Look With Thine Ears:
Remixing Shakespeare for the Public Domain at LibriVox.org
Audio versions of Shakespeare’s plays contribute an interesting perspective to the ongoing debate within Shakespeare Studies over whether early modern playgoers went to hear or see a play. Andrew Gurr contends that “poets wrote plays as poetry rather than spectacle” (105) and argues for characterizing early modern playgoers as an audience, rather than as spectators, since their priority was “listening, not viewing” (1). Critical studies of the senses and sensory perceptions in early modern drama have likewise tended to privilege hearing over seeing, although there have been significant challenges to this perspective. Mark Robson, for example, writes, “I have no desire to deny the strength of orality within early modern culture, but I do wish to complicate a model which seems all too easily to privilege the spoken over the written” (11). Audio Shakespeares render this debate moot, to a certain extent, since they negate the visual and emphasize the aural. Yet they engage simultaneously in a different, albeit related, debate surrounding the merits of contemporary audiobooks. As Matthew Rubery points out, listening to an audio version of a written text has come to be characterized as “not reading” in contemporary Anglo-American culture, and critics of audiobooks perceive them as “posing a threat to concentrated attention or … ‘deep reading,’ the unhurried, meditative immersion in the language of the printed page” (3). Self-proclaimed “audiobibliophiles” like William Irwin critique the perception of listening as “passive” and defend the audiobook as a form, arguing that they “have been mistaken as a refuge for the illiterate and lazy. They have become a guilty pleasure about which we have no reason to feel guilty” (Irwin 363, 367).
Contemporary audio versions of Shakespeare, such as radio performances or recordings like the Arkangel Shakespeare series, reveal that these debates are not simply about whether text or performance matters more to the interpretation of Shakespeare’s plays. Audio Shakespeares place listeners in a unique situation: unlike early modern playgoers (or modern ones, for that matter), they are not receiving visual information, such as gestures, facial expressions, and staging choices, along with the language. However, they are also not reading the written word in a traditional manner, even if they are following along in a printed text while listening to the audio performance. (As Rubery points out, this is relatively rare for contemporary audiobook listeners, who tend to listen while doing other activities, such as driving.) The only visual elements present for a listener must be created imaginatively, and while some see this as a “limitation … [which] preclude[s] simultaneity” (Kliman 276), others see it as stimulating “active listening: [the listener] must wait, and she may then be more likely to wonder and protend and reflect” (Wittkower 225). James Parker, in an article for the New York Times, similarly celebrates the audiobook as appealing to “my mind’s ear – the invisible tympanum against which every sentence must be tested. There’s a kind of honesty involved in the act of giving voice to the written word . . . And there is liberation, too. Children will tell you that being read to frees them up, licenses their imaginations, unhooks them from the fussy horizontal crawl of the printed page” (Parker n. pag.). Further, some see audio Shakespeares in particular as a way of accessing “the ‘invisible essence’ of the play” beneath layers of “visual conceit” (Oesterlen 40).
Producing audio Shakespeare for the website LibriVox.org highlights these debates and conflicts about the merits and limitations of sound over sight. It also extends them by adding in complicating factors: these are amateur audio Shakespeare recordings produced by a global cast of volunteers. Unlike radio versions or professionally-produced audio recordings, LibriVox versions of Shakespeare are communally recorded, edited, and cataloged by site volunteers and released into the public domain, downloadable for free in MP3 and OGG formats for anyone in any country who has access to the internet. LibriVox is defined by its volunteer ethos and prioritizes an open-access and welcoming community: as the “Volunteering for LibriVox” page notes, “We’re all volunteers, and we’ve flipped traditional hierarchy upside down. The most important people in LibriVox are the readers” (https://librivox.org/pages/volunteer-for-librivox). Although a tiny fraction of the site’s over 8,000 readers are professional actors, audiobook narrators, and voice-over artists, most contributors are amateurs: both in the sense of being bibliophiles with a passion for the printed page, and in the sense that they have never been paid for indulging that passion.  Michael Dobson has noted the difficulty of defining and separating “amateur” from “professional” performance: “Once one starts examining the permeable boundaries of what might count as non-professional performance, it is surprising what a high percentage of theatrical activity over the last four centuries might fall within them. The category ‘amateur’ being as unstable as it is, and perpetually defined against the changing and equally shaky category of ‘professional,’ there can be no seamless unitary history of amateur performance” (8). Michael Hancher writes of LibriVox, “The range of accomplishment that different readers bring to their task is remarkable and often impressive – as is the global and social range of English accents . . . [T]he range of reading skills may approximate what many listeners were accustomed to hearing in the eighteenth and nineteenth centuries, when reading aloud was a common social activity” (199).
