definition of direct presentation in literature

Direct Characterization

Definition of direct characterization.

Direct characterization means the way an author or another character within the story describes or reveals a character, through the use of descriptive adjectives , epithets , or phrases . In other words, direct characterization happens when a writer reveals traits of a character in a straightforward manner, or through comments made by another character involved with him in the storyline.

Direct characterization helps the readers understand the type of character they are going to read about. For instance, in Arthur Miller’s The Crucible , he describes his character John Proctor in this way: “He was the kind of man – powerful of body, even-tempered, and not easily led – who cannot refuse support to partisans without drawing their deepest resentment.”

Examples of Direct Characterization in Literature

Example #1: the most dangerous game (by richard connell).

“The first thing Rainsford’s eyes discerned was the largest man Rainsford had ever seen – a gigantic creature, solidly made and black bearded to the waist. … ” ‘Ivan is an incredibly strong fellow,’ remarked the general, ‘but he has the misfortune to be deaf and dumb. A simple fellow, but, I’m afraid, like all his race, a bit of a savage.’ “

The above passage shows a good example of a direct characterization. Here Zaroff has explicitly described another character Ivan in the story The Most Dangerous Game , leaving readers with no more questions about him. Ivan is a muscular, huge man, having a long black beard. He is deaf and dumb, yet strong, Zaroff says.

Example #2: The Old Man and the Sea (by Earnest Hemingway)

“The old man was thin and gaunt with deep wrinkles in the back of his neck. The brown blotches of the benevolent skin cancer the sun brings from its reflection on the tropic sea were on his cheek … Everything about him was old except his eyes and they were the same color as the sea and were cheerful and undefeated.”

Hemingway uses the method of direct characterization to describe the old man’s personality traits, especially the vivid eyes of his main character, the old man, Santiago in his novel .

Example #3: Hedda Gabler (by Henrik Ibsen)

“MISS JULIANA TESMAN, with her bonnet on a carrying a parasol, comes in from the hall, followed by BERTA, who carries a bouquet wrapped in paper. MISS TESMAN is a comely and pleasant- looking lady of about sixty-five. She is nicely but simply dressed in a grey walking-costume. BERTA is a middle-aged woman of plain and rather countrified appearance…GEORGE TESMAN comes from the right into the inner room … He is a middle-sized, young-looking man … He wears spectacles, and is somewhat carelessly dressed in comfortable indoor clothes.”

In this excerpt, Henrik Ibsen has described three characters: Miss Tesman, Berta, and George Tesman. He has clearly shown their personalities and mannerism through direct characterization.

Example #4: Pride and Prejudice (by Jane Austen)

“Mr. Bingley was good-looking and gentlemanlike; he had a pleasant countenance, and easy, unaffected manners. … he was discovered to be proud, to be above his company, and above being pleased; and not all his large estate in Derbyshire could then save him from having a most forbidding, disagreeable countenance, and being unworthy to be compared with his friend.”

Mr. Bingley, the romantic interest of Jane, and his friend, Mr. Darcey, are described in this excerpt through direct characterization. She has admired Mr. Bingley for his pleasant countenance, comparing him to Mr. Darcy.

Example #5: The Canterbury Tales (by Geoffrey Chaucer)

“He yaf nat of that text a pulled hen, That seith that hunters ben nat hooly men, Ne that a monk, whan he is recchelees… His heed was balled, that shoon as any glas, And eek his face, as he hadde been enoynt. His eyen stepe, and rollynge in his heed, That stemed as a forneys of a leed; His bootes souple, his hors in greet estaat.”

Through monk’s portrait, his physical and social life, readers see a satire of the religious figures that should live a proper monastic life of hard work and deprivation. This is the achievement of the description of Chaucer that he has described a character through direct characterization.

Function of Direct Characterization

Direct characterization shows traits as well as motivation of a character. Motivation can refer to desires, love, hate, or fear of the character. It is a crucial part that makes a story compelling. Descriptions about a character’s behavior, appearance, way of speaking, interests, mannerisms, and other aspects draw the interest of the readers and make the characters seem real. Also, good descriptions develop readers’ strong sense of interest in the story.

Related posts:

• Characterization
• Direct Object
• 10 Best Characterization Examples in Literature

Post navigation

Direct vs indirect characterization: How to show and tell

There are two main ways to reveal characters: direct characterization, and indirect characterization. What defines these two characterization types, and what are the strengths and weaknesses of each?

