1. Introduction

In June 2021, the University of Alberta SpokenWeb research team gathered on Zoom, in-person gatherings being limited at the time due to pandemic restrictions, to take part in a collective “unboxing” of archival audio materials. In the video recording of the event, SpokenWeb UAlberta Project Leader Michael O’Driscoll removes the lid of the filing box and, to the delight of the team, uncovers myriad archival treasures: original scripts, hand-written interview notes, and taped interviews. The archival materials, which can be seen stacked behind O’Driscoll’s chair in Figure 1, were donated by Jars Balan, past Director of the Canadian Institute for Ukrainian Studies. The boxes arrived in the SpokenWeb team’s hands in part due to the “enterprising” skills of then-graduate research assistants, Ariel Kroon, Nicholas Beauchesne, and Chelsea Miya, who tracked down and reached out to Balan in the process of investigating the origin of a campus radio show in the University of Alberta archives and the identity of the speakers on tape, of which more will be said further on.

Figure 1

Michael O’Driscoll “unboxes” archival tapes with the SpokenWeb team (2021).

What do we mean by “unboxing” the archive? What does “unboxing,” as a critical and archival practice, entail, and how do we approach these processes with care? Our use of the term playfully invokes the performative social media phenomenon of unboxing, which soared to popularity on YouTube during the COVID pandemic. In unboxing videos, YouTubers typically film themselves unpackaging an online order and describing and reacting to the product inside. (For more on the unboxing phenomenon, see Lieber 2019; Craig and Cunningham 2017; Mowlabocus 2018.) For content creators, the box itself is integral to the experience: the tearing of the cardboard and rustling of the paper builds anticipation and creates sensorial pleasure for the viewers, who themselves participate in the activity through their visual and auditory engagement. The unboxing genre is bound up with capitalist ideologies of consumption and accumulation, with creators sponsored by brands to showcase the latest products and gadgets. Companies have capitalized on the popularity ofs unboxing videos, redesigning packages to enhance the experience, and entire sub-industries like monthly subscription packages and meal kits have arisen around box-based consumption (Kim, Self, and Bae 2018; Spence, Youssef, and Levitan 2021; De la Peña 2023). At the same time, media scholars David Craig and Stuart Cunningham argue that “unboxing” is more than just a marketing tactic, pointing out its origins in “hobbyist” and fan culture and its unique role in community building (Craig and Cunningham 2017, 83). Sharif Mowlabocus similarly argues that unboxing is a distinctive creative practice deserving of serious scholarly attention, homing in on the unique “affective intensities and tactile pleasures” afforded by the genre (Mowlabocus 2018, 565).

Our case study is an experiment in “critical unboxing” as archival method. T.L. Cowan calls out the tendency within both academic and digital modes of production to privilege “open access, speed, and preservation” over other considerations (Cowan 2020, 47). Along these lines, the authors of this article reimagine “unboxing” as a way to celebrate slow and reflexive archival practices, taking time to “pause” and “contemplate the outcomes” of digitization (Cowan 2020, 47). (Cowan’s work exists in conversation with a larger cultural movement, whose proponents advocate careful and thoughtful engagement in various forms, from “slow academia” to “slow computing,” such as Berg and Seeber 2016; Kitchin and Fraser 2020.) Critical unboxing is not just “slow,” but playful and performative. Filming archival work in practice, capturing footage of the research team participating in and reacting to the reveal of newly donated, never-before-opened archival material, invites us to reflect on how archivists engage with archives on an affective, emotional level. In the aforementioned clip, the team members break into grins and exclaim with delight as each object is lifted up to the camera. In being guided by and inspired by our personal, affective response to the collection, our study also underscores what Michelle Caswell and Marika Cifor term the archivists’ “affective bond” to the archival objects and, in turn, the creators of that content (Caswell and Cifor 2016). Finally, by unpacking the organizational systems that surround and support how archival audio is consumed and experienced—from the rights management statement to the website architecture—the authors show how the notion of privacy and its associated law, ethics, and rights are predicated on what Zoë Sofoulis terms “container technologies” (Sofoulis 2000), of which more will be said further on.

Sound archives have existed for almost as long as audio recording technology, dating back to at least the 1890s (Schüller 1991, 204). Recently, sound scholars like Caswell and Cifor, Josh Sheppard, Carolyn Birdsall (Birdsall 2016), and others have drawn attention to the role of sound archives and archivists in circulating, activating, and re-activating discourse, as well as the political and ethical connotations of this work, in terms of which voices and “codes” are amplified and which are suppressed. PennSound, Ubuweb, and the Library of Congress’s Radio Preservation Task Force are just a few examples of archival audio projects, and researchers across the SpokenWeb Network of higher educational institutions are likewise engaged in listening to, documenting the metadata of, and digitally archiving the recordings held in their care.

The act of “unboxing” is integral to the SpokenWeb Network’s origin story (see Eastwood 2023). The SSHRC-funded research partnership is comprised of a cross-institutional and multi-disciplinary collaboratory of “researchers who are dedicated to the discovery and preservation of sonic artifacts that have captured literary events of the past to activate these artifacts in the present” (SpokenWeb 2023). The local University of Alberta branch began with the discovery of a stack of cardboard boxes of reel-to-reel tapes. Many of the magnetic reels, which had been gathering dust for decades and were slated to be disposed of, turned out to contain original recordings of author readings and performances dating back to the 1960s. Since then, the audio collections within the Network continue to grow, and the extensive holdings are currently in the process of being catalogued and made available online.

Jason Camlot and Katherine McLeod (Camlot and McLeod 2019) describe how textual meaning is continuously being re-negotiated and generated anew through processes of remediation or “migration” across different media forms. The scholars use the term “unarchive” to refer to the ways that digitization transforms and remakes archival artifacts, and in the process creates possibilities for new types of public engagement and meaning-making. In doing so, unarchiving “stages a resonant confrontation between structures of the past and the present” (Camlot and McLeod 2019, 9). Rick Altman likewise famously posits that recordings are “double, marked both by the specific circumstances of recording and by the particularities of the reproduction situation” (Altman 1992, 13).

