Notebook: Listening Session
In 2024 record numbers are doing their reading by ear. What’s going on, and what does it mean for writers and readers?
Listening, as a supplement or alternative to reading words with one’s eyes, has more and more become a feature of modern consciousness. I can remember my first encounters with audiobooks. Peggy Miller, the boss’s legendarily stately secretary at my first publishing job years ago at Farrar Straus & Giroux, used to click off once a month on her tidy heels to record books for the blind. I would work sometimes in the Andrew Heiskell Braille and Talking Books Library in Chelsea, which had big reel-to-reel listening devices for books. (The Heiskell Branch website notes, “when applying for a patent for his tinfoil phonograph in 1877, Thomas Edison listed ‘phonograph books, which will speak to blind people without effort on their part’ as one of the ten potential uses for his invention.”) I had friends who had multi-cassette copies of (respectively) Suetonius’s Lives of the Caesars and Shelby Foote’s book on the Battle of Vicksburg (his voice still in the air from Kevin Burns’s epic first series on the Civil War) strewn about their cars. (The earliest big audiobook producers—Books on Tape, Recorded Books, and Blackstone Audio—were founded by commuters.) But now that we carry gadgets around in our pockets that connect directly to our brains, these reading voices seem ever more intimately to permeate daily consciousness.
The audiobook market has been growing and growing since the introduction of smartphones and has been a lifeline for publishers and writers, expanding steadily by over 50 percent since 2018 and now accounting for around 10 percent of book sales. (There was a bit of a dip during the pandemic lockdown, alongside a rise in old-fashioned reading, likely because so many people stopped commuting.) In 2023 audiobook sales reached $2 billion, and more than half of all adults reported they had listened to an audiobook. Library patrons, who set a record in 2024 of over a million digital checkouts, borrowing worldwide over 739 million ebooks, audiobooks, and digital magazines through digital library book distributor Overdrive’s Libby and Sora apps, checked out 278.3 million audiobooks (up 19 percent over the previous year) compared to 366.2 million ebooks (up 7 percent).
Sales of audiobooks rose 27 percent in the first nine months of 2024, an acceleration fueled by a new audiobook entrant, the music subscription service Spotify. In 2023 Spotify added fifteen hours of audiobook listening to its standard music subscription, which readers could bump up with $10.99 installments to read more, apparently boosting total audiobook sales by about 10 percent. In spite of early fears, authors seem to be earning through Spotify: novelist George Mahood told The Guardian that, although Spotify’s royalty rates are lower than other audiobook platforms’, he’s showing more returns from Spotify. Michael Cader wrote in Publishers Marketplace that Spotify “has been putting real money into the hands of publishers and authors, marking one of the few initiatives in our world from a big tech player that actually seems to have grown the market.” Robert Thomson, CEO of HarperCollins’s parent company News Corp, “has repeatedly called Spotify a ‘game changer’” for audiobooks, according to Edward Nawotka at Publishers Weekly. Edward Nawotka notes that Spotify’s recommendation system, which draws on readers’ musical tastes as well as deliberate book recommendations, is contributing to discovery and, with Spotify’s vast audience of music listeners, drawing new and younger audiences to books. Spotify reported last spring that one quarter of Spotify Premium users had “engaged” with audiobooks; of its 602 million monthly active users worldwide, 236 million are paying subscribers; users aged 18 to 34 accounted for 57 percent of its audiobook listeners. Jane Friedman in her newsletter The Hot Sheet quoted Nir Zicherman, global head of audio and gated content at Spotify, telling audiences at a FutureBook conference in the fall that Spotify “isn’t focused on taking a share of existing sales; they’re targeting casual listeners and people who have never listened to audiobooks.” “The opportunity is to grow the pie,” he said.
An important difference between audiobooks on Spotify and other platforms is the departure from the “credit model” that has been pervasive in audiobook listening: on Amazon’s Audible and other platforms, readers buy a whole audiobook at once, either paying for it directly or using a “credit” that they receive with a paid monthly membership. Subscribers on Spotify can sample around in different audiobooks and listen for a while and drop away without making a financial commitment. Publishers and authors are paid some if you make it part-way through a book on Spotify, and the full royalty if you read the whole thing.
Nir Zicherman said at the FutureBook conference that “convenience, choice, and discoverability are the three critical factors in growing the audiobook market.” Jane Friedman says, “Spotify is absolutely correct that the credit model has warped how consumers buy audiobooks.” The “credit worthy” problem means that “consumers are less likely to choose shorter books, children’s books, or any book that doesn’t feel ‘worth’ their monthly credit.” Spotify “is offering subscribers the flexibility to mix it up, explore a range of books, and pick the right book for any moment.” Studies of audiobook listening in Europe, where a Spotify-like subscription model is prevalent, show that the model favors “backlist” listening, to already-published books, where the “credit” model favors new books, where discovery is more reliant on a book’s publicity and marketing. In Europe some authors have seen higher earnings with the boost that the subscriber model gives to their backlist, according to Jane Friedman. Another audiobook executive, Nathan Hull, said at the FutureBook conference that “publishers ought to be experimenting more often, given what we do know about audio consumption, which tends to happen in small chunks during morning and evening commutes. Yet publishers still focus on mirroring the book and producing a product that’s nine or twelve hours long.” (Such observations mirror studies showing that younger people move between reading and other platforms throughout the day [see our “Book Note” a couple of years ago].)
