In this blog, I’ve had to focus on books written in English, because blogs are short, and English is the language I know best. I intend no slight to speakers and readers of other languages by doing so.
I often hear Anglophone readers of fantasy and science fiction complain about the “pronunciation guides” that lurk at the beginning of most such books. Although no one needs to be quite as compulsive as Tolkien, good world building includes inventing at least parts of some new languages, enough so the author can create consistent sets of personal and place names. One of the most important parts is the phonemic system, that is, what we usually call its “sounds.” And that’s the problem for so many readers. “Why,” they cry, “why can’t you have names and things that are spelled just like English?”
Well, first off we have to ask, “Which English?” What we call the English language is actually a collection of dialects. American, Australian, British, and India English are each spoken by millions of people each, and they all sound different. We mustn’t forget the minor dialects as well, such as the Anglo-Celtic group and the New Zealand with its back vowels. Within each of the major dialects there are differences. The British English of middle class London sounds very different from the English of working class Glasgow, for one example. For another, consider this bit of doggerel by a writer named Coogler. In his American Deep South dialect, the lines rhyme.
“Alas for the South, her books have grown fewer/ She was never much given to literature.”
All of these forms of English are written in the same way in the same alphabet, which is why we can all read the same books and emails and understand each other. (If we ever get a pure version of spelling reform in our language, World English is going to be in a hell of a mess.) We all know that English spelling is, to be polite, eccentric. It all made perfect sense at the end of the 14th century, but alas, languages always change over time. The problems with the spelling system lie at the heart of our pronunciation guide difficulties. Let’s consider the reasons.
The human voice can create an enormous range of sounds. A language that included all of them would take several lifetimes to learn, so all human languages select a small number of them to act as phonemes, that is, sounds that are distinguished from one another by carrying meaning and that aren’t bound to a specific range. The voice of the speaker can be low or high, the tone can be angry or flat — speakers of the language will recognize that “sound” as a phoneme. Someone shouting ‘dog!’ in a high voice will be as easy to understand as someone whispering the word in a dark voice. Sometimes a pair of sounds will form a single phoneme. Consider the sounds in loop and pool — we’d say that the P’s and L’s are the same, but if you listen carefully, they differ. It doesn’t matter. We recognize them as carrying the same meanings. That is, “pool” is not “poop”, no matter how the last letter is sounded.
The number of phonemes varies from language to language. English, for example, has between 44 and 47, depending on dialect. Some Caucasian languages have 87. In an ideal language, one letter, or grapheme, would represent one phoneme. The English alphabet has 26 letters. English has over 40 phonemes. Oops. Which letter represents which phoneme can vary by dialect, particularly when it comes to the vowels. Double oops. Conveying our new pronunciations is not going to be easy.
The standard spelling of English contains a number of work-arounds that were mostly brought into being by the scribes of the English royal chancery around 1400, before a phenomenon known as the Great Vowel Shift. When the spelling began to be codified, to give one example, the letter we call “e” was pronounced more like “ay,” as in “hay” or “day”. A word like “read” or “lead” would have been pronounced to rhyme with our modern-day “raid”. A word like “red” or “led” would have been pronounced much as it is in modern American or British, however. To show the difference, the scribes designated vowels like the “e” in “lead” as long vowels. To show these letters took more time to pronounce, they added that “a” in. Perfectly clear or perhaps clayer. Except things changed drastically a few years later.
So what happens when we try to represent the words in a made-up language with English spelling? We’re forced to use analogies with other words that may well be pronounced in several different ways by different readers of our work, assuming of course they have some understanding of spoken English and didn’t just learn a specialized reading version in graduate school. To call this system imperfect doesn’t begin to describe the problems. It does explain, though, why no, we can’t make our imaginary language read “just like English.”