Digitizing historical recordings, re-presenting the audio event in a new context for a new listening audience, destabilizes meaning in ways that can be productive and generative, but also expose the subjects represented in the archive to potential and unanticipated harms. Cowan reminds us of the value in “mov[ing] slowly” (Cowan 2020, 47) and taking time to pause, reflect on, and account for the potential risks and harms of making collections publicly available, particularly when the communities represented in those collections are from marginalized and vulnerable groups. In addition to Cowan, other notable theorizations of ethical curation include that of Deanna Fong and Felicity Tayler (Fong and Tayler 2022); Spy Dénommé-Welch, Jean Becker, and Cecilia Garcia Vega (Dénommé-Welch, Becker, and Garcia Vega 2022); Michelle Caswell and Marika Cifor (Caswell and Cifor 2016); Alexander Freund (Freund 2016); and Dietrich Schüller (Schüller 1991). Our paper draws on and owes a great debt to their work, exploring how digitizing historical audio collections can create opportunities to open up a dialogue between scholars and artists, but at the same time also introduce new complexities around issues of privacy and consent. (Other examples of critically engaged listening practices include Ouzounian 2023; Robinson 2020; Stoever 2016.) As we have found, “care-full” archival work, to use Cowan’s term (Cowan 2020, 46), requires us to engage with and imagine past, present, and future media, as well as past, present, and future listeners.

The retrieval of antiquated sound objects from dusty boxes, and their remobilization in digital form, invokes complex questions that cultural scholars have been grappling with for some time. For example, technologist scholar Sofoulis, in her theorization of “container technology,” points out how media function as containers or systems of storing, ordering, and processing, citing “books, photographs and albums, telephone directories, the television, the stereo, cassettes and CDs” as examples (Sofoulis 2000, 189–190). Sofoulis touches on, as well, the affective function of media, as containers not just for information but “experiences,” pointing out how listening to a tape can “open up a whole concert or aural landscape of feelings” (Sofoulis 2000, 190). The following paper uses the “unboxing” of historical audio by members of the SpokenWeb team at the University of Alberta, which was then repackaged and re-presented as an online audio exhibit, as a case study of care-oriented critical archival practice. As shown, the process of digitizing historical audio involves rethinking the boundaries of archival objects, whose meanings are negotiated through a networked web of past, present, and future technological “structures” and their attendant affordances. Archival recordings resound within specific contexts, which includes not only media devices, but other techniques and practices of information management. Along with recording and playback technologies, we consider present-day structures, such as rights management standards and web design practices, and how they work together to reframe and remediate the audio event.

To “unarchive” is to understand archival practice as a process of activation rather than preservation, with the goal being to open up these collections to the public and to artists, who are, in turn, engaged in the process of reflecting on and responding to these works and, in this way, invigorating them with new meaning. Along these lines, the SpokenWeb project at the University of Alberta has undertaken to digitize and disseminate historical audio in a praxis of unarchiving in several ways: by hosting digitized audio files in streamable format on the university library’s Aviary platform; by creating podcasts about and featuring clips of historical audio; and through the creation and maintenance of a public-facing website, showcasing the accessible audio in the SpokenWeb UAlberta collection (SpokenWeb UAlberta 2023). Our paper is part of a larger conversation around the logistical as well as legal and ethical challenges that researchers who work with historical audio collections face, and how archival work fits within the scope of Information Access and Protection of Privacy (IAPP) and copyright law in Canada. Our case study details the unique challenges that the digitization of audio ephemera can present to sound researchers and archivists around issues of obtaining consent and ascertaining ownership. As we explain, the development of SpokenWeb UAlberta’s rights management practices was ultimately guided by its rights holders: the creators and subjects of the content (or their families or estates), which includes but is not limited to the poets and authors who came to the University of Alberta to give talks or performances of their works. We use the example of the “unboxing” of an audio event in our collection, an ongoing process that unfolded across multiple media forms (library repository, rights management document, digital exhibit, and podcast), as an invitation to think through how container technologies have historically transformed archival work and engage with key questions such as these: what best practices can be established, in terms of disseminating audio recordings online, not only to make sure that researchers adhere to legal requirements but also respect the privacy of the individuals and communities captured in these records (Sections 1 and 3)? How can sound researchers be good caretakers of archival audio data, especially when it is of a personal or “fragile” nature and might contain sensitive information (Section 2)? Finally, how do we locate the listening and sounding subject within these “unboxing” processes, and take steps to better understand and centre their role in making meaning from the sounded event; in other words, in continued “resoundings” (Section 4 and Conclusion)?

2. Privacy and container media

Privacy is conceptually entwined with the notion of containment, and one’s physical, external self and inner life and identity have historically been protected and defined in relation to various containers. Initially, the right to privacy was primarily concerned with “physical space” (Franks 2023, 3). At a time when most families lived together in single-room dwellings, privacy was a privilege enjoyed only by the rich. Only the elite upper classes could afford grandiose houses with special rooms, such as a drawing room or study, specially designed as secluded or “contained” areas where one could retreat for peace and quiet or hold a small, private gathering away from the servants (Franks 2023, 3).

Our contemporary understanding of privacy, not as a privilege for the elite few but a human right that should be guaranteed to all, can be traced back to the nineteenth-century Industrial Era. Much like today, early advocates of privacy were reacting against technological changes, which enhanced the ease with which personal information could be surreptitiously recorded and circulated, often via the press. In their 1890 Harvard Law Review article “The Right to Privacy,” lawyers Samuel Warren and Louis Brandeis (Warren and Brandeis [1890] 1994) called attention to the need to establish limitations on what could and could not be published about an individual’s personal life. As they pointed out, libel law only protected against slander and did not address the harm caused by “gossip” (Warren and Brandeis [1890] 1994, 76). The duo argued that one did not have the right to expose the intimate details of someone’s life simply because it was true. In their view, some information was too “sensitive” and personal to be publicized without causing undue “pain and distress” to the individual in question (Warren and Brandeis [1890] 1994, 77; see also Smith 2000, 122).

As Robert Ellis Smith points out, the widening definition of privacy, as not just concerning the body, but having deeper social, moral, and emotional connotations, was rooted in certain technological innovations (Smith 2000, 122–124). Chief amongst these was the Kodak “snap” camera, which hit the market in 1888 (Smith 2000, 123). Previously, taking someone’s photograph required the sitter’s consent, as he or she had to remain in position for several seconds or even minutes while the chemical-coated plates set. Thanks to George Eastman’s invention, the process became much more user-friendly. The lightweight camera came with pre-loaded film and automated shutter speeds, hence the term “snapshot” (Thompson 2014). Now, a person’s image could be captured on the fly without her permission or even awareness (Smith 2000, 124).

