A
Literary Saloon
&
Site of Review.
Trying to meet all your book preview and review needs.
| Main |
 |
| the Best |
| the Rest |
| Review Index |
| Links |
to e-mail us:
support the site
People's Choice Literature
by
Tom Comitta
general information | review summaries | our review | links | about the author
| Title: | People's Choice Literature |
| Author: | Tom Comitta |
| Genre: | Novel(s) |
| Written: | 2025 |
| Length: | 576 pages |
| Availability: | People's Choice Literature - US |
| People's Choice Literature - UK | |
| People's Choice Literature - Canada | |
| from: Bookshop.org (US) |
- The Most Wanted and Unwanted Novels
- Return to top of the page -
Our Assessment:
B : fun concept-novel games
See our review for fuller assessment.
The complete review's Review:
People's Choice Literature pairs two novels in one volume.
Here, Tom Comitta's ambition is to do for literature what Komar and Melamid (in)famously did for painting: create works that, based on what a large sampling of people say they want, are the most, and the least, appealing to them.
Comitta polled 1,045 Americans in December, 2021, asking them seventy-six questions in his 'National Literature Survey' about what they wanted in a novel, and used the results as the basis for these two -- 'The Most Wanted Novel', trying to give the poll-respondents what they most wanted from and in a work of fiction, and 'The Most Unwanted Novel', its elements and attributes everything the poll-respondents wanted least.
(The poll-results are included in full near the end of the 'The Most Unwanted Novel'.)
Beyond the poll, Comitta also relied on Jodie Archer and Matthew Jockers' The Bestseller Code (2016) -- "an algorithmic study of what makes a book a bestseller", as they created: "a computer program to read and analyze 500 books from New York Times bestseller lists" -- to fine-tune figuring out what readers wanted.
He also used an LLM, OpenAI's Playground, in the writing of the novels -- concluding also that such: "public-opinion driven books felt particularly poised to engage with such a technology" -- though also noting that: "Ninety-nine percent of the time, the LLM's outputs needed a lot of help to turn its often-tinny language into something readable".
People's Choice Literature isn't a 'paint-by-the-numbers' work.
The poll results offer only a sort of guidance, with Comitta using them as -- to use the LLM-lingo -- 'prompts' for his own (and his LLM's) creative spin and input.
This is more apparent in the second half of the book, as 'The Most Unwanted Novel' has the authorial presence popping up repeatedly -- "Hi there. Tom Comitta again. Couldn't help but chime in" --, addressing the audience directly and commenting on the text as he goes along.
So, for example, he notes that:
When asked what activities Americans liked to read about, "playing a sport" lost, with only 14 of 1,045 respondents selecting it (1.34%); when asked which sport they do want, tennis received no votes.So, of course, 'The Most Unwanted Novel' features a lengthy tennis-playing scene .....
'The Most Wanted Novel' is, unsurprisingly, the more traditional-looking novel, as almost three-quarters of respondents preferred a realistic novel to an unrealistic one, and almost two-thirds preferred novels to be more traditional rather than more experimental as far as plot and language go. Comitta's far-ranging poll also asked about preferred book-length, with just over a third of respondents opting for 200-300 pages (and another third for the 300-400 page range; interestingly, only eight per cent preferred novels shorter than 200 pages), and so 'The Most Wanted Novel' is on the shorter side and 'The Most Unwanted Novel' on the more bloated side.
The top response to both the question of what kind of novel respondents would like to have if they could commission their favorite author to write one is, as well as simply what their favorite type of novel is was 'thriller/mystery' -- with over twenty per cent saying it was their favorite kind, and over thirty per cent saying it was their second-favorite kind. (Romance was only the favorite of twelve per cent of respondents -- and, rather surprisingly, classic literature was the lowest-ranked, with barely over four per cent ranking it tops; the general American public apparently has really little interest in classic literature (having perhaps gotten enough (too much ...) of it at school ?).)
'The Most Wanted Novel' then is very much a thriller, featuring twenty-five-year-old twins Alix and John Finn, who both work for Genera, "the most advanced tech company in the world". John is actually: "a novelist at heart"; he's not looking to make his career at Genera, but rather works security to fund his true passion; he is working on a novel which he has high hopes for. Alix is one of the six assistants of Genera's CEO, D.J.Wylde -- though, despite her proximity to the visionary in charge, even she doesn't know what the company's big, about-to-be-launched project, Quanta is.