Traditional publishing has long feared subscription models like Netflix, where readers potentially receive universal access to books without paying for a particular work. At the Penguin Random House (PRH) / Simon & Schuster merger trial in 2020, PRH CEO Markus Dohle characterized subscription models as a mortal threat to the industry, and when PRH reached a deal with Spotify (after having pulled their books from all subscription services in 2020), it was a signal that advantageous terms were on offer. A “Coalition of Concerned Creators” has demanded more transparency from Spotify, registering concern about the impact that the all-access model “will have on author compensation, the value of books, and the literary industry more broadly.” Coalition member Kim Scott told The Guardian that “on average, people only read 50 percent of the book, and so that’s going to cut author pay by 50 percent … the system is paying what’s easiest to measure instead of what’s most valuable.” But an Authors Guild statement outlining the terms they could identify found them to be comparable to existing audiobook compensation. Consultant Andrew Rhomberg wrote a year ago that the negotiated terms were “a triumph of large English-language publishers resisting the Netflix-style model and convincing Spotify to embrace a-la-carte pricing”—i. e., you as a listener can move in and out of audiobooks as you can music, but Spotify pays authors and publishers for each book accessed. By December Jane Friedman was wondering if Spotify would have to introduce credit purchasing to make audiobooks sustainable for them, whether Spotify were taking a loss to build a program and a customer base. If Spotify is in effect subsidizing these robust returns, the threat that an all-subscription model might yet corrode author earnings as it has musicians’ on Spotify may yet materialize.
At the very least, Spotify has broken the stranglehold on audiobooks that has until now been enjoyed by Amazon’s Audible. (Though there are other credit-based audiobook providers! Best is to order audiobooks from Libro.fm, which channels sales through independent booksellers; you can also buy audiobooks from Kobo, Barnes and Noble, Google Play, and others.) The science fiction author and digital rights advocate Cory Doctorow offered a gruelling account of trying to get around Audible/Amazon to sell an audiobook of his novel Red Team Blues directly to readers in 2023. He wrote, “I believe we simply can't afford to have all our audiobooks under the control of any single company, much less one that is as committed to wringing margin out of its suppliers as Amazon is.” Competition has its benefits: publishers have noticed that competition from Spotify is already compelling Amazon to put more marketing effort behind its audiobooks.
Another shock to the world of audiobooks is the arrival and dramatic improvement of AI narration, making possible inexpensive production of audiobooks for smaller publishers, independent authors, and more obscure books. Writer platforms like Galatea allow self-published authors to generate instant audiobooks using AI voices. The Washington Post had an interesting story by audiobook reviewer Katharine Powers about tensions within the world of audiobook narrators over whether or not to participate in this development, the same year, 2023, that Apple and Google Play announced tools to generate AI-powered narration of books on their platforms. The debate resembles the dilemma in journalism over whether to strike deals with AI companies or sue them for copyright infringement. Human narrators are expected to remain in demand for high-profile projects. Quixotically, as Katharine Powers notes, the poems of the anthology I Am Code: An Artificial Intelligence Speaks were read for its audiobook by the distinctly actual voice of Werner Herzog.
Readers used to debate whether audiobook listening counts as “reading.” Not so much any more. I listen to a lot of audiobooks and definitely consider it reading. It helps me fit more books into my already book-larded day. I’ve listened to a number of big substantial works of social science and reportage that I probably would not have carved out time for otherwise. On the other hand, I made a conscious effort at a certain point to move away from news and listen to fiction, hoping to retrieve some reflective mental space that I had lost to the cacophony of the days’ alarming events. I am nothing but grateful for the things I have learned and the experiences I have had listening to books.
I do wonder, though, what starts to happen if reading words on a page (or screen) becomes widely replaced by oral experience. A correspondent asked the “spiritual advice columnist” at Wired if there is anything wrong with listening to books instead of reading them: “Part of me wants to read more, but I find it much easier to listen,” they wrote. “Easier”? Reading is an act of engagement and active interpretation. Psychologist Cody Kommers wrote in Psychology Today that “reading is something you do, where listening is something that happens to you.” (Studies have identified various reasons why print reading seems most amenable to retention.) Wired’s spiritual advice columnist reflected in their reply that listening might “eliminate the reader’s responsibility to interpret things like irony, tone, and inflection,” but I think it goes deeper. When you read a sentence to yourself it is up to you to identify its logic and internal structure. As audiobook director Christina Rooney told the authors of a consideration of labor issues in audiobook narration in the Los Angeles Review of Books, “you’re not casting a voice” to read an audiobook, “you’re casting a brain.” I wonder if people who grow up listening will begin to lose the cognitive muscle that tracks meaning in a sentence. I have sometimes found AI reading hard to follow because the subtle syntactic inflection following the logic of sentences and paragraphs is not registered by the reading voice in the ways we have learned to expect in the spoken language. Wired’s spiritual advice columnist noted the ancient lineage of oral culture but did not consider that oral literature is composed to be received and preserved orally: the passage from oral traditions to written ones follows the passage from poetry to prose. Neil Postman (whom I seem to keep bringing up) is eloquent on the extent to which modern intellectual life is rooted in print literacy.
