Are speaker tags always necessary when multiple people are in the conversation?
I understand that if you’ve got two people speaking, you often don’t need speaker tags because it’s assumed that both people take it in turns unless otherwise specified. This keeps the conversation flowing
I often have three or four people in a conversation and adding a tag to every single line breaks up the rhythm and flow. As far as I can, I try to make dialogue specific to the speaker. However, sometimes what they say is too short to give it sufficient personality. I am also concerned that even if I give clues, doing so on consecutive speeches might leave the reader awkwardly trying match up who said what.
For example, four people looking at a moving blob in the distance:
‘Does it know we’re here?’
‘Well we know it’s there so I expect so.’
‘An animal?’
‘What sort of animal moves like that?’
‘A fae animal?’
‘I doubt it’s fae. It would use its invisible state. What it’s doing is most inefficient.’
‘Should we go back? I mean, what if it’s not friendly?’
‘I’m not going back through these accursed demon woods.’
‘Nothing’s friendly here, it would be advisable to get used to it. And the woods are neither cursed nor demonic, thank you.’
I suppose at the moment it’s too ‘screenplay’ - I guess I’m imagining four people peering at this thing, trying to work out what it is and what to do. I’ve tried adding tags and it’s terrible. It’s slightly better if I add action beats, but they slow the whole conversation down, which doesn’t work in context.
Perhaps the answer is that if it doesn’t matter who said what, the conversation could be summarised, but I’m reluctant to do that because I want to give a sense of their collective uncertainty.
If you have multiple people in a conversation, does it matter if it’s not clear exactly who said what? And if it does matter, how can I make the speaker obvious without the conversation grinding to a halt?
2 answers
You are accessing this answer with a direct link, so it's being shown above all other answers regardless of its score. You can return to the normal view.
Essentially, you do need a speech tag if the reader can easily tell who is speaking. Under what conditions can the reader easily tell who is speaking:
There are only two people in the conversation and we know who spoke last, so the new speaker must be the other person. But note that you can't carry this on for too long or the reader will lose track, unless there is some other way for them to tell.
The words could only be spoken by one character. This is by far the most powerful technique. If you have established to goals, manners, and speech patterns of each character so that in this particular conversation, only that character could possibly be the one to say those words, then you don't need a speech tag.
There are other contextual clues to indicate who spoke (such as one character naming another, as has been mentioned). Inserting these artificially to avoid using a speech tag, however, would be a bad idea.
Personally, I think that item 2 above is so important that it is a good test of the quality of your dialogue, and of your character building, that you can omit speech tags most of the time. It ought to be the case that every line of dialogue could only possibly be spoken by one characters. Making sure that that is true for your story is the best way to limit your reliance on speech tags.
0 comment threads
Sometimes it doesn't matter who said what. In your example, where everyone has the same goals and is working together, it seems fine to leave most or all of them out -- the focus here is more on the group discussion than on individual speakers. This approach wouldn't work in cases where the speakers are less aligned, though, like if three people are arguing with each other and one is trying to calm them down.
You can tag speakers without saying "said" or equivalent. You've got a wall of dialogue; interspersing small actions can provide the signals you're looking for without bogging down. Consider this adaptation of your text (my additions are in bold):
‘Does it know we’re here?’ Steve twitched nervously.
‘Well we know it’s there, so I expect so.’ Sheldon rolled his eyes. *Why is Steve always so illogical?*
Kyle squinted. ‘An animal?’
‘What sort of animal moves like that?’ Caleb tightened his grip on his bow.
‘A fae animal?’
‘I doubt it’s fae, Kyle. It would use its invisible state. What it’s doing is most inefficient.’
‘Should we go back? I mean, what if it’s not friendly?’ Steve glanced at the path behind them longingly.
Caleb reached for an arrow. ‘I’m not going back through these accursed demon woods.’
‘Nothing’s friendly here, it would be advisable to get used to it. And the woods are neither cursed nor demonic, thank you.’
Caleb turned toward Sheldon, contemplating his next shot, but returned his attention to the creature.
I assumed this was the start of the dialogue. It's a good idea to establish speakers on their first utterances, somehow. Let's walk through what I did:
I've portrayed Steve as nervous -- first he asks what it is, and later he suggests retreat.
All of the "know-it-all" passages seemed to be coming from the same person; Sheldon's eye-roll establishes that trait and makes it easier to identify his future utterances.
When Kyle asks if it's an animal and Sheldon responds (note I didn't tell you it was Sheldon but you knew, right?), it seems reasonable that the revision of the question ("a fae animal?") would come from the same speaker (Kyle), which is reinforced in Sheldon's response naming him. (Sheldon seems like the sort who talks down to people, and people who do that sometimes inject people's names much like a parent or teacher does.)
I connected Caleb's utterances through his bow; this is safe because if a paragraph combines both dialogue and actions, the actor is expected to be the speaker.
In the last paragraph, Caleb turning toward Sheldon reinforces that Sheldon was the previous speaker, even if it wasn't clear from the dialogue. (If you don't actually want that strife in your group, you could safely leave that part out.)
As I demonstrated here, you can use a combination of associated actions, direct address, and implied speakers (not identifying people if the utterance itself makes it clear) to convey who is speaking without paragraph after paragraph of "so-and-so said".
0 comment threads