perversa (www.santaperversa.com) and is currently working on her first solo performance entitled Whipped!
—
Featured image: Screenshot from “Ramona (1928) – Brunswick Hour Orchestra”
—
REWIND!…If you liked this post, you may also dig:
Afecto Caribeño / Caribbean Affect in Desi Arnaz’s “Babalú Aye” – reina alejandra prado saldivar
SO! Amplifies: Shizu Saldamando’s OUROBOROS–Jennifer Stoever
SO! Reads: Dolores Inés Casillas’s ¡Sounds of Belonging! – Monica de la Torre
Speaking American

This is something you just don’t hear about in the history books.
–Alex Hutchinson, Creative Director, Assassin’s Creed III
What does it mean to speak American? According to Laura Amico, a teacher in the Cliffside Park school district in New Jersey, to speak American is the opposite of speaking Spanish. As suggested by this video, captured by students via Snapchat before they walked out in protest, to be American is the same as to speak American. Evidently, to speak American is to speak English, a historical curiosity, given that English was not spoken on the North American continent until the 17th century, preceded by many Native American languages and, ironically, Spanish. In simplest terms, “American English” is not a homogenous dialect either. What kind of “American English” does Laura Amico exalt as acceptable for patriotic conduct? How does an American patriot speak?

Colonel Tarleton, painted by Sir Joshua Reynolds, 1782, British National Gallery, CC license
In The Patriot (2000), Mel Gibson plays Benjamin Martin, a hero of the French and Indian War, who is forced out of his South Carolina plantation-managing life to avenge the deaths of his sons at the hands of the British during the American Revolution. The main British villain of that film, the fictional Colonel Tavington,–played in sneering style by Jason Isaacs—is notable for his barbaric conduct (among other things, he traps civilians in a barn and burns them alive). British audiences were noticeably aghast at this characterization, which implicitly compared British command of the time period to Nazi atrocities in World War II. Tavington may in part have been based on Banastre Tarleton, a notorious and brutal, misogynist dragoon whose victory at Monck’s Corner was among the bloodiest battles in the dirty war fought in the south at the end of the Revolution, or Lord Rawdon, a serial rapist and generally unpleasant English officer.
Nevertheless, Tavington’s barn-burning exploits are more likely based on American General John Sullivan’s slash-and-burn campaign against British-allied Mohawks, Senecas, Onondagas, and Cayugas, a subject little discussed in American history. This culminated in a 1782 massacre at Gnadenhutten in Ohio, where white Americans murdered 96 Delawares who had converted to the Moravian faith (pacifists). Evidently, in The Patriot, the conflict was not a civil war, and character was constructed in part through dialect (as detailed below, dialect here refers to all the clues in the way a person speaks, not just his or her accent).
Although Tavington’s conduct in The Patriot manifested visibly onscreen, his dialect said much about him. As noted in a recent Atlantic Monthly article, entertainment—such as Disney cartoons—can “take [bias] and pour concrete over it.” Rosina Lippi-Green, a linguist, notes that such biases are then difficult to remove: “They etch it in.” Another linguist, Chi Luu, suggests that with English RP (see below), the effect is more subtle; such accents go beyond “pure evil” by shading villains as more educated and intelligent—but still bad. Relating these findings to The Patriot, viewers understand Tavington as educated, intelligent, and evil through his English RP dialect; moreover, in black-and-white terms, his dialect makes tangible that he is completely different from the rough-and-ready, courageous Benjamin Martin. An aural line is drawn between American “good guys” and British “bad guys.” The British are fighting just to be sadistic or greedy; the Americans are fighting to defend their way of life.
Compared to media about other American conflicts, there has been little recent dramatic depiction of the American Revolution. Historically, one reason dramatizations of the American Revolution have historically remained rare is the tradition of American reverence for the “Founding Fathers,” causing their depiction as human beings to be circumscribed. Just as visual conventions have been built up regarding the antagonists of the conflict, so, too, have aural conventions in audio-visual representations such as a general use of a Standard North American accent for American (patriot) characters (whether they hail from northern or southern colonies) and English RP for all (villainous) British characters. The effect of this is epitomized by the example from The Patriot above: “good guys” versus “bad guys.”