• Post author By Jordan
• 26 Comments on Direct vs indirect characterization: How to show and tell

Characterization describes the way a writer or actor creates or implies a character’s personality, their inner life and psyche. Two main ways to reveal your characters are direct characterization and indirect characterization. What are these character creation techniques? Read on for examples of characterization that illustrate both:

Guide to direct and indirect characterization: Contents

What is direct characterization, direct characterization example, what is indirect characterization, indirect characterization example.

• Eight tips for using direct vs indirect characterization

Let’s delve into using both characterization devices:

To begin with a definition of direct characterization, this means the author explicitly tells the reader a character’s personality .

For example, explicitly telling the reader a character is kind, funny, eccentric, and so forth.

Here’s an example of direct characterization from Virginia Woolf’s To the Lighthouse (1925).

Woolf explicitly shows what characters think of one another . In the example, an artist staying with the Ramsay family, Lily Briscoe, thinks about Mr Ramsay whom a man Mr Bankes has just called a hypocrite:

Looking up, there he was – Mr. Ramsay – advancing towards them, swinging, careless, oblivious, remote. A bit of a hypocrite? she repeated. Oh no – the most sincere of men, the truest (here he was), the best; but, looking down, she thought, he is absorbed in himself, he is tyrannical, he is unjust… Virginia Woolf, To The Lighthouse (1927), p. 52.

This is direct characterization – through Lily, Woolf describes Mr. Ramsay’s traits directly.

It’s telling (direct characterization typically is), but because we read it as one character’s opinion of another, it also shows us how Lily feels, whether or not she agrees with the statement that Mr. Ramsay is a hypocrite.

Through Lily, we learn Ramsay is ‘absorbed in himself’ or self-absorbed, tyrannical – we read direct statements about Ramsay’s personality that help us picture him and how he comes across to others.

‘Indirect characterization’ shows readers your characters’ traits without explicitly describing them.

To give simpler examples of direct vs indirect characterization, for direct you might write, ‘Jessica was a goofy, eccentric teacher’.

For indirect revelation of Jessica’s character, you might write instead, ‘Jessica had named the stick with a hook on the end she used to open the classroom’s high windows Belinda and would regale her children with stories of Belinda’s adventures (even though they were fourteen, not four)’.

In the second example of characterization above (the indirect kind), it is inferred that Jessica is goofy and eccentric. She names inanimate objects and tells teenagers stories of make-believe that would probably be better-suited to younger children.

Indirect characterization invites your reader to deduce things about your characters, without explicitly telling them who they are.

Here, John Steinbeck in  The Grapes of Wrath (1939) shows a character’s personality indirectly.

Steinbeck doesn’t say that hitchhiker Joad is a down-and-out, blue-collar worker. Instead, the author creates indirect characterization through the items a worker in this context would perhaps have: whiskey, cigarettes, calloused hands:

Joad took a quick drink from the flask. He dragged the last smoke from his raveling cigarette and then, with callused thumb and forefinger, crushed out the glowing end. He rubbed the butt to a pulp and put it out the window, letting the breeze suck it from his fingers. John Steinbeck, The Grapes of Wrath (1939), p. 9.

Types of indirect characterization

What types of indirect characterization are there?

Any writing that helps us infer or deduce things about a person’s psyche, emotions, values or mannerisms. For example:

• Dialogue-based inference: From the way your character speaks to others in the story, your reader may deduce that they are kind, cruel, gentle, etc.
• Implying through action: What your character does (for example jumping on a beetle to squash it) implies their character (in this case, it may imply that a character is cruel).
• Fly-on-wall description: Although what visual description implies may differ from country to country, culture to culture, neutrally-worded description may cause your reader to make specific assumptions based on what you’ve shown. We might assume, for example, an extremely pale-skinned character is reclusive or agoraphobic, like the reclusive Boo Radley in Harper Lee’s  To Kill a Mockingbird .

So how do you use direct and indirect characterization well? Read tips for each (and our complete guide to description for more examples):

8 tips for using direct and indirect characterization

Avoid overusing direct characterization, be direct with key details, support direct character statements with scenes, imply character through action and reaction, tell direct details that serve concision, use dialogue to characterize indirectly, let narrative voice give character, read examples of direct and indirect characterization.

