The process of writing a book doesn’t just involve thinking of a subject, writing the whole book and then finding someone who’ll publish it. Typically you come up with an idea, discuss that with an agent—in my case, Doug Young at Pew Literary—and then write an outline, which the agent shops that around various publishers. Sometimes you also do a beauty contest where you go and talk to the publishers about what they’d do in terms of marketing. For me and Doug, the clear winner was OneWorld.
But Doug also negotiated a contract with Audible, Amazon’s audiobook brand. Once the book had been written, the question naturally arose of who would narrate/record it.
I had no ideas at all, and was waiting for Audible to suggest someone—I was imagining someone like Roger Hallam, whose liquid tones seem to be everywhere—when one of the Audible team suggested I give it a stab.
This was quite a surprise to me. I’ve always thought I had a face for radio and a voice for print, but the Audible folk said that listeners actually like the audiobook to be voiced by the author: they know the nuances of what they’re saying, which can mean complex topics are easier to understand because the emphasis falls on the right part of the phrases and sentences. At least, in theory. For novels, it’s different: you want a proper voice talent (such as Hallam) who can bring the words alive.
So I recorded a short test piece at home, reading some of the introduction. Success! I got the gig. Now all that was needed was to find the studio and some time to record, and, well, record it.
Thus it was that on 31 May 2021 I headed over to Headline Studios in Harston, near Cambridge, to start recording. Five days were scheduled. I honestly had no idea what I was letting myself in for.
24 tracks and one thing on
Headline is a proper studio: there’s a 24-track mixing board, plus a big space with a grand piano (which they were getting rid of—takes too much space) for bands to rehearse/play/record in, plus a couple of smaller rooms for vocalists. And, it turns out, people recording audiobooks. Piers, who runs the studio (and, it also happens, plays a mean guitar), was ready for us. We were still following Covid protocols, hence his mask. Despite that marvellous array of buttons and faders, when you’re recording an audiobook there’s a very specific number of tracks that you need, and a very specific volume fader:
You guessed it. One track, and its associated fader. Pick any of them. The keyboard goes to a computer where he’s recording it, and keeping up with the text I’m reading.
I was despatched to one of the lone vocalist rooms.
You’re in there on your own, with some suitably placed water; I’d also brought a proof copy of the book and an iPad with the corrected copy, all on the table. The studio had supplied an iPad loaded with the corrected copy. From my seat, I couldn’t see anyone else. In my headphones, I could hear Piers in the control room, and down the line there was Toby in the Audible studios.
There’s a Really Proper Microphone to record it all:
Our mission: to get me to read the words of the book without hesitation, mispronunciation, gabbling, horrible accents, speaking too slowly, or speaking too fast, or speaking at an uneven speed measured over the course of the whole five days. Easy, right?
So we begin: me and an iPad and two people hanging on my every word, quite literally. It’s 10am (just after) and we’re scheduled to do this for five days, working until 5.30pm each day. Should be easy, right?
Have you ever listened to a newsreader on the radio and thought it sounds pretty simple just reading aloud some words written down in front of you? Let me tell you, it is surprisingly difficult. Try doing it for the words in those two short paragraphs above, checking yourself for stumbles or mispronunciations. If you get it first time, well done! Now you can continue through the next 90,000-odd words, 310+ pages, of the book. If it takes you a couple of times, well, don’t worry. They can edit it. Digital recording means you can fluff your lines many times and all it costs is people’s time and patience. (Luckily Piers and Toby had colossal amounts of this.)
Onward, onward ran the sentences
Reading out sentence after sentence without stumbling, rushing or dragging (thanks Whiplash) turns out to be hard work. You have to concentrate. And quite quickly, a few things happen.
First: accents. Not thése thîngs, but speaking accents. If you’re reading an audiobook and there’s a quote from, say, an American, should you adopt an American accent? The potential to get it horribly wrong arises at once, and if you get to more unusual accents (or more horribly parodied ones) then things could spin out of control. My recollection is that we decided that I could go for a “soft” American accent for, say, things said by Mark Zuckerberg, but probably shouldn’t go further.
Similarly for quotes from women: should I try to adopt a slightly higher voice, softer tone? The feeling was no.
