The Question

The controversy around the Hugo committee’s handling of whether Mary Robinette Kowal’s story counted as a story or a dramatic work has brought up the question of just how much audiobooks are separate works from the stories they present.

I wasn’t sure. I thought I’d do an experiment to clarify my views.


I’ve actually got two different hypotheses. I started with one and added another when I realized there might be another explanation for the initial results. I suppose that technically makes it two separate experiments.


First off, if an audiobook is the same as a novel, I’d think that once I’d consumed one I’d be no more inclined to consume the other than I would normally be to re-read that same story. This is the way it is with things like hard covers and mass market paperbacks. Just because a book comes out in a second format I have no sudden urge to buy and read it again. For that, I usually want to wait a year or two so that I forget some of what happens and the story isn’t too familiar.

However, when a movie based on a book comes out, I’ll be inclined to see the movie. Having already read the book doesn’t preclude that - I have a good idea what it will be about but the dramatic presentation is so different that the experiences don’t have a lot of overlap. Often-times I’ll even re-read the book right before or after the movie so that I can better appreciate the differences. They’re so different that having just read the book doesn’t detract from seeing the movie.

So, if an audiobook is more like a written book I’ll not especially enjoy listening to it right after reading the written book. I’ll need to wait longer before I really enjoy it again. If it’s more like a movie then doing one right after the other should be fine.


Another attribute of a separate work is that different people are involved. We concentrate on the author of books, but editors and people who prepare the book for publication also have a hand. The roles of typesetter and cover artist are minimal compared to the author and editor, though. The cover art and typesetting doesn’t determine whether I’ll enjoy a novel. I’ve slogged through poorly OCRed text on a Wheel of Time re-read (Thorn Merrilin? Really?) and cursed mass market paperbacks with missing pages or chapters all because the story was worth it. Likewise I’ve pitched perfectly well put together trade paperbacks because the story didn’t suit me.

In the case of audiobooks, there’s huge differences between narrator performances. They’re far more noticeable than changes to font and page layout. Barring truly incompetent performances (akin to illegible printing), if narrator preference is the determining factor in whether or not I enjoy an audiobook, then it must be a separate work from the written book.

The Experiment

To put this to the test I bought 4 different audiobooks.

  • Rosemary and Rue: by Seanan McGuire, narrated by Mary Robinette Kowal. I read this when it first came out and have not re-read it. I’ve avidly consumed the rest of the books in the series, so this seems like a good test for a book that has rested a while but is ripe for re-reading.

  • Throne of the Crescent Moon: by Saladin Ahmed, narrated by Phil Gigante. I read this about a year ago, some months after it was first released. I really liked it (it got my vote for the Best Novel Hugo), but it’s certainly marginal for a re-read so soon.

  • The Human Division: by John Scalzi, narrated by William Dufris. I read this as it came out week by week. I loved it, but it’s even more recent than Throne, so definitely not normally eligible for a re-read.

  • Old Man’s War: by John Scalzi, narrated by William Dufris. I read this several years ago, but not when it first came out. It’s definitely eligible for a re-read based on timing. As an added bonus, it’s read by the same narrator as The Human Division so it can help rule out narrator preference effects.

So, I have 4 different books that I liked. By all accounts they’re all good audiobooks. I set about listening to them while commuting, exercising and doing work around the house.

The Results

Rosemary and Rue

I loved this book when I first read it and going back to it again was wonderful. Mary Robinette Kowal’s narration was a joy (even if I had to adjust to the correct pronunciation of Luidaeg) and I thought her voicing of the different characters really added to the experience. I’ve already bought the second book as an audiobook since I think re-consuming them this way is going to be great.

Throne of the Crescent Moon

This was straight-up awesome. Phil Gigante does an incredible job narrating a very diverse cast of characters. As with Kowal, he’s very much voicing a full cast. I read this much more recently, but I still found myself having forgotten various plot points. While the dialog seemed familiar I wasn’t mentally quoting it.

The Human Division

This originally came out as a serial and I read or listened to each episode as it came out last spring. This time around I got through the first episode and called it quits. I kept thinking “and this is going to happen next” because it was too fresh in my mind. Also, I didn’t like the narration as much. William Dufris doesn’t do as much of a range of voices as the other narrators and I found it less entertaining.

This difference in narrative style made me doubt my initial hypothesis. What if I wasn’t enjoying this audiobook because of the narration, rather than how recently I’d read the book? Thus was born my second hypothesis about audiobooks being separate works. To test for this, I added a fourth book to the experiment.

Old Man’s War

This is narrated by William Dufris in very similar fashion to The Human Division. It’s an older book I hadn’t read for several years. This is a good control for whether I disliked The Human Division because of over-familiarity or because of the narration style.

I was enthralled. I looked for excuses to listen more. I quickly adjusted to the narration and it became transparent in a way that the narration on Rosemary and Rue and Throne of the Crescent Moon didn’t.


I find that the quality of narration can add or subtract from the joy of listening to an audiobook, in much the same way that font, typesetting and binding can all add or subtract from the joy of reading a novel. However, the primary determinant seems to be the qualities of the underlying story, in much the same manner that the quality of the story determines how much you enjoy reading a novel. Obviously you want a good performance because that makes it more pleasant, just like having a good ereader does. However, they’re still conveying the same story and to me are still the same work as the print version.

I think that audiobooks should be treated the same as printed works and grouped by length for purposes of Hugo nominations. I’d like to suggest that audiobook narrators deserve their own awards category, though. There’s real art going on there that deserves recognition, just as the artists who do cover art deserve it.