Direct characterization is useful shorthand. Instead of pages showing how a character is mean, you could start with ‘He was mean.’ Balance is key, though. Overusing direct characterizing may skew the balance towards telling, not showing. Tweet This

If, for example, you wrote, ‘He was mean. He was petty and generally unkind, so that neighbors crossed the street when he passed,’ that mixes some indirect characterization with the direct type. Neighbors crossing the street is a visual that indirectly implies avoidance and discomfort or possible dislike.

If you were to only tell readers about your characters’ traits without weaving in illustrative showing (which give indirect inference about who your characters are), the effect would be:

• Hazy visuals : Crossing the street in the example above gives a more specific visual than simply saying ‘he was disliked by the community’.
• Lack of depth and color: If you tell your reader who your characters are exclusively with minimal showing or inferring, it may read as though you have a private understanding of your characters you are summarizing for the reader, rather than showing them a fuller, more detailed picture.

Make a Strong Start to your Book

Join Kickstart your Novel and get professional feedback on your first three chapters and story synopsis, plus workbooks and videos.

Example of blending direct and indirect character detail

The opening of Toni Morrison’s powerful novel Beloved characterizes a house that is haunted by the ghost of an infant.

Note how Morrison moves from the direct characterization of the first sentence to specific, visual details:

124 was spiteful. Full of a baby’s venom. The women in the house knew it and so did the children. For years each put up with the spite in his own way, but by 1873 Sethe and her daughter Denver were its only victims. The grandmother, Baby Suggs, was dead, and the sons, Howard and Buglar, had run away by the time they were thirteen years old – as soon as merely looking in a mirror shattered it (that was the signal for Buglar); as soon as two tiny hand prints appeared in the cake (that was it for Howard). Toni Morrison, Beloved (1987), p. 17.

The trick to effective direct characterization is to reserve it for key details you want to establish upfront.

In the example of blending indirect and direct character description above, Morrison starts with direct, broad detail. A sense of spite that drives boys in the family from a home filled with the ghosts of a corrosive, violent history.

If you were to write a retelling of Charles Dickens’ A Christmas Carol you might start with ‘Scrooge was stingy’ and then similar flesh this key detail out with the illustrative, supporting detail.

The indirect characterization you then add on to key details gives further texture, color, specificity to your characters. It helps, of course, to know your characters inside out:

The example above from Beloved shows how indirect characterization supports direct descriptive statements .

The boys Howard and Buglar fleeing from mirrors that seem to shatter by themselves or tiny hand prints appearing in a cake, for example. These specific images and incidents support the suggestion that the home at 124 is haunted by a ‘spiteful’ (or rather, determined-to-be-known) presence.

If you tell your reader a character is kind, think of dedicated scenes as well as passing moments that support the direct revelation.

Maybe your character gives up a seat on public transport for an elderly person. Maybe they help a neighbor get a pet that has run out of an open gate into a busy road to safety.

Indirect characterization is useful because it shows your reader the type of actions your character is likely to take .

This in turn enables your reader to make educated, qualified guesses about how your characters’ might react in situations whose outcome is not yet known. Through this, one ‘gets to know’ characters as though they were real people.

Action and reaction provide useful ways to tell your reader who your characters are indirectly.

For example, Sarah has a vase that belonged to her grandmother that she cherishes, and her hyperactive son knocks it over and breaks it. Does she scold him to be careful? Lash out? Show a mix of anger and understanding?

Think about what you want your reader to infer about a character from the way they react, even in incidents or situations that are trivial or secondary to your story’s main plotline . In this way every scene, every incident, will contribute toward building your characters’ personae.

One of the benefits of direct characterization is that it allows you to be concise.

Direct characterization is useful, for example, when a narrator is recapping prior events that are useful to the present story but not its main focus. For example, in the first page of Nick Hornby’s Slam , a novel about a sixteen-year-old skater named Sam:

So things were ticking along quite nicely. In fact, I’d say that good stuff had been happening pretty solidly for about six months. – For example: Mum got rid of Steve, her rubbish boyfriend. – For example: Mrs Gillet, my art and design teacher, took me to one side after a lesson and asked whether I’d thought of doing art at college. Nick Hornby, Slam (2007), p. 1

At this point in the story, the reader doesn’t need lengthy exposition about why Steve was a rubbish boyfriend. So the direct, telling characterization suits the purpose of this part of the story – catching the reader up on what has been happening in the teenaged protagonist’s life.

