FOR EDITORS, PROOFREADERS AND WRITERS

Writers can place dialogue tags before, between and after speech – there’s no right or wrong way to do it. Tag-first speech does have a different feel to it though, particularly when the construction is used frequently. This post explores the impact on your novel.

Read on to find out more about the following:
​

• What a dialogue tag is
• Back-loaded tags
• Mid-loaded tags
• Front-loaded tags
• Impact level, psychic distancing and lyricism in front-loaded tags

A dialogue tag is the short piece of text that tells a reader that a character is speaking, and which character is speaking. For example:
​

• Are you enjoying reading that blog post?’ Louise said.
• ‘Are you enjoying reading that blog post?’ she asked.

In the above examples, the tags are shown in bold and comprise the subject (someone’s name or their pronoun) doing the speaking, and the verb from which the reader can infer that the action of speech is taking place.

Commonly used effective verbs include ‘said’, ‘asked’, ‘replied’, ‘whispered’, ‘muttered’, ‘yelled’, ‘continued’ and ‘added’.
​
Ineffective dialogue tags use verbs that bring to mind action that’s not related precisely to speech but to some other behaviour. Examples include ‘sneered’, ‘grimaced’, ‘laughed’, ‘harrumphed’, ‘huffed’, ‘sighed’, ‘snarled’ and ‘urged’.

There is no right or wrong position for a dialogue tag. Authors can mix them up as they choose. Tags can even be omitted when it’s clear who’s speaking.
​
So where might they go?

Dialogue tags can be front-loaded, mid-loaded and back-loaded.

Back-loaded dialogue tags
Back-loaded tags come after the speech and are used commonly. Consider using them in the following circumstances:

• ​Length of dialogue: Your character’s dialogue is a short burst and you want to ensure the reader’s attention is focused on what’s being said straightaway, rather than who’s saying it.
• Impact level: The dialogue is relevant but low key. There’s no punchline that might be flattened by a tag.

EXAMPLES

Mid-loaded dialogue tags
Mid-loaded tags come between the speech. They, too, are a popular choice for writers. Consider this option in the following circumstances:
​

• Length: The dialogue stream is longer and you want to ensure your readers don’t wait too long to discover who’s speaking.
• Rhythm and context: You want to introduce a natural pause so that the speech doesn’t turn into a monologue. You can also supplement the tag with descriptive action that grounds the dialogue in the environment and helps the reader picture the scene.
• Impact level: You’re written a witty, suspenseful or impactful punchline into the dialogue and don’t want it interrupted or flattened by a tag.
• Interrupted speech: You’ve written speech that’s interrupted abruptly and don’t want the tag interfering with the interrupter’s speech.

EXAMPLE 1
This first example is from David Rosenfeld's Collared.

Notice how in the above example of single-character dialogue, which comprises a total of 51 words, the impact point is with the closing sentence: ‘Sign the damn form and send it in’. Because the tag’s located earlier, that dialogue gets to shine.

​Compare the original with this version, which I’ve given a back-loaded tag:

​Back-loading the tag strips ‘Sign the damn form and send it in’ of its oomph.

EXAMPLE 2
Here’s an example from Linwood Barclay’s Parting Shot that shows how mid-loaded tags can protect the flow of interrupted dialogue.

Notice how by mid-loading the dialogue tag, Ms. Plimpton’s interruption – indicated by the closed-up em dash at the end of Duckworth’s speech – feels more authentic because it’s given the space to flow.
​
Front-loaded dialogue tags
Front-loaded tags come before the dialogue. This position is the one least used in most commercial fiction, and there’s a good reason for that: reader focus.

Those familiar with advice on writing for the web will know that web copy needs to be front-loaded with relevant keywords. This means that the important stuff comes first. That’s because visitors to websites are busy and scanning for solutions to their problems. When they don’t get them fast, they become frustrated and are more likely to jump to another site.

If the novelist’s done their job well, readers will invest way more time in soaking up their prose than if they were shopping for a new duvet cover. Even so, every word in a piece of fiction needs to count, and readers should still be focusing on the most important stuff.

And so if you’ve written great dialogue, most of the time you’ll want to ensure your readers are focusing on it as soon as possible. Front-loading the speech, rather than the tag, helps achieve that.

That’s not to say that front-loaded dialogue tags don’t have their place. They do, and they can be extremely effective when used purposefully.

Let’s have a look at how a front-loaded dialogue tag can be used to superb effect when used purposefully.

Impact level
A front-loaded dialogue tag can function in the same way as a mid-loaded one when it comes to speech containing impact points. Again, we’re talking about dialogue that’s witty, suspenseful, or closes with an impactful line that you don’t want to flatten with a tag.

EXAMPLE
Let’s return to the excerpt from Collared. Although Rosenfelt uses a mid-loaded tag, he could have opted for a front-loaded one and preserved the oomph in his closing sentence. Here’s how it might look:

Psychic distancing
If you want the prose to feel more expository so that the reader is less closely connected to the character, front-loading the tag might be just the ticket. Doing this widens the psychic (or narrative) distance.

A tag tells of speaking whereas dialogue shows what’s being said. By placing the tag first, you draw the reader away from the character’s voice and give the prose a more objective feel.

EXAMPLE
This excerpt is from Jens Lapidus’s Life Deluxe.

Lapidus has front-loaded dialogue tags and thoughts in this excerpt, and it’s an excellent example of psychic distancing in action. The centring of the characters rather than the speech gives the prose a detached, clinical feel that shows rather than tells mood.

Jorge is a drug-dealer operating in Stockholm’s shady underworld. He’s only just out of jail but already he’s frustrated with a life of honesty. In fact, he’s got only one thing on his mind: easy money.

The wider psychic distance means we get to see the world through Jorge’s eyes but without getting too close to him. Perhaps Lapidus doesn’t want us to empathize with him too much. Instead, he widens the psychic distance just enough that we can make up our own minds about whether Jorge deserves the trouble coming his way.

Lyricism
Repeated use of front-loaded tags with short bursts of dialogue can introduce a lyricism into prose whereby the tags function as more than just indications of who’s speaking. They become part of the poetry.
​
This approach can work particularly well with parody, satire and comedic prose.

EXAMPLE

​Notice how the multiple front-loaded dialogue tags are performing anaphorically. Anaphora is the purposeful repetition of words or phrases at the beginning of successive clauses.

It’s often used in poetry and speeches. When it’s used in novels, that repetition draws the reader's eye and can show rather than tell mood – boredom, monotony or, as in this case, disinterest. The tags are therefore key to the lyricism, and as important as the speech. 

Front-loading dialogue tags is something most authors tend to avoid. However, as is usually the case when it comes to line craft in fiction, there are no rules.

​The key is to consider what purpose your tag is serving and how it can best amplify the speech, evoke mood, and improve rhythm.

• ​Barclay, l., Parting Shot, Orion; 2017 (p. 380)
• Boyd, W., Solo: A James Bond Novel, Vintage, 2014 (p. 260)
• Crosby, SA, Blacktop Wasteland, Headline, 2021 (Chapter 1, Kindle edition)
• Lapidus, J, Life Deluxe, Pan, 2015 (Chapter 1, Kindle edition)
• Rosenfelt, D, Collared, Minotaur Books, 2017 (Kindle edition)​

• Editing Fiction at Sentence Level (book)
• Fiction editing and writing resources (online library)
• How to Line Edit for Suspense (multimedia course)
• How to Write the Perfect Fiction Editorial Report (multimedia course)
• Narrative Distance: A Toolbox for Writers and Editors (multimedia course)
• Preparing Your Book for Submission (multimedia course)
• Switching to Fiction (multimedia course)
• What is anaphora and how can you use it in fiction writing? (blog post)

Does ‘said’ come before or after the subject in a dialogue tag? Does it matter? This post explores the options and offers practical recommendations that serve writers seeking to build their author platform.