While the way characters sound is important in most creative works—from the written dialect of Dickens’s characters to the chilling, mask-assisted tones of Darth Vader—it is no more so than in audio drama. Richard J. Hand and Mary Traynor have argued in the The Radio Drama Handbook (2011) that the elements of audio drama are words, music, sound, and silence. Arguably the most important of these to the audio drama are words. Alan Beck’s work on radio acting further argues that to create character through words and sound, refinement of dialect is needed. In this context, dialect refers to more than just a regional pronunciation: it includes details on a character’s geographical region, social class, gender, age, style, and subgroup.
In terms of The Patriot, probably the highest profile pop cultural depiction of the American Revolution until Hamilton, accents mattered. “A well-educated British accent,” in American films at least, “had come to serve as a sufficient shorthand for villainy” as Glancy pointed out, or “British, upper class, and a psychopath, happily burning people alive,” as a 2010 Daily Mail article put it. Whatever his villainy, would a historical Colonel Tavington, have sounded the way he did as depicted in The Patriot? How did those on the North American continent in the 18th century sound?
Richard Cullen Rath has convincingly argued that sound was incredibly important in colonial America, but what role did speech in English have? David Crystal, in discussing the Globe Theatre’s 2004 production of Romeo and Juliet in original pronunciation (OP)/Early Modern English (EME), was most frequently asked one question: how do we know? How do we know what Shakespeare’s English sounded like? For EME, spelling, contemporary accounts of the language, and the use of rhythms, rhymes, and puns illuminate this area. Crystal noted that EME would not sound like any modern accent but would have the pronunciation of “rs” in common with “rural” accents, as in Ireland and the West Country (Somerset).
This is important because the association of Shakespeare with a certain kind of English accent—the RP referred to below—has long had the effect of making his works less accessible than they might have been. If Shakespeare’s actual working dialect had more in common with “rural” accents like Irish and West Country, the latter of which is often still designated as a shorthand for stupidity in common with American southern dialect, this has a liberating effect on Shakespeare in performance. It may sound “strange” at first, but it suggests that Shakespeare can/should be performed in a vernacular, at the very least to make it more accessible to a wide variety of audiences.
Crystal has suggested that, “It is unfortunately all too common to hear accents on stage or film in which—notwithstanding the best efforts of dialect coaches—accuracy and consistency throughout the performance leaves, shall we say, a little to be desired.” Sky 1’s recent big-budget effort in historical drama, Jamestown, set in 1619, is diverse and rich in dialect, with characters speaking with Irish, northern English, southern English, and Scottish accents—nor are they uniform in suggesting that one accent will signify the character’s goodness or badness (though it has to be said most of the southern English-speaking characters are, at the very least, antagonistic). Most of the Native American characters have so far not spoken English rather than racist “Tonto talk,” and white English-speaking characters must speak Powhatan (or use sign language) in order to communicate. As actor Jay Tavare points out, early Hollywood was not concerned with the accurate portrayal of indigenous languages, given that Navajo extras were most often used in Westerns (meaning the Diné language of the southwest was heard regardless of where the film was set). More recently, even such films as Dances with Wolves (1991), which has generally received praise for its accurate portrayal of Sioux/Lakota culture, included Lakota dialogue which was linguistically imprecise. Clearly, there is still a long way to go.
Suppose we decide we cannot ever know how such historical actors sounded, so we decide to depict 18th century English speakers based on other criteria. How do directors, actors, and voice coaches bring the viewer/listener as close to the author’s intent as possible in fictional depictions of the years surrounding the American Revolution? How is the 18th century made accessible through dialect? As Glancy suggests above, British officers in film, television, and radio depictions of the American Revolution are most commonly associated with Received Pronunciation (RP), though the term was not coined until 1869, and, as Crystal points out, the “best speech” heard in the English court of the 16th century was “nothing like” RP. In fact, it was probably influenced by the Devonshire accents of Sir Francis Drake and Sir Walter Raleigh (an English so far in Jamestown associated with servants . . .). As Edda Sharp and Jan Haydn Rowles point out in How to Do Standard English Accents, RP has almost disappeared in England, “now confined to a very small section of society, the older upper and upper middle classes, older actors and broadcasters.”