There is still balance between indirect and direct characterization in this example. The second example Sam gives tells us (through Mrs Gillet’s action) that the teacher is caring and sees artistic potential in Sam, without saying so explicitly. The part or unique incident suggests the whole of the teacher-student relationship.

Dialogue is a fantastic device for characterization because it may move the story forward while also telling your reader who characters are.

If, for example, there is banter and characters tease each other, it may imply an ease and familiarity (compared to stiff formality between strangers). Note, for example, how Hornby creates a sense of how awkward Rabbit is (an 18-year-old skater at Grind City, a skate park Sam frequents) in the dialogue below:

‘Yo, Sam,’ he said. Did I tell you that my name is sam? Well, now you know. ‘All right?’ ‘How’s it going, man?’ ‘OK.’ ‘Right. Hey, Sam. I know what I was gonna ask you. You know your mum?’ See what I mean about Rabbit being thick? Yes, I told him. I knew my mum. Hornby, pp. 11-12.

In this brief exchange, we see through the awkward, stop-start flow of conversation how Rabbit lacks social graces and awareness and (in the ensuing dialogue) reveals he has a crush on Sam’s mother.

Another useful way to use indirect characterization is to give an involved narrator (a narrator who is also a character in the story) a personality-filled voice .

In the above example of characterization via dialogue, for example, Sam’s asides to the reader (‘Well, now you know’ and ‘See what I mean about Rabbit being thick?’) create the sense of a streetwise, slightly jaded teenaged voice.

Think of ways to inject characters’ personalities into their narration. What subjects do they obsess over (it’s clear Sam loves skating from the first few pages of Slam )? How do they see others (Sam appears fairly dismissive and a little cocky, from referring to his mom’s ‘rubbish’ boyfriend to his blunt description of Rabbit as ‘thick’).

Use language in narration your character would use based on demographic details such as age, cultural background or class identity.

The casual, clipped language Sam uses in the example above suggests the awkward and ‘too cool’ qualities of a teenaged boy.

To really understand the uses of direct and indirect characterization (and how to blend to two to show and tell, describe and imply), look for examples in books.

You could even write out the descriptions you love, to create your own guide to dip into whenever you’re creating characters.

Create believable, developed characters. Finishing a book is easier with structured tools and encouraging support.

Related Posts:

• Indirect characterization: Revealing characters subtly
• Direct characterization: 6 tips for precise description
• Writing advice: Show, don't tell: or should you?
• Tags character description , characterization

Jordan is a writer, editor, community manager and product developer. He received his BA Honours in English Literature and his undergraduate in English Literature and Music from the University of Cape Town.

26 replies on “Direct vs indirect characterization: How to show and tell”

Well explained and helpful

Thank you, Lexi. Thanks for reading!

Thanks for this, I’ve been back a few times now but failed to post a comment. ? This is going to help a lot during this revision!

It’s a pleasure, Robin. Glad you managed in the end. We’ve been migrating our blog to https (more secure) which may have been the cause. Good luck with your revision!

Ah! Okay. Thank you! First books are hard so much to learn. I feel like I could have written two other books while fixings this one. xD (I want to, I enjoy shaping the old chapters to how I write now. ^;^) I also found a program to help speed this up; bibisco. I like it way more than the complicated expensive writing programs out there. IMO.

Where to sign up to get updates for this blog? I don’t want to forget about your blog. (I need reminders for everything. lol. A newsletter is a good way to do that.)

If you sign up for a Now Novel member account, you get subscribed to our blog newsletter too. Alternatively, drop us a line at help at now novel dot com with the email address you’d like to use to get updates and I’ll have our email guy add you to our mailing list. Thanks!

Can you please ,include a section about dynamic and static characters? Thanks for your precious help

Hi Abdou, thank you for the suggestion! I’ll add it to the list for revision ideas, thank you.

You are welcome.

This is such a great website offering very useful tools for writers. I’ve been Googling for days now about everything I wanted to learn in novel writing and I can’t believe I just found this site.

Thanks, Alexa. I’m glad you’ve found our website helpful 🙂

You shared some excellent tips on characterization. I think all writers can benefit from this blog.

Thank you so much, Derrick. I’m glad you’re finding our blog helpful! Thanks for reading.

This is very helpful and I Aced my quiz on something i’m not that good at cool when you lookat the paper it looks long but when you start reading you get lost

Glad you aced your quiz, Kimberly.