One high-profile example, cited in Smith, is the case of Abigail Roberson, a “shy” teenager whose photograph was used without her consent in ads to sell flour (Smith 2000, 140). Though Roberson lost her case in the New York Court of Appeals, failing to block the ads or claim damages, the controversial ruling brought further attention to the need to revise existing laws to include an expanded definition of privacy. The well-publicized case, which sparked much sympathy and outrage on Robertson’s behalf, helped generate the public support needed to pass America’s first privacy laws, which came into effect in New York State in 1903 (Smith 2000, 139–141).

Audio technologies, like visual technologies, have redefined our understanding of privacy. Magnetic tape recorders were first developed in Germany in the 1930s. However, it was the portable, more affordable version of these recorders, which hit the market in the early 1960s, that sparked legal challenges to existing privacy legislation. The Nagra handheld recorder, as well as equivalent models like the Uher, revolutionized journalism and sparked, once more, a fierce debate around privacy. There was some legal pushback against reporters who could, for the first time, take these recorders into the field, engage in “ambush” interviews, and even record conversations secretly (Hewa 2021, 348–351; Aucoin 2005, 78–79). In 1963, the US Supreme Court briefly flirted with the idea of conversation as “property,” which could be protected from being “seized” or intruded upon under the Fourth Amendment (Middleton 1979, 290–292). However, this statute was struck down a few years later as unconstitutional. What replaced it was the “eavesdropping” rule. In Canada and the US, audio recordings are subject to one-party consent, meaning that the recordist or interviewer can record a conversation of someone else, even if that person is not aware or consenting at the time (Criminal Code of Canada, R.S.C. 1985, c. C-46, section 184; see also Bowal 1998). However, it is illegal to engage in eavesdropping or wiretapping; in other words, to record a conversation that the recordist is not part of, such as the instance of two people talking in their homes.

The development of ever more sophisticated tools and techniques for capturing, processing, and distributing information—from “snap” cameras to portable recorders to smartphones—has continued to introduce new legal and ethical challenges to privacy (Franks 2023, Module 1, 3, 5–7; Module 6, 14–17). Today, the right to be “left alone” is protected in Canada under specific Access to Information and Protection of Privacy legislation: privacy was included in Part IV of the Canadian Human Rights Act in 1977. The Privacy Act, which applies to public bodies or government organizations, was introduced in 1982, with the Personal Information Protection and Electronic Documents Act (PIPEDA), which specifically applies to the private sector, following in 2001.

Online privacy law has since expanded to include the Right to Be Forgotten (RTBF), or the right of individuals to remove their personal information from the web. The RTBF has been protected under the European Union’s General Data Protection Regulation (GDPR) since 2014, the California Consumer Privacy Act (CCPA) since 2018, and China’s Personal Information Protection Law since 2021 (Google Spain SL and Google Inc. v. Agencia Española de Protección de Datos (AEPD) and Mario Costeja González 2014; CCPA 2018, sec. 1798.105; Personal Information Protection Law of the People’s Republic of China 2021, art. 47). Until recently, Canadians have not enjoyed similar protections, since the right to erasure is not explicitly included in PIPEDA (Belbin 2018; Ni et al. 2022). That may soon change: a Federal Court of Appeal Decision in 2023 ruled that Google is required to “delist” personal information if requested by an individual, laying the groundwork for similar complaint-based RTBF legislation (Fine 2023). (Google LLC v. Canada [Privacy Commissioner] 2023, FCA 2023). The expansion of online privacy protections, particularly the right to be forgotten, has significant implications for online collections, as many institutes are unprepared to respond to requests. A 2017 study of Association of Research Libraries (ARL) member institutes, for instance, found that only 40 percent of institutes who took part in the survey had takedown policies, and, of those, fewer still had made copies of their policies available online (Dressler and Kristof 2018, 978).

3. SpokenWeb UAlberta case study

Even as policy makers continue to grapple with the connotations of the world wide web for privacy rights, many academic institutes have in recent years adopted open access strategies and taken steps to make research material more available to the public. In keeping with the desire to improve the accessibility of its audio collections, in 2019, the University of Alberta SpokenWeb team began digitizing its collection of archival audio holdings. At first, the research team concentrated its digitization efforts on poetry readings and artist performances. More recently, however, SpokenWeb UAlberta has expanded the collection to include audio works that defy easy categorization.

The University of Alberta has in its collection, for instance, three campus radio shows—Voiceprint, Celebrations, and Paper Tygers—which aired on the Edmonton-based community radio station CKUA from 1980 to 1984 (Department of Radio and Television 2007). The shows were of particular interest to the authors because they were produced by graduate students and featured interviews with celebrated Canadian literary stars: everyone from Rudy Wiebe to Margaret Atwood. The host of the show, Jars Balan, whose donated materials the authors “unboxed” in the opening paragraphs of this paper, was at the time pursuing a master’s degree in the English department.

After being contacted by the SpokenWeb UAlberta team decades later, Balan was amazed to learn that the tapes had been rediscovered and confessed that at the time of making the show he did not consider its lasting impact or potential future audiences. Before digitization, radio stations had to gauge audience metrics through traditional methods like surveys and interviews—barring that, series like Voiceprint were, as Balan put it, broadcast “out into the ether” (Balan 2022, personal communication). Along with the edited programs, material that never made it to air was recovered during the research process. These include raw interviews and “private” conversations between poets captured on “hot mics,” before the official poetry reading began.

In addition to the radio collection, SpokenWeb UAlberta also acquired a collection of “classroom visits.” The University of Alberta had a robust writer-in-residence program. Often, these authors were asked to take part in discussion sessions with students. Unlike official readings and performances, which were open to the wider university community, these events were part of a class—the speaker was not on stage but speaking to a closed group of students—and as such, they were much more intimate affairs. The identity of the author was easy to ascertain, since they were often introduced to the class by the instructor at the start of the recording. However, the recordings also contained many anonymous voices: those of the students. Indeed, as heard on the tapes, the tone of the conversation was driven by the students’ contributions: their comments, questions, and reactions steered not only the subject matter but the energy and mood of the room. In comparison to a planned performance, the authors’ classroom visits unfolded in a much more freewheeling and unpredictable way. The students’ remarks range from pedagogical queries to laughter to what Zachary Morrison names hostile or “paranoid” comments or questioning (Morrison 2023). The casual speech and candid reactions captured in these tapes serve as an example of the difficulties of transcribing and categorizing the subtleties of informal speech within the bounds of an archive (even a digital one).

