Why Independent Bookshops Still Set the Canon
A bookshop is a machine for making decisions in public. The table by the door, the shelf labelled with a phrase rather than a category, the handwritten card under a novel nobody has heard of: these are editorial acts, performed by someone who will be in the shop tomorrow and can be argued with.
The table as an argument
The front table is the most valuable real estate in the shop, and there are two ways to fill it. One is to accept payment from publishers, a practice the trade calls co-op, which puts the books with the largest marketing budgets at eye level. The other is to choose, which means reading, arguing and being wrong in public.
A table is an argument about what matters this month, and the argument is legible to anyone who looks. Six novels with nothing in common except excellence is a different claim from six novels from one imprint, and a customer can feel the difference without being able to name it.
The backlist as a slow filter
The mid-list is the part of publishing that sells a few thousand copies over ten years: serious novels, translated fiction, books about a single subject by an author with no platform. It is invisible to an algorithm, because the algorithm optimises for the median and the mid-list is defined by being unusual.
A bookshop keeps the mid-list because someone will ask for it, and because the bookseller has decided it should exist. That decision is a subsidy, paid out of the margin on bestsellers and stationery, and it is the mechanism by which a book stays available long enough to become a classic.
How a canon is made
A canon is not a list handed down. It is a set of books that keep being reprinted, taught and recommended, and recommendation requires a recommender. The shop is the place where a stranger’s judgment is available cheaply: a card under a novel that says this is the best thing I read this year, in handwriting, with a name attached.
That is a different transaction from a review, because it is local and accountable. The bookseller will see the customer again, and will hear whether the recommendation was good. Over years, that accountability produces a body of judgment that is more reliable than any bestseller list, and it is the only reason certain books survive their own publication.
It also produces a category of book that cannot be marketed: the odd one, the translated one, the one that is neither a novel nor a memoir and therefore has no shelf in a chain. These books are found by accident, in shops with a person in them, and their survival depends on that person’s stubbornness. A chain can stock a bestseller; only a shop can keep a book alive.
The economics of small
A bookshop runs on a margin of thirty to forty per cent, on a product with a fixed price, in a building with a rent. The arithmetic does not work without events, coffee, stationery or a landlord who is patient. The danger is the obvious one: becoming a café with books, where the shelves are decoration and the till is the point.
The shops that survive tend to specialise. A shop that knows its subject deeply can charge for that knowledge, run events that people travel for, and build a mail-order list of customers who trust the selection. The specialisation is not a marketing strategy; it is a consequence of having a point of view.
Search and encounter
The difference between a search box and a shelf is the difference between answering a question and asking one. Search requires a person to know what they want, which excludes everyone who does not. A shelf presents adjacencies: a book about forests next to a book about paper next to a novel set in a logging town, and the connection is made by the arrangement rather than by the reader.
That is serendipity, and it is not mystical. It is a design property of physical space, and it depends on someone having decided what belongs near what. An algorithm can imitate it, badly, by recommending what other people bought. It cannot reproduce the judgment of a person who has read all three books and put them there on purpose.
The canon is made by people who choose in public and are accountable for the choice. That is a small, unglamorous, badly paid activity, and it is the only one that produces a culture rather than a market. It also happens at a speed a marketplace cannot match: one reader, one table, one year at a time.
A shop that closes takes its table with it, and the table was where the argument was being made. What goes with it is not the books, which are available everywhere, but the argument about which ones deserve the space.