Thank You great job!

Thanks, Anna. Thank you for reading our blog!

Very useful.

Thank you so much,

Thank you for your feedback, Aleix. It’s a pleasure, thank you for reading our articles.

Thank you for a clear explanation. It is most useful.

It’s a pleasure, Vivienne. Thank you for reading our blog.

An author employs indirect characterization to avoid explicitly announcing a character’s attributes by revealing those aspects to the reader through the character’s actions, thoughts, and words. Using the phrase “John had a short fuse” as an example would be direct characterisation, but the phrase “John hissed at the man without any prior warning” would be indirect portrayal.

Thank you for sharing your example of indirect and direct characterization and for reading our blog.

[…] is direct characterization is so […]

Thank you for providing this complete reference on direct and indirect characterization. It is often difficult to strike a balance between showing and explaining in writing, and your examples and advice are quite helpful. I really like the focus on utilizing both strategies sparingly, as well as the reminder that indirect characterization may frequently result in a more detailed and compelling picture of characters. I’ll keep these tips in mind as I strive to hone my own writing style.

Thanks so very for your comment. We’re so pleased that you found it helpful. All the best with your writing!

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Pin It on Pinterest

Direct Characterization

Definition of direct characterization.

Direct characterization means the way an author or another character within the story describes or reveals a character , through the use of descriptive adjectives, epithets, or phrases. In other words, direct characterization happens when a writer reveals traits of a character in a straightforward manner, or through comments made by another character involved with him in the storyline.

Direct characterization helps the readers understand the type of character they are going to read about. For instance, in Arthur Miller’s The Crucible , he describes his character John Proctor in this way: “He was the kind of man – powerful of body, even-tempered, and not easily led – who cannot refuse support to partisans without drawing their deepest resentment.”

Examples of Direct Characterization in Literature

Example #1: the most dangerous game (by richard connell).

“The first thing Rainsford’s eyes discerned was the largest man Rainsford had ever seen – a gigantic creature, solidly made and black bearded to the waist. … ” ‘Ivan is an incredibly strong fellow,’ remarked the general, ‘but he has the misfortune to be deaf and dumb. A simple fellow, but, I’m afraid, like all his race, a bit of a savage.’ “

The above passage shows a good example of a direct characterization . Here Zaroff has explicitly described another character Ivan in the story The Most Dangerous Game , leaving readers with no more questions about him. Ivan is a muscular, huge man, having a long black beard. He is deaf and dumb, yet strong, Zaroff says.

Example #2: The Old Man and the Sea (by Earnest Hemingway)

“The old man was thin and gaunt with deep wrinkles in the back of his neck. The brown blotches of the benevolent skin cancer the sun brings from its reflection on the tropic sea were on his cheek … Everything about him was old except his eyes and they were the same color as the sea and were cheerful and undefeated.”

Hemingway uses the method of direct characterization to describe the old man’s personality traits, especially the vivid eyes of his main character , the old man, Santiago in his novel.

Example #3: Hedda Gabler (by Henrik Ibsen)

“MISS JULIANA TESMAN, with her bonnet on a carrying a parasol, comes in from the hall, followed by BERTA, who carries a bouquet wrapped in paper. MISS TESMAN is a comely and pleasant- looking lady of about sixty-five. She is nicely but simply dressed in a grey walking-costume. BERTA is a middle-aged woman of plain and rather countrified appearance…GEORGE TESMAN comes from the right into the inner room … He is a middle-sized, young-looking man … He wears spectacles, and is somewhat carelessly dressed in comfortable indoor clothes.”

In this excerpt, Henrik Ibsen has described three characters: Miss Tesman, Berta, and George Tesman. He has clearly shown their personalities and mannerism through direct characterization .

Example #4: Pride and Prejudice (by Jane Austen)

“Mr. Bingley was good-looking and gentlemanlike; he had a pleasant countenance, and easy, unaffected manners. … he was discovered to be proud, to be above his company, and above being pleased; and not all his large estate in Derbyshire could then save him from having a most forbidding, disagreeable countenance, and being unworthy to be compared with his friend.”

Mr. Bingley, the romantic interest of Jane, and his friend, Mr. Darcey, are described in this excerpt through direct characterization . She has admired Mr. Bingley for his pleasant countenance, comparing him to Mr. Darcy.