This essay places the LibriVox project within the concept of electracy, a term coined by Gregory Ulmer to “distinguish the emerging apparatus [digital media] from the established one [print culture]” (28). Electracy “is to digital media what literacy is to print,” and is an emergent apparatus that is being continually produced and re-produced with and through the media itself (Ulmer xii, 7). Jeff Rice and Marcel O’Gorman acknowledge the shift in media studies from literacy to electracy, while keeping in mind that the latter term “integrates [both] orality and literacy” (5). They likewise characterize electracy as a process rather than a static set of products, noting that Ulmer and his fellow practitioners “jump right in and shape the electronic apparatus by inventing new modes of discourse that take both critical theory and digital media for granted” (6). Given these descriptions, it is evident how the LibriVox project fits into electracy. It uses digital media – the internet, digital audio listening technology, and digital recording/editing software—to organize, produce, and transmit its recordings. Electrate users employ this same technology to consume LibriVox products. LibriVox recordings must be in the public domain, which means that most audiobooks are based on texts that were written before 1923. Many are them are, therefore, “classics”: works that have entered the academic canon, and are taught in schools at various levels. The recordings participate in an older discourse of literacy (acquainting listeners with classic texts) through a new electrate medium. The entirely volunteer-run LibriVox forum is a kind of “virtual civic sphere,” as Ulmer describes it, where participants can work together to create these works in a process-driven community (xiii).
I focus on LibriVox productions of Shakespeare in this essay for several key reasons. It helps to narrow down the extraordinary range of audiobooks produced on LibriVox to a manageable subset. It also allows us to concentrate on an author who is still often seen as the center of the Western literary canon: someone whose works are not only “classics,” but have been characterized as essential to the concept of literacy and human identity formation. Most importantly, however, since most of Shakespeare’s works are plays, they allow us to examine the process of producing the works, which are more fully community-driven than most LibriVox projects, due to the complex coordination and editing that dramatic works demand. I will argue that LibriVox Shakespeares are “remixes” of the literary texts, using one of the key terms that Rice and O’Gorman associate with new media. To create a remix, they argue, one must “stop thinking of media in terms of permanent and stable production. In addition, stop thinking of media in terms of authorial creations. Place yourself outside of the authorial concept and allow yourself to become a media-being, one who is a remix as well as re-mixes. Instead of producing ‘true’ texts, consider the alternative, the out-take, the remake, and the remix as new media divergences” (14). LibriVox Shakespeares allow the amateur volunteers to become producers of media and meanings as significant as those generated by Shakespeare’s plays.
I include myself as a member of the “virtual civic sphere” generated by LibriVox, and this essay draws on my own experiences working with the community created by their forum. I volunteered for LibriVox from 2007-2015, and was a member of their administrative team from 2009-2014. I was deeply involved in producing audio versions of Shakespeare during my time at LibriVox. As a reader, I contributed to forty-four productions of plays by Shakespeare, in addition to reading The Sonnets, Venus and Adonis, and The Rape of Lucrece, either in whole or in part. I have edited and served as Book Coordinator for twenty-one of those productions. Given my experience as a volunteer, I am very aware of both the intense rewards and frustrating limitations of LibriVox Shakespeares, which are produced collaboratively by volunteers all over the globe who have no real-life contact with each other. The ambivalent and hybrid nature of these recordings constitutes a unique entry in the catalog of audio Shakespeares, one that has hitherto not been subjected to critical analysis. Using examples primarily from my own experience working with LibriVox, I will analyze the ways in which the volunteers remix Shakespeare, within the limitations and liberations provided by the online and digital process.
Shakespeare occupies a relatively small but significant place in the LibriVox catalog; as of October 2016, all of his plays and poems have been recorded at least once. (Fourteen of his plays have been recorded at least twice.) Of the 425 plays in the LibriVox catalog, 61 (about 14%) are by Shakespeare. No other dramatist comes close to this many titles; the runners-up are Anton Chekhov (fourteen plays completed or in progress), George Bernard Shaw (fourteen), Henrik Ibsen (fourteen), Oscar Wilde (ten), and August Strindberg (eight). Of the other early modern dramatists, Christopher Marlowe has the next-highest number of projects, with six plays completed. Although other early modern playwrights, such as John Webster, Thomas Middleton, Francis Beaumont, John Fletcher, John Ford, Ben Jonson, and Thomas Kyd, have been recorded for LibriVox, Shakespeare remains dominant, reflecting the perceived centrality of his work that is pervasive in Anglo-American culture. There is also a hierarchy within the Shakespearean canon that is perceptible in the sequence of LibriVox recordings of the plays. The first Shakespeare play produced by LibriVox was King Lear in 2006, followed by well-known and often-taught texts such as A Midsummer Night’s Dream (2008), Macbeth (2008) and Hamlet (2009). It is unsurprising that the “lesser-known” and disputed plays have been the last to be recorded.
Since the process of recording dramatic works for LibriVox is radically dissimilar from other forms of audio drama, it deserves to be described at some length. All recordings for LibriVox start with the working forum. There is a separate forum category for Dramatic Works, and projects-in-progress are posted as separate threads. The Book Coordinator (BC) chooses a project, starts a new thread, and provides all the information that prospective readers need to know: where to find the online text, what roles are available to read, how to sign up, and how to pronounce character and place names (if necessary). Readers post to the project thread to request roles, which are usually assigned on a first-come, first-served basis. There are no auditions for any LibriVox recordings; site policy emphasizes that all voices are welcome, and that the only factors precluding the use of an audio file are absolute unintelligibility and/or sound problems. Unlike other audio Shakespeares, LibriVox recordings do not have a director, nor is there a collective rehearsal period. Although the BC actively oversees the project, s/he does not usually provide direction for the readers or propose an interpretation of the play, simply because this is quite difficult to do. The readers record their roles separately, reading their character’s lines with several seconds of silence in between, and the sound files are then edited together (usually by the BC using digital sound-editing software). There is no guarantee that all contributors will have read the play outside of their own role, precluding the possibility of all readers being “on the same page.” There is also a proof-listening (PL) aspect to the project: usually each character is PL’d individually, and then finished acts are PL’d once they are edited. (Sometimes the BC serves as DPL – dedicated proof-listener – for the project, and sometimes that role is taken by another volunteer.) Finally, the recordings cannot, for the most part, use music or sound effects due to copyright issues. (There are exceptions, which will be discussed below.) Dramatic works use a “narrator,” who reads the stage directions and provides intros and outros for the sound files. 