The officer class of the British Army in the 18th century was in fact largely composed of Scots (as per Stuart Reid and Paul Chappell)—though how much they retained their Scottish identity (including accent) is of course impossible to know. As Kevin Phillips in 1775: A Good Year for a Revolution points out, within England, support for the colonies during the American Revolution was generally strongest in the major cities and in the eastern and southern counties—the same places where during the Civil War, support had been strongest for Parliament and Cromwell. Support in Scotland was limited to a small fringe. By contrast, statistics from Revolutionary America 1763 to 1800: Almanacs of American Life on the ancestral origins of the white population of Virginia in 1790 make Scotch-Irish (11.7%) and Scottish (5.9%), the second-largest groups after English. References to the Scottish officer class are made in Cooper’s The Last of the Mohicans, where Colonel Munro is referred to as “the Scotsman” (played memorably in the Michael Mann film by Maurice Roëves). Beal suggests that guides to “correct” pronunciation in the 18th century were often written by and for Scots—to pass as “British.” The Scottish people have had a long history of wishing to disassociate themselves from rule by the English (as evident in the 2014 Scottish Referendum). The Wars of Scottish Independence in the 13th and 14th centuries were fought between English and Scottish forces, at the conclusion of which Scotland retained its status as an independent state. However, in 1707, the Act of Union unified England and Scotland under one government.
With these many regional British accents potentially to be heard in the British Army in America in the 18th century, and with approximately 92.2% of white colonists in Virginia hailing from the British Isles in 1790 (including England, Scotland, Ireland, and Wales), it is perhaps curious that there has been so little variation in dialect in media depictions such as films, TV, and radio. Traditionally, for example, George Washington has been depicted with a Standard North American accent and not a modern southern American accent. This is also true in the case of Colonial Williamsburg re-enactors.
Besides being generally a very sound depiction of the Revolutionary period, HBO’s John Adams (2008), based on David McCullough’s book, is also unusual as far as accents go. John Adams gives the British King George III, played by Tom Hollander, Standard Neutral English, but it does something very unusual with its American characters. Virginian characters (including Washington and Jefferson) have suspiciously Somerset-sounding accents. John Adams introduces a clear regionalism in Anglo-American accents that goes beyond the singling out of the Boston accent (which is generally personified in John Adams himself). It isn’t clear why this decision is made; the accent does not belong exclusively to Virginian characters, as it is also the accent of Benjamin Franklin, proudly Pennsylvanian, and Alexander Hamilton himself. Is the shorthand present to aurally distinguish between Loyalists, and American rebels? What affinity, one might ask, does the Somerset accent have with a depiction of the American character?
Standard North American speech is rhotic (people who say an ‘R’ whenever it is written), as is the Somerset accent (and, as Ben Zimmer explains, was associated with country speech in England from the 17th century). This contrasts with Neutral Standard English Accent, which is non-rhotic (people who only say an ‘R’ if there is a vowel sound spoken after it). Perhaps the filmmakers had in mind the best guesses about how “proper” Anglophones spoke in the 18th century, as in John Walker’s Critical Pronouncing Dictionary (1791), where he normatively condemns rhotic English. Then again, perhaps it is merely a storytelling expedient; look at the actors assembled to play Founding Fathers:
- Washington – David Morse, American, born in Massachusetts
- Jefferson – Richard Dillane, born in London
- Hamilton – Rufus Sewell, born in southern England of Welsh and Australian parents
- Franklin – Tom Wilkinson, born in Leeds, grew up in Canada
- Sam Adams – Danny Huston, American, born in Rome
Standardization may have made practical sense, but the use of the rhotic sends a subtly different message than North American Standard English, and perhaps hints at the influence of David Crystal’s work on OP/EME.

First page of Noah Webster’s Compendious Dictionary (1806)
Even among linguists, North American 18th century English has not been examined to the extent that EME has, with Beal going no further than to suggest “the contact between the various regional dialects of the English-speaking colonists and the languages of the Native Americans and of other European colonists” would have influenced pronunciation, but she is not conclusive as to what extent. Noah Webster in 1789 envisioned “a language in North America, as different from the future language of England, as the modern Dutch, Danish and Swedish are from the German or from one another” (cited in Mühleisen). From the beginning, too, American fiction was concerned with pronunciations, with Cooper sensitive to those which “disclosed class or origins,” as Schachterle noted. The word “Americanism,” in fact, was coined by John Witherspoon in 1781. Its entry in Webster’s Compendious Dictionary (1806) is “a love of America and preference of her interest.” As David Crystal points out, this was the first dictionary written in English to include words like butternut, caucus, checkers, chowder, constitutionality, hickory, opossum, skunk and succotash. It was also the first dictionary to give now-familiar American spellings to words like color and defense.
