Digitizing: book to audiobook

A couple of months ago I listened to the audiobook of the science fiction novel Sleeping Giants by Sylvain Neuvel. It’s a fun read! This giant mechanical hand is discovered on earth, which it turns out was left on the planet by aliens. What makes the novel interesting however, and what makes it a fascinating audio experience, is that nearly the entire thing is told through interview transcripts. The novel uses different texts to mark different speakers, with the interviewer in bold and the person being interviewed in regular text. For example, the first chapter of the novel begins like this:

Excerpt from first chapter of Sleeping Giants by Sylvain Neuvel (2016, p. 9). Photo credit: https://www.amazon.com/Sleeping-Giants-Themis-Sylvain-Neuvel/dp/1101886692

photo credit: http://www.neuvel.net/

In adapting this novel, the audiobook producers chose to use a full cast. Without anything syntactic like “she said” etc. to indicate the speaker, the audiobook would have become a jumble with just one reader, even considering most narrators’ ability to change the pitch and accent of their voice.

In their article “Conceptualising the Audiobook Experience” Pederson and Have (2012) describe the printed book's “material mode," which involves sight, touch, and a “liberty to organize,” i.e. the reader can read at the pace they want and can jump around in the text (p. 86-7). In contrast, they note that “the audiobook experience differs in almost every respect from that of the printed book:” It involves sound rather than sight, and has little “liberty to organize” - listeners have to read at the pace of the reader, and can’t skip through the text easily (p. 87).

In our reading for last week, Sperberg McQueen (1991) notes that computers can’t contain numbers or texts, only electronic charges, and that computers only process representations of those forms of data (p. 34). He adds that representations “are inevitably partial, never disinterested…” and “reveal their authors’ conscious and unconscious judgements and biases” (p. 34). Although Sperberg McQueen is referring specifically to text encoding, a visually readable representation, I think many of his observations about digitization can be applied to audio forms too. Audiobooks are representations of data, and are therefore unavoidably interpretative. 

To bring these two articles together, as Pederson and Have note, there is in fact a huge change in form or “modality” from physical book to audiobook, and there is therefore that much more opportunity for Sperberg McQueen's interpretive aspect of representation to come into effect. Audiobooks have a narrator who interprets them, who must make decisions about many things, such as speed and emphasis. The choice of narrator is itself also an interpretation (male or female? British or American?).

This necessity for interpretation is even more true with a full cast audiobook like Sleeping Giants, because each voice must be cast and each cast member is another conduit for interpretation. It is essentially a theatre production without visuals. This theatrical element is true of all audiobooks, but a full cast makes it much more obvious.

The audiobook interpretation of Sleeping Giants reveals many aspects of the original we might have taken for granted, and even adds many things that perhaps weren’t originally there at all. For example, the readers add their interpretations of characters’ emotions in the way they voice the dialogue. Also, some characters have accents that, while mentioned in the text, become much more a part of their character when constantly present in the reader’s voice. At one point a character is investigating something when she suddenly screams loudly and the chapter ends; this audible scream adds suspense.

No matter how perfect Sleeping Giants seems to be for the audiobook format, it still fails in some places. In particular, there was a chapter that described a military air deployment that included time stamps. The narrator read each time stamp, which made the listening experience very clunky. I think this problem relates to Sperberg McQueen’s 6th Axiom, "texts are both linear and hierarchical." Texts have a linear stream of characters, as well as hierarchies such as chapters, paragraphs and sentences, and these hierarchies sometimes overlap (p. 40). But in physical texts these things can exist parallel to each other too, I think. For example, books can have footnotes or marginalia. Audiobooks however can only accommodate a single linear system smoothly, since their words must be read out loud in sequence.

This means that audiobooks aren’t ideal for a lot of genres — textbooks for example, or any dense non-fiction book that uses footnotes. Anything with a visual aspect, too, such as graphic novels, or books for which text layout is important, like the Alice in Wonderland mouse tail poem from class, obviously aren't ideal for the audiobook format. But audiobooks can add a lot to fiction through the added interpretative aspects, as long as the reader doesn’t mind a significantly changed experience. Sleeping Giants's dialogic format works nicely as an audiobook digitization, but even so the novel is originally a visual textual object, and so the audio adaptation still fails in places, like when it has to deal with the multiple linearities of time and text in the section described above.

References

Have, I., and Pedersen, B. S. (2012). Conceptualising the audiobook experience. Sound Effects, 2(2), 79 - 95. http://www.soundeffects.dk/article/view/6967

Sperberg-McQueen, C. M. (1991). Text in the Electronic Age: Textual Study and Text Encoding, with Examples from Medieval Texts. Literary and Linguistic Computing 6(1), 34-46. doi: 10.1093/llc/6.1.34

Sylvain, N (2016). Sleeping Giants. New York: Del Rey.