Roundtable

The Power of Flawed Lists

How The Bookman invented the best seller.

By Elizabeth Della Zazzera

Monday, July 27, 2020

An advertisement for the Bookman. On a green background, a drawing of a fish reading a copy of the Bookman dangling from the hook on a fishing line. The ad is for June and says "For Sale Here" and "Price 20 cents"

Advertisement for The Bookman, by M.E. Norton, 1896. Library of Congress, Prints and Photographs Division.

Lapham’s Quarterly is running a series on the history of best sellers, exploring the circumstances that might inspire thousands to gravitate toward the same book and revisiting well-loved works from the past that, due to a variety of circumstances, vanished from the conversation after they peaked on the charts. We are also publishing a digital edition of one of these forgotten best sellers, Mary Augusta Ward’s 1903 novel Lady Rose’s Daughter, with a new introduction, annotations, and an appendix. To read more about the project and explore the other entries in the series, click here.


Before the New York Times best-seller list, there was The Bookman. Founded in 1895, the illustrated monthly literary journal was the only place to find out what were supposed to be the country’s top-selling books and remained so until 1912, when Publishers Weekly began producing its own best-seller lists. The Bookman’s first editor, Harry Thurston Peck, a Latin professor at Columbia, took his inspiration for the lists from a monthly British magazine combining industry news and book reviews with poetry, essays, and serialized books that was also called The Bookman, founded in 1891. Subtitled A Monthly Journal for Bookreaders, Bookbuyers, and Booksellers, it served as a kind of model for its American namesake. Published by the major New York publishing house Dodd, Mead, and Company, The Bookman offered its readers bookish news, profiles of prominent authors and their work, both long and short reviews of new books, and, of course, lists of best-selling books. These lists, with all their vagaries and inaccuracies, began to shape discussions about popular literature almost immediately. Books that appeared on these lists, whether or not they were truly the top-selling books of their day, became best sellers because the lists said they were.

Once invented, the best seller could be discussed in literary journals, trade publications, social circles, and book clubs, solidifying a popular conception of what it meant to be a best seller and what it meant to read one. In 1914 the Washington Herald published a short feature, “What Women Should Read,” in which a father advises his son against marrying women who read best sellers, saying, “There have been more domestic disturbances over the best-seller habit than any other one thing.” He claims that women taken up with reading best sellers would neglect their housework—the same argument critics made about reading novels in general one hundred years before. Women had long been considered especially susceptible to the engrossing power of novels, which would both distract them and inflame their passions, potentially leading to immorality. In 1799, in her book on women’s education, religious writer Hannah More wrote that novels were “daily becoming vehicles of wider mischief,” dressing up vice as virtue and teaching “that no duty exists which is not prompted by feeling.”1 Discussions about best sellers as a coherent subset of books also helped establish the authority of the lists as the final word on what people were reading, even as critics questioned the methodologies and advertisers appropriated the term best seller to promote titles, however dubious the claim to the label.

But neither “popular fiction” nor “best sellers” are a genre in the way we usually use that term; they have no distinct generic conventions that distinguish them from other kinds of books. No clear list of characteristics defines the typical best seller, and in fact there is no such thing as a “typical best seller”—even if all can be blamed for distracting men’s wives from cooking them dinner.

Although modern book marketing might have us believe otherwise, it is very difficult to predict what stories will speak to a large number of people at once. Authors who have set out to write a best seller on purpose almost never succeed. As publishers have lamented for at least 120 years, the code that might unlock the secret behind this category is not only elusive, it is ever changing.