Read on to find out more about the following:
​

• What a dialogue tag is
• Why tags should support, not supplant dialogue
• Tagging: Positioning the verb in relation to pronouns
• Tagging: Positioning the verb in relation to nouns
• Standing out in ways that serve reader expectations

A dialogue tag, or speech tag, is the short piece of text that tells a reader that a character is speaking, and which character is speaking. For example:

• ‘Are you enjoying reading that blog post,’ Louise said.
• ‘I’m interested in learning about dialogue,’ ze replied.
• ‘Can I stick to non-fiction editing?’ he asked.

Effective dialogue tags use verbs from which the reader can infer that the action of speech is taking place. Examples include ‘said’, ‘asked’, ‘replied’, ‘whispered’, ‘muttered’, ‘yelled’, ‘continued’ and ‘added’.

Ineffective dialogue tags use verbs that bring to mind action that’s not related precisely to speech but to some other behaviour. Examples include ‘sneered’, ‘grimaced’, ‘laughed’, ‘harrumphed’, ‘huffed’, ‘sighed’, ‘snarled’ and ‘urged’.

In this post, I’m going to focus on a question that beginner fiction authors and editors often ask about: where to locate ‘said’ and other effective tagging verbs.
​
First, though, here’s a quick recap about why ‘said’ is such a popular dialogue tag.

‘Said’ is the most popular speech tag in English-language writing because it’s virtually invisible. Readers are so used to seeing it that they ignore it and instead focus on its mechanical function – to verify who’s doing the speaking.

It’s become so conventional that when authors go out of their way to replace every instance of ‘said’ with alternatives, they risk creating prose that feels laboured.

Showy speech tags in particular stand out, and that means they pull the reader’s attention away from the dialogue and push it towards the speech tag.

That’s not a good reader experience because when characters communicate through speech, that’s where the action is in that moment. The tag should support that action, not supplant it.

You can find out more about how to tag dialogue in Editing Fiction at Sentence Level, including:

• Showy speech tags and underdeveloped dialogue
• Showy speech tags and double-telling
• Non-speech-based dialogue tags and the reality flop
• Alternatives to showy speech tags – more on action beats
• Using proper nouns and pronouns in dialogue tags
• Omitting dialogue tags

Now let’s look at where to place ‘said’ and other tagging verbs.

Regardless of which pronouns are in play, it’s conventional in contemporary commercial fiction to place ‘said’ or any other tagging verb after the pronoun. Take a look at what’s on your bookshelf. Examples are easy to find. Here are some from my collection.

​Let’s have a look at those examples when the verb is placed before the pronouns.

​There’s actually nothing grammatically wrong with any of the edited examples, though I wasn't able to retain the past perfect in the second. What has gone awry is the narrative setting.

The dialogue is written in contemporary English but the verb placement evokes a distinctly historical feel that’s inappropriate.

So how about when nouns rather than pronouns are in play?

Authors published by mainstream presses tend to follow the same convention: the verb follows the noun. Once more, a visit to your bookshelf will confirm the frequency in contemporary commercial fiction. Here are some more examples from my collection.

Now let’s move the verb so that it sits before the noun.

Again, there’s nothing grammatically problematic about this structure. Furthermore, the switch doesn’t jar in the same way as when the subjects are pronouns.
​
So how should independent authors (and the editors who support them) approach the structure of a tag?

I recommend my authors follow the noun-first convention, not because it’s right, or best, or grammatically correct, or the only one true way, but because when it comes to contemporary commercial fiction:

• lots of potential buyers of their books have come to expect this construction
• and it's the approach that mainstream presses tend to take.

When readers are presented with stories that break with convention, there’s a risk that the story’s no longer the standout feature. Instead, those with strong preferences focus on minutiae that challenge their expectations.

That focus isn’t serving a writer who’s trying to build their author platform.

Mick Herron has built enough trust with his readership with his brilliant Jackson Lamb series that no one will bat an eyelid when they read the following in Slough House:
​
‘Ian Fleming,’ said Diana Taverner. ‘Means “Death to spies”.’
​
Nevertheless, I scoured the first chapter of that book and found that in the main he still favours placing the verb after the subject. 

Placing ‘said’ and other speech-related verbs after the subject isn’t a ‘rule’ that must be applied in order to ensure a dialogue tag is grammatically correct.

However, it’s such a common convention that I recommend indie authors of contemporary commercial fiction follow it.

When the subjects in a tag are pronouns, this structure is essential for retaining a contemporary feel. When the subjects are nouns, placing them first gives the pedants one less thing to gripe about.

• Coben, H., Fool Me Once, Arrow, 2016 (Chapter 1, Kindle edition)
• Connelly, M, The Dark Hours, Orion, 2021 (Chapter 1, Kindle edition)
• Crosby, SA, Blacktop Wasteland, Headline, 2021 (Chapter 1, Kindle edition)
• Crouch, B, Dark Matter, Pan, 2017 (Chapter 1, Kindle edition)
• Harnby, L, Editing Fiction at Sentence Level, Panx Press, 2020
• Herron, M, Slough House, John Murray, 2021 (Chapter 1, Kindle edition)
• McDermid, V, 1979, Little, Brown, 2021 (Chapter 1, Kindle edition)

Fiction editing training courses

• How to Write the Perfect Fiction Editorial Report
• Narrative Distance: A Toolbox for Writers and Editors
• Narrative Style: Viewpoint, Tense and Voice
• Preparing Your Book for Submission
• Switching to Fiction

Your novel’s written in first person. Here are some tips for how to ensure your narrative doesn’t become overloaded with ‘I’ but remains immersive.

• Why reduce the ‘I’
• Why ‘I’ still has a place front and centre
• Focusing on the exterior rather than the interior
• Reducing the use of filter words
• Removing speech and thought tags
• Applying the principles of free indirect speech
• Taking the 'I' out of introspection
• Balancing ‘I’ with ‘we’

We might think that the mention of ‘I’ would always make prose more immediate, and draw the reader in closer to the viewpoint character. But sometimes the opposite is true.

Too much ‘I’ is a tap on the shoulder, one that says to the reader, ‘Just in case you’ve forgotten who the narrator is, here are lots of reminders.’
​
The consequence is that readers are pulled away. And that can actually increase rather than reduce narrative distance.

I confess to being a huge fan of first-person narrations. When done well, the pronoun is almost invisible, even if it’s used frequently. Certainly the books I’ve borrowed excerpts from here allow ‘I’ to take centre stage.

However, they don’t rely on a first-person pronoun to convey experience, thought, speech and action. Below, I'll show you some examples – ones that ensure the intimacy of the narration style is left intact.

And so while we don’t want to obliterate ‘I’, because avoiding it completely would render the prose awkward, inauthentic and overworked, too much ‘I’ can be repetitive and interruptive. What’s required is a balance.
​
This post aims to offer you choice – fitting alternatives that retain intimacy and immediacy when you’re concerned you’ve overdone it.