Example #5: The Canterbury Tales (by Geoffrey Chaucer)

“He yaf nat of that text a pulled hen, That seith that hunters ben nat hooly men, Ne that a monk, whan he is recchelees… His heed was balled, that shoon as any glas, And eek his face, as he hadde been enoynt. His eyen stepe, and rollynge in his heed, That stemed as a forneys of a leed; His bootes souple, his hors in greet estaat.”

Through monk’s portrait, his physical and social life, readers see a satire of the religious figures that should live a proper monastic life of hard work and deprivation. This is the achievement of the description of Chaucer that he has described a character through direct characterization .

Function of Direct Characterization

Direct characterization shows traits as well as motivation of a character . Motivation can refer to desires, love, hate, or fear of the character . It is a crucial part that makes a story compelling. Descriptions about a character ’s behavior, appearance, way of speaking, interests, mannerisms, and other aspects draw the interest of the readers and make the characters seem real. Also, good descriptions develop readers’ strong sense of interest in the story.

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

4 – Characterization

4.1 defining characterization.

Quick Links

The worlds of prose fiction are not only made of events arranged into plots and environments arranged into settings. To have a story, there must also be characters. The arrangement of characters in the story is called characterization . But what are characters? Why are they so necessary for narrative? What kinds of characters do we find in fiction stories? How are they characterized and represented? These are some of the questions that will occupy us in this last chapter dedicated to the formal elements of story .

A character is any entity in the story that has agency , that is, who is able to act in the environments of the storyworld. Characters are most often individuals (e.g. Ivan Karamazov in The Brothers Karamazov , Werther in The Sorrows of Young Werther , or Henry Jekyll in Dr Jekyll and Mr Hyde ), but there are some special cases where we find collective or choral characters (e.g. Thebans in Oedipus Rex , or the group of neighborhood boys in The Virgin Suicides ). Characters are most often human beings, but they can also be nonhuman animals or other entities who behave like humans (e.g. the White Rabbit in Alice’s Adventures in Wonderland — Figure 4.1 — or the robots in I, Robot ).

Ben’s Bonus Bit – Literary Devices

Two related devices are relevant here: personification and dehumanization . Nonhuman characters who effectively function with agency in prose fiction may be said to be ‘personified’ or ‘humanized.’ Abstract forces or ideas may also be personified, such as Lady Liberty, Lady Justice ( lustitia in Latin), or Wisdom’s characterization as a woman in Proverbs 8:1 – 9:12. Whether these figures have agency (and are therefore characters) or whether they serve as symbols varies from text to text. Curious that women are so often dehumanized this way, transformed into abstractions for the admiration, objectification, or condemnation of male audiences 1 .

Personification as a characterization device is distinct from personification that functions as imagery (“The wind howled. Lightning danced across the sky”), which adds life, energy, or personality to setting without granting it true agency. These examples are implied metaphors , as readers understand the wind seems to howl (auditory imagery) and the lightning appears to dance (visual imagery). When writing essays about personification, keep this functional difference in mind: that personification becomes a characterization device only if the object/idea personified may take action in the storyworld.

Similar to dehumanization is zoomorphism , where human characters are given the qualities of animals. Zombies occasionally operate this way in speculative fiction , but we also frequently find zoomorphism and its inverse anthropomorphism in didactic children’s literature ( The Giving Tree , If You Give a Mouse a Cookie , Winnie-the-Pooh , Three Little Pigs , The Ugly Duckling , and so on), folklore (see Brer Rabbit, a Bugs Bunny progenitor in Joel Chandler Harris’ Uncle Remus: His Songs and His Sayings), and in the literature of the Harlem Renaissance (Ralph Ellison’s Invisible Man or Richard Wright’s Native Son ). When society treats a racial or ethnic group as less than human, authors often draw attention to that inequity via characterizing devices.

Because of the popularity of folklore and fable, be on the lookout for tricksters and tricksterism whenever zoomorphism or anthropomorphism appear. Besides the animal tricksters from classic cartoons (Bugs Bunny, Wile E Coyote, Woody Woodpecker, Tom & Jerry, and so on), consider that trickster god Loki from Norse mythology transforms into a horse and a wolf, that Reynard in medieval French allegory is an anthropomorphic fox, that Anansi, the Trickster god in Igbo culture, is a spider, that Dionysus, god of wine and madness in Greek mythology, is a shapeshifter, that in Japanese Yōkai folklore, kitsune are intelligent, shape-shifting foxes, and that Māui, a Polynesian trickster figure, is given shape-shifting powers a magic hook Macguffin in Disney’s 2016 film, Moana . There are many other examples. In this way, stories about transgression, transformation, androgyny, rebellion, justice, or disruption of social order often employ anthropomorphism, zoomorphism, or trickster characters.