The novel opens with a death -- the first but not the last case of a self-driving car put to use for nefarious ends. (Ford Fusion-model cars are the killer-vehicles of choice in the novel, though I suspect that any current-day (May, 2025) reader-poll would have wanted and expected the vehicles to be Teslas .....) Meanwhile, it seems that John has stumbled on some significant information , and so:
Suddenly the weight of the past week hit him like a colossal wave. The sleepless nights. The exhaustion of keeping the biggest secret in the world. The fear. First finding that document outside his Genera security booth. The desperately working to create an untraceable copy and sneaking up to San Francisco in this ridiculous disguise.The document is on a USB drive -- containing information that some parties clearly desperately want to get their hands on. Indeed, the FBI seems to have reached out and made John a good offer, a pile of cash in exchange for the drive, and he sets out to make the trade. Still, good that there's a copy ... if anyone can find it.
As mysterious deaths and disappearances mount, Alix also gets drawn into the goings-on -- and gets to know FBI Agent Jason Stone. She's worried about John, and not sure she can trust Stone (though there's certainly some physical attraction there ...), while Stone faces a variety of difficulties in his investigations, not least, eventually, because he also is convinced that, as he tells Alix: "you and John are innocent, two twins caught in the wrong place at the worst possible time"
It builds, of course, to the launch of Quanta -- a product which is, of course, ... not good. Believe it or not, however -- things work out. (Surprisingly, less than sixty per cent of poll-respondents said they wanted their novel to end 'resolved', with forty-one per cent preferring 'unresolved', but in a thriller of this kind surely pretty much everyone wants a decent amount of resolution.)
It all adds up to a passable thriller -- definitely B-grade, of the airport or supermarket-rack type, but readable. Fast-paced, with short sentences and chapters (sixty-seven over some two hundred pages), it doesn't hum along like a really solid thriller, but does the job.
Nice touches include having many of the opening paragraphs to chapters provide a brief informational-overview of, for example, the locales where the action is taking place (as a fair number of thriller-writers like to do in their books), e.g.:
Just south of Alcatraz Island lies a forgotten jewel of the bay, Treasure Island. A haven for day trippers, fishermen, and those seeking unparalleled vistas of the Golden Gate Bridge, its tranquil façade belies a rich history. Once home to the Golden Gate International Exposition, a World’s Fair, and then a vital naval base, Treasure Island was created by dredging a shipping channel from the bay to the Port of Oakland. When the U.S. Navy decommissioned the island in the 1960s, the sprawling wonder fell under the stewardship of the City of San Francisco. Today, the island contains seventy-six acres of parkland and 2,500 rental units.Beside making John a would-be author, Comitta uses the City Lights bookstore ("a historic treasure nestled between Chinatown and North Beach in San Francisco") as an important locale -- and nicely slips in some favored-author names, including:
She reached the Is. Rachel Ingalls. Christopher Isherwood. Kazuo Ishiguro. Then the Js. Fleur Jaeggy. Alfred Jarry. James Joyce (sleep aid).(The Joyce-dig is a nod to another thriller *master*, as Comitta notes in his Introduction that among his observations from his research into bestselling-fiction is that: "James Patterson regularly hates on James Joyce’s Ulysses")
When she saw the Ks, she stopped dead in her tracks. Franz Kafka. John Keene. Ken Kesey. She’d gone too far.
Is this the 'most wanted' American novel ? It checks off many of the boxes from Comitta's poll-results, and puts the familiar formulas into practice reasonably well, but it's hard to see this as an all-satisfying novel that the wider public would clamor for. Part of the difficulty in creating the most-wanted is, of course, that readers like a wide variety of things and writing-approaches -- as is reflected in the poll results, where few responses get an overwhelming majority of support. Even where they do, it's not necessarily helpful -- sure, 92.06 of respondents answered 'No' to the question: "Do you prefer reading about characters of a particular race/ethnicity ?" but where does that get you (or rather, Comitta) ? So also, when the top response to the question: "What kind of conflict are you most interested in reading about ?" is -- at 44.78 per cent -- "It depends" ... well, that doesn't really narrow things down much .....