With a first-person narration, what’s reported must be through the lens of the narrator. Since their presence is a given, we don’t always need to be reminded that ‘I’ is involved.

A little peppering in a more objective report will suffice because the reader knows that it’s coming from the narrator, and only the narrator. It has to be.

And while writers can make space to explore the viewpoint character’s emotional behaviour, the exterior world is what grounds their experience in the novel’s physical world. It gives the novel substance, and the reader something to bite into.
​
Instead of focusing on who’s doing the reporting, shift the prose towards what’s being reported.

What and who else is in the scene? Why are they there? How do they behave? What do they look like? This information can be reported without ‘I’ so that the reader experiences the physical world within which the narrator is operating.
​
Here’s an example from To Kill a Mockingbird (Harper Lee, Pan, 1974, p. 11).

Notice the (almost) absence of ‘I’. Scout – our narrator – tells us about the town she lived in: Maycomb. The recollection is hers certainly –  ‘when I first knew it’ anchors it as such. It’s therefore intimate.

And yet because there’s only one I-nudge, we’re allowed enough emotional distance to step back and pan, like a roving camera, across Maycomb’s vista. We’re dislocated from Scout’s doing the experiencing and encouraged instead to focus on what she’s experiencing.
​
What’s happening here is a shift from the subjective to the objective.

​Here’s an example of a short excerpt that’s subjective. The focus is on the I-narrator. ​

​And here’s the real excerpt from David Rosenfelt’s Play Dead (Grand Central, 2009, p. 19). Now the focus is objective, yet in no way does this distance us from the centrality of the first-person narrator’s experience. We’re still deep in his head.

Filter words are a clue that an interior rather than exterior focus is in play. They’re verbs that increase the narrative distance, reminding us that what we’re reading is being told by someone rather than experienced, or shown, through the eyes of the character.

Examples include noticed, seemed, spotted, saw, realized, felt, thought, wondered, believed, knew, and decided.

Filter focus the reader’s gaze inwards (interior focus) on the manner through which the viewpoint character experiences the world – the how.

They come with a pronoun: I saw, they believed, we decided, she knew, he noticed.

By removing filter words, the reader’s gaze is shifted outwards (exterior focus) and onto what is being experienced. That can make for a more immersive read. Plus, the omission means we say goodbye to their accompanying pronoun: 'I'.

Here are a few examples to give you a flavour of how you might recast in a way that avoids first-person filtering.

Dialogue tags are what writers use to indicate which character is speaking. Their function is, for the most part, mechanical. If the reader can keep track of who’s saying what in a conversation, you can omit dialogue tags.

This will work best if there are no more than two characters. Most writers don’t extend the omission for more than a few back-and-forths before they introduce a reminder tag or an action beat.

Watching out for unnecessary tags is good practice regardless of narration style, but with a first-person narration it’s a particularly efficient way to declutter ‘I’-heavy prose.

​Take a look at this excerpt from David Rosenfelt’s Play Dead, pp. 194–5. There are two characters in this scene: Andy Carpenter, the protagonist and narrator, and Sam Willis, the non-POV character on the other end of the phone.

The exchange involves 19 speech elements within the thread, but only 3 speech tags, and only one of those marks our first-person narrator.
​
At no point do we lose track, and at no point are we distracted by repetitive ‘I said’s.

If you’ve played with free indirect speech (also called free indirect style/discourse) in third-person narratives, call on your craft for first-person narration.

In a nutshell, free indirect speech offers the essence of first-person dialogue or thought but through a third-person viewpoint. The character’s voice takes the lead, but without the clutter of speech marks, speech tags, italic, or other devices to indicate who’s thinking or saying what.
​
Here’s an example of third-person narration. Notice the filter words ‘glanced’ and ‘noticed’, the italic present-tense thought, and the thought tag:

Let’s change that to a first-person narration. The filter words are still there and there’s a thought tag with the ‘I’ pronoun.

Here’s what the third-person version could look like in free indirect style. The filter words and tags are gone. It feels like a first-person thought but the base tense and third-person narration remain intact.

And now the first-person version. All I’ve done is swapped out the pronoun ‘his’ for ‘my’. 

There’s nothing wrong with contemplation and introspection. Authentic characters ruminate just like real people.

However, when prose is littered rather than peppered with constructions such as I wasn’t sure if, I didn’t know whether, I wondered if, it can feel muddled and be laborious to read. The reader might respond: Well, of course you’re wondering. Who else could it be? You’re the narrator.

Worse, readers might think the narrator’s rather self-absorbed and unsure of themselves. While that might be necessary now and then, it’s problematic if it’s a staple because a narrator who’s always focused on themselves, and who never instils confidence in us, can’t tell the story as effectively.
​
Look out for ‘I’-centred introspection and experiment with statements and questions that allow the ‘I’ to be assumed.

Here are a few examples to show you how it might work.

Another option is to consider whether your narrator’s lived experience at particular points within the novel involves others.
​
This is an opportunity to frame the narrative around ‘we’ rather than just ‘I’.

Here’s an excerpt from To Kill a Mockingbird (p. 162) in which Scout, Harper Lee’s first-person narrator, frames the recollection around not just her own experience but those of the people she was hanging out with.

The effect is powerful because we’re shown rather than told a sense of her belonging, of her being in a group, of the togetherness of that experience. And that intensifies our immersion in her world.

There’s nothing wrong with ‘I’, but a first-person narrator can tell a story without relying on their pronoun all the time. Since they’re the ones doing the reporting, the ‘I’ can often be assumed.
​
Try recasting sentences that start with ‘I’ more objectively, so that the focus is on the what – the emotion, the object, the person, the action and so on – rather than the sense being used to experience it or the I-narrator doing the experience.

Use the principles of free indirect speech to reduce your ‘I’ count. It’s a tool that encourages a narrowing of narrative distance to such a degree that the reader feels deeply connected to the viewpoint character – more like we’re reading a thought than straight narrative.

As for speech and thought tags, you might not need as many as you think. The speaker can usually be identified without them if there are only two people in the conversation. Removing redundant tags is worth considering whichever narration style you’re writing in.

• ​3 reasons to use free indirect speech
• ​6 ways to improve your novel right now
• Author resource library (includes links to free webinars)
• ​Editing Fiction at Sentence Level: A Guide for Beginner and Developing Writers
• ​Filter words in fiction: Purposeful inclusion and dramatic restriction
• Making Sense of Point of View: Transform Your Fiction 1
• What is head-hopping, and is it spoiling your fiction writing?
• ​Switching to Fiction: Course for new fiction editors

Novel dialogue is not like reality, where much of what we say is of little consequence to the bigger picture of our lives. Here’s how to check that all your dialogue needs to be there. Then remove the mundane!

Artful dialogue requires balancing realism with engagement and ensuring that every word spoken by a character pushes the novel forward rather than making the reader feel like they’re eavesdropping on a mundane conversation at the bus stop

Every line of dialogue should have a purpose. If it doesn’t, it probably shouldn’t be in your book.

It can take courage to remove words you've spent an age crafting but as I hope to show in the example below, readers don't need every little detail. Less is so often more!

My favourite way of assessing whether dialogue is working is to think in terms of voice, mood and intention. 

When we focus on those three things, we avoid dull dialogue – conversations about the weather, how someone takes their tea or coffee, and courtesy statements such as ‘Hi, how are you?’

Let’s look at an example of dialogue that represents the kind of conversation one would expect to hear in real life. It includes the polite chitchat that people indulge in before they get down to business.

