Let The (Techno) Sculpture Speak Machines and Orality in Recent Peruvian Sound Art

Standing at the center of a room, we see seven small robotic figures placed upon stands. We can hear them speak with synthetic voices. Each one emits different messages, sounding either on their own or in unison. The voices are generated by a software that renders text into audible speech, known as text-to-speech or TTS. In this case, it reproduces the dialogues and arguments that take place in the comments sections of three popular Peruvian songs on the YouTube platform: Cuando pienses en Volver (When you think about coming back) by Pedro Suarez Vertiz, Todos Vuelven (They all come back) by Los Morochucos and Esta es mi Tierra (This is my land) by Augusto Polo Campos, three songs that talk about migration and the nostalgia of those who are far from their place of origin.

The dialogues, transformed into synthetic voices, have been redistributed across the seven speakers, causing an effect of disjointed phrases that overlap, in which it is possible to distinguish topics commonly related to Peru’s national identity, such as political corruption, the gastronomic boom, the ancestral past, and to the displacement of a generation that traveled far from their country in search for better life opportunities. This sound installation, El Eterno Retorno (The Eternal Return) (2012), by Peruvian artist José Luis Martinat was presented as part of the exhibition Remesas: flujos simbólicos, movilidades del capital (Remittances: symbolic flows, capital mobilities), curated by Rodrigo Quijano, which acutely emphasized the constant migratory flow and participation of local artists in the international art circuit as a as a source for the production of new themes on the local scene. The thought underlined by Remesas, however, was that these trips to the centers of international artistic production  did not undermine the desire to overcome dominant currents, and that this need for displacement was the starting point “to disarm an ancillary and   condition, a condition from which it is often difficult to breathe with across regional difference.”[1] The exhibition was held at Centro Fundacion Telefonica in Lima, a cultural space that is the property of a transnational telephone company  an important role in promoting and validating a new artistic scene in which art and technology are deeply related.

Pleno (2016), another recent work by Martinat, takes the form of a vinyl LP record that contains the oaths made by Peruvian congressmen elected in 2006 and between 2011-2016, when assuming office. We listen to the oaths, once again reproduced by synthesized voices using the TTS software, one after the other (in País con futuro (2016) the artist used the same method to reproduce a series of slogans from electoral campaigns). Additionally, the vinyl has been pressed in such a way that it makes the needle jump and return to the same groove, generating a loop in its reproduction. The idea is to present the piece as a set of vinyls played simultaneously and continuously on multiple turntables, each looping randomly on a different oath.

The piece reminds us that in the political sphere, the voice has been, and still is, a fundamental element of legitimation, as the law is not complete until the vocal staging that makes it public (1). [2]What Martinat does in Pleno (and in País con Futuro) is to decompose that operation. Automated artificial reproduction ends up turning messages into mechanized information. The piece shows that these messages can be reproduced and repeated as many times as desired, like a mantra, but by doing so it becomes devoid of its meaning.

These records adhere to the tradition of using recordings of political figures as a means of propaganda, although here in a parodic way, by replacing their voices with an artificial reconstruction that results in their standardization. It is a gesture that affirms, in a disbelieving and cynical way, how the promises of those who exercise political power are perceived. The work is not mere skepticism embodied in a form of artistic appropriation, but rather a questioning of the staging of the voice – in this case an artificially produced substitute.

From the technological side, this is possible thanks to the development of TTS software. This technology began to spread massively in the early 80s and played an important role in the development of so-called assistive technologies (AT) that expand the capacity of people with disabilities to interact with the wider world. TTS software has also been used for the creation of avatars within the culture of videogames, and in the development of automated translators, amongst other uses, with a wide range of systems that have developed from the   of recorded samples of real voices to the use of synthesizers and artificial intelligence.

