Attempts to address setting too frequently focus on description, but if we read a piece on characterisation that stopped at how to evoke X’s roman nose or Y’s greying hair, we might suspect something is missing. Settings and characters make similar demands on the writer, and it is true that a good setting can function structurally much like a character. When you start a novel, short story or play you have to ask how many characters you include (from zero up), their relationships, histories, story-arcs and temperaments. Likewise, will there be one setting or many? That choice can significantly alter the mood of your book. And if more than one, what is the relationship between the different places? What has occurred there in the past? What is the mood of this meadow or that street? What will change there during the course of the narrative?
Location can be an afterthought to character and plot. These choices are so interwoven that it is tempting to create a character and then merely have them inhabit a world that stereotypically suggests itself—the lawyer’s office, the stay-at-home wife’s suburban house. This is not just lazy, but misses an opportunity; by starting with ‘where’ rather than ‘who’ we might imagine more interesting answers to both queries. There is a mistaken assumption in making characters more foundational than places. It assumes a somewhat static, atomised abstracted notion of people, who shape their world but are never shaped by it. Beginning a novel with dialogue that lacks a context can be tricky for this reason—the reader wants to be able to place the conversation because people talk differently in different settings. Without geographical roots, it is also harder for your character to have a meaningful history. Where someone was raised, the history of that place, will tell you a lot about them. Who would inhabit the dilapidated house where we find Tyler Durden in Chuck Palahniuk’s Fight Club? Who would find herself shut up in the Narrator’s family apartment in the fifth volume of Marcel Proust’s In Search of Lost Time?
We should not only ask how characters and different settings relate, we must also ask how characters relate to their settings. And we need to ask how these relationships change during the course of a novel. Someone’s home may have disturbing symbolic resonances; for example, Robert Louis Stevenson’s revelation that while Dr. Henry Jekyll inhabits the luxurious rooms of a house, Mr. Edward Hyde resides in the murky laboratory below. The more one analyses characters and settings, the more curious parallels can be discovered. Structurally important absent characters (such as the titular of Samuel Beckett’s Waiting for Godot) can find their double in absent places (such as Medea’s previous home of Colchians, present-day Georgia, in Euripides’ play). Unreliable narrators (such as Kazuo Ishiguro’s Mr. Stevens in The Remains of the Day) can find much in common with places hiding secrets (think Thornfield Hall in Charlotte Brontë’s Jane Eyre or the woodshed in Stella Gibbons' Cold Comfort Farm).
There is a great diversity of types of settings. From moveable (e.g. trains, cars and boats) to abstract (virtual worlds and dreamscapes); from stable (the perfectly preserved family home) to unstable (an office during an earthquake or fire). There are also liminal places, the reality and stability of which are under constant suspicion: alternative realities or our own reality with its basic assumptions hewn away. Domain-specific behaviour is a term for the way particular people are expected to conduct themselves in particular settings. The philosopher Nassim Nicholas Taleb uses the example of someone taking an elevator to exercise at a gym. In Austin Wright’s novel Tony and Susan a harrowing scene in which a group of men harass a family, culminating in kidnap, is made worse by the social assumptions of the scene’s protagonist. The roadside location becomes an unreal nightmare space, robbed of basic expectations about safety and civilisation.
One of the advantages I found in studying utopia (a word with etymological connotations of nowhere) was that it is a genre rooted in strange notions of place, and few stranger places can be found than those of H. G. Wells’ metafictional A Modern Utopia. In Wells’ novel two characters, the Owner of the Voice and the botanist, imaginatively build an ideal society around them that later collapses under the weight of their arguments, the world’s internal contradictions and its thinness. In part, A Modern Utopia is a critique of a bad fictional setting. Many utopias, including Thomas More’s Utopia, also point to the centrality of their settings by making it their book’s title. A lot of speculative fiction is rooted firmly in its setting; from H. P. Lovecraft’s ‘At the Mountains of Madness’ to Frank Herbert’s Dune. Speculative fiction has a long history of commenting on the nature of reality itself, which cannot be done without a firm (or deliberately slippery) setting. But even an introspective, psychological character study needs to know where to situate its character, just read Virginia Woolf’s short story, and the basis for her novel, ‘Mrs Dalloway in Bond Street’.
When I taught the short story at university, and every time I edit a new manuscript, I think carefully about the worlds of my students and clients. A great world, replete with its own identity, performing a role in the story, adds a lot. Sketchy, inconsistent or clichéd places can be as deleterious as (and are likely to result in) sketchy, inconsistent or clichéd characters.