Timestamping—indexing different audio events to make the information in them more findable—is an integral part of the audio curation process, one which has arguably arisen as a kind of technological redress to the problems caused by unboxing. The process of digitization, creating a computer-readable copy of an analog signal, extracts duration material from its container, preparing the audio event to be re-packaged and re-presented anew, according to myriad standards.

Timestamping involves a high degree of interpretive work, not only when it comes to describing the content or subject matter of the reading, but also different types of speech. An audio recording is divided into chapters, often based on natural pauses in the event. For instance, a poetry reading might have different sections for the opening and closing remarks, and an individual poem that was read would likewise have a separate heading. The SpokenWeb team developed a Timestamping Style Guide to this effect, with specific instructions for how to label speakers (e.g., speaker, host, audience), texts (e.g., titles of books and individual poems), and topics. Categorizing these sonic events, creating standards that could be applied across our collection, proved challenging, not only because of the diverse types of audio events in our collection, but the nuanced and complex ways in which sound is interpreted by the listener. As Alexander Freund observes, even absences or silence convey contextual information (Freund 2016, 253), and the SpokenWeb team is often confronted with the dilemma of how we choose which silences to include/encode or to exclude/cut. What hierarchies of silence exist, and how do we describe the specific contexts in which these silences are embedded and to which they speak?

The meanings that we ascribe to sounds are highly contextual, and these contexts are shaped not only at the time of recording but at playback. In “The Material Heterogeneity of Recorded Sound,” Altman uses the example of musical notation to demonstrate how information systems create the illusion that sound is “single, discrete, uniform, and unidimensional” (Altman 1992, 16). In actuality, sound is both spatially and temporally oriented with various factors, such as construction of the room and the location of the speaker, such that “no two versions of the middle C are identical” (Altman 1992, 16). As Altman points out, how we process and derive meaning from sound is dependent on our relation to the “sounder.” Our ears can distinguish between “direct” sound and “reflected” sound or sound that has reverberated off of other surfaces. We sense sounds that are aimed at us more acutely, that is, when the speaker is facing towards us as opposed to away (Altman 1992, 21–22). Audio-visual capture technologies transform sonic signals in ways that can disrupt and confuse our ability to locate ourselves in relation to the sounder. Video conference technology, for instance, has a homogenizing effect, with all speakers facing and sounding in the same direction. The inability to filter out non-essential sonic information and determine if a speaker is, in fact, speaking to us, could contribute to the phenomenon of “Zoom exhaustion” (see Bailenson 2021).

The timestamper has to listen not only for subject content but for other sonic cues that reveal the nuances of the speaker-audience relations. Jean-Francois Augoyard and colleagues challenge the tendency to treat “distortion” as information loss (Augoyard et al. 2006, 39). In fact, in our own experience, imperfections and flaws in the recording can reveal much about the circumstances of the recording event. There are various reasons, for instance, that a speaker’s words might be “inaudible.” She might be physically sitting further from the recorder, or facing a different direction to address another audience member. She might be nervous and speaking quietly, or someone else might be interrupting and speaking over her voice. Radio broadcasts were traditionally tuned to a frequency that better suits male rather than female voices. Likewise, modern audio compression algorithms disproportionately distort higher-frequency audio, which means female voices sound “thinner” and less rich (Tallon 2019).

These imperfections in the recording can alert us to the presence or “positionality” of the recordist and draw attention to the materiality and features of the tapes and recording equipment. In many of the recordings that SpokenWeb UAlberta has digitized, speakers check that their mics are working, make jokes with the audience or host about the recording equipment and their facility with them (or lack thereof); in other examples, recordings contain feedback sounds, are muffled, record the voices of audience members nearby, or otherwise store extraneous sounds, all of which the timestamper must note as extra to the subject content of the recording (see Thompson [1975] 2023; Engel [1978] 2023). In one singular example, there is construction work being done outside of the room in which the poet, James Reaney, is conducting a classroom lecture. Reaney spontaneously incorporates the metallic clanking of the jackhammer into the lesson. O’Driscoll, in his analysis of the recording, remarks on the ingenuity of the improvised moment: “most lecturers would be off-put by the force of such overwhelming, machinic noise—one could well imagine cancelling the lecture, especially one focused on attentive listening. Reaney, however, simply absorbs the intrusive jackhammer into the performance of his classroom delivery” (O’Driscoll 2020, 66-67). The harsh and relentless staccato of the drill, which could have distracted from or disrupted the lesson, instead becomes its focus: a real-world example of sonic collage in action. As Reaney quips to his students, “As a matter of fact, you could call this hour ‘Collage with Jackhammer’” (Reaney [1969] 2023, 1:55–1:59). O’Driscoll points out how Reaney, in “engaging” with the “noise,” invites his audience to reflect on how “sonic violence” is demarcated socially and politically.

Listening to the playback, there is another layer of “affective tension” (O’Driscoll 2020, 67), to use O’Driscoll’s phrase, produced by a hyper-awareness of the always already mediatedness of the event. The classroom itself functions as a type of container media, its walls marking the division between wanted and unwanted or invasive sounds; and the industrial chatter of the drill is not the only “noise” on the track. The track is distorted in other ways, which remind the listener of the presence of the recording device. When Reaney’s voice dips in volume mid-sentence and we hear what sounds like chalk scratching, the listener can infer that Reaney has turned away from the students (and the recorder) to write notes on the blackboard. Nadja Wallaszkovits describes phenomenon such as this as “artefacts” of signal processing, the material traces left behind by the contexts of recording, storing, and replaying audio information (Wallaszkovits 2018, 288–289). In other words, what the listener hears is not only an impression of an event but of multiplicities of containers, the contours of which are signalled by the imperfections in the sonic record.

• CLIP 1: James Reaney Reading (Reaney [1969] 2023), on Aviary (0:00–43:50). https://ualberta.aviaryplatform.com/r/3r0pr7nr1k.