It is probably obvious from this description of the LibriVox process that there are significant limitations arising from the voluntary and public domain nature of the project. Perhaps most importantly, the process precludes any possibility of readers in a dramatic project actually being in the same room and speaking to each other to record their roles and create their characterizations. It is a strange process: the reader must imagine the response of the other readers in the silences between the lines. This diverges from most modern acting methods, which rely on listening and responding to the other actors, even when one is not speaking. Therefore, it occasionally happens, when a LibriVox drama is edited together, that there are characterizations very much out of sync with each other, or that there are individual moments where it becomes all too clear that the readers were not actually speaking to each other. However, as Maire Steadman points out in her analysis of radio productions of Shakespeare’s plays, it is common for individual performers to “have neither any idea of the final sounds that will represent their voice, their interpretations nor of the final structure and progression of the play’s text heard by the audience” (106). Librivox Shakespeares can likewise sound surprising (in either a positive or negative way) when the many different pieces are assembled.
Theoretically, the BC could ask for the reader to re-record their role, either entirely or in parts, to improve the performance; however, as I know from experience, this almost never happens unless there is an actual mistake (such as a mis-pronunciation or mis-reading). The reasons for this arise primarily from the volunteer-driven ethos of the site; as a community-minded project, the first rule of the forum is “be nice.” Criticism of individual readers’ performances is not permitted, and the most important thing is to make readers feel welcome to read and contribute. Further, dramatic works can take a long time to complete at LibriVox, particularly Shakespearean recordings. This is due to their length and the number of speaking roles involved, but also because some readers either do not follow through on their commitment to read, or take months to submit their roles. Some roles are subject to continual re-casting if a reader disappears and never records their part, and in the worst-case scenario, projects can take years to complete. (The first version of Hamlet probably holds the record for Shakespearean recordings; it was started in July 2006 and cataloged in November 2009.) As the first recording of King Lear proves, however, the process does not have to work this way: BC David Barnes finished the recording in a week, and the project was cataloged for the 400th anniversary of the play’s first performance on December 26, 2006. More recently, volunteers produced “flash mob” recordings of The Taming of the Shrew, The Comedy of Errors, and The Merchant of Venice in two weeks, each as part of the celebrations of the 400th anniversary of Shakespeare’s death in April 2016. This kind of speed is very rare, however; in my experience, Shakespeare takes a minimum of three to four months to produce for LibriVox.
Another potential problem area for LibriVox Shakespeares arises from textual issues. LibriVox policy states that readers cannot change the texts that they choose to record: “We present the text as it is written: no additions, omissions, or substitutions” (http://wiki.librivox.org/index.php/ Recording_%26_Text_Policies). The concept of recording a text as it was “published” or “written” is problematic when it comes to early modern works, especially Shakespeare, given the unstable nature of these texts, which often exist in multiple published versions. It has become standard practice for editors of King Lear, for example, to include both the Quarto and Folio texts in their editions so that readers can see the significant differences between them. Some include a conflated text that combines these two versions as well, as in the Norton Shakespeare complete works. Deciding which text to use is a problem that besets every BC who takes on a Shakespearean project. The most immediate question is: where do I get the text? Most LibriVox recordings use Project Gutenberg (http://www.gutenberg.org), an online treasure trove of public domain books, as textual sources, with a secondary source provided by the Internet Archive (http://www.archive.org). With early modern plays, however, the BC must be aware of formatting issues that might pose problems for readers. The Project Gutenberg text of The Two Noble Kinsmen, for example, is rife with archaic spelling and character name abbreviations, which pose significant challenges for readers who are not used to reading early modern drama. On another level, privileging the “whole text” reduces the possibilities for interpretive engagement with Shakespeare that is nearly always a factor in other audio Shakespeares (and indeed, in live and recorded Shakespearean performance generally). Because of this policy, the text cannot be cut, rearranged, or shaped in any way, nor can music or sound effects be used to further artistic or interpretive ends. The policy is certainly justifiable, given the public domain nature of the project. As the website states, “Adding music or other effects makes it harder for us to achieve our objective by introducing complicated copyright issues” (http://wiki.librivox.org/index.php/Music_Rules). The only place where it becomes permissible to use music “is when music is specified in the text”; the policy then recommends that the reader sing/play the song using a public domain tune. For example, in Shakespeare’s Henry VIII, the 3.1 song was written, sung, and played by a volunteer, ezwa.