El Eterno Retorno, Pleno and País con Futuro all rely on speech technologies that sound robotic and artificial to us. They are works that deal with the machinic return of orality proposed by Félix Guatari, “capable of engendering mutant subjectivities, projections and fantasies of a human/machine interstice.” [3]

Hierarchies of speech
In Callejón Oscuro – Concierto para una mente y 52 parlantes (Dark Alley – Concert for One Mind and 52 Speakers) (2013) by Sergio Zevallos, one hears a variety of voices that come from 52 speakers, arranged like the sidewalls of a corridor. Callejón Oscuro (Dark Alley) refers to a boys’ game, popular in Peru, in which a corridor is formed by two lines of people who then beat and attack another person who has been forced, as punishment, to walk between the lines. An act replete with shame (that arose within military academies) here serves as a metaphor for the dynamic of certain relationships of power. In the notes that accompany the album that includes the soundtrack of this installation, the author notes:

The different voices were chosen according to the texts they read – young, adult, male, female voices. The texts form discursive groups in a semantic counterpoint, where they complement each other’s content. There is a group that consists of excerpts from military speeches, statements from a convicted human rights violator and critical texts on heterosexuality. Another group is made of popular texts about nationalism and identity. Finally, a third group assembles purely subjective texts, such as a list of insults from the local language, with sexual expressions used to discriminate, and short stories of amoral content. [4]

These voices simultaneously represent imperatives upon which subjectivity is built and ways in which control is exercised.

There are various aspects to point out regarding Callejón Oscuro. On one side, the staging of the 52 speakers serves to present the piece in its whole dimension, operating in the manner of an experimental radio composition. The piece is an ensemble work that reminds us of the common experience of someone who walks down certain streets in the central district of Lima, stunned by the cacophony of sounds that emerge from peddler’s speakers, shops and public transport horns; a chaotic sound scenery of the established ways of informal economy, where a competition to be the loudest and most insistent, as a means to attract possible customers, takes form. It is the law of the strongest, of the one who attacks the most persistently, that seems to be reproduced in Zevallos’ work.

This frequent use of technological devices for the amplified representation of orality, as well as its manipulation through audio montages and the use of TTS, marks a tendency towards a machinic orality that undoubtedly feeds on a new technological imaginary, in which interfaces and prostheses that expand beyond standard human capacities are common.

The use of these machinic oralities may also respond to a growing trend in local contemporary art, due to practices inherited from neo-conceptualism, which find in sound and speech a field for exploration. However, the emphasis put on speech and its ability to be reproduced is also a consequence of socialization and the widespread use of the new audio reproduction technologies that have been transforming the world of visual arts for almost two decades, as well as life itself: we have become ever more nomophobic—afraid of being without our mobile phones—as we live attached increasingly to a device that reproduces our own voice and amplifies the voice of the other. What is remarkable is that this use of technologically mediated orality in recent Peruvian sound art poignantly stresses certain aspects of what we could define as a social dysfunctionality.

The vocal piece by Luz María Bedoya, Me tiemblan mis labios (My lips [my] tremble) (2009), references directly in its title the prejudice against Quechua-accented Spanish speech, by the use of an incorrect syntax (the double possessive “Me-mis”), a linguistic structure frequently used by Andean immigrants in the capital of Lima. The piece is composed of a radio broadcast that repeatedly utters key dates from recent Peruvian history, which have the letters contained in the work’s title removed. The artist states:

The first text is the succession of a series of recent dates that indicate certain flaws of Peruanity: the day on which a national newspaper publishes, with an accusing tone, the photo of a Quechua-speaking congresswoman holding her notes with errors in Castilian Spanish writing; or the day that the 1090 legislation, which affects the use of Amazonian lands, is promulgated; or the day that the reuse of disposable surgical instruments is authorized in a state hospital, for example. The second text is the phrase “My lips [my] tremble,” which is not heard but exists when the first is wounded. [5]

Luz María Bedoya’s piece alludes to the complex problem of centralism in Peru, which has exacerbated economic and social inequalities, and population disparities. This is, undoubtedly, also expressed in the hierarchization of linguistic varieties in Peru, where Spanish is the most used and the “language of the capital,” therefore, the one that will generate the greatest job opportunities, leaving Quechua in a subaltern position. [6]

Writer and anthropologist Jose María Arguedas is a figure key to understanding the complexity of this issue. Through his work he preserved and strengthened Andean oral traditions, and reinvigorated Quechua by his own practice of oral poetry. It is nonetheless interesting to point out that Arguedas’ work comes along with the emergence of a mass culture, in the context of which he became aware of the importance of technology to the preservation and transmission of that oral culture.[7] To imagine a new radio, then, is to imagine a new system of social relationships that puts an end to the centralized ways in which power constitutes itself.