The Reaney recording is perhaps one of the most explicit instances of how technology has shaped this archive of literary sound, but it is by far not the only one. The affordances and limitations of media capture devices not only enabled specific types of recording, but also empowered the recordists to travel to the subjects of their interviews instead of having to bring them into an agreed-upon space because of technological constraints (e.g., setting up bulky equipment that does not travel well). In an interview for the SpokenWeb Podcast “Academics on Air” episode, for instance, former audio engineer for the show Terri Wynnyk described the equipment she used while working:

We had a Nagra, which is a Swiss-built small recorder that took small reels, but it was portable so we could take that out into the field. And I have a really strong memory of going in the dead of winter with my arm in a cast and this heavy tape recorder, trudging through the snow from the Biological Sciences building to the Humanities to interview Rudy Wiebe, and it took me forever to get there and get my parka off and get the reels done. Poor Rudy. But he was such a prince, such a king of a man, you know. He gave me this fantastic interview. (Wynnyk in Kroon, Beauchesne, and Miya 2022)

Balan has, in turn, explained how the unobtrusive nature of the handheld recorder (as shown in Figure 2) had a disarming effect, creating an environment for more substantial, in-depth interviews. He would often turn the recorder on before the “official” interview began, making casual conversation, “general stuff, get them comfortable talking,” by way of easing into the harder-hitting questions (Balan 2022, personal communication).

Figure 2

Nagra portable audio recorder, University of Alberta Archives.

The “unboxing” and repackaging of the audio event was part of the initial editing process. Balan and Wynnyk not only worked in the field and in-studio, conducting interviews and writing and voicing the scripts, they also worked behind the scenes as audio editors and sound designers. As described in the same “Academics on Air” SpokenWeb Podcast episode, Voiceprint was known for its unique, collage-like format. The polyphonic vocal effects were created entirely by hand, through a cut-and-paste like process. As Balan says, “these were the days when editing the shows … was a razor blade operation” (Balan 2022, personal communication). Despite the meticulous and labour-intensive nature of the work, Balan relished the “tactile,” hands-on aspect of editing reel-to-reel and the creative freedom that it afforded him (Balan 2022, personal communication). The technical and material demands of manually cutting the clips together shaped the recordings in other ways. As Balan explains, he learned to refine his interview technique, to “get in and out,” so that the version that made it to air required less editing (Balan 2022, personal communication).

As articulated by Altman (Altman 1992) and Birtwistle, and as is demonstrated in these examples, audio recordings have a materiality, both in terms of their measurable sonic features, like volume and pitch, and the technical devices used for audio storage, organization, and playback (Birtwistle 2021, 196). At the same time, sound is also experienced subjectively and relationally, in accordance with the positionality of the listener and her relation to the sounder. With the help of computer technology, these campus radio and classroom lecture recordings are (once more) being remediated and, for the first time, made available online. Close and “care-full” unboxing and unarchiving of these events involves a triangulation of sounder(s), listener(s), and media device(s), often over multiple chronotropic planes.

The digitization of pre-internet analog recordings, which were never intended to be consumed on an online platform, can conflict with key privacy tenets, such as the need to adhere to a stated purpose and to obtain “meaningful” consent, as articulated in Canada’s Personal Information Protection and Electronic Documents Act (OPC 2019, Principles 2, 3). As pointed out by Cowan (Cowan 2020), Fong and Shearer (Fong and Shearer 2022), and Sarah Cipes (Cipes 2022), each reflecting on queer and feminist approaches to archival practice, ephemeral material, once re-published online, can expose the subject to greater scrutiny. As Fong says, in a transcribed exchange with Karis Shearer, there are “different levels of what people are willing to publicly disclose … even conversations around poetry be really, [sic] very intimate and so in that sense there might be ethical impetus to keep those things private” (Fong and Shearer 2022). An interviewee might be more candid and less self-conscious in a radio interview, which was intended to be aired for a brief period, as opposed to a digitized recording that can be slowed down, played back, and transcribed. The conversation might be taken out of context or represent views that the subject no longer holds. She might also reveal information about third parties, including sensitive information, such as “ethnic and racial origins, political opinions … and religious or philosophical beliefs,” that is subject to a higher degree of protections (OPC 2019; see also OPC 2022).

From a legal standpoint, in choosing the proper course of action, it is important to determine whether the records in question fall within the scope of IAPP legislation, as well as how federal regulations relate to regional or special statutes. The administrative activities of public bodies, including post-secondary institutions, fall under Alberta’s Freedom of Information and Protection of Privacy Act (FOIP) and would apply to, for instance, student records. However, there are exemptions for personal information that was collected for “research” or “teaching” activities, as per Sections 4(1)(h) and (i) of FOIP (Government of Alberta 2000). Though what counts as “research” is not clearly outlined in Alberta’s provincial legislation, the Office of the Information and Privacy Commissioner has, in their interpretation of FOIP, deferred to the definition used in Ontario, which is that of “a systematic investigation designed to develop or establish principles, facts or generalizable knowledge, or any combination of them, and includes the development, testing and evaluation of research. The research must refer to specific, identifiable research projects that have been conceived by a specific faculty member, employee or associate of an educational institution” (Ontario IPC 2008, par. 16; Alberta OIPC 2022, par. 18).

Of course, the exemption for research under FOIP does not mean that these activities are unregulated. As per the Tri-Council Policy Statement, studies conducted at post-secondary institutes involving human research participants must be approved by an ethics committee, which at UAlberta would be the Research Ethics Office (REO). There are, however, certain types of academic work that do not fit the aforementioned definition. Research-creation is defined by the Social Sciences and Humanities Research Council as “an approach to research that combines creative and academic research practices and supports the development of knowledge and innovations through artistic expression, scholarly investigation, and experimentation” (SSHRC 2021). In some cases, research-creation involves working with personal data. Though “creative practices” are exempt from ethics review (Government of Canada 2022, art. 2.6), Oliver, Hamilton, and Ingram underscore the importance of communicating with and empowering participants, and outline various ways for artist-researchers to “minimiz[e] harm” (Oliver, Hamilton, and Ingram 2022, 7; for more on informed consent, see University of Michigan 2024).

Archival work, like research-creation, differs from a traditional research study, in that the information is not collected for a “specific” purpose. Archival records might, nonetheless, fall into a different category of exemption, depending on the nature of the information. For instance, according to FOIP, information can be disclosed if the records are more than 75 years old or more than 25 years old and would not constitute “an unreasonable invasion of personal privacy” (Government of Alberta 2000, 43[1][a][i–ii]); and likewise if the individual in question has been dead for more than 25 years (Government of Alberta 2000, 17[2][h][i]).