Were you enthralled by the welcome and refreshments section, or did you just wish we could get to the point? I think I know the answer!

Now let’s look at how author David Rosenfelt actually wrote this excerpt from Play Dead (Grand Central, 2009, p. 175), and beautifully too:

Rosenfelt knows that none of his readers care about the weather, the coffee, or whether people say hello to each other or not.

And so he leaves all of that out and lets the reader imagine that this stuff took place. And it’s enough. In the published novel, as opposed to the version I butchered, the first line of speech is “And you believed him.”

With that, we’re straight into Kevin’s incredulity and concern, and his desire to understand what the team is dealing with in regard to Petrone.

Meanwhile, Andy has his lawyer hat on. His initial reply is succinct, so that we are left in no doubt about his belief that Petrone was telling the truth, and that he is determined to reassure Kevin.

This is no-messing dialogue that focuses on story, not whether the speech is what we might actually hear – in its entirety – in real life.

It’s an excellent example of an author ensuring that every word counts and that there’s no bus-stop-talk filler.

To declutter dialogue and make every word count, ask yourself the following:

• Is every line relevant to the story?
• Is the character speaking with purpose or taking up ink/pixels on the page?
• Can mundane chitchat be removed without damaging sense and flow?
• Could the dull stuff be replaced with speech that deepens character?

Want the booklet version of this post? It's available on my Dialogue resources library. Click on the cover below to hop over to the page. Once you're there, choose the Booklets icon.

• Editing Fiction at Sentence Level
  
• Making Sense of ‘Show, Don’t Tell’
  
• Making Sense of Point of View
  
• Making Sense of Punctuation
  
• How to Write the Perfect Fiction Editorial Report
  
• Switching to Fiction
  
• The Differences Between Developmental Editing, Line Editing, Copyediting and Proofreading (free webinar)
  
• How to Punctuate Dialogue in a Novel (free webinar)
  
• Resource library: Dialogue and thoughts

Give your novel a sentence-level workout. Here are 6 common problems, and the solutions that will improve the flow of your fiction and make the prose pop.

​Review your novel for 6 common problems. None involve major rewriting, just relatively gentle recasts that will improve your prose significantly, and make your reader's experience more immersive.

1. Assess invasive adverbs
2. Remove redundant filter words
3. Take the spotlight off speech tags
4. Pick up dropped viewpoint
5. Trim anatomy-based action
6. Turn intention into action

​Not all adverbs and adverbial phrases are bad. Suddenly, slightly, slowly, nervously, calmly, quietly can be effective if used now and then.
​
However, overuse is often a symptom of an author telling us what’s already been shown, which means the adverbs are repetitive and cluttering. In the two examples that follow, they can be ditched because 'fidgeted' shows the nervousness, and the apology in the dialogue shows the regret.

​Even when adverbs are telling us something new, consider elegant recasts that use stronger verbs but still keep readers in the moment.

When readers are told of doing being done by a viewpoint character, filtering is in play. Realized, knew that, wondered, thought, saw, and decided are just a few examples.

The reader is already experiencing the story through a viewpoint character. For that reason, we often don’t need to be told that they realize, see, think or feel anything. We’re already in their heads.

​It’s telling what’s already been shown.

​Filtering pulls us out of the deep, limited viewpoint. Worse, it’s repetitive and obvious. Jane has already looked at the screen so we know her eyes are doing the work; telling us that she saw as a result of her glancing is redundant. 

In the example where Matthew’s the narrative viewpoint character, we needn’t be told he feels the thumping in his temples, since if he weren’t feeling it he couldn’t report it. Nor do we need to know he’s thinking about that third glass because we’re already in his head.

'Said' is almost invisible when it comes to dialogue tags. A smattering of 'asked', 'replied', 'whispered', and 'yelled' can also work well.

Sometimes tags aren’t even necessary because it’s obvious who’s speaking. Other times, we can replace a tag with an action beat than conveys movement and emotion.

Readers should be focused on the dialogue. If a showy tag is necessary to convey a character’s voice or mood, the speech might need a rethink.

In the examples below, the speech tags do the following:

1. tell what the punctuation’s already shown
2. tell what the dialogue’s already shown
3. tell us what the dialogue could have shown but doesn’t
4. express non-speech-related behaviour

Narrative viewpoint is a big topic so I’ve focused on two common sentence-level slips:

• The viewpoint character reports what they can’t know.
• The reader is given access to non-viewpoint characters' internal experiences.

The viewpoint character reports what they can’t know 
Reporting what can’t be known often comes with filter phrases such as 'could tell (that)' and 'knew (that)'.

​In this example, John is the viewpoint character. We experience the story though his senses.  

Actually he can’t tell any such thing. It might seem that way, but for all John knows, she could be hiding it because she has another agenda. Telling us that’s not the case removes any underlying suspense – stops us asking the question.

This might seem like a small slip but it’s the kind of thing that turns over all the power of a limited/deep viewpoint to an all-knowing narrator and rips apart the tight psychic distance between reader and the viewpoint character.

Here are two recasts that avoid the viewpoint drop:

Non-viewpoint characters’ internal experiences
Here we’re talking about head-hopping. It’s when readers are able to access emotions, mood and thoughts of a non-viewpoint character. In the example that follows, Reja is the viewpoint character.

The solution is to recast the text so that these emotions, mood and thoughts can be inferred or accessed externally – for example, through movement or speech – by the viewpoint character only. Here’s a possible recast.

Prose is more immersive when readers aren’t told what they can assume is being done by body parts that are associated with particular actions – holding with hands, gazing with eyes, standing to their feet, kneeling on their knees, nodding heads, and shrugging shoulders.

In the example below, we might remove the obvious body parts and focus more specifically on the part of John's legs doing the kicking and the impact of his action. As for the gun-toter, the hands have been ditched.

Sometimes the reader needs to know what a character wants to achieve from a particular action. This is about the why of an action.

However, when an author means to show the how of an action but tells of intention to act, there’s a problem. The red flag to watch out for is 'to'.

We can check whether the focus is on point by asking a question: What action do we want to show the reader (via Jane)?

​​If we want to show the reader that Jane can lift her wrist – because that’s what the first example below is showing us – we can leave as is.

​However, that's rather dull; it's more likely that we want to show that Jane is checking the time, and so a leaner alternative is more effective.

None of these 6 tweaks are rules! Think of them instead as suggestions to consider, ideas that can help you smooth and tighten up your prose.

And don't worry about them at first-draft stage. Use that space to get the words on the page. Put your sentence-level editing craft in play with a later draft, and once your story's structure, plot and characterization have been fully developed.

Want to develop your line-editing skills? Check out these resources:

• Author resource library (includes links to free webinars)
• Editing Fiction at Sentence Level: A Guide for Beginner and Developing Writers
• Making Sense of Point of View: Transform Your Fiction 1
• Making Sense of Punctuation: Transform Your Fiction 2
• ‘Playing with sentence length in crime fiction. Is it time to trim the fat?’
• ‘Playing with the rhythm of fiction: commas and conjunctions’
• ‘What is anaphora and how can you use it in fiction writing?’
• ​'2 ways to write about physical pain'
• 'What is head-hopping, and is it spoiling your fiction writing?'
  

​Not sure where to place your speech tags? This guide shows you how to tell readers who’s speaking, not based on a set of rules but in respect of clarity, suspense, invisibility, and rhythm.