Rarely are the characters of literary fiction animals or other entities without human features (e.g. the white whale in Moby Dick or the monoliths in 2001: A Space Odyssey ). In our discussion of character, therefore, we will assume that the characters of any story we’re studying are human or human-like individuals, although there are notable exceptions.

That characters are important for narrative fiction can be seen from the fact that the titles of many short stories and novels are taken from the proper names of their main characters ( protagonists ). These are sometimes called eponymous characters and are quite common in the history of literature. Some of the most famous novels are named after their protagonists, such as Don Quixote , Robinson Crusoe , Jane Eyre , Madame Bovary , or Anna Karenina . Even when characters do not appear in the title, they are still the most relevant existents in a great majority of short stories and novels. This seems to be a consequence of the nature and function of narrative. As we saw in the introduction, narrative is fundamentally a way for people to give meaning to our world. And what is more important to us than ourselves and others like us?

Fig. 4.1 Illustration of Lewis Carroll Alice in Wonderland (1865). By John Tenniel, Public Domain, https://en.wikipedia. org/wiki/Alice%27s_Adventures_in_ Wonderland#/media/File:Alice_par_ John_Tenniel_02.png

While recognizing the relevance of characters in narrative, we should not forget the intimate connections between characters and the other two existents of the story, events and environments (see Fig. 3.1, in Chapter 3 ). Stories are not simply made of characters acting in an environment. All the existents of the story are indispensable to the recreation of a convincing storyworld, just as they are in our own lifeworld. Thus, characterization, plot, and setting work together to effectively sustain narrative discourse and contribute to meaningful communication between authors and readers.

In this chapter, we will start by discussing how the nature of characters changes when we analyze them at the level of narrative , discourse , or story . We will then consider the notion of individuation to show that characterization in prose fiction is generally aimed at constructing fully individuated characters, but often also produces typical and universal characters . When analyzing fictional characters in psychological/ realistic terms, it is common to distinguish their degree of individuation ( flat vs. round characters ), as well as their degree of personal development throughout the plot ( static vs. dynamic characters ). After looking at these typologies of character, we will discuss the most common approaches to representing them in narrative: indirect and direct characterization . One vital method of direct characterization is dialogue , which will be the topic of the last section in this chapter.

4.2 The Actants of Narrative

The nature of characters varies depending on what level of the semiotic model we position ourselves in (see Figs. 1.5 and 1.7 in Chapter 1 ). At the level of narrative, characters may be seen as figments of the author , who endows them with certain features or qualities drawn from her imagination or observations, which are then recreated by readers in every reading. At the level of discourse, however, we can see characters as a construct of the text , a sort of ‘paper people’ whose features are exclusively constituted by the descriptions found in the text and the inferences that can be made from textual cues. 2 In this sense, characters are incomplete creatures, mere actants with no life beyond the text and no reason to exist other than to fulfill their function in the plot. 3 Harry Potter, for example, might appear as an almost real individual for many readers, but at the level of discourse he is simply the hero of an adventure story whose ‘life’ does not extend beyond the events narrated in the eponymous novels.

Things look different when we analyze characters as existents of the story. At that level, characters may be seen as individuals who inhabit an alternative world, the storyworld . It is a matter of some debate whether the existence of characters in the alternative world of the story should be regarded as complete or limited to text-based inferences. Here, we will assume that characters are endowed with at least a potentially complete existence in the storyworld. Of course, this existence is ultimately dependent on the narrator of the story (a figure of discourse). But within the confines of the storyworld created by narrative discourse, characters are generally agents endowed with an identity , social and personal relationships, feelings, desires and thoughts, just like any of us in our own lifeworld . Thus, Harry Potter might not be, nor could ever be, a real person in the world of readers. But, in the storyworld created by J. K. Rowling’s novels, he is a heroic and charismatic young wizard, with a multifaceted life, which includes the adventures narrated in the plots of the novels, but also, at least potentially, many other events, big and small, of which we may never hear.