Sound file: Song from Henry VIII. [This song – “Orpheus With his Lute” – is from Shakespeare and John Fletcher’s Henry VIII, Act 3 Scene 1. Performed by ezwa as the singer, Elizabeth Klett as Queen Katherine, and Tricia G as the narrator. The rest of this recording can be accessed here: https://librivox.org/henry-viii-by-william-shakespeare/]
This is a rare endeavor, however, for reasons similar to those indicated above: time, ability, and commitment on the part of volunteers. Implicit in the LibriVox textual policy is a presumed objectivity about the nature of the “whole text” that breaks down under scrutiny. The suggestion is that if we record the entirety of a Shakespeare play we will be somehow accessing the “thing itself.” Yet, as Shakespeare scholars know, there is no true, authoritative text, only the version one happens to be reading. To suggest that LibriVox volunteers and administrators are unaware of these issues is to do them a disservice, however. Catalog pages of finished projects always display a link to the text used in recording, partly to prove their public domain status, but also as a record of what the readers used, variants and all. Similarly, there have been recordings of different versions of the same text, such as the two different versions of Marlowe’s Doctor Faustus.
While the process of creating audio Shakespeares for LibriVox has significant limitations, there are also substantial benefits for electrate volunteers working within the parameters provided by the forum and by digital media. LibriVox volunteers have remixed Shakespeare in a variety of creative ways, becoming producers of meanings through their active and process-driven engagements with his texts. I will look at four different kinds of remixes that demonstrate this point, with representative examples: projects that use audio editing to interpret scenes where multiple characters speak together (as in Coriolanus); projects that convert Shakespeare’s poetry into “dramatic readings” (as in Venus and Adonis); projects that juxtapose purposefully selected excerpts of Shakespeare’s plays to create curated collections (as in the monologue and dialogue collections); and projects that generate participant interactions on the forum to produce a more directed and focused final production (as in the second version of A Midsummer Night’s Dream).
When I decided to serve as BC for LibriVox's Coriolanus, I knew that it would be an arduous project; as one of Shakespeare’s lesser-known tragedies, it could be more difficult to communicate to listeners what was going on. It has over three dozen characters, many of whom have complicated Roman names; for example, “Coriolanus” is an honorific surname given to a successful Roman general, Caius Martius. Achieving unanimity of pronunciations alone would prove to be a laborious process. However, as I discovered once in the thick of the project, editing the crowd scenes would also be a serious challenge. The play dramatizes Caius Martius’ failed bid to turn his success in battle into a political career. Contempt for the common people proves to be his undoing, and Shakespeare includes several key scenes of the citizens mutinously challenging governmental authority. The opening scene of the play sets the tone:
Enter a company of mutinous Citizens with staves, clubs, and other weapons.
FIRST CITIZEN: Before we proceed any further, hear me speak.
ALL: Speak, speak!
FIRST CITIZEN: You are all resolved rather to die than to famish?
ALL: Resolved, resolved!
FIRST CITIZEN: First, you know Caius Martius is chief enemy to the people.
ALL: We know ’t, we know ’t!
FIRST CITIZEN: Let us kill him, and we’ll have corn at our own price. Is ’t a verdict?
ALL: No more talking on ’t; let it be done. Away, away!
SECOND CITIZEN: One word, good citizens. (1.1.1-13)
Our recording of the scene includes readers who had agreed to take on the roles of the individual unnamed citizens, but it also contains a medley of volunteers who separately recorded the lines assigned to “All” in the text. I copied and pasted their lines into separate tracks so that they overlapped each other to convey the effect of a crowd of people speaking at the same time, but not in unison. The finished product sounds like this:
Sound file: Coriolanus crowd scene #1. [Coriolanus, Act 1 Scene 1. Performed by Chuck Williamson as First Citizen, Patti Cunningham as Second Citizen, Joshua Letchford as Third Citizen, Diana Majlinger as the narrator, and other volunteers as the crowd voices. The rest of this recording can be accessed here: https://librivox.org/coriolanus-by-william-shakespeare/]
As you can hear, I did not strictly adhere to the text in how I chose to edit the scene. For example, I had Third Citizen alone read the line “No more talking on’t; let it be done,” and then brought in the rest of the citizens to shout “Let it be done!” and “Away, away!” However, if you listen closely, you can hear that some of the citizens are simultaneously shouting, “No!” and “Wait!” This anticipates Second Citizen’s objection to First Citizen’s proposal that follows.
I made similar choices in editing the violent crowd scene in Act 3 Scene 1 of the play, in which the tribunes, Sicinius and Brutus, incite the citizens to arrest Coriolanus for treason, over the pleas of two patricians, Menenius and Cominius. There are disagreements over how to render parts of this scene, particularly some of the lines attributed to the “rabble of plebians” that enter in response to the tribunes’ summons. For example, these lines – “Tribunes! Patricians! Citizens! What, ho! Sicinius! Brutus! Coriolanus! Citizens!” (3.1.-234-235) – are variously attributed to the whole crowd, or to the Second Senator, depending on the edition. These lines obviously can have very different effects depending on whether they are spoken by a single person desperately trying to fend off a crowd, or by a large crowd with a diversity of foci and agendas. Here is how it sounds in the LibriVox recording:
Sound file: Coriolanus crowd scene #2. [Coriolanus, Act 3 Scene 1. Performed by thebicyclethief as Coriolanus, Bob Gonzalez as Cominius, Algy Pug as Menenius, Ric F as Sicinius, Ron Altman as Brutus, Chuck Williamson as First Citizen, Diana Majlinger as the narrator, and other volunteers as the crowd voices. The rest of this recording can be accessed here: https://librivox.org/coriolanus-by-william-shakespeare/]
I chose to have all those lines spoken variously by crowd members, drowning out any individual voices. Yet, as you can hear, I chose to interpret a key line assigned to all the citizens in the text—“True, / The people are the city” (3.1.250-251)—as spoken in a low, intense manner by an individual citizen, with a few other citizens simultaneously echoing “True! True!” This was to assure that the line received particular emphasis in the context of the scene. While my audio editing skills and equipment are recognizably those of an amateur, my interaction with the text was heavily process-driven, including: soliciting recordings of crowd voices from project volunteers, selecting which voices to use where, deciding which version of the text to present and how, and mixing down the edited voices into a single audio track. The end result is a remix of Shakespeare’s Coriolanus, one that conveys an interpretation that moves beyond the scope of the so-called “original text,” using the voices of volunteers from around the globe.