I also worry that our increasing reliance on extemporaneous forms like podcasts migrates the culture away from the logical and evidentiary demands of writing. (Book Post author Hugh Eakin has written about the lack of factchecking infrastructure in podcasting.) Listening to nonfiction audiobooks at least keeps in our linguistic diet arguments that are reflected upon, syntactically rigorous, developed at length, in an environment where podcasts are accustoming listeners to improvised thinking and conversational diction. Reading sentences on the page and interpreting them for our inner ear obliges us to recognize within ourselves how ideas are rooted in language, the material of thought.
For me nonfiction that has narrative pull and (conversely, perhaps) fiction in the realistic tradition fits most comfortably into audio experience. I’ve listened with pleasure and benefit to Saul Bellow, Hilary Mantel, Philip Roth, James McBride, Hernan Diaz’s Trust. Writers who were more lyrical, or layered, like Jesmyn Ward, Marilynne Robinson, Virginia Woolf, demand (of me) to be read on the page. I listened to Sigrid Nunez’s The Friend as an audiobook and completely missed its montage-ish paragraphing, a loss. Unlike most, apparently (including Cody Kommers in Psychology Today), I dislike audiobooks with multiply-voiced characters, or even narrators who take on too-distinct voices for characters. I think it’s part of the experience of reading to refract different voices through oneself. If readers become so used to listening that they lose—or never develop—the ability to process these more complex, layered, syntactically demanding readerly experiences, literature will have lost out, our appreciation of works of the past will be flattened, and our intellectual apparatus will weaken. Or they already have.
Just now I’m listening to Russell Banks’s great* Cloudsplitter (*I haven’t finished it, so perhaps I should hold back on the conclusive adjectives), an excellent book for listening. It’s an accident that I am coming upon it just now. I listened to Continental Drift a few weeks ago and was left so uneasy by it that I needed to keep going with Russell Banks. In the novel, the main character, who is John Brown’s son Owen, is faced over and over with the question of whether an extreme wrong demands an extreme response. The novel does not ask the question and try to answer it at one go: it requires that we live through the dilemma in as many iterations as life offers, and respond to it through the multiple emotional permutations and wavering convictions that a person can contain. A timely story to be feeding into my mind through a little cord at just this moment. I’m glad, or grateful, to have it in the air for me, not just tucked in the pages of a book.
Ann Kjellberg is the founding editor of Book Post. You can read her other Notebooks on publishing, writing, and the life of ideas here.
Book Post is a by-subscription book review delivery service, bringing snack-sized book reviews by distinguished and engaging writers direct to our paying subscribers’ in-boxes, as well as free posts like this one from time to time to those who follow us. We aspire to grow a shared reading life in a divided world. Become a paying subscriber to support our work and receive our straight-to-you book posts. Some Book Post writers: Adrian Nicole LeBlanc, Jamaica Kincaid, Marina Warner, Lawrence Jackson, John Banville, Álvaro Enrigue, Nicholson Baker, Kim Ghattis, Michael Robbins, more.
Community Bookstore and its sibling Terrace Books, in Park Slope and Windsor Terrace, Brooklyn, are Book Post’s Winter 2024 partner bookstores. Read our profile of them here. We partner with independent booksellers to link to their books, support their work, and bring you news of local book life across the land. We send a free three-month subscription to any reader who spends more than $100 at our partner bookstore during our partnership. To claim your subscription send your receipt to info@bookpostusa.com.
Follow us: Instagram, Facebook, TikTok, Notes, Bluesky, Threads @bookpostusa
Interesting read! Audiobooks are amazing for folks who may not be able to engage with a text some other way, but for me, I lose too much listening—I can't see the page layouts or punctuation, and these things feel crucial to me in trying to get what an author is doing. I also resent the insertion of another mind/interpretation between mine and the author's—even when it is the author's own voice, their intonation, breath, and pacing add a layer between me and the words that I don't want; there is often so much to be found in a text beyond what an author intends that gets obscured if someone else is speaking it. If a work was specifically written to be performed or read aloud, that's different, and I imagine a fair few books now are written with audio audiences in mind. Podcasts aren't meant to be read, and I don't mind listening to a play. But I haven't found audiobooks to be much use to me as a reader.
I used to be an avid consumer of books on tape and on CD/DVD when I regularly drove 500 miles a week to and from work, but since I retired, I avoid recorded books because I dislike being led around by the nose. Reading is an active experience. One pauses, rereads, or skims at a roaring pace depending on reactions to the text. Listening is at at the pace of the recorded voice, not my pace.
The voice is also present in reading. I and--I suspect--everyone else can spot AI generated text within a few lines because the AI voice is garbled and incoherent; an average, not an intellect.