Writing Dialog, Part 3

(Original post: http://blogs.officezealot.com/spiller/archive/2005/09/19/8051.aspx)

In the previous two blogs, I wrote about the grammar and punctuation of writing dialog and about how much is enough. In this blog, I’ll examine how to make the characters sound different from each other.


In my earliest attempts, I felt that my characters all sounded like me. In fact, most of what I write sounds like me, and I’ve worked hard to hone that. Now I need to unhone it to get some diversity among the voices in my fiction. So how do I make my characters sound like themselves or the people after whom they’re modeled? (I admit that I have someone specific in mind when I create a character—often it’s a combination of people, so don’t go looking for exact replicas of yourself.)


First, I considered the elements that make someone’s speech distinctive: syntax, dialect, and rhythm.


Syntax is easy to analyze. You can easily discern where the nouns and verbs and things are, and what kind of punctuation is used. In my own writing, I use lots of interruptive punctuation and conjunctions (and, with, or, etc.). I hardly ever write a simple sentence, like the first sentence in this paragraph. I’m big on prepositional phrases and adjectives; I use them to provide a specific type of atmosphere in the individual sentences. Prepositions provide direction and adjectives provide color and shape, so you’ll see lots of both in my work. I’m not big on verb pile-ups, passive voice, or writing everything in the present tense. That’s part of what my sentences look like. Now let’s look at what I sound like.


I choose words partially for how they sound, partially for their surface-level of meaning, and partially for their alternative meanings. That’s right. I’m big on puns (it’s all my mother’s fault), and in my fiction writing, I’m happiest about the result when the words can be interpreted in several ways, or that the meaning broadens after the reader sits with it for a while. I love metaphors and similes, and in revision, that’s the sort of thing that I tend to insert or refine. But that’s also part of what makes my characters sound like me.


I spent a lot of time reading the dictionary and playing Scrabble in my youth. No kidding. Even today, every time I dip into the dictionary to find something specific, I end up finding all sorts of interesting things along the way. (Today, for instance, I found dactyl on the way to something else. Isn’t that a delicious word? Now, go look it up.) I may not remember everything I find, but I am accumulating a really huge vocabulary. That’s another way my characters sound like me. Sometimes they wield unusual words, or they just use them in an unusual way because I know the second or third meanings.


You have to think about who your characters are in order to write plausible dialog for them. Is it a scientist, a slow-witted child, an angry parent, or your own sibling? In order to think reasonably about what kind of language such a person would use, you have to fill in a few stereotypes about them. Are all scientists nerds? Shy? Quiet? Smooth talkers? What do they talk about? Are all their words polysyllabic and are they all Latin derivatives? Do they have trouble talking about non-academic subjects? If your character doesn’t fit all the stereotypes, how are they different? Once you’ve thought about it a bit, you’ll see that it’s a lot like defining your audience. You imagine who they are and then you can write like them.


The next step is to look at dialect. Where are you from? Where is your character from? What do these places have in common? Midwesterners and French people are often nasal—can you put that into your dialog without it sounding hokey? I can’t. I tried, but mostly I felt like the stereotype police were going to come collect me for their wax museum. So dialect warranted another approach for me.


I took a look at writers’ work that includes dialect. Mark Twain is probably the master of the American south when it comes to conveying the way people really sound and still managing to produce intelligible sentences. The closer I looked at it, the more I realized that I couldn’t replicate it. I’ve never spent any time in the south, and anything I could produce would sound hackneyed or stylized and would distract from what the characters were trying to say. I looked at a few others (Flannery O’Connor does this well, Stephen King does it occasionally, and Margaret Atwood does it periodically). Nope. I might be able to do it with considerable practice, but I’m not there yet.


Some of my characters come from other places, and I want them to sound as though they do. If I can’t write cahn’t to illustrate that my character comes from New England without feeling deeply self-conscious about it, how DO I make them sound different from my Californian granola? The answer is rhythm.


Fair warning: if you’re not a fan of poetry, you’ll have trouble with this bit. Words have rhythm. It’s not just whether they’re one syllable or two; it’s also which syllable is accented, whether one syllable is longer than the other, and whether the vowels are diphthongs or not. (Diphthongs are vowels that make more than one sound—a long I is actually pronounced “ah-ee,” for instance.)


You can count the syllables without my help, and you can probably figure out which syllable gets the accent. But whether something is longer in duration than something else, that might take a little examination. Let’s look at a list of rhymes.







Rat hole



The first couple of rhymes are obvious—they have similar spellings. But then they start wandering off into variant spellings, multiple syllables, two words and even a not-so-close rhyme. If you thought to yourself “toll, noble,” you wouldn’t think they rhyme at all. But in context of the rest of the list, you can hear it, right?


The reason these all rhyme is the rhythm. The vowel sounds are the same, in fact, except for noble, the “ol” or “oh” sound is at the end. In the word noble, you can hear that the two sounds have been separated into syllables with an intervening consonant, but the drawn out long O sound is predominant, just like the other examples.


You can hear that the vowel sound is elongated—it takes longer to say those single syllables than it does to say these: that, cup, mote, or tight. The reasons are the vowel sounds themselves and the curbing effect of the consonants. When you’re paying attention to the rhythm of the language, you notice details like that. (Singers do this all the time.)


So if I wanted my speaker to drawl a little, I’d look for those elongated kinds of vowel sounds. Listen:


I don’t think it will rain today.

I think it won’t rain today.

No rain today, missy.

The rain won’t come today.


The last two of those are the shortest to say. Yes, if you count syllables, they’re the shortest as well, but if you were to set a metronome (or otherwise find a steady beat), you’d see that the first two take nearly the same time to say, even though the number of syllables is different. The third one takes less time to say than all the rest. If your character is quick, beetle-like, and prickly, he probably speaks in little jerks too. If your character is pompous and likes to hear himself speak, he’ll choose both the most words and the longest words to say.


Now you have another tool. In addition to knowing what subjects you want your speakers to cover, you can play with the aural quality of the words they use. This is probably the hardest step to master, but you can do the same thing I did in my example. Just think through all the various ways of saying something and choose the one that suits your character best.


You can find these blogs, a little information about my editorial services and me, and a collection of pages about my “real” life on my Web site, www.MelanieSpiller.com.

Posted: September 19th, 2005, 3:03pm CEST by Melanie Spiller

Datos del Post