Creating a LibriVox version of Venus and Adonis was a much smaller-scale undertaking than Coriolanus, as it involved only three volunteers including myself. Yet it constituted a different kind of Shakespearean remix: a “dramatic reading” of the narrative poem. I prepared the text for this project, taking an online version of the play and cutting and pasting it into a Word document via Google Drive. I marked up the text to indicate speaking roles for three different readers: myself as the Narrator, Arielle Lipshaw as Venus, and Bob Gonzalez as Adonis. I color-coded the text, highlighting Venus’ lines in green, Adonis’ in red, and leaving the narrator’s lines black. A sample from the text featuring all three looks like this:
And now Adonis with a lazy spright,
And with a heavy, dark, disliking eye,
His louring brows o'erwhelming his fair sight,
Like misty vapours when they blot the sky,
Souring his cheeks, cries, 'Fie! no more of love:
The sun doth burn my face; I must remove.'
'Ay me,' quoth Venus, 'young, and so unkind!
What bare excuses mak'st thou to be gone!
When recorded and edited together, the scene sounds like this (with the rest of Venus’ speech included):
Sound file: Venus and Adonis. [Lines 181-222 from Shakespeare’s Venus and Adonis, performed by Elizabeth Klett, Bob Gonzalez, and Arielle Lipshaw. The rest of the poem can be accessed here: https://librivox.org/venus-and-adonis-by-william-shakespeare/]
Remixing the text in this fashion makes it into something new: a work more akin to a play than to a poem. This kind of project takes advantage of electracy to forge a new digital form, one that has proven very popular with LibriVox volunteers. My version of Venus and Adonis is only a small example of how the dramatic reading form has been used at LibriVox. Volunteers have taken lengthy novels, by authors such as Jane Austen, Charlotte Bronte, Charles Dickens, and Elizabeth Gaskell, marked up the texts to indicate the speaking roles, and cast these prose works with different voices reading the characters. While LibriVox volunteers did not invent this format—there are, after all, professional “full cast” readings of novels available on Audible and elsewhere —they have embraced it with enthusiasm and fervor, producing 79 completed works, with 13 more in progress as of October 2016.
Another kind of remix project very much in the spirit of the LibriVox endeavor is the production of short works collections. LibriVox has an entire Short Works forum, which includes collections of short stories, poetry, and essays, as well as the weekly poetry project, in which a single poem is recorded by as many readers as possible and then cataloged together. The Dramatic Works volunteers have created a number of similar projects, including one-act play collections and dramatic reading collections. They have produced two different kinds of Shakespearean short works projects: monologue collections and dialogue collections. There are thirteen completed monologue collections in the catalog, each containing between fifteen and twenty-five speeches from Shakespeare’s plays, selected and performed by individual volunteers. There are no restrictions on what can be recorded for these collections, except the number in each collection, and the fact that an individual collection cannot contain two or more recordings of the same monologue. While the contents of these collections often feel random (generated as they are by the desires of individual readers to perform their favorite speeches), there are interesting juxtapositions that nonetheless emerge. For example, there are two multilingual monologue collections that compare the same speech in English and a different language. In Collection 12, Leanne Yau reads Hamlet’s Act 2 Scene 2 soliloquy (“O what a rogue and peasant slave am I!”) alongside Fabiola’s rendering of the same speech in Italian. Here is what they sound like next to each other:
Sound file: Hamlet Act 2 Scene 2 soliloquy, in English and Italian. [English version read by Leanne Yau; Italian version read by Fabiola. These monologues are from the following collection: https://librivox.org/shakespeare-monologues-collection-vol-12-by-william-shakespeare/]
While this juxtaposition was clearly serendipitous, others are purposeful. For instance, in the tenth monologue collection, Patrick Wallace performs John of Gaunt’s speech from Richard II in both English and German, allowing listeners to hear what they sound like together:
Sound file: Richard II Act 2 Scene 1, in English and German. [Performed by Patrick Wallace. These monologues are from the following collection: https://librivox.org/multilingual-shakespeare-monologues-collection-010-by-william-shakespeare/]
These kinds of remixes relate directly to LibriVox’s mission to make texts available in a wide variety of languages, either through the abilities of multilingual readers, or by bringing together volunteers from different parts of the world in the “virtual civic space” of the forum.