The artist Eliana Otta also conceives the possibility of altering the message of a Peruvian song with wide mass circulation. In Con P de… (2016), she crafts an audio piece from a text written by the artist in 2011, which extracts the well-known phrase “Con P de Patria” (With P for Homeland) from the Peruvian waltz Y se llama Perú (And it’s called Peru), composed by Augusto Polo Campos in 1974, at the request of the government of Velasco Alvarado, which has become an official symbol of Peruvian identity. The artist turns the text into a polyphonic reading to which poets, singers, writers, activists and artists lend their voices. In this way, Otta’s exercise suggests two ways of seeing Peru: one that hints at the patriotic idealization that is read through the lyrics of the famous waltz, contrasted with another definition of the “Peruvian” that the artist proposes as desire and demand.

Memoria Colectiva (Collective Memory) (2010), by Elena Tejada-Herrera, is a sound piece that works as an audio microdocumentary. Composed using voices and slogans from protest marches held in Lima, it forms a collage that unfolds the voices of various groups of women protesting against the Fujimori dictatorship. In Estado de emergencia (State of emergency) (2014), by Luisa Fernanda Lindo, one hears the sound of a siren, against which the artist repeats the phrase “Repetir palabras hasta hacer del espacio un lugar común” (Repeat words until space turns into a common place). At first, the voice is unintelligible due to digital processing effects, presenting cuts and interventions, but it becomes increasingly intelligible, insistent despite the permanent presence of the continuously-shrieking siren. Both Lindo and Tejada-Herrera’s pieces are symptomatic of a relationship of tension with the inhabited space, where orality is what allows the defusing of that tension. Beyond the words themselves, the fact that the words exist in that auditory spectrum, whether as noise, processed screams or as collage, speaks to a new repertoire that this machinic orality is making possible.

Spectra of the voice
In the sound installation La realidad ausente (The absent reality) (2015), Santiago Pillado-Matheu builds an auditory piece composed with recordings of testimonies from the time of the internal conflict in Peru, which are reproduced through several loudspeakers placed at various points in a fully darkened room. These testimonies denounce the violation of human rights and abuses that were inflicted on many indigenous populations during this time. Experimental musical compositions intertwine with gasps, excavation sounds and radio transmissions of old popular songs as a way of delving into the ghosts of an era, while contrasting it with the Peruvian radio stations on which nostalgia reigns – little more than museums of songs that have been in rotation for decades, symbolizing a haunting past.

Fig. 1 La realidad ausente (The absent reality) (2015), Santiago Pillado-Matheu. Photo by Juan Pablo Murrugara

The public hearings held during the work carried out by the Comisión de la Verdad y Reconciliación en el Perú (Peru’s Truth and Reconciliation Commission), charged with preparing a report on the violent internal conflict experienced in Peru between 1980 and 2000, confronted the country with a series of oral testimonies from the victims who recounted the abuses committed during that period. Oral testimony in itself became a tool for clarification of historical memory. As Esther Espinoza points out, “Said in front of an audience, the testimonies intend to break the silence, a silence that has been part of the tragedy, not only because of the fear of speaking due to intimidation but because there was no interlocutor who would respond meaningfully […] The public and nominal character of the hearings is the face of the law that now could have a defined image for the victims.” [8]

