'Naturalistic' Dialogue

We writers often try to write dialogue in what we regard as a naturalistic style, but this can be fraught with difficulties. The result may not be what we intended.

We know that the spoken word is very different from the written word. We rarely speak in nice, complete, neat, grammatically correct sentences. Our speech stops and starts; words slide together; we leave the beginnings and endings off words; we pronounce words in particular ways; we change direction in mid sentence. It is the ‘words slide together; we leave the beginnings and endings off words; we pronounce words in particular ways’ that causes the difficulties.

The problem with depicting a character as speaking in a particular accent is that we all have accents, although we seldom realise it of ourselves. If we choose to represent the speech of (say) our Scottish or Irish characters in a particular way, why don’t we do so for all the other characters? The result is usually somewhat clichéd, and some of the characters verge on becoming caricatures. If we wish to represent a character as being less well educated—for example by dropping Hs or leaving the ends off words—we are being unfair to them, because we probably also do that ourselves!

Consider this sentence: ‘Are you going to eat that or not?’ Now say it out loud. How many of you said ‘you’ and ‘going to’? I imagine most of us (brilliantly well educated or not) said something like: ‘Are y’ gonna eat that or not?’ I suggest that when we write a sentence of dialogue, we can assume that the reader will ‘hear’ the latter (or something like it) without our having to spell it out. When it is spelled out it can become difficult to read (and downright annoying). It is also incredibly difficult to maintain such patterns consistently. If I write: ‘Whad’re y’ havin’ f’ dinna’, am I going to do this every time the character speaks? When I as an editor come across something like this, should I be going through and removing the Gs off the end of all this character’s words, because I can guarantee there will be plenty of them. It is as much a nightmare trying to edit this into some kind of consistency as it is to write it consistently in the first place.

Then there is the question of how you actually represent these sounds. Short of using formal phonetic notation, it is very difficult to properly represent pronunciation. Consider this sentence which I came across recently: ‘I do luv a good soup.’ Why ‘luv’ rather than ‘love’? How am I meant to pronounce it? Is the vowel sound the same as in ‘put’ or ‘cup’? These are different sounds in my particular accent. The vowel sound in ‘cup’ (for me) is identical to that in ‘love’. On the other hand, ‘put’ is a different sound from ‘putt’ (which, for me, is like ‘love’ and ‘cup’). So ‘luv’ is unhelpful. How I hear it will depend on where I come from, which is the same for ‘love’ in any case, so why bother spelling it differently? My mother, for instance, in her Birmingham accent, will say ‘love’ as if the vowel is the same as in ‘put’ rather than ‘putt’, but as far as she is concerned she is saying ‘love’ as it is meant to be said, and who am I to tell her otherwise?

Note that an accent is different from dialect. A dialect uses different words, rather than just pronouncing the same word differently. If you can use authentic dialect (and not just what outsiders often wrongly consider to be authentic dialect) that is great.

I always think that a better way to alert the reader to a particular speech pattern is to point it out in the narrative: ‘I struggled to understand his strong Scouse accent’ (for example). ‘Fred always misplaced his Hs when he was speaking.’ Then you don’t have to write it all the time. The reader will hear it. If the reader doesn’t know what a Scouse accent (Liverpool) sounds like, nothing you write on the page will help!

I would never say never write out these ‘naturalistic’ passages of dialogue. On occasion it might be helpful if you want to create a particular effect. But if you do, do so sparingly and with great care. Ask yourself: Will this sound to the reader the way I want it to sound? To some it will; to others it won’t.