The Shakespearean dialogue collections create different kinds of juxtapositions from the monologue collections. I coordinated two dialogue projects in my time at LibriVox, both of which were conceived and produced as thematic collections. The first one brought together fifteen scenes around the theme of “Wooing, Wedding, and Repenting” (adapting a line from Much Ado About Nothing). The idea was to compare wooing scenes in different genres; there are scenes from comedies, tragedies, histories, and romances. By placing these scenes side-by-side, listeners have the opportunity to identify interesting points of convergence between the selections. A number of the scenes involve coercion, for instance (as in The Taming of the Shrew, Richard III, Measure For Measure and Cymbeline), while others utilize cross-dressing by one character to create a queer context for the courtship (as in As You Like It and Twelfth Night). Parental opposition is often an issue for young lovers (as in Romeo and Juliet, The Winter’s Tale and The Tempest), while more mature lovers are impeded by social and political concerns (as in Much Ado About Nothing and Antony and Cleopatra). The second collection sought to draw on the numerous female contributors to the Dramatic Works forum by bringing together “Women’s Worlds” from Shakespeare’s plays. (All the readers for the finished collection were women, except for Martin Geeson, who delightfully read Volumnia in a scene from Coriolanus.) As with the first collection, the project juxtaposed the scenes to highlight similarities and differences. Perhaps most prominently, almost none of the scenes would pass the Bechdel Test; while the dialogues are populated by named women talking to each other, it is almost always about men. Only the scenes from history plays—Henry V and Richard III—transcend this concern, as you can hear in these recordings:
Sound file: women’s dialogues from Henry V and Richard III. [Henry V, Act 3 Scene 4, performed by Tiffany Halla Colonna as Katherine and Elizabeth Klett as Alice; Richard III, Act 4 Scene 4, performed by Caprisha Page as the Duchess, Arielle Lipshaw as Elizabeth, and Elizabeth Klett as Margaret. These dialogues are from the following collection: https://librivox.org/shakespearean-dialogues-collection-002-by-william-shakespeare/]
These collections remix Shakespeare by creating themed playlists, of a kind. Volunteers engage in a curatorial project so that listeners can engage with Shakespeare’s texts in a new and electrate way.
Perhaps the most process-driven remix projects I have coordinated for LibriVox are those in which I have tried to use the online forum to replicate, as closely as possible, the conditions of theatrical rehearsals. I first tried this idea with a non-Shakespearean project, Henrik Ibsen’s The Master Builder, because it seemed small-scale enough to accommodate a more intensive process. For the play’s many scenes of two characters talking to each other, I had one reader record their role first with lots of space in between lines, so that the second reader could listen to the first while recording her lines. So, for example, as Hilda Wangel I was able to listen to and answer Bruce Pirie’s reading of Halvard Solness, allowing me to be “in the moment” with my responses to him. This was a satisfying—if one-sided—experience and produced effective finished results.
Sound file: excerpt from The Master Builder. [This excerpt from Act 3 of Ibsen’s play is performed by Bruce Pirie and Elizabeth Klett. The complete recording can be accessed here: https://librivox.org/the-master-builder-by-henrik-ibsen/]
However, I discovered that this makes an arduous production process even more labor-intensive, and is really only feasible with very small casts. I decided that I would never try to make this work with a Shakespearean recording, which can have anywhere from twenty to sixty different roles.
Our production of The Master Builder incorporated thematic analysis of the play on the forum by all the readers, and I adapted this process when I coordinated A Midsummer Night’s Dream in 2012. This was the first “version two” of a Shakespearean play for the catalog. I decided to see if it would be possible to record a version of the play with a coherent interpretation that all readers would discuss and apply to their individual roles. Along with the project thread, I set up a separate discussion thread dedicated to readers’ analyses of the text. I also provided a number of conditions for participation in the project in the working thread. “If you would like to participate in this project, you agree to do the following: read the text in its entirety (not just your role . . .), participate in the discussion thread for this project, and record your role with the discussion and interpretation in mind.” The interpretation I chose was a “dark” approach to the play, articulated in the discussion thread as follows:
While this play is, of course, a comedy in all the traditional ways (focusing on courtship and love, ending with multiple weddings) the more recent performance history of the play has often emphasized not-so-comic elements, such as the following:
• The violence inherent in many of the central relationships.
• The irrational, fickle, often unpleasant and mobile nature of love.
• The darkness and menace of the forest settings of Acts 2-4, and the potential malevolence of the fairies.
• The perhaps brutal eroticism of the various romantic couplings (particularly Titania and Bottom).
I limited the number of readers to twenty by combining some of the roles (Titania/Hippolyta, Oberon/Theseus, Puck/Philostrate). As I told the readers in the discussion thread, however, the doubling was not necessarily only for expediency, but rather to convey the connection between the fairy and human worlds of the play. Yet the doubling raised another issue that we debated throughout the process: “How can we effectively convey an interpretation with only our voices? In a stage production, there's physical action and visual cues that can signal an interpretation to the audience; yet we have only audio. How might doubling roles function in a production where the audience can't see the characters?” I am not sure we ever resolved this issue, or came to a consensus of how to practically address it through our readings. Thematically, the discussion between the readers raised productive points, like the following comment from Bruce Pirie, who read Egeus:
I suppose the doubling of the actors in the original production would have drawn attention to the subtle variations in the two sets of relationships. In the case of Theseus/Hippolyta, we see the couple that has been through their battle; the dust has settled, and the victor decided. Order — of a kind — seems established in their world. When we slip into the dream-world, however, we find ourselves in the thick of the fray: Oberon and Titania, played by those same two actors, are strategizing and struggling, and the outcome of their battle is up for grabs. Until it is resolved, the world suffers the disorder described in Act 2, scene 1. Interesting that a play about a bunch of lovers is actually quite a bit about war, beginning with the recollection of a military victory, and proceeding to various battles: battles between generations, between lovers, and between fairies.