 

Best-seller lists appeared on the scene as more and more books were competing to be best sellers. By one count, eight thousand volumes were published in the United States in 1901 and thirteen thousand were published in 1910. This growth of the book industry occurred alongside the growth of the reading public; cheaper production, new copyright laws, the expansion of wealth, free elementary education, and a rise in literacy all collaborated to put a larger variety of books and other reading material in more people’s hands. The novel, perhaps more than any other genre, benefited most from this expansion of both literacy and the book trade. The Bookman did not distinguish between fiction and nonfiction bestsellers—that was a later innovation, developed by Publishers Weekly—and novels dominated its monthly lists.2

Read more about How to Write a Best Seller

The lists were complicated to compile, especially in an era before instant communication or centralized distribution networks, and relied on methodologies that were often opaque and unsound.3 The list makers responsible for these supposedly reliable snapshots of the reading public’s taste have always known that best-seller lists are flawed, and so have publishers and booksellers. In 1899, upon debuting its annual list of best-selling titles, The Bookman explained that “although these lists must be regarded as largely arbitrary, it may be claimed that the results are fairly approximate and represent the books that are most popular at the time.” “Largely arbitrary” and “fairly approximate” are not the enthusiastic endorsements of best-seller lists one might expect from the magazine that brought them to the United States. Despite their many faults, best-seller lists persist because, as sociologist Laura J. Miller explains, they have become important marketing tools for the book industry, especially with the rise of mall book chains in the 1960s and ’70s and the dominance of the New York Times best-seller list, introduced in October 1931. Books that make that list are advertised as New York Times best sellers, displayed prominently in stores, and often discounted by large book retailers. These factors increase sales and keep books on the list—a self-sustaining cycle of best-sellerdom. Publishers’ and booksellers’ use of best seller as a selling point in advertising goes back to the beginning of this story, when, as Miller notes, overgenerous application of the label made supposed best sellers of books that had never graced a list. Although many advertisements touting books as best sellers did have the lists to back up their claims, one 1895 ad’s assertion that Betsey Jane on Wheels is the best-selling book ever printed seems dubious. Even if “everyone [was] talking about it,” the book doesn’t appear on any of The Bookman’s lists for 1895 or 1896.

Ad for Bestey Jane on Wheels claiming it is the best-selling book ever printed.

 

Best-seller lists, with their perennial connection to sales and marketing, emerged at a moment of expansion and increasing commercialization for both the book and magazine trades. As a literary magazine published by a major press, The Bookman was part of that commercialization. The scholar Richard Ohmann argues that mass culture in the United States truly emerged in the 1890s with the growth of the popular monthly magazine like The Cosmopolitan and Ladies’ Home Journal, which, financed primarily by advertising, sold for as little as ten cents; the decade before Harper’s Monthly and The Atlantic Monthly had sold for thirty-five cents. Many of these magazines, aimed at the white middle class, were owned by large publishing houses, where they served as badges of prestige, a marker of the house’s brand, a means to publicize books, and only sometimes a source of revenue. A press with its own magazine could offer a novelist serialization in its monthly followed by a book deal, keeping the profits from both in-house.

Although some book publishers embraced this pivot to magazines as an expansion in reading generally, many were concerned that the magazine boom would cut into the market for books (which it did). But magazines about book culture—like The Bookman and its rivals Book Buyer and The Literary World—directly encouraged readers to think about, buy, and consume books. Yet not everyone was convinced that periodicals like The Bookman, which published book reviews and essays but also industry news and gossip, helped the book industry. Walter Hines Page—the Page of the publisher Doubleday, Page, & Company—remarked that such periodicals paid unceasing attention to the latest best sellers while simultaneously complaining about the commercialization of the book industry, making the whole enterprise of publishing seem degraded. “Did it ever occur to you,” he wrote in his 1905 book A Publisher’s Confession, “that in the ‘good old days’ of publishing there were no magazines that retailed the commercial and personal gossip of the craft?” At the very best, Page thought, these magazines might be benignly irrelevant. He insisted that judgments about books by the “professional literary class” have “no practical importance” and cannot affect the sales of a novel. A review might inform the public that a book has been published, but that, he insisted, was the extent of its power.

But magazines and especially their lists did have a profound impact on the world of publishing. Soon after the lists’ invention, critics and publishers, well aware of both their deficiencies and their power, called for reform. In 1932 M. Lincoln Schuster of Simon & Schuster called for an overhaul of all best-seller lists in the form of “an impartial, nationwide, and corroborated best-seller list,” arguing that the current unsystematic process misled readers and created a flawed record for posterity. He faulted the lists for their reliance on booksellers to accurately report book sales, as well as their exclusive focus on new titles. “Why not a supplementary classification of best sellers from the backlist?” he asked. “Why not show the trend among the enduring books of yesteryear? There are still some people who resent the ‘keeping up with the Joneses’ emphasis on fashion among new books. Some of them follow Emerson’s advice and buy an old book every time a new one is announced.”

