Writers of urban fantasy, near future sci-fi, even mystery, historical, and other genres have an important setting decision to make: real or fictitious city? The answer to the question offers a variety of opportunities and potential problems.
There are several benefits to choosing a real city as a setting. First and foremost, the city is already named. Real cities also have maps, politics, locations, notable people, and histories already “pre-made” as it were. This can be great for writers as it saves a lot of the creation process. All the writer has to do is personalize the city to fit his/her worldbuild and genre (e.g. add magical sites, intervening history, tech modifications, fictitious agency offices). Additionally, real cities, particularly famous ones, have reputations, a feel, and form preconceptions in the readers’ minds: compare expectations of New York or Las Vegas to London or Beijing. William Gibson made especially good use of the city’s feel in his foundational cyberpunk works, employing Tokyo his primary inspiration. Ilona Andrews’ magic post-apocalypse Atlanta works well in this respect as well.
On the other hand, using a real city requires a significant amount of research and even a number of visits or long term visits to the city. Invariably, if a real city is chosen, there is going to be at least one reader who lives in or is very familiar with the city. Said readers will jump on errors in description of places, the city’s feel, and the like. Michael Scott and his Nicholas Flamel series is a great example of the level of research necessary, as he explicitly states that he repeatedly visited San Francisco, Paris, and London before writing his series (the vast majority of which occurs in the three cities). Using a real city can also open the writer up to a certain amount of ridicule, depending on the plot and genre. Case in point, the number of jokes circulating the web about the most recent run of Dr. Who and its use of London–ex. a Time Lord can travel anywhere in space and time, but he seems to spend an awful lot of time in early-21st century London.
Fictitious cities have the benefit of being tailored to the writer’s wishes and needs. The writer is not limited to a real place’s history, politics, or geographic limitations. Nor are there reader preconceptions. Fictitious cities also allow for secret societies and hidden communities, such as Eureka’s town Eureka and Sanctuary’s Old City. In both cases, the writers are given free rein to do whatever they wish with the city, in terms of initial design.
On the other hand, fictitious cities take a lot more work in some ways. They have to be designed from the ground up, including mapping, history, businesses and other sites, politics, famous people, cultural feel, everything. Even woth fictional cities, many writers take short cut by borrowing from real cities for some elements, or being inspired by real cities. This even comes into play with maps, ex. Ed McBain’s 87th Precinct series, where he mapped his city by taking NYC and giving it a quarter turn. Another effect of fictional cities can be ambiguity of location, which can be a pro or con depending in one’s perspective. Case in point, Charles de Lint’s Newford, which American readers generally think is Canadian while Canadian readers tend to think Newford is an American city. De Lint says it was inspired by both, but tends to use U.S. laws.