​What is a dialogue tag?
Dialogue tags, or speech tags, are complementary short phrases that tell the reader who’s talking. They’re not always necessary, particularly if there are only two speakers in a scene, but when they are used, this is what they look like:

• ‘Dump that corpse and don’t ever mention it again,’ the hooded guy whispered.
• “Thanks for holding the gun,” Tom said. “Now pull the trigger.”
• Marg turned. Smirked. Said, ‘Tomorrow. If you’re late, he’s late. Geddit?’

Said is often best because readers are so used to seeing it that it’s pretty much invisible and therefore less interruptive.

What’s the rule about where tags go?
Dialogue tags can be placed after, between or before dialogue. Authors sometimes ask which position is best or whether there’s a rule.

There is no rule. All three positions have advantages and disadvantages, depending on what you want to achieve.

Position: After dialogue
Readers are so used to seeing speech tags like said at the end of dialogue that they’re almost invisible. That allows the dialogue, rather than the speaking of the dialogue, to be the focus.

Below is a wee example from Recursion (p. 292). The speech takes centre stage; the doing of speech (screaming, in this case) comes afterwards.
​​
Furthermore, when the tag comes after the dialogue, it can roll seamlessly into any supporting narrative, as shown in the example from The Ghost Fields (p. 194).

There are a couple of potential disadvantages:
​

• In longer chunks of dialogue in scenes with more than two speakers, the reader will have to wait until the end of the speech to find out who’s saying what.
• Placement at the end of speech can flatten a one-liner or suspense point in dialogue.

Position: Between dialogue
Placing speech tags between dialogue is also common and unlikely to jar the reader. Here are three reasons why it works:

• The tag breaks up longer streams of dialogue, which is especially handy if a monologue’s rearing its head.
• We’re given an early indication who’s speaking. If there are more than two speakers in a scene, and the reader’s likely to be confused, placement between the speech is an effective solution.
• One-liners, suspense points and shocks get to take centre stage. Adding the speech tag at the end could flatten the tension.

Here are two examples in which the mid placement of the tag means the suspense isn't interfered with. The first is taken from The Ghost Fields (p. 194); the second is something I made up.

​

In the first example, not having the speech tag at the end of the dialogue focuses the reader on one question: what’s the clue? Not: Frank’s the speaker.

In the second example, rejig the sentence so that Tom said comes after all the speech, and notice how this makes the wallop vanish from the line about pulling the trigger.

Position: Before dialogue
Placement of the tag before the dialogue isn’t a no-no but it is a less common option and more noticeable.

A tag tells of speaking; dialogue shows character voice, mood and intention. When the speaker’s announced first, it’s a tap on the shoulder that draws attention to speaking being done. It expands what author and creative-writing expert Emma Darwin calls the ‘psychic distance’ between the reader and the speaker, which can flatten the mood.

And, yet, this can also be its advantage. That tap introduces a more staccato rhythm that can stop a reader in their tracks.
​
In this extract from Recursion (p. 292), the placement of the tag before the dialogue induces an acute sense of resignation – that dull thump in the pit of one’s stomach when the proverbial’s hit the fan.

Not placing tags: Omission
There’s no need to include a speech tag if it’s adding nothing but clutter. In the following example from Recursion (p. 125), the author has omitted them because there are only two speakers. He lets the dialogue, and its punctuation, inject the voice, mood and intention into the scene rather than telling us who’s speaking and how they’re saying it.

Summing up
​Placement of dialogue tags isn’t about rules. It’s about purpose:
​

• Varying rhythm
• Respecting mood and suspense
• Clarifying who’s speaking
• Avoiding unnecessary clutter

For that reason, mixing up the position of speech tags can be effective.

​Let’s end with an extract from Out of Sight (pp. 135–7), which demonstrates the varied ways in which author Elmore Leonard handles his tagging: beginning, between, end, and omission.

Cited sources and further reading

• Editing Fiction at Sentence Level, Louise Harnby, Panx Press, 2020
• How to Punctuate Dialogue (free webinar)
• Out of Sight, Elmore Leonard, Weidenfeld & Nicolson, 2017
• Recursion, Blake Crouch, Pan, 2020
• The Ghost Fields, Elly Griffiths, Quercus, 2015

This post explains when and how to indent your narrative and dialogue according to publishing-industry convention.

The purpose of first-line indents
Each new paragraph signifies a change or shift of some sort ... perhaps a new idea, piece of action, thought or speaker, even a moderation or acceleration of pace. Still, the prose in all those paragraphs within a section is connected.

Paragraph indents have two purposes in fiction:

• Readability: They help the reader identify the shifts visually.
• Connectivity: They indicate a journey. Indented paragraphs are related to what's come before ... part of the same scene.

First lines in chapters and new sections
Chapters and sections are bigger shifts: perhaps the viewpoint character changes, or there's a shift in timeline or location.

To mark this bigger shift in a novel, it’s conventional not to indent the first line of text in a new chapter or a new section. You might hear editorial folks refer to this non-indented text as full out.

• This is standard with narrative and dialogue.
• The convention applies regardless of your line spacing.

NARRATIVE LAYOUT
The following example is taken from Part 5, Chapter 2, of Christopher Priest’s Inverted World (p. 287, 2010):
​

• Paragraph 1 is the first in the chapter.
• The first line is not indented.
• The first lines of the paragraphs that follow it (2) are indented.

And here's an example from Part 2, Chapter 6, p. 147, which shows how the layout works the same after a section break:
​

• Paragraph 1 is the first in the section.
• The first line is not indented.
• The first lines of the paragraphs that follow it (2) are indented.

Even if an author chooses to include a design feature such as a dropped capital (sometimes called a drop cap), it's standard for that letter to be full out, as shown in the following example from To Kill a Devil (John A. Connell, p. 6, Nailhead Publishing, 2020):

• Paragraph 1 is the first in the chapter.
• The capital letter on the first and second lines is not indented.
• The first line of the paragraph that follows it (2) is indented.

DIALOGUE LAYOUT
​The same applies even if the chapter or section starts with dialogue, as in this excerpt from David Rosenfelt's Dog Tags (p. 192, Grand Central, 2010):

• Paragraph 1 is the first in the chapter.
• The first line is not indented.
• The first lines of the paragraphs that follow it (2) are indented.

Body text: dialogue and narrative
The example below from Blake Crouch's Recursion (p. 4, Macmillan, 2019) shows how the indentation works in the body text when there's a mixture of dialogue and narrative.

• Regardless of whether the prose is narrative or reported speech, the text is indented.
• The convention applies regardless of line spacing.

IMPACT OF LINE SPACING
Even if you've elected to set your book file with double line spacing (perhaps at the request of a publisher, agent or editor), the indentation convention applies. Here's the Recursion example again, tweaked to show what it would look like: 

Indenting text that follows special elements
Your novel might include special elements such as letters, texts, reports, lists or newspaper articles. Authors can choose to set off these elements with wider line spacing, but how do we handle the text that comes after?

Again, it's conventional to indent text that follows this content, regardless of whether it's narrative or dialogue. That's because of the connective function; the text is part of the same scene.

​Here are some examples from commercial fiction pulled from my bookshelves.