This is less of an issue with the other half of the book, because respondents tended to have stronger opinions about things they didn't like -- in the sense that very few asked for these things. So, for example, Comitta asked: "Which perspective do you prefer when reading novel ?" and while third person (he, she) beat out first person (I) by a comfortable margin, 55.60% to 41.91%, the very clear winner (for unwanted purposes), i.e. the big loser, was the extremely unpopular second person (you), with just 2.49% voting for that. Bingo ! And so much of 'The Most Unwanted Novel' is written in the second person. Only four respondents wanted the novel to cover only a single day of action ? Bingo ! (The question: "Over which duration of time do you prefer your novels to take place ?" was another one of those unhelpfully dominated by the "It depends"-response (47.18%), with more than twice as many opting for that over the runner-up duration (multiple years, at 19.62%).)
(Clear answers regarding what was unpopular were, however, also not always entirely helpful: when asking: "How much dialogue do you prefer in a novel ?" the clear most popular answer -- "A moderate amount of dialogue", with 59.58% -- could be easily avoided, but how to choose between the deeply unpopular choices: "A great deal of dialogue" (3.93%) and "A little dialogue" (0.48) ?)
'The Most Unwanted Novel' begins with a section addressing the reader directly (in which Comitta explains what he's doing here), before segueing into the novel-proper, with 'you' -- 'Lord Tickletext' -- the protagonist. It's Christmas Eve (pretty much for the whole novel, then). And the novel is set ... on Mars. ('Extraterrestrial' wasn't even an option for the poll question "What is your preferred setting for a novel to take place ?" but Comitta seems to have made a safe (i.e. appropriately unpopular) choice with this locale). A nice title-page gives also a lengthy subtitle and content-summary in best overdone classic-novel (several centuries back) style.
With the epistolary form only favoured by 2.68% of respondents, Comitta offers chunks of the novel as 'neural text exchanges'. It also features a book-within-the-book -- the collection (of "Tales of Madness, Death, & Other Perils"), 'Lesser of Two Devils' --, as well as a ballad, and a script-excerpt from Days of Our Tentacled Lives, among much else.
As Comitta sums up at one point, he took the results from the question what kinds of experimental fiction readers preferred and:
To make sure the greatest level of displeasure was reached, I relied on the three most-rejected methods at every chance: encyclopedic sections, author commentary/metafiction, and long meandering sentences.And, indeed, there's a fill of these. (Still, even the *unreadable* six-page list of (real) books 'you' find on the bookcase in VR room 438 -- while also presented in varying font-sizes ... -- is a damn good list (and also makes for a convenient reference list of many of the well-read Comitta's real literary influences (some of which he also mentions elsewhere); many, many of these titles are under review at the complete review).)
Throughout -- and as part of the novel --, Comitta also reflects on his undertaking -- including wondering:
Can a poll actually produce the most unwanted book ? Will “Ball Boys,” the earlier erotic scene between a humanoid cat and a 157-year-old disk jockey, or the extreme violence found throughout this book make this work unpublishable ? If you happen to be reading this book in bound or digital form, then the answer is surprisingly and unequivocally No. And in some ways, this will feel like a minor failure. Could it have gone further ? Are there realms of unwantedness and unpublishability that remain to be seen ?(Among the revealing titbits is also Comitta's behind-the-scenes commentary regarding said sex scene, as he notes that, after composing the first few paragraphs he used an LLM to generate what followed (editing that then into presentable form) -- noting that: "The only reason it was even possible for the LLM to write such vivid, pulpy sex is because it was writing in 2021, when the Playground was only available to selected applicants. At that time, Open AI had yet to censor the LLM’s ability to write porn. If you try to get ChatGPT, the Playground’s cousin, to write a similar scene today, you’ll get nowhere.")
There is some story to 'The Most Unwanted Novel' -- or rather, a variety of stories in this (intentionally) messy heap of a novel -- as well as, as noted, the complete poll both these novels were based on. Obviously not offering the satisfactions of traditional fiction, it nevertheless has considerable appeal, especially in its metafictional games and Comitta's reflections on and in the work.