 

While critics of best-seller lists never managed to overhaul the system, the lists did change over time. The Bookman tried out a few ways of presenting the data they collected from booksellers, never quite explaining what that data was or how it was turned into a ranked index of book titles.

Initially, The Bookman offered twenty city-specific best-seller lists that it gathered from “leading booksellers” in nineteen cities (New York City merited two lists—one for uptown and one for downtown).4 The first issue tells us that between January 1 and February 1, 1895, the top-selling book in eleven of those cities, plus downtown New York, was Trilby by George du Maurier, which had been serialized in Harper’s Monthly the previous year and published as a book in September 1894.

Uptown New York City, along with Buffalo, New York, preferred Francis Marion Crawford’s The Ralstons, the sequel to his Katharine Lauderdale. The number-one book in Worcester, Massachusetts, was The Prisoner of Zenda by Anthony Hope; for St. Louis, Missouri, it was Round the Red Lamp, a story collection by Arthur Conan Doyle. Over time The Bookman’s editors expanded to gather what they called “authentic reports sent in by the leading booksellers in some forty of the more important cities” across the United States and Canada. Sometimes one city would have two lists with no explanation. It’s possible that while most months The Bookman collected data from one bookstore per city, it would occasionally get sales records from two stores and decide to publish both; yet there are also some indications that they may have polled multiple stores per city for every list.5

Page 64 of the first issue of the Bookman, showing the best-seller lists.

The lists had no way to accommodate sales of multiple editions, which now seems an obvious flaw and ensured that they would include only new books. Critics at the time complained that the lists overemphasized small cities and relied too heavily on the word of booksellers, who might not keep good track of their sales and might even have incentive to lie in order to drive up demand for books in oversupply, like restaurant waitstaff told to push the fish when the kitchen orders too much. In late 1897 the Bookman editors added a monthly national summary aggregated from the city-specific lists, creating for the first time the kind of national best-seller data that today’s lists offer us. But because the six best-selling books in the country were determined “according to the above lists,” aggregation only replicated and even exacerbated the city-specific lists’ methodological issues. Contemporary best-seller lists are mired in similarly flawed and opaque methodologies, as well as being subject to potential manipulation and fraud. All these faulty lists provide only limited snapshots of the habits of the reading public.

The Bookman made annual lists a standard feature beginning in 1899, crediting the idea to an intrepid subscriber, Dr. W.H. Martin, who had used The Bookman’s list to aggregate the six best-selling books for each year since 1895. Because the annual list was aggregated from The Bookman’s monthly national lists, it oversampled books that were briefly very popular while neglecting those that sold in smaller numbers consistently over the year. Best-seller lists favor books that blaze for a moment over those that shine steadily. The Bible may be the best-selling book of all time, but it never sells enough in a given window of time to appear on a best-seller list.

The American humorist and publisher Burges Johnson quipped in 1921 that best sellers should be called “quick sellers.” Sales for a best seller evaporate within six months, he argued, but volumes of Dickens continue to sell in large numbers forty-five years after his death. Johnson understood that best-seller lists record only the “flash in the pan” reading habits of a small segment of the American reading population.

Mr. So-and-So’s society scandals make a very loud noise and then they die. Any publisher in the land, if offered a choice between the works of best-seller So-and-So and the works of Joseph Conrad, for instance, would choose the latter because Conrad’s works are a better property in their second year.

Best sellers (or quick sellers), while temporarily profitable, are also highly disposable. The popularity of books is like the popularity of any fashion or trend—it seems to come from nowhere, is generally unpredictable, and can pass into obscurity just as quickly as it came to prominence in the first place. Trilby sold 200,000 copies in the United States, spawning a long-running play, four silent films, and a new name for a hat, but fell out of fashion so quickly that its publisher was reportedly left with a large number of copies suddenly valuable only as paper pulp.