Ben’s Bonus Bit – Fan Fiction, Fandoms, Negotiated Meaning, and Essay Writing

That fictional characters may carry on full lives within the storyworlds created by their authors takes on surprising cultural significance when we consider that a text’s ultimate meaning is negotiated between an author and readers . While authorial intent is a matter of great interest among readers and scholars who pour over an author’s notes, interviews, or drafts for clues about theme or message, fiction is written to be read. Because words are signs (representations) of thoughts or emotions and consist of symbols (letters), they must be translated from page to brain. Interpretive wiggle room exists in that grand game of telephone, as thousands or millions of readers bring their own biases, connotations, worldviews, and backgrounds to any text and then begin discussing, arguing, or appropriating texts for their own purposes.

Because authors cannot anticipate all potential readings (and may have no interest in appealing to wide audiences), readers are largely responsible for a text’s enduring value in society . Shakespeare’s plays, for example, could never be appreciated by modern audiences without real leg-work on the part of readers to interpret the author’s themes, helping his texts move beyond the limitations of the time period in which they were written (archaic language, lost stage directions, or female parts played by men, for example). Modern actors, directors, stage managers, and theater goers may claim real authorship over meaning in any production of Hamlet . Does stage direction of “Gertrude’s Closet” mean our scene occurs in a private room, in her bedroom, or on her bed(!) in Act 3, Scene 4? Is Lawrence Olivia wrong to suggest an Oedipal drive for Hamlet in his influential 1947 portrayal? Sigmund Freud was born nearly 300 years after Shakespeare. We must decide. What clues does the text give us?

When certain storyworlds are especially well constructed (or when they receive marketing support from large publishers) fandoms may arise with significant claim to ownership over the intellectual properties created by individual authors. In this way, Superman or Luke Skywalker or Harry Potter may take on cultural significance that outgrows the intentions, ambitions, or politics of their creators, spawning reams of fan-fiction, non-canonical sequels, and thriving, active fan communities on the Internet. If this happens during an author’s lifetime, struggle is possible between a writer and a fanbase over who controls ultimate meaning in an intellectual property. This is the case with Harry Potter, as JK Rowling’s comments on transgender rights caused significant controversy in June of 2020. For many fans, Potter’s story is about an excluded outsider who finds family and community in his new identity. This remains true despite the author’s individual politics, largely because readers play a vital role in making meaning for the text.

As students of fiction, it is therefore useful to consider meaning within a text as multiplicitous . There is indeed a matrix of plausible meanings within any single text. A specific reading is plausible if it is supported by the preponderance of textual evidence . Author’s biography or cultural studies may also contribute to plausibility, but for students of fiction, finding, reporting, and analyzing textual evidence is key to creating compelling, persuasive essays that advance the critical conversation about a piece. There may not be one, singular, authoritative reading that your essay must uncover, but positioning your argument within the matrix of plausible interpretations requires respect for the text as an authoritative source of information about meaning and a rigorous commitment to providing proof of your claims.

As existents in the storyworld, all characters have in principle the same importance. In J. K. Rowling’s fictional world, to continue with the same example, Harry is not more important than Hermione Granger or Neville Longbottom. But narrative discourse, by arranging events, environments, and characters into a plot, necessarily establishes distinctions amongst the characters, just as it does amongst the events and environments. Thus, Harry Potter becomes more relevant than all the other characters, taking on the role of the main character ( the protagonist or hero) in the story, while the rest appear as secondary characters . Some of these secondary characters, like Hermione or Ron, have a prominent role next to Harry, while many others, like Angelina Johnson or Bertha Jorkins, only appear fleetingly and play minor supporting roles in the plot. A few other characters from this storyworld, like Draco Malfoy or Dolores Umbridge, are cast as antagonists to Harry and his friends in the conflicts that drive the plot of the novels, even if, under a different arrangement of events and environments, they might have been cast in a different role. John Updike’s Gertrude and Claudius , for instance, places Hamlet’s villains in starring roles.

4.3 Individuation

Whether characters are central to the story or only play a secondary role, their characterization generally requires the narrator to directly or indirectly ascribe to them certain characteristics or properties that identify them as individuals. This is what we call individuation . In principle, a primary character will be more individuated than a secondary one. And we can expect the characters that are least relevant for the plot to be also the least individuated. But this rule has, in fact, notable exceptions. It is not uncommon to find secondary characters with characteristics so well defined that they become at least as individuated in the minds of the reader as the protagonist