One of the unique problems (or joys, to some of us) of writing about invented worlds is the need to make up names for everything. Authors who set their stories in the real world (even if it’s fantasy or SF) can simply scan through lists of forenames or surnames, or even the local phone book. It isn’t quite that simple, of course (would Harry Potter be as popular if he’d been called Dudley Dursley? I doubt it, though it’s hard to define why) but the basic building blocks are all there for the author to play with.
Fantasy requires new names for a huge range of components: not only people, but also countries, cities, villages, rivers, seas, forests and mountains. And, just as with the real world, the choice of name for each is going to affect the reader’s response.
There are a number of approaches to this. Many authors use RW names, and some are very successful with it, but I personally dislike this approach. It tends to jerk me out of my suspension of disbelief if I come across something too familiar, especially if it has other associations. George R. R. Martin, for instance, has a town called Ashford and a character called Jon Snow. Both perfectly good names, unless you’re more used to them as a station on the Eurostar and a TV newsreader. That makes it harder to take them seriously.
Using RW names for a fantasy world is very difficult to get right, since names have general associations, even if not specific ones. A common mistake is to mix cultures. If, for instance, your hero has a classical name like Demetrius, why would his brother have a Norse name like Thorvald? There might be a specific reason, of course – parents from different cultures, deliberately using a foreign name etc. – but, without that, it’s going to feel wrong to a lot of readers. And it can be ridiculous if you use names with too modern a flavour. I could buy a sword-and-sorcery hero called Edmund or Richard, if the story was good enough, but not one called Wayne.
I find it far more convincing if the names are all invented, but that too has its pitfalls. Inventing names doesn’t mean you have a completely free hand – they must be believable within the context of the story and the world, and readers must be able to relate to them.
A common piece of advice is to keep invented names short and simple. That has some merit, but it doesn’t always work. In a one-off story, the best approach is to use shortish, familiar names for the main character and the people and places s/he knows best, growing more exotic in proportion to their strangeness to him/her. The reader’s going to be a lot more tolerant with the far-off land of Kh’revt-zhorq if the main character’s having problems with the name too.
It’s not always that straightforward, though. I use a world that stretches to seven continents and five thousand years, and I have stories that focus on many different bits of it. The exotic in one story might be the familiar in another, and vice versa. It would clearly be absurd if every name were short and simple. This certainly isn’t true of our own world, where cities, for instance, can be anything from Lima to N’djamena to Antananarivo, and the current wicket-keeper of the Sri Lankan national cricket team is Hewasandatchige Asiri Prasanna Wishvanath Jayawardene.
It’s inevitable that, at some time, the “familiar” culture must have people and places with names that are less easy for the reader. Perhaps the simplest strategy to make this less of a problem is to give characters shortened versions of their names. I’ve done this, for instance, with my recurring characters, Karaghr and Failiu, who call each other Kari and Fai. Although their full names occur frequently in the stories, readers who feel uncomfortable with them can think of the characters by their short names.
Ultimately, though, I try to make the names convincing by giving them linguistic integrity. Now, this doesn’t have to mean inventing the languages themselves – even Tolkien would have problems with the number of languages I’m dealing with – but it does mean having a broad understanding of three factors for each language I use as a source for names.
One is the language’s structure, which will often dictate the form the name will take. It’s not accidental, for example, that so many Roman names end in -us (although it’s a good deal more complex in reality) or that Chinese names tend to be composites of single-syllable elements. Languages can be isolating (words don’t change according to function), agglutinative (words have prefixes or suffixes “glued on” to indicate function) or synthetic (words change their form to indicate function). Very rarely, they can be polysynthetic, but that’s strictly for the fearless.
Secondly, it’s necessary to know what a language sounds like. No-one familiar with European culture could possibly mistake Schultz for an Italian name or Verona for a German one. German words tend to have a lot of consonants clustered together and to end with a consonant (or occasionally an -e) while Italian words tend to alternate consonant and vowel and to finish with a vowel. German has a harsh, guttural sound, Italian a flowing, musical one. That’s not “good” or “bad”, but each language has its own sound.
Thirdly, decide which sounds your language uses. Different languages have very different requirements, ranging from some languages that use barely more than a dozen sounds to others that require alphabets far longer than ours. (Come to that, English really needs a longer alphabet.) Some extra sounds can’t really be represented without making the names look like a linguistic text, but many can.
One much-maligned example is the apostrophe. The use of apostrophes in fantasy names is often dismissed as self-indulgent decoration – and, to be fair, that’s how many authors use it – but that’s unjust. The most common meaning of an apostrophe (other than for grammatical use) is to indicate a glottal stop. This is a sound that doesn’t happen to be used (officially) in any of the languages our alphabet was created for, but it’s as vital as A or B in a great many others. It’s a kind of active pause or gulp, as when a Cockney pronounces bottle as bo’al.
This would be recognised by any fans of Stargate SG1, where the character Teal’c has a glottal stop in his name. From time to time in the show, some character (usually an unsympathetic suit) will mispronounce his name as Tealc, and it sounds completely wrong. The difference is that the glottal stop has been left out.
As an example of my method, I have a character in my novel, At An Uncertain Hour, called Drezhe-va. Although a fairly major character, she’s fairly exotic – her country’s a bit of a mystery to the lands around, and it lies half a world from where the viewpoint character was born, so I didn’t feel the need to make her name “familiar”.
My knowledge of her language stretches to half a dozen names, none of whose meanings I know. This name, however, does reflect several things I do know about the language.
1) It’s an agglutinative language, which uses suffixes to indicate noun-classes. Quite a few RW languages have these – they’re a little like gender in some European languages, but with a lot more variety and rationality. In this case, the suffix -va indicates that the word belongs to the person class (as all personal names would). There’s also a place class, indicated by -ne in the place-names.
2) Words are typically two or three syllables, plus the suffix.
3) There are few consonant clusters. The only one in this name is dr- which is fairly straightforward. The zh is simply the voiced version of sh and is pronounced like the s in treasure.
4) Words (excluding the suffix) tend to finish with a vowel.
The other names – Amlo-va, Ario-ne, MaƤni-ne and Enla-ne – reflect this structure, and hopefully give a genuine sense of cultural unity. This is even more so with cultures and languages I’ve written about more than this one.
Everyone has their own way of creating fantasy names, and many of them work. Perhaps the most important thing is to believe in the names yourself. After all, if you can’t believe in them, how are the readers going to?