1. REPORT: The Outsider, Stephen King, Hodder, 2018, p. 252
2. LIST: Life of Pi, Yann Martel, Canongate, 2002, p. 146
3. TRANSCRIPT: Snap, Belinda Bauer, Black Swan, 2018, p. 36
4. RECORD: Ready Player One, Ernest Cline, Arrow, 2012, p. 300

It's not the case that full-out text is never used, or can't be used, but fiction readers are used to conventions. When a paragraph isn't indented, they assume it's a new section, which creates a tiny disconnect.

That's what I think's happened in the example below from Kate Hamer's The Girl in the Red Coat (p. 325, Faber & Faber, 2015). Of course, it took me only a split second to work out that the narrator is referring to the preceding letter, but it's a split second that took me away from the story because I'd assumed I was looking at a section break.

​My preference would be to indent 'I touch my finger [...]' because that text is part of the scene, not a new section.

How to create a first-line indent in Word
Let's finish with some quick guidance on creating first-line indents. 

​Avoid using spaces and tabs to create indents in Word. Instead, create proper indents. There are several ways to do this.

• Open the Home tab (1).
• Select your text.
• Move your cursor to the ruler and select the top marker (2).
• Drag it to the position of your preferred indent.
• Right-click on the style in the ribbon (3).
• Select 'Update Normal to Match Selection'.​

OR​

• Open the Home tab (1).
• Open the Styles pane via the arrow icon (4).
• Select your text.
• Move your cursor to the ruler and select the top marker (2).
• Drag it to the position of your preferred indent.
• Go to the Styles pane (5) and right-click on the style (6).
• Select 'Update Normal to Match Selection'.​ ​

OR

• Open the Home tab (1).
• Open the Styles pane via the arrow icon (4).
• Go to the Styles pane (5) and right-click on the style (6).
• Select 'Modify' to open the Modify Styles pane (A).
• Click on the Format button in the bottom left-hand corner (B).
• Select Paragraph to open the Paragraph pane (C).
• Make sure you're in the Indents and Spacing tab.
• Look at the Indentations section in the middle. Make sure 'First line' is selected under 'Special:' (D).
• Adjust the first-line indent according to your preference (E).
• Click OK (F).

Create a new style for your full-out paragraphs using the same tools.
​

• If using the ruler, ensure the markers (2) are aligned, one on top of the other.
• If using the styles pane, adjust the indent spacing (E) to zero.

If you need more assistance with creating styles, watch this free webinar. There's no sign-up; just click on the button and dig in.

Not sure if contractions are a good fit for your fiction’s dialogue? Here’s why they (nearly always) work.

Next time you’re in the pub with friends, at the dinner table with your family, or travelling on the bus, do a bit of people-listening. They’ll speak with contractions: you’re, they’re, I’m, don’t, hadn’t, can’t and so on.

Contractions are a normal part of speech. They help us communicate faster and improve the flow of a sentence.

Watch the following videos and listen to the spoken words. They feature people with very different backgrounds, sharing different stories, and in situations that demand varying levels of formality.
​

• A philosopher presenting a TED Talk about why the universe exists
• A fictional lawyer questioning a witness
• A child reading a poem
• A drug dealer describing his criminal activity
• A reporter’s voiceover on a news programme
• A vicar giving a sermon in a UK church
• A queen broadcasting a message to the nation

The people talking have one thing in common: they use contractions when they speak ... most (though not all) of the time.

That’s why when we want to write natural dialogue – dialogue that flows with the ease of real-life speech – contractions work.

How contractions affect the flow of dialogue
Take a look at this excerpt from No Dominion by Louise Welsh (Kindle edition, John Murray, 2017).

Here’s the contraction-free recast. It reads awkwardly, and leaves us unconvinced that what’s in our mind’s ear bears any relation to what we would have heard had this been real speech. And that means we’re questioning the authenticity of the story rather than immersing ourselves in it.

Contractions and genre
Some authors avoid contractions because of the genre they’re writing in. You’re more likely to see this in historical fiction than contemporary commercial fiction but it’s not strictly genre-specific and is more an issue of authorial style.
​
Here’s an excerpt from a contemporary psychological mystery, The Wych Elm (Penguin, 2019, p. 71). Tana French uses contractions in the narrative and dialogue. The speech sounds natural; the narrative that frames it is informal.

Compare this with Our Mutual Friend (Wordsworth Editions, 1997, p. 235). Dickens, writing literary fiction in the 1860s, still uses contractions in dialogue, although he avoids them in his more formal but wickedly tart narrative.

Now look at Ambrose Parry’s The Way of All Flesh (Canongate Books, 2018). The authors (Parry is the pseudonym used by Chris Brookmyre and Marisa Haetzman) don’t avoid contractions completely but the sparing usage does give the dialogue a more archaic feel. Here’s an excerpt from p. 259.

I doubt anyone would be surprised when I say that the setting is 1847. And so it works. However, if the novel were set in 2019, there’d be a problem. We’d consider those characters unrealistically pompous and the dialogue overblown.

​Contractions, pace and voice
The decision to use or avoid contractions is a tool that authors can use to deepen character voice.

Specific contracted forms might enable readers to imagine regional accents, social status and personality traits such as pomposity.

P.G. Wodehouse is a master of dialogue. Bertie Wooster is a wealthy young idler from the 1920s. Jeeves is his savvy valet. The dialogue between the two pops off the page and Wodehouse uses or avoids contractions to make the characters’ voices distinct.
​

• Jeeves’s speech is often uncontracted. This moderates the pace of the dialogue and helps to render his voice as serious, formal, patient and long-suffering.
• Bertie’s speech is often contracted. This accelerates the dialogue and helps to render his voice as flighty, foolish and careless.

You can see it in action in The Inimitable Jeeves (Kindle version, Aegitus, 2019).

In Oliver Twist (Wordsworth Classics, 1992, p. 8), Dickens contracts is not (ain’t) and them (’em) to indicate the low social standing of Mrs Mann, who runs a workhouse into which orphaned children are farmed.

Contractions and their impact on stress and tone
You can use or omit contractions in order to force where the stress falls in a sentence.

Compare the following:
​

• ‘I can’t believe you said that,’ Louise said.
• ‘I cannot believe you said that,’ Louise said.

By not using a contraction in the second example, the stress on ‘cannot’ is harder. The change is subtle but evident. With can’t the mood is one of disbelief tempered with a whining tone; with cannot the disbelief remains but the tone is angry.

Here’s another excerpt from The Way of All Flesh (p. 140). The speaker is a surgeon, Dr Ziegler.

Parry’s dialogue doesn’t just evoke a historical setting. The style of the speech affects the tone too. The voice is compassionate, the mood stoic. However, the lack of contractions renders the tone precise and careful.

On p. 142, Raven – a medical student turned sleuth – talks with the matron about medical charlatanry:

• ‘Aye, though there is worse,’ said Mrs Stevenson.

Using there is forces the stress on is, and in consequence the tone is resigned. With there’s the stress would have fallen on worse, and we might have assumed a more conspiratorial tone, as if she were about to divulge a secret.

Contractions and narration style
If you’re wondering whether to reserve your contractions for dialogue only, consider who the narrator is.

Let’s revisit Tana French (p. 1). The viewpoint character is a privileged man called Toby, the setting contemporary Dublin. What’s key here is that the narration viewpoint style is first person.

It is Toby who reports the events of the mystery; the narrative voice belongs to him. His narration register is therefore the same as his dialogue register – relaxed, colloquial – and the author, accordingly, retains the contractions in the narrative.