 

Book distribution faced significantly more obstacles in the nineteenth and early twentieth centuries than it does now. The variety of titles continued to increase, but each book commanded a smaller portion of the overall market. Page wrote that if a book sold ten thousand copies, it seemed logical to believe that another ten thousand people might buy it if they knew about the book and could access it—but the reality of book distribution networks meant that only 2 to 3 percent of the U.S. population was in convenient proximity to a bookstore or the book section of a department or general store. Colporteurs, traveling booksellers who usually sold by subscription, reached places bookstores could not, but their books were much more expensive and more specialized (heavy and highly illustrated) than those available in bookstores.6 Readers in remote locations had to order many books directly by mail. Subscription, mail-order, and likely department-store sales had no impact on best-seller lists. Compiled only from bookstore data, the lists ignored the reading habits of the vast majority of Americans.

Distribution problems affected more than the accuracy of the lists, of course. In 1913 the publisher George P. Brett wrote in The Atlantic Monthly that the demand for books was not growing in proportion to the rise in population and literacy. At the time four out of every five new titles lost money, and the meteoric rise of cheap magazines, along with book piracy and distribution problems, did not help the stagnant book market. The increase in available titles without a corresponding increase in overall sales seemed to negatively affect the lifespan of best sellers. In 1902 The Bookman noted that “the period during which a popular novel enjoys favor is growing shorter all the time,” measured in weeks not months. As Brett put it, “the outpouring of novels is so great that a recent authority states that the life of a ‘best seller’ novel is now little longer than a month, as compared with a period of popularity extending over several years, when the vogue of the ‘best seller’ first became a feature in book publishing.”

A painting of a man standing on top of a ladder in a room covered in built-in bookshelves. He is holding a book open and appears to be reading intently.

In March 1912 Publishers Weekly remarked that although U.S. publishers had released 151 novels since January, it would be “hard to find a legitimate raison d’être for a dozen.” The article expressed concern that the publishing industry, while releasing more books than ever, was failing to justify that increase in output by maintaining a “reasonable standard of excellence.” It asked, “Is publishing as a profession to simply become a method of providing our friends, the printers and bookbinders, with work, regardless of possible buyers—or readers of the product?”

Many publishers and critics were also concerned that the books everyone wanted to read—the best sellers—were the wrong books. They worried that the public read trash while true works of literature, if they managed to get published in the first place, gathered dust on bookstore shelves while waiting to be returned to their publishers. The concern that, if left to their own devices, people would choose the wrong books to read was not new; it stretches back to at least the eighteenth century and the belief that reading novels was destroying women’s minds. But beginning in the early twentieth century critics could point to published lists of best sellers to back up their complaints. Page opened his chapter “Why ‘Bad’ Novels Succeed and ‘Good’ Ones Fail” with a quote from a reader’s evaluation of a book manuscript: “This novel is bad enough to succeed.” A Bookman review of best-selling author Meredith Nicholson’s The Lords of High Decision complained that his stories would be better if he didn’t simplify them to fit the mold of a best seller. Brett lamented that “much of the current fiction is written with a view to supplying just the sort of thrills the public demands,” and that novels of true merit rarely make their way onto lists of best sellers. For Brett and others a lack of success might even signify literary worth and, conversely, best-sellerdom might indicate bad writing. Brett ends his Atlantic article by quoting an unidentified critic and author: “I should consider myself disgraced if I had written a book which in these days had sold 100,000 copies.” A 1905 Bookman article predicting the best-selling novelists for the next few years was careful to note that “most of these men and women are doing excellent work and when we speak of them as ‘Best Sellers’ it is without any intention of disrespect.” Yet even today many books—especially books by women—are seen as losing their literary sheen once they become too popular. Page had more sympathy for the taste of the average reader than many critics, contending that while best sellers may not have literary merit, they have “some very genuine and positive quality.” He elaborated:

They have construction. They have action. They have substance. A series of events come to pass in a certain order, by a well-laid plan. Each book makes its appeal as a thing built, finished, shapen, if not well-proportioned, substantial. It is a real structure—not a mere pile of bricks and lumber. The bricks and lumber that went into them are not as fine nor as good as somebody else may have in his brickyard and his lumber pile. But they are put together. A well-shapen house of bad bricks is a more pleasing thing than any mere brick pile whatever.