While the discussion board stimulated collegiality, promoted intellectual engagement, and contributed to the LibriVox mission of a welcoming and open community, it did not necessarily lead to a coherent and focused interpretation of the play in the final product. We never really resolved the issue of how to “translate” the theoretical into the practical. Also, not all of the readers participated in the discussion thread, nor did I, as BC, enforce participation, again in deference to the volunteer spirit of the site and the busy lives of the many readers.
The final version of the play did reveal interesting choices that were connected to the “dark” reading of the play, however. Jeff Schwab, reading Francis Flute, decided to emphasize the serious nature of Flute’s engagement with the role of Thisbe in the final interlude. Initially he used a very high-pitched falsetto voice when Flute was playing Thisbe that was clearly awkward for Flute to maintain. As the interlude went on, Flute began coughing, until finally he relinquished the fake voice altogether and played Thisbe’s final moments in his own voice. Jeff conveyed the sense that when Flute discarded artifice he really started to connect with his role; the final moments are slow, serious, and quite moving.
Sound file: A Midsummer Night’s Dream, Act 5 Scene 1. [Jeff Schwab as Flute in A Midsummer Night’s Dream. The entire recording can be accessed here: https://librivox.org/a-midsummer-nights-dream-version-2-by-william-shakespeare/]
Although I have seen performances of the play where this scene was played seriously (the 1994-95 Royal Shakespeare Company production, for instance), Jeff successfully used his voice here to bring another dimension to Flute’s engagement with his role. Similarly, the readings of Dan Raynham and Elizabeth Barr as Lysander and Hermia successfully translated the serious nature of the play to the “mind’s ear.” Elizabeth contributed a lengthy analysis of her character to the discussion forum, including the following observation:
Hermia's situation is, perhaps, the most bleak at the beginning of the play. She is in the unfortunate position of being in love with someone who her father despises (for no distinguishable reason) and, because she refuses to bend to his will, she faces a horrifying choice, required by law: be put to death or become a nun. Or, of course, she could behave like a ‘proper woman’ and marry her father's choice. I think the actual horror of this situation is pushed aside in most productions- can you imagine being dragged by your father in front of the Duke so that he can invoke a law that could result in your death? TRAUMATIC!
Her analysis was clearly apparent in her reading, which comprised anger, indignation, horror, and even trauma. While Dan’s appraisal of Lysander was limited to observing that he seemed like a “typical bloke,” his reading clearly conveyed the character’s sexual motivations once he had Hermia in the wood, and lingered cruelly over his horrific insults of her, as in this clip from 3.2.
Sound file: A Midsummer Night’s Dream, Act 3 Scene 2. [Elizabeth Barr as Hermia, Dan Raynham as Lysander, Bear Schacht as Demetrius, and Kristingj as Helena in A Midsummer Night’s Dream. The entire recording can be accessed here: https://librivox.org/a-midsummer-nights-dream-version-2-by-william-shakespeare/]
Although there are undoubtedly limitations inherent in recording Shakespeare for LibriVox, the liberating effects created by projects like the “remixes” I have described certainly outweigh them. While the fact that the readers are all literally all over the map problematizes the recording process, it affords the opportunity for bringing different voices and perspectives together to work on the same project. It also offers the listener a singular opportunity to suspend disbelief, in ways that are different from the suspension of disbelief that is inherent in live theatrical performance; the listener must enter into the fiction of the audio recording and believe that the readers are really talking to each other. Although I am undoubtedly biased in my perception, I find this remarkably easy to do, despite my consciousness of the highly constructed nature of these recordings (which take many hours to cut and paste together). When I edit a dramatic work, I am constantly surprised and delighted at the relationships that emerge between the readers and their characters. The variety and richness of our volunteers’ voices and experiences create a finished product that is simply unavailable anywhere else.
As Arielle Lipshaw notes of her experience editing Titus Andronicus, “I was thinking about how wonderful it was that all these people from all over the world had gathered together to read Shakespeare” (https://librivox.org/2012/01/22/librivox-community-podcast-122/). Given the global nature of the site, LibriVox Shakespeares are interestingly positioned with respect to local and national identities. Unlike the amateur performances traced by Dobson in his cultural history, we cannot locate a LibriVox recording’s relationship to local or national identity and culture. No reader or group of readers can “claim” a recording or a play for a particular nation, thus detaching Shakespeare from the service of British national identity, in particular. Eve-Marie Oesterlen’s article on a radio version of King Lear recorded live at the Globe Theatre in London and broadcast globally by the BBC analyzes how such a production could be mobilized in the service of British nationalism. The LibriVox recording of King Lear, by contrast, was produced by a global cast of volunteers in a variety of voices and accents, effectively undercutting local/national identities. Further, in contrast to professional recordings like the Arkangel Shakespeare series, LibriVox is completely detached from market concerns. Aside from a fundraising drive in 2010 that met the site’s minimal financial needs, LibriVox has never solicited funds, and even allows others to profit by the recordings. (LibriVox files are in the public domain and can, therefore, be downloaded and used for a variety of purposes. Some eBay users burn the files to CD and sell them, for example.) Trudi Rosenblum’s article on the Arkangel series reveals how closely the recordings were tied into marketing the Penguin print editions of Shakespeare’s plays, and quotes Penguin’s director of audio as saying, “Penguin’s strength in educational marketing is going to make these tapes … the first choice of schools, colleges, and libraries” (30). While on one level the Arkangel recordings were designed to sell as many tapes/CDs and books as possible, LibriVox volunteers are “anarchists of the free market,” in Michael Hancher’s terms (201): proudly, almost defiantly, non-profit. Yet ultimately, rather than differentiating between the two, LibriVox Shakespeares trouble the line between amateur and professional, between text and performance, and continually renegotiate the problematic tension between volunteerism and artistic interpretation. They are, therefore, indebted to the liberating possibilities created by the digital media with which they work. They also work to continually challenge and transform the possibilities of those media, for electrate volunteers and listeners alike.