For sections written in third-person limited, the narrative voice would likely mirror the viewpoint character’s style of speaking. However, if you shift to a third-person objective viewpoint, where the distance between the characters and the readers is greater, the narrative might handle contractions differently to dialogue. It's a style choice you'll have to make.

Guidance from Chicago
Still a little nervous? Here’s some sensible advice from The Chicago Manual of Style, section 5.105:

Evaluate the sound
Here are three ways to help you evaluate the effectiveness of contracted or contracted-free dialogue:

1. Read the dialogue aloud: Is it difficult or awkward to say it? Does it sound unnatural to your ear? Do you stumble? Does it feel laboured, like you’re forcing the flow? If so, recast it with contractions. If the revised version is smoother, and the integrity of the setting is retained, go with the contracted forms.

2. Ask someone else to read the dialogue: Objectivity is almost impossible when it comes to our own writing. If the plan is that no one but you will ever read your book, write your dialogue the way you want to write it and leave it at that. If, however, you’re writing for readers too, and want to give them the best experience possible, fresh eyes (and ears) will serve you well.

You could even give your readers two versions of the dialogue sample – one with contractions and one without – and ask them which flows better and reads most naturally.

3. Head for YouTube: Dig out examples of speech by people whose backgrounds, environments and historical settings are similar to those of your characters. Watching characters in action will give you confidence to place on the page what can be heard from the mouths of those on the screen.

Summing up
Whether to use contractions or not in dialogue is a style choice. There are no rules. However, a style choice that renders dialogue stilted and unrealistic is not good dialogue.

Here's how to use quote marks (or speech marks) according to publishing convention in your fiction writing. The guidance covers both US English and UK English conventions.

In this post, I cover the following:

• What quote marks are used for
• Omitting a closing quote mark in dialogue
• Whether to use single or double quote marks
• Whether to use straight or curly quote marks
• Where the closing quote mark goes in relation to other punctuation​
• When not to use quote marks​

What quote marks are used for
Quote marks are used in 3 ways in fiction:

• Character dialogue
• To distance the narrator from what's being reported 
• ​To denote song titles and other works

Character dialogue
Quote marks show that we’re reporting what someone else is saying or said. 

​Each new speaker's dialogue should appear on a new line and include opening and closing quote marks.

To distance the narrator from what's being reported 
The tone of the distancing rendered by the quote marks will depend on narrative intent. Perhaps the voice is sarcastic. Or the author might want the reader to suspend belief by indicating that a character considers a word or phrase unreliable.

Imagine the character is saying so-called or supposed or allegedly before the word in quotes.

A word of caution: Don't be tempted to differentiate distancing terms in the narrative from dialogue by using an alternate style. If there are double speech marks around the dialogue, there should be double marks around the distancing words.

To denote song titles and other works
Quote marks are also used to identify certain published works such as song titles and book chapter titles.

So, for example, if a writer is referring to an album or book title, this is rendered in italic. However, when it comes to a song on an album, or a chapter in a book, it's conventional to use quote marks.

Omitting a closing quote mark in dialogue
There's one occasion where it's acceptable to omit the closing speech mark in dialogue: same speaker, new paragraph.

So, if you want your dialogue to take a new paragraph while retaining the current speaker, use a quotation mark at start of the new line but omit the closing one at the end of the previous paragraph.

​Single versus double quote marks
There’s no rule, just convention.

There are lots of Englishes: US, UK, Canadian, Australian, New Zealand, South African, Indian, etc. Each has its own preferences and idiosyncrasies.

Focus on which English your audience will expect, and punctuate your writing accordingly. Whichever style you choose, the main thing is be consistent. 
​

• In the UK, it’s more common to use single quote marks. And if there’s a quote within the quote, that’s a double. You might hear quotes within quotes called nested quotes.
• In US English it’s conventional to use double quote marks with nested singles. 

If you choose double quote marks, use the correct symbol, not two singles.

Straight versus curly quote marks
Curly quote marks are more conventionally known as smart quotes.

​It’s conventional in mainstream publishing to use smart or curly quotation marks, not unidirectional ones.

Changing straight quotes to smart quotes is one of the initial clean-up jobs an editor will carry out when they start work on a file. To prevent the problem occurring from the minute you begin typing:
 

• Go to FILE and select OPTIONS
• Select PROOFING, then click on the AUTOCORRECT OPTIONS button
• Choose the AUTOFORMAT AS YOU TYPE tab
• Make sure there’s a tick in the "STRAIGHT QUOTES" WITH “SMART QUOTES” box
• Click on OK

 
If you’ve pasted material into your book from elsewhere, or you didn’t check autocorrect options before you began typing, there might be some rogue unidirectional marks in your file. To change them quickly, do a global find/replace:
 

• Select CTRL+H on your keyboard to open FIND AND REPLACE
• Type a quotation mark into the FIND WHAT box
• Type the same quotation mark into the REPLACE WITH box
• Click on the REPLACE ALL button

The closing quote mark in relation to other punctuation​
In fiction, punctuation related to dialogue is placed similarly whether you're writing in US or UK style: within the quote marks.

Here are some examples:

​However, there's a difference when it comes to distancing or cited works. Note the different placement of the commas and full stops in the US and UK examples. In US English, the commas come before the closing quotation marks; in UK English, they come after.

When not to use quote marks
There are 2 issues to consider here:

• Thoughts 
• Emphasis

Thoughts
CMOS at section 13.43 says you can use quote marks to indicate thought, imagined dialogue and other internal discourse if you want to. However, I recommend you don't. For one thing, I can’t remember the last time I saw this approach used in commercial fiction coming out of a mainstream publisher’s stable.

But the best reason for not putting thoughts in quote marks is because it might confuse your reader. The beauty of quote marks – or speech marks – is that they indicate speech. Let them do their job!

Emphasis 
It can be tempting to use quote marks in your writing to draw attention to a word or phrase, but it’s rarely necessary and could even have the opposite effect to what you intended. It works instead as a distancing tool, as discussed above.

If you’re tempted to use quote marks for emphasis, imagine saying the sentence out loud, and making air quotes with your fingers as you speak. Would your character/narrator say it like that? If the answer's no, leave out the quote marks. Italic will work better. Or recast your dialogue so that the reader can work out where to place the stress themselves.

Summing up
If in doubt about how to use quote marks for your book, consult a style manual. I recommend the Chicago Manual of Style, the Penguin Guide to Punctuation and New Hart’s Rules, all of which offer industry-standard guidance.

Fancy listening instead?
If you'd prefer to listen to the advice offered here, Denise Cowle (a non-fiction editor) and I chat about how to use quote marks in all types of writing on The Editing Podcast. You can listen right here or via Apple Podcasts, Spotify or your favourite podcast platform

Is your dialogue pushing your novel forward or making the reader feel like they’re eavesdropping on a mundane conversation at the bus stop? Here’s how to ensure your dialogue pops.

What good dialogue isn’t
Good dialogue is the icing on the cake of a well-structured novel. Bad dialogue will mar a well-structured narrative and bury a story that is barely hanging in there. Before we look at what makes great dialogue, let’s look at what dialogue isn’t: 

3 components of effective dialogue
‘Dialogue should be the character in action,’ says John Yorke in his must-read Into the Woods (p. 151). Yorke’s talking about the art of screenwriting but the advice is just as pertinent for novelists. I recommend you read it even if you have no intention of writing for the screen because it’s a masterclass in storytelling, whatever the medium.

When we stop thinking about dialogue as words spoken – as conversations – and instead frame it in terms of characters, we create something that’s fit for a novel.