Second: reading aloud requires you to continually read ahead as you’re speaking, so you don’t get tripped up by an unusual word or sentence construction. Sure, I wrote it all, but that had been a while ago. Quite often I’d be amazed at how long sentences were: they seemed to go on forever and ever and ever. Who the hell wrote this, I’d think to myself, before remembering: oh yeah, me. All these faults are mine. I’d written my book to be read silently, not aloud. If you listen to news bulletins, the sentences tend to be short, with obvious inflexion; learning to write for them is a skill in itself. Reading aloud is a slower process: this article, for example, is expected to take about 8 minutes to read, but 11 minutes to speak aloud.
The first hour or so went fine. By the end of two hours, though, I was starting to flag. Read and read while scrolling and scrolling and pausing and going back over stumbles, retaking (“let’s take it from the start of that sentence.. [three takes later] OK we can cut it in”) turns out to be tiring. You’re in a version of a sensory deprivation tank, seeing nobody, hearing voices in your head which aren’t all your own, struggling not to talk too fast or slow yet also accurately. Fun!
Happily, Headline has a wonderful kitchen, which even though it’s up a short flight of stairs is a lovely haven. One wonders about all the bands who must have lounged around on the sofas, making strong instant coffee at ungodly hours while trying not to turn into the Troggs. There’s a shop a short drive away (not open at ungodly hours) from which you can get sandwiches and other refreshment.
And then it’s back to the word mines. We made progress. Good progress. I’ve no idea how far we got by the end of day one. But we had done well.
On and on and on
Slogging on, we made good progress. Quite often, pronunciation questions would come up. Is the terror group pronounced Al-ky-Ada or Al-ky-EEda? Is Zeynep Tufekci’s surname pronounced Tufek-CHEE or Tufek-SEE? Toby, for Audible, would reliably check such things, and I’d try to wrap my mouth around the correct version.
Far more often I’d trip over what seemed like straightforward phrases. Try saying “algorithmic amplification” four or five times; it’s a phrase that occurs a lot in the book. At first it was amusing, as I stumbled over it like a brick on the pavement. Then it cropped up again, but wasn’t easier. Then again. I grew to slightly hate and then mildly fear that phrase because of the way my verbal momentum somehow collapsed in the valley between the two words.
The gradual buildup of overlong sentences, mouthstumbling phrases and tricky pronunciations has a strange effect. After a while you sense a faint resentment towards whoever came up with this damn stuff. Look, which idiot wr—oh yeah, me. You surrender to the slog, and after a while you start to feel the underlying rhythm of the words, the sentences, the pages, the chapters. You take breaks. You eat. You go back. You finish the day. You come back again and do more. It’s really as simple as that. Some sentences seem to be impossible, and you retake them four or five times until there’s a voice in your ear telling you to start from the middle, because they’ll edit the beginning and end together.
We pushed on, surprisingly efficiently, so that towards the end of day four—Thursday, if you’re counting—we were approaching the final pages. But having nearly completed this Grand National of a challenge, there was a linguistic equivalent of Becher’s Brook waiting in the last chapter.
First question: are you going to try pronouncing that word in a German accent? I think not. (It would be like some awful World War 2 movie.) Second question: can you say the word in isolation? Third question: can you say it in the midst of that sentence, which (if you read it) is pretty long—36 words in six clauses. I had a few goes. And on, I think, the third attempt, I cleared it like (to switch metaphors) a high jumper finally managing to get over the bar. Garnered rather less applause, though.
Anyway, with day four finished, there was really very little left to do. Day five dawned, and there was just some tidying up—a few retakes, a few more pages, the book acknowledgements, and we were done. Wrapped up. Now it was all up to Audible to edit it all together into an audiobook.
Time for a celebratory photo! Since we’d determined none of us had Covid, we dispensed with the protocols for this photo.
So what did I take away from the experience? First, that you should have a try at these things. You never know if you might be capable of doing it. Second, reading aloud is hard work, and all respect is due to newsreaders who don’t get second takes, and to the actors who read novels inserting accents and voices and so on. Third, if you’re going to write writing a book that you suspect you might have to narrate, use shorter sentences.
And avoid the phrase “algorithmic amplification”.
I have tried creating audio files of my writings. I learned that the more my writing is like real speech, the clearer my writing becomes. And the more speaking is like my writing, the more understandable I become. I am considering producing an audio version of each of my Substack posts. Mainly they are long form pieces. Great post. Thanks.
What a great insight into the process. I look forward to enjoying your book in audio :)