Page added that there’s nothing a publisher can do to make the public buy a book they do not want, and a book they do want they will discover “by some freemasonry”—a publisher could not stop them from buying it even if he tried, he insisted. Yet even Page thought best sellers were ephemeral compared to their more literary counterparts. He believed that while a reader might buy a best seller for entertainment on a train ride, the books they really wanted for their collections, to preserve in their libraries, were books of substance, of which there never seemed to be enough published. But Page insisted this is a fate that the public and book publishers have arrived at together.

For the publisher and even the author are but human after all; and in the mood that has possessed us all for a decade or two—since presses and paper became so cheap—we have perhaps worshipped mere numbers. I have published some books only because thousands and thousands of persons would read them. You have read them simply because thousands of other people were reading them and for no better reason. Perhaps our sins have not been heinous. But if you are so stubbornly virtuous as to cry shame at me, I promise you this: I will reform on the day that you yourself reform; but you must first signify repentance. For you—the public—are after all our masters.

 


1 It was this anxiety about the effects of novel reading on women that Jane Austen satirized in Northanger Abbey when her heroine Catherine Morland, her imagination fueled by gothic novels, becomes convinced that a murder has taken place. The same anxiety led Samuel Richardson, the best-selling author of Pamela and Clarissa, to believe that his large female audience had been too distracted by the sentimental plot and characters of his epistolary novels to fully grasp their true lessons. In 1755 he published as a corrective a four-hundred-page book of moral maxims with quotations and page references to his books.

2 As the scholar James D. Hart argues, the novel’s domination of the book market might be explained by the limits of American education: an increasing subset of people were educated enough to read for pleasure even if they couldn’t dissect a philosophical treatise or annotate a scientific tract. The turn of the twentieth century brought not only more educational opportunities but also increased access to books as communities built more public libraries, book-distribution networks improved, and bookselling became a solid business venture.

3 While the term best seller now almost always implies a best-selling book, before The Bookman a best seller was any item that sold better than other items in its category. Its association with books exclusively grew out of the publishing and book-selling industries at the turn of the twentieth century, when use of the term increased significantly and publishing became increasingly commercialized and mass-market.

4 The cities were Albany, Buffalo, New York City (uptown and downtown), and Rochester, New York; Boston and Worcester, Massachusetts; Chicago; Cincinnati and Toledo, Ohio; Hartford and New Haven, Connecticut; Indianapolis; Kansas City and St. Louis, Missouri; Louisville, Kentucky; Portland, Maine; Portland, Oregon; St. Paul, Minnesota; and Washington, DC.

5 To the March 1897 best-seller lists, The Bookman appended a note that one bookseller in Los Angeles, C.C. Parker, sold two hundred copies of The Rubáiyát of Omar Khayyam, Edward FitzGerald’s loose 1859 translation of medieval Persian quatrains, because the store had created an elaborate window display of the book’s many editions and advertised the display in local papers. But The Rubáiyát does not appear on the list of Los Angeles best sellers, presumably because sales were spread out over all these editions, which cost anywhere from 20 cents to $25 (The Bookman list always indicated the publisher and price of each title). This note gives us a rare glimpse into the inner workings of the lists and seems to suggest that The Bookman gathered sales information from more than one Los Angeles bookstore, referring to C.C. Parker as “one of the stores.”

6 Page was convinced that in the face of these distribution issues, publishers wasted their money on trying to advertise a book into a success. He wrote, “When the public has bought a certain number of copies of a novel—of one novel it may be 1,000 copies, of another 100,000 copies—there is nothing that can be done to make it buy another 1,000 or 100,000. It seems to know when it has enough. Take more it will not.”