​What does your dialogue tell readers about who your characters are, how they’re feeling, and what their motivations are?

Unreliable dialogue
What a character expresses through dialogue need not match their true voice, mood or intent. Unreliable dialogue is powerful precisely because it jars the reader by masking the truth (which the characters themselves may even have buried).

Imagine this scenario: John has been kidnapped by Jane. They met in a club where she spiked his drink. He started to feel unwell and she offered him a ride to the Tube station. He never made it. He’s been held captive for several days, during which time he’s been physically abused and deprived of food. He’s frightened out of his wits, and weak to boot.

The dialogue between John and Jane could go as follows: John raves and rants, telling Jane her behaviour is monstrous, that Jane’s going to pay for her actions and that he’s going straight to the police as soon as he’s escaped. Jane responds in fury, telling John he deserves it all and how there’s no way he’ll ever escape.

Or the dialogue could be unreliable. John might be polite, sycophantic even, as he thanks her for the water she provided, compliments her on her appearance, or asks her about her life. Through that speech, we are shown his desperation. It’s about keeping her on side and calm in order to save himself.

And Jane’s verbal response might be chipper, seductive even. Through that dialogue, we are shown her psychosis.

The result is a sinister verbal exchange that allows us to explore the inner workings of the characters’ minds without it being forced down our throats via an all-to-obvious narrative that’s centred around the viewpoint character.

Breaking free of viewpoint limitations
Most novelists opt to hold third- or first-person narrative viewpoints. That means the story in a chapter plays out through one person’s perspective.

When authors drop viewpoint, readers end up playing a game of narrative table tennis in which they bounce from one character’s head to another. We know who everyone is (voice), what everyone’s feeling (mood), and what everyone wants (intention) all of the time. Readers become disengaged because they don’t have time to immerse themselves in any one character’s experience.

Good dialogue allows writers and readers to break free without head-hopping. Through dialogue, readers can intuit the voice, mood and intention of multiple characters, yet the singular narrative viewpoint remains true throughout. That keeps the reader engaged and the writing taut.

Purposeful dialogue in action
Here’s an excerpt from Lee Child’s Never Go Back (pp. 457–8). I chose it because I’ve also seen the movie, which allowed me to compare my experience of the screen dialogue (and the advice Yorke gives) with the novel’s, and because there are no action beats, only two speech tags and no narrative. It’s just dialogue between a teenage girl and Jack Reacher. 

WHAT WE LEARN

• Viewpoint: Reacher is the viewpoint character. However, through the dialogue, readers are able to access information about him and the teenager.
• Voice: Both characters speak in a no-nonsense fashion. Reacher might think he holds the balance of power, given his job and age, but the girl’s streetwise tone suggests she thinks otherwise; they’re equals.
• Mood: Reacher is calm (we’d expect nothing more) and measured, but there is emotional engagement; he seems keen not to give too much away (probably in respect of the fact that he’s talking to a child, perhaps his daughter). As for the girl, she’s almost disinterested. I imagine her stuffing fries into her mouth, as focused on her food as she is on the conversation ... until she thinks her mother might be in trouble. Child doesn’t tell us, but I can see her in my mind’s eye, mouth full, looking at him, just a twitch of alarm registering in her expression. There’s almost a sense of the traditional mother/daughter relationship reversed; this is a kid who’s used to looking after herself. There’s movement in this dialogue; the speech is infused with its own action beats.
• Intent: Reacher wants to find out who the mother is. This exchange is all about building trust with the girl in order to achieve that. The girl needs someone she can turn to because bad people are watching her and her mother. She wants to know how Reacher might know Candice; Reacher wants to keep the possible sexual encounter and his resulting paternity to himself. The girl’s not stupid and picks up on his vague inference to a sexual encounter. As the conversation and the chapter close, Reacher and the readers discover that Candice and our protagonist have never met, and that the girl can’t be his daughter.   

This is characters in action, expressed through speech. Child makes every line count towards the chapter denouement. If he was tempted to introduce narrative and action beats that ensured we’d get it, it doesn’t show. In fact, they would have been interruptions and slowed the pace. Instead, he trusts us to do the work because the dialogue gives us everything we need.

Here’s an excerpt from The Poison Artist by Jonathan Moore (pp. 244–5). I chose this because of the contrast between Kennon and Emmeline’s speech.

WHAT WE LEARN

• Viewpoint: Even though Emmeline isn’t the viewpoint character, we get to access her transgressor headspace through the dialogue, and the result is powerful and sinister.
• Voice: Kennon doesn’t say much, but what he does say is what we’d expect from a trained inspector, though it’s offered quietly. Emmeline, however, is pathologically measured. We’re left in no doubt that she’s dangerous.
• Mood: We sense Kennon’s fear and desperation through his truncated, whispered speech. There’s almost exhaustion in play. Emmeline, in contrast, is seductive (offering him a drink as if she were hosting a dinner party). Given the torture in evidence, her dialogue is obscene.
• Intent: Kennon is trying to save himself, Caleb and the twitching man, but he’s playing by the rule of law, warning Emmeline that he will fire his gun. Emmeline wants to finish her game, and her desire to play is evident in her speech. Of course, this may be unreliable; it could be masking a deep-seated fear that she’ll be harmed or killed.

This, too, is characters in action, expressed through speech. Moore draws us deep into the transgressor’s mind – even though she’s not the viewpoint character. Her dialogue, juxtaposed with Kennon’s exhausted near-silence, generates a powerful scene that oozes with sickly tension.

Here’s a third example from Harlan Coben’s Run Away (pp. 68–9):

WHAT WE LEARN

• Viewpoint: Simon is the viewpoint character but through Ingrid’s speech we learn how she uses her professional mindset to manage stressful situations. 
• Voice: Simon’s voice is emotional and verbose; Ingrid’s is precise and clinical.
• Mood: Simon’s disgust and distress are evident. You can almost feel the words falling out of his mouth. Ingrid’s speech exudes clinical detachment. She’s in doctor mode.
• Intent: Simon wants to unload. Ingrid wants to understand.

Again, through speech we see the characters in action. The contrasting voices and moods show us Simon and Ingrid’s different intentions. There’s no need for more than a peppering of supporting narrative.

Summing up

1. Make your characters’ speech count. Use it to show who the character is (voice), how they feel (mood) and what they want (intention).
2. Play with unreliable dialogue if it will enhance our understanding of characters’ emotions and motivations. Those deliberate juxtapositions will deepen our engagement.
3. Realistic everyday speech, while authentic, is dull, invasive and will disengage your reader. Remove it! 

​Free dialogue enrichment tool
To help you think about your characters' voices, moods and intentions, and how these will enrich your dialogue, download this tool. It's a fillable PDF with ready-made examples and space for you to record your own decisions. ​

A vocative expression is one in which a person is directly referred to in dialogue. It needn’t be someone’s name; it could be a form of address that relates to their job or position, or a term of endearment, respect or disrespect. Here’s how to work with them.

Purpose of vocative expressions
Vocatives serve several purposes in fiction writing:
​
1. They help readers keep track of who’s saying what to whom. This is especially useful when a character is talking to two or more people.

2. They can enrich characters’ emotions by conveying a deeper sense of urgency, frustration, surprise or patience.

3. Readers can learn quickly about how characters relate to each other. Does one rank higher or defer to the other? Perhaps they’re friends, lovers, or loathe each other.