The authority to disclose personal information is contingent upon the consent of the donor, as suggested in UAlberta’s University Archives Records Acquisition Guideline and Procedure (University of Alberta Libraries 2015). However, there are additional considerations involving affected third parties. Donors cannot supersede the privacy rights of other individuals who are captured on tape or transcript, and this likely also applies to information indirectly disclosed about said individuals. The UAlberta Archives Records Acquisition Guidelines, to the knowledge of the SpokenWeb team, do not clearly stipulate how consent might be impacted by the digitization of archival materials. Would consent need to be reobtained, for instance, if records were made available online?

In order to maintain the trust and goodwill of the literary communities that we engage with, the SpokenWeb UAlberta team formed a Rights Management Task Force, led by Research Associate Ariel Kroon, Governing Board Member Michael O’Driscoll, and Digital Curation Librarian Sean Luyk, whose objective has been to reevaluate SpokenWeb UAlberta’s archival policies and establish special provisions to account for the impact of new tools and techniques, including the digitization of recordings, as a way to ensure that the standards for consent continue to be met. These provisions include seeking and re-obtaining consent and installing take-down measures, which give individuals the ability to request that identifying information be removed.

To begin the work of rights management, O’Driscoll and Luyk drafted a recommendation with a proof of concept rights management process that would apply to the University Archives in the case of SpokenWeb. The work undertaken by the Rights Management Task Force, which involved intellectual property considerations, fair dealing analysis, and engagement with deeper ethical questions, ultimately resulted in the creation of the two rights statements in use by SpokenWeb UAlberta today (one addressed to the rights holder; the other to the estate, depending). SpokenWeb UAlberta does not have a legal right to distribute the recordings by ourselves, and so we are engaged in the work of seeking permission from rights holders in order to make these audio objects accessible in their entirety to scholars, students, and interested members of the public for the purposes of research and study. When asked, O’Driscoll noted that “[the Copyright Office] endorsed it and were appreciative of the work and thoughtfulness that went into it. I think they were most happy with the logic of the position we took, care for the content creators, clear communication with users, and examples of other standards” (O’Driscoll 2023, personal communication). The SpokenWeb Project is committed to the ethical conduct of its curatorial responsibilities, and we try to reach out to, communicate, and work with the creators and their estates to ensure the well-being of everyone involved before going forward with publicization of these recordings.

4. Rights management process

In the experience of the UAlberta’s Rights Management Team (RMT) the most effective rights statement is one that non-academics and non-lawyers are able to read and understand. According to RightsStatements.org, an international consortium that strives to provide rights statements to cultural heritage institutions, rights statements can also be described simply as “simple and standardized terms … to summarize the copyright status of objects in [a] collection, as well as how those objects may be used” (RightsStatements.org 2023), as well as the terms of their re-use and permissions when sharing. Most of the rights holders that the RMT works with are lovers of language, but not legal experts, and so clear language and avoidance of jargon (or explaining what is meant where possible) is key for us. The official email that SpokenWeb sends to rights holders strives to clearly outline what SpokenWeb is and what the RMT at the University of Alberta are doing, explain the process of digitization of archival material, and state the ultimate goal of making these recordings available to researchers, students, and interested members of the broader community. Please see Appendix A for the full text of the email. The email describes the RMT take-down policies and ends with a call to get in touch with the RMT team, either through the official SpokenWeb email account, or by contacting Dr Michael O’Driscoll directly, with any concerns or questions.

The ethical connotations of silence take on another aspect when applied to digital communications: sometimes, emails are never answered or there is an automatic “bounce-back” since the domain no longer exists, and so the conversation halts abruptly, as if a statement or question were met with silence. As Freund observes, researchers and research subjects can imbue silence with many possible meanings, such as a “breakdown of rapport,” the need to self-censor, or even the desire to disclose (Freund 2016, 253). However, while researchers who encounter silence in the course of an oral history interview can fill in the gaps in the audio record with contextual clues—like body language, gestures, or even written notes—digital silences are ambiguous and sometimes left unresolved while the RMT team is left to wonder at the significance: Does an unanswered email indicate a passive form of rejection, or is it simply an indication that the rights holder does not check their inbox?

The moniker “rights holders” refers to the people whose original voices have been preserved on the audio recordings that SpokenWeb UAlberta is digitizing. Figure 3 illustrates a small portion of the research process involved in tracking them down. The RMT wants to be able to speak directly with these rights holders, either personally or, if they are deceased, with whoever holds the rights to their literary works; sometimes this is family, sometimes it is a representative of their estate. In some cases, it is necessary to go through an agent or publisher; however, the RMT tries to avoid those situations, as SpokenWeb UAlberta is attempting to create a community of which the poets, writers, and speakers on these original audio tapes are an integral part.

Figure 3

A flow chart representing a simplified version of the preliminary research process involved with contacting rights holders.

Many rights holders we have reached out to have forgotten that they did these recordings in the first place (named rights holders were approached for permissions; approval was not needed for those who remain strictly anonymous):

I’d forgotten about this reading with the League. Dorothy and I were the first women co-founders in 1966, so this 1970 reading was one of the earliest, after U of Toronto in 1968. How wonderful that U of Alberta recorded it at the time. … Yes, this is a great project for our literary history. (Lakshmi Gill 2023, personal communication)

[The interview] must have been conducted in connection with a sort of celebration of the short story that the university held in 1974 or 1975. [My co-presenter] later became a very good friend, but I didn’t know him at that time and have no recollection of doing an interview with him. (Rights Holder 1 2023, personal communication)

The odd thing is that although I remember the conference very well, I can’t remember my reading there. I have been convinced for years that I never met [my co-presenter]. Of course, I remember the story, and it was published—though not in book form, but I am surprised I had the courage to read it in front of such an august group of writers. The recording shows that a few of the audience members (none of them writers) were at least mildly shocked by the story. … The story would not be considered shocking these days. And the discussion at the end could stimulate more discussions. (Rights Holder 2 2023, personal communication)

Similarly, in cases where the speaker is deceased, we have reached out to and spoken with surviving family members who are thrilled to hear the voices of their loved ones:

My family and I were so touched at hearing [the speaker’s] voice again and it was wonderful!! … Hearing the actual voices of the writers is so much more meaningful and empowers the writing for readers who have not had the opportunity to hear those voices of the writers from the past. … [Our] family is grateful … for giving our loved one some recorded “life.” (Family Member of Rights Holder 3 2023, personal communication)

Dad died 15 years ago and would have loved to know this project was in the works. … This project involving his voice elicits lots of memories and emotion for us … what an honour. Thank you for including him and reaching out. (Family Member of Rights Holder 4 2023, personal communication)

5. Unboxing Voiceprint for online audiences

The “right to know” seems to be at odds with the right to privacy, in a sense to be “let alone” or “forgotten”: one is concerned with making information available, the other is concerned with safeguarding or containing it. However, as Information Commissioner John Reid pointed out in 2001, these two rights are “not inherently contradictory,” and in fact together serve the same purpose (Reid 2001, 2–3). Our own experience has shown how access and respect for privacy can, in fact, complement one another.