There is another group of pieces in which, more than orality (in the sense of texts that communicate meaning) we find vocality, illegibility and abstraction. Regarding the gesture of listening, Vilém Flusser argues that speaking voices are listened to differently than music. The former are heard while deciphered, and they are read, which is why the deaf can read lips, but not music. Although deciphering also occurs when listening to music, since music is codified sound, “It is no “semantic reading,” no deciphering of a codified meaning. […] The human body is permeable to sound waves, and in such a way that these waves set it in vibration, affect it. […] the gesture of hearing is essentially an adaptation of the body to an acoustic message” concludes Flusser. [9] This is the type of relationship found in a work such as Speaker Tower (2017) by the artist Nicolas Kisic, which is a sound sculpture for public spaces, built with two large rotating speakers made of recycled material, that can be used in different ways. For example, it has sometimes served to amplify recordings of a vocal sound piece, created in collaboration with Shipibo-Conibo artist Rawa. Conceived as a machine to accompany protests, the author explains: “I created the Speaker Tower, but the Speaker Tower is also creating a different protester in me. The Speaker Tower and I are producing new ways to amplify voices that would otherwise be silent. Its bold presence on an urban scale claims the most civic dimension of public space, while projecting and amplifying our voices in the air.” [10]

Another example is Gráfico Sagrado 1 (Sacred Graphic 1) (2015) by José Luis and José Carlos Martinat, a piece composed of a large suspended canvas with a magnetic tape reproduction system attached. Tape reels over the canvas to form an Andean cross or chakana with its path. Through the complex device, an audio recording of a Quechua-speaking voice is reproduced in a loop, which is amplified and distorted by the device’s default filters, generated by its reproduction heads, resulting in an enveloping effect of abstract sound that invades the listeners’ senses. 

Fig. 2 Gráfico Sagrado 1 (Sacred Graphic 1) (2015) by José Luis and José Carlos Martinat.

In the same way, artist Paola Torres Núñez del Prado has been building tactile sound devices that she calls Textile Controllers (2016). Made up of quipus and Amazonian Shipibo textiles of geometric design, she uses her creations to establish patterns that are decodified and sonified into audio.[11] The artist uses the textiles as an interface to control a great range of samples that are activated and manipulated through her interaction with the fabric. Among the embedded samples there is a recording of Hanaq pachap cussicuinin (Joy of Heaven), the first registered polyphonic composition written in the Americas using the European musical tradition, sung in the Quechua language. She also uses recordings of indigenous songs by the Shipibo Conibo singer Auristela Brito Rengifo-Valles. Both the quipu and the Amazonian fabric are understood as instruments and technologies that refer to the past, but at the same time, conserve their validity in the present life of various indigenous groups.

In a recent project called Aielson (2020), which reached its culmination during the pandemic, Núñez del Prado takes quipus as a reference and trains a Machine Learning tool by feeding it with poems by Jorge Eduardo Eielson, Cecilia Vicuña (two poets and artists known for the use of quipus within the context of contemporary art) and herself, so that this Artificial Intelligence produces new poems and learns to speak by imitating Eielson’s voice (a haunting imitation, surprisingly quite intelligible and natural, but at times interrupted by errors and glitches, that the artist decided to retain in order to remind us of the impossibility of AI attaining the naturalness of a perfect imitation). With this work, Torres suggests a commentary on the differences in meaning and forms of translation. What are the interpretations of these objects of the Andean world and what remains of them? What common place can be found?

Fig. 3 Paola Torres Núñez del Prado, Textile Controllers (2016). Photo by Rupa Flores. Courtesy of the artist and Proyecto AMIL.

Radio Futura
The appearance of technologies for the production and reproduction of speech has, without a doubt, been one of the greatest triggers of the appearance of a wide range of artistic practices that have voice as a central element. This is either because such practices refer to the different incarnations of popular speech or song, or because they are conceived within the forms of sound poetry, radio art and the various explorations of vocal art in general, including what we call talking sculptures or vocal sound installations. Thinking beyond the use of these technologies as an expansion of the repertoire of aesthetic possibilities for the local artistic scene, what should be remembered is that the voice exists in a network of amplifications and reproductions that define a social and mediatic space, and that the production of this new machinic orality is generating questions regarding what kind of participation we have, or want to have, in it. What is truly transformative is not only the adherence to a technological discourse, but rather an increasing consciousness of the idea that having a voice is not only a matter of phonic capacity, but of political participation; that is to say, participation in the audible spectrum of that mediatic space.