 Catalog statistics as of March 28, 2017 show that there are 8,016 LibriVox readers. There are 10,598 completed works in the LibriVox catalog, of which 1,402 are non-English works, and which are read in 36 different languages.
 There is a standard LibriVox intro, identifying the file as a public-domain LibriVox recording, which must be included on all files. The outro is simply “End of Act 1” and the like. Dramatic works narrators read only what is written in the text, unlike some narrators of early twentieth century radio versions of Shakespeare, who, according to Bernice Kliman, provided interpretive material to aid audience comprehension (279-282).
 An example of a professional dramatic recording of prose works is Philip Pullman’s His Dark Materials trilogy, read by the author as the “narrator” and with a full cast reading the individual characters.
Dobson, Michael. Shakespeare and Amateur Performance: A Cultural History. Cambridge: Cambridge University Press, 2011. Print.
Gurr, Andrew. Playgoing in Shakespeare’s London. Third edition. Cambridge: Cambridge University Press, 2004. Print.
Hancher, Michael. “Learning from LibriVox.” Audiobooks, Literature, and Sound Studies. Ed. Matthew Rubery. New York: Routledge, 2011. 199-215. Print.
Irwin, William. “Reading Audio Books.” Philosophy and Literature 33.2 (October 2009), 358-368. Print.
Kliman, Bernice W. Hamlet: Film, Television, and Audio Performance. Rutherford, NJ: Fairleigh Dickinson University Press, 1988. Print.
Oesterlen, Eve-Marie. “Lend me your 84 million ears: Exploring a special radio event – Shakespeare’s King Lear on BBC World Service radio.” The Radio Journal 6.1 (2008): 33-44. Print.
Parker, James. “The Mind’s Ear.” The New York Times, 25 November 2011, <http://www.nytimes.com/2011/11/27/books/review/the-minds-ear.html>. Accessed 17 January 2017.
Rice, Jeff, and Marcel O’Gorman. “Getting Schooled: Introduction to the Florida School.” New Media/New Methods: The Academic Turn from Literacy to Electracy. Ed. Jeff Rice and Marcel O’Gorman. West Lafayette, IN: Parlor Press, 2008. 3-18. Print.
Robson, Mark. “Looking with ears, hearing with eyes: Shakespeare and the ear of the early modern.” Early Modern Literary Studies 7.1/Special Issue 8 (May 2001): 1-23, <http://purl.oclc.org/emls/07-1/robsears.htm>. Accessed 17 January 2017.
Rosenblum, Trudi M. “For Penguin AudioBooks, the Play’s the Thing.” Publisher’s Weekly (April 6, 1998): 30. Print.
Rubery, Matthew. “Introduction: Talking Books.” Audiobooks, Literature, and Sound Studies. Ed. Matthew Rubery. New York: Routledge, 2011. 1-21. Print.
Shakespeare, William. The Norton Shakespeare. Third edition. Ed. Stephen Greenblatt et al. Norton, 2015. Print.
Steadman, Maire J. “Audio Shakespeare.” Sh@kespeare in the Media: From the Globe Theatre to the World Wide Web. Ed. Stefani Brusberg-Kiermeier and Jörg Helbig. Peter Lang, 2004. 103-114. Print.
Ulmer, Gregory L. Internet Invention: From Literacy to Electracy. New York: Longman, 2003. Print.
Wittkower, D. E. “A Preliminary Phenomenology of the Audiobook.” Audiobooks, Literature, and Sound Studies. Ed. Matthew Rubery. New York: Routledge, 2011. 216-231. Print.
Elizabeth Klett is Associate Professor of Literature at the University of Houston – Clear Lake. Her teaching and research bring together Shakespeare and literary studies, drama and performance studies, women's and gender studies, and cultural studies. She is the author of Cross-Gender Shakespeare and English National Identity: Wearing the Codpiece (Palgrave, 2009) and articles on adaptations of Shakespeare in theatre, film, television, and dance. She has been published in the academic journals Theatre Journal, Shakespeare Bulletin, Literature/Film Quarterly, Shakespeare, Early Modern Studies Journal, and Borrowers & Lenders, as well as the collections The Oxford Handbook of Shakespeare and Dance (2017), Shakespeare’s Hamlet in an Age of Textual Exhaustion (2017), and Shakespeare Re-dressed: Cross-Gender Casting in Contemporary Performance (2008). She is also an audiobook narrator and voice artist who releases both professional recordings through Audible.com and volunteer projects via LibriVox.org and Onlinestage.org.