'The Most Unwanted Novel' is much more 'aware' of what it is and does than 'The Most Wanted Novel', and for those who enjoy that in their fiction there's a lot of good fun to be had here; indeed, I'd argue that, while 'The Most Unwanted Novel' is considerably more challenging (and sometimes frustrating) and messy that the easy- (and fine) but disposable- and forgettable-reading 'The Most Wanted Novel', it is the book that would be more 'wanted' by more readers, because it's more fun and offers considerably greater rewards. (It must be remembered, however, that Comitta's remit is to please (or displease) the entire American population, rather than just more dedicated readers; it seems safe to assume that the most casual or only-occasional reader would certainly prefer the undemanding and modestly entertaining 'The Most Wanted Novel' over the heavy-going and often baffling 'The Most Unwanted Novel'.)
The 'National Literature Survey' that the novels are based and built on is presented in full here, so that readers can see what respondents were asked and how they answered -- an interesting body of data in and of itself. It is worth noting (indeed, emphasizing) that the survey was one of the general public -- i.e. not specifically readers. So, for example, almost half of the respondents reported reading a mere 1 to 4 novels a year, and almost five per cent said they read none.
As someone who is deeply suspicious of polls -- not least because, while I love taking them, it would basically never occur to me to answer truthfully -- I also have some concerns about relying on one for an undertaking such as this one. Comitta apparently put considerable thought and effort into designing this one, but makes clear how hard it is to get as much information as possible -- noting, for example, that he has to consider (i.e. limit) the size of the poll, as: "poll fatigue is a real thing".
As to specific questions (and answers), I found some of the differentiations hard to judge: in asking how much violence respondents prefer in their novels options included: "Some violence" and "A moderate amount of violence", which seem pretty much the same to me; similarly, regarding dialogue, choices included: "Some dialogue" and "A moderate amount of dialogue". More problematically, some (significant) options were missing (the poll was almost entirely multiple-choice) -- including regarding dialogue, where respondents could choose from five options, but were not offered the choice of: 'No dialogue' (surely something people might want from their novels) -- while, for example "No violence" was (sensibly) one of the options regarding how much violence respondents preferred. In many categories, the poll could only list so many options -- and one can argue about the ones included and the ones that are missing.
Some categories also have follow-up questions, asking for more detailed responses to specific responses -- but here the pool of respondents is often very small: the poll asks: "In the books you prefer reading, how do the main characters spend most of their time ?" and only fourteen respondents said: "Playing a sport"; one of the follow-ups to this question asks the slightly different: "Which kinds of sports would you most want to read about in a novel ?" -- but apparently only asks the fourteen who reported 'sports' as the main activity of the main characters in their preferred reading; it's unclear why we wouldn't want everyone to respond to this second question. (That is the question where tennis got zero votes -- but zero out of fourteen isn't necessarily that terrible; four other sports only got a single mention each; more significantly, only eight sports are given as possible answers (and, for example, neither car- nor horse-racing -- which seem like they would be big in the US -- are among the options; neither are, for example, cycling and track and field).)
People's Choice Literature is a fun and interesting exercise. 'The Most Wanted Novel' is an acceptable thriller, and 'The Most Unwanted Novel' engages entertainingly with its premise (since the narrative repeatedly reflects on the actual poll data and how it is worked into the novel, the novel itself can't be separated from the 'constraints' Comitta worked under -- unlike 'The Most Wanted Novel', which can also arguably stand on its own, despite being built on a similar foundation).
People's Choice Literature also offers a somewhat frightening glimpse of our future: there have always been authors who wrote books under a (previous-)success formula, especially in popular series-type fiction, but the much greater amounts of data now available, and the abilities of LLMs, mean we're entering a whole new world. The limited 'National Literature Survey' and the early-days LLM Comitta worked with are rudimentary compared to the tools and data that will soon be available to *writers* -- who, rather than creatively-writing will be prompting and editing, reshaping past successes, dressed up in new ways, as they try to crack the 'bestseller code'. A few experimental variations are all good and fun -- especially if, like here, it is all done and presented very self-awarely -- but once the flood is unleashed .....
- M.A.Orthofer, 4 May 2025
- Return to top of the page -
People's Choice Literature:
- Columbia University Press publicity page
- The Most Wanted Paintings on the Web by Komar and Melamid
- See Index of Contemporary American fiction
- Return to top of the page -
American author Tom Comitta was born in 1985.
- Return to top of the page -
© 2025 the complete review
Main | the New | the Best | the Rest | Review Index | Links