As explained, the SpokenWeb UAlberta research team often had limited information about the tapes in our archive; though the CKUA materials had quite a bit of information extant, it was often difficult to decipher, and the EFS archival audio is very sparsely contextualized. The handwritten liner notes, scribbled on the paper inserts or sleeves of the tape, were in some cases the only clues as to the content and context of the audio event at hand. In other words, when the team played back the tapes for the first time, it was not always clear precisely what (or who) was being listened to, and identifying the voices on tape required some investigation.

In the process of researching the campus radio show Voiceprint, Kroon, Miya, and Beauchesne, who were at the time graduate research assistants with SpokenWeb, tracked down Balan, the original host and creator of the show. As mentioned, Balan produced the show in the early eighties, while he was still a graduate student and poet in UAlberta’s English and Creative Writing program. The student-led team, who were working out of a makeshift studio in the basement of the University of Alberta biology lab, built a small following and created a space for cutting-edge literary talent in Canada. Over the show’s three-year run, Balan produced over fifty episodes and interviewed some of Canada’s most well-known literary stars, from Margaret Atwood to Robert Kroetsch.

In the course of Kroon, Miya, and Beauchesne’s investigation, they discovered that Balan was still living and working in Edmonton, and in fact continues to practice journalism. Balan not only agreed to be interviewed by the SpokenWeb research team, he also put the SpokenWeb team in touch with other Voiceprint contributors: Wynnyk, the former co-producer and sound technician of Voiceprint who worked on the show while a social science student; and Jo-Ann Kolmes, a human rights lawyer who was featured in the Voiceprint episode “Women’s Language and Literature.”

• CLIP 2: “Academics on Air” (Kroon, Beauchesne, and Miya 2022), from SpokenWeb Podcast (0:00–50:55). https://spokenweb.ca/podcast/episodes/academics-on-air/.

After tracking down the creators of Voiceprint and learning their stories, the interviews with Balan and Wynnyk became the basis of The SpokenWeb Podcast episode “Academics on Air.” Balan, in turn, generously donated boxes upon boxes of supplementary material including scripts, handwritten interview notes, and more tapes, which are still in the process of being digitized. Wynnyk, for her part, contributed candid photos of the Voiceprint team from her personal collection, which she gave us permission to share on the SpokenWeb UAlberta digital exhibits site. These images, which capture Balan and other members of the Voiceprint production team in action, are themselves valuable historical objects, which offer a behind-the-scenes glimpse of a campus radio studio. What began as a fact-checking exercise grew into a much larger research project. In the process, the SpokenWeb UAlberta archive itself was enriched and expanded. Not only were the research team able to confirm and correct details in our metadata, the audio objects themselves were enhanced and contextualized through a wealth of supplementary material.

The creation of digital exhibits has been fundamental to our project of re-presenting and unboxing our aural/oral archive and offering new ways for the public to engage with these materials. The audio objects in our collection are primarily housed in the UAlberta Library Aviary audio-visual repository. Aviary is an excellent system for managing AV collections, both in terms of the customizability of the metadata schema and the ability to annotate and index audio events. At the same time, Aviary is intended to be a repository, not a research website, so its affordances are based around that purpose. The SpokenWeb UAlberta website was created, in part, as a curatorial tool. The digital exhibits pages guide the reader-listener through a series of subcollections, offering a glimpse into the making of these materials. Rather than a stand-alone collections site, the research site was created to complement and enhance the main repository. While browsing the exhibits, visitors can listen to playlists of related audio material. Clicking on the clip title opens an objects page, which invites the visitor to explore the collection in full on Aviary. The Aviary objects page, in turn, contains links to the exhibits site.

Alex Gil, in his work on minimal computing as critical practice, argues that academics have become “disconnected” from the means of scholarly production, with traditional forms of research dissemination like journal articles mediated through editors and publishers (Gil 2015; see also Risam and Gil 2022; Sayers 2016; Rockwell 2010). Campus radio shows like Voiceprint were, in a sense, early experiments in minimalist, DIY approaches to scholarship, with students like Balan and Wynnyk closely involved in each step of the production process. In creating the online exhibit, we took inspiration from the hands-on approach of the campus radio shows in our archive. Platforms like Omeka and Wordpress offer the convenience of ready-made templates, but also depend on stacks of applications, and so require continual updates and subscriptions to be maintained and kept secure. Static site-generators, used to transform Markdown content into HTML pages, arguably offer a more lightweight and sustainable alternative.

That said, the learning curve for scholars inexperienced with web design can be steep, as Quinn Dombrowski points out in “Minimal Computing Maximizes Labour” (Dombrowski 2022). To create our local site, the SpokenWeb team adapted the Wax workflow (Nyrop and Gil 2023), originally developed for image-based exhibits, to work with audio collections. We customized layouts for the different page types, which included events, exhibits, and audio object pages, and built a media player with playlist functionality. The metadata for the curated collections was scraped from Aviary and stored as a CSV file, which can be updated as needed. Once the initial back-end framework was in place, the textual content could be edited via Markdown, either on GitPages or locally; Markdown being a word processing language that is simple and easy to use, even for researchers with limited technical skills.

Minimal computing as a web development strategy forwards the practice of slow, reflexive archival unboxing in multiple ways. In our case, investing the time and energy to build an audio exhibit using a minimal computing approach was worthwhile, not the least because of the creative freedom it afforded us. Static sites are built with pure CSS, JavaScript, and HTML, and as such they are endlessly customizable. The look and feel of the site, designed by our talented Graduate Research Assistant, Tejas Ambarani, and implemented by developer Ryan Chartier and research fellow Chelsea Miya, evokes the materiality of the audio objects in our collection. For instance, the red, white, and black colour scheme and font style, as seen in Figure 4, were inspired by Sony’s 1970s reel-to-reel tapes. The individual collections are further enriched by extra-audio material, such as archival newspaper articles and photographs.

Figure 4

The “splash page” of SpokenWeb UAlberta’s digital exhibits site.

Furthermore, unlike with dynamic sites, which generate content on the fly by pulling information from databases in response to user requests, static sites are pre-generated. In other words, what you see is what you get. Our website code is freely available to the public and can be copied and downloaded in its entirety. Choosing minimal computing, a process which ensured that the underlying architecture of our website was visible and accessible to the public, was a conscious strategy of unboxing: diverse communities of users can examine our “digital box” and even adapt or transform our code in order to create new audio exhibits of their own.

6. Conclusion

The SpokenWeb Network began “with the opening of a box” (Eastwood 2023) and has led our research team to engage with deeper questions around how to approach “unboxing” with care, through responsible and ethical engagement with the rights holders represented in audio collections. Our case study details how myriad forms of “unboxing”—through activities as diverse as rights management practices and web design—can function as acts of radical unarchiving, in which archives themselves (the proverbial “box”) are posited as vital, growing, in some ways uncontainable. As we have found, attuning to the audio record with care and respect for the voices in it is not about preserving a moment in time so much as active, engaged and continued listening, an openness and attentiveness to the ways that archives are nuanced, contradictory, and very much alive, waiting for the right listener to reawaken them.

In the course of this paper, we first briefly contextualized the origins of the right to privacy as bound up with and contingent on container media. We then reflected on how digitization has transformed archival practices, and we brought new attention to the ethical connotations of audio processing; as well as how audio techniques, like close and careful listening, can be utilized as acts of care, for both the archives and the subjects represented within them. We reflected on how the creation of a rights management statement and policy, rather than boxing in an organization, can open up new avenues of interpretation and creative engagement. Finally, we reflected on the architecture of online exhibits as a media container and minimal computing as a strategy of conscious and careful unboxing.

It has not escaped the authors of this paper that as the SpokenWeb UAlberta archive is time-bound, so too is the work of dearchiving time-bound! The recordings that we are digitizing now are, to quote O’Driscoll and Luyk’s copyright proposal, “stored on obsolete media formats at risk of degradation and loss, and … otherwise inaccessible for research and educational purposes, [and so] the only reasonable way to provide access is through digital means” (O’Driscoll and Luyk 2022, 6). Yet even digital files can become inaccessible due to the retirement of software, upgrading of operating systems and subsequent backwards-incompatibility, or access issues. In twenty years, will the materials in the SpokenWeb repository at the University of Alberta need to be unboxed and repackaged once more in order to keep these files accessible? Perhaps most pressingly, once the project comes to the end of its lifecycle, who will take on the responsibility of caring for and managing these myriad containers and, in turn, the attendant relationships we have built with the rights holders? These are likely scenarios, and one that the SpokenWeb research team as a whole wants to plan for, though we cannot predict in what ways archives will change in the future.

As the “boxes” for storing, organizing, and delivering audio information continue to be reinvented and reshaped, the boundaries of archival objects will keep expanding in new and unforeseen ways. So, too, will archivists and researchers need to continually reflect upon and adapt their rights management best practices and guidelines to accommodate and adequately address these new forms of archival activation.

Appendix A: SpokenWeb UAlberta Rights Management Task Force Email Template

Dear [rights holder name],

I’m writing as a member of the University of Alberta SpokenWeb Project with some exciting news. We are the stewards of a collection of audio recordings made during the 1960’s – 1980’s, [a number] of which feature you and your work. We are eager to make these recordings available to researchers, students, and members of the public through the University of Alberta Libraries Aviary Repository, and we’d like to explore that possibility with you. I’ve attached more information about SpokenWeb below.

Out of respect for the creative labour of the authors involved, we’re reaching out to the individuals who have made these recordings possible to inform them of our plans and solicit their feedback. We would like to make these recordings available to you for review, and invite you to inform us if you have any concerns about making them more publicly available.

Recordings in Aviary will include a rights statement that stipulates that they are for research and educational purposes only; all rights are reserved and future uses (e.g. commercial) require permission from the rights holders. The recordings will be available publicly for free through streaming only. In the event that a researcher or teacher wishes to download the file, any request would be evaluated on a case-by-case basis and would be restricted in use.

Our collection includes the following recordings that feature you and your creative work:

1. Title and Link.

2. Title and Link

3. Title and Link.

You can access these recordings by clicking on the link above and streaming the file on your computer, tablet, or phone. Each file has been made exclusively available to you for a thirty day period.

At the conclusion of that thirty day period, our hope is to make the recordings available as we’ve described, unless we receive from you a request to restrict access to a file, or any portion of a file. In the event that we do not receive a response from you, we will follow up and attempt other means of contact. If we are unable to, or do not receive a response, we will make the recordings available accompanied by a public take-down notice that will allow anyone to inform us of copyright or other concerns.

As we continue our work with this collection, we may discover further recordings that feature you and your work. If you wish, we can contact you again regarding any additional recordings. Please note that other institutional partners in the SpokenWeb network may also be in possession of recordings involving you, and may also be in touch. If you would like us to extend your consent to other recordings across the SpokenWeb Project, please inform us and we can do so at this time. Our hope is that you’ll find the prospect of making these recordings available as exciting as we do.

We are eager to hear back from you, whether to express your enthusiasm or to offer any concerns. We are very happy to address any questions you might have. Please contact us at the email address above, or feel free to reach out directly to Michael O’Driscoll, University of Alberta Professor and SpokenWeb Governing Board Member, at mo@ualberta.ca.

Thank you, and best regards,

[name and position]

Acknowledgments

We are deeply grateful for the support, feedback, and camaraderie of our colleagues at the SpokenWeb. We would especially like to acknowledge Michael O’Driscoll and Sean Luyk: not only was their careful and attentive reading of our article draft instrumental in shaping its focus, without their leadership in establishing Rights Management practices at SpokenWeb UAlberta, we would not have this case study.

Competing interests

The authors have no competing interests to declare.

Contributions