It is probably not a surprise that many of the words above have their origins in Native American languages. Black and Native voices (and use of dialect) are somewhat rare in media depictions of the Revolutionary period, notable exceptions being The Last of the Mohicans, with its heroes (“Hawkeye,” Uncas, and Chingachgook, played by Daniel Day-Lewis, Eric Schweig, and Russell Means, respectively), as part of an extended, multi-racial family unit, speaking with a uniform Standard North American accent:
and Ubisoft’s smash video game, Assassin’s Creed III (2012).
The hero of this Xbox/Playstation game is Connor/Ratohnhaké:ton (Noah Watts), half-British, half-Mohawk, who speaks Mohawk as well as English (again, no Tonto talk here), as taught by the project’s language consultant, Akwiratekha’ Martin. However, again ignoring historical realities like the Sullivan campaign, the villains are British redcoats, echoing the storyline set up by The Patriot (Connor’s village is burned and his parents are killed by a British officer who sneers, “You are a nothing, living in the dirt like animals.”) In a making-of featurette, the creators of Assassin’s Creed III note that it has more than two hours of “cinematics,” a concept that suggests that the game will be a more potent window into history for its users than film, television, radio, or books. While it is positive that the game inaugurates a Native American action hero role model, it is disappointing that it recycles “the British are baddies” dialect trope.
In media like cartoons, video games, and audio drama, dialect supplies us with many clues about characters and what we should think and feel about them. Class is still very much implicit in accents in Britain, and every time we open our mouths we display characteristics about ourselves, whether feigned or not. When mass media-makers produce historical drama, they make choices. Despite the United States’ and the United Kingdom’s “special relationship,” the “British are baddies” trope remains. Its continued use and the future choices made by media-makers have much to say about Anglo-American relations and conventions in fiction today.
If, as suggested by several linguists identified in this article, children very quickly pick up conventions of who is “bad” via what they sound like, it behooves us to pay careful attention to the way mass media uses voices and what norms these establish, for listeners young and old. Also, what accents filmmakers use for past accents may profoundly influence who can hear themselves as “American” in the present moment–one important reason why Lin-Manuel Miranda’s Hamilton resonated with contemporary audiences. If teacher Laura Amico in the Cliffside Park school district in New Jersey has established that her pupils should “speak American,” what does this mean (besides the fact she expects them to speak English not Spanish)? Should they, for example, speak like President Trump, who uses a New York Queens accent which, according to Jeff Guo, connotes “competence, aggressiveness and directness” despite the fact its speaker comes from a privileged background? As Guo suggests, the way we speak is at least as important as what we say. In response to reports that Trump mocked the Indian Prime Minister, Narendra Modi, by imitating his accent, Congressman Raja Krishnamoorthi published a statement saying, “Americans are not defined by their accents, but by their commitment to this nation’s values and ideals.” That, perhaps, is speaking American?
—
Featured Image: Arno Victor Dorian, Assassin’s Creed Unity: Leon Chiro Cosplay Art
—
With thanks to Kenneth Longden and Matthew Kilburn
—
Leslie McMurtry has a PhD in English (radio drama) and an MA in Creative and Media Writing from Swansea University. Her work on audio drama has been published in The Journal of Popular Culture, The Journal of American Studies in Turkey, and Rádio-Leituras. Her radio drama The Mesmerist was produced by Camino Real Productions in 2010, and she writes about audio drama at It’s Great to Be a Radio Maniac.
—
REWIND!…If you liked this post, you may also dig:
The Magical Post-Horn: A Trip to the BBC Archive Centre in Perivale–Leslie McMurtry
Speaking “Mexican” and the use of “Mock Spanish” in Children’s Books (or Do Not Read Skippyjon Jones)–Inés Casillas
Sounding Out! Podcast #40: Linguicide, Indigenous Community and the Search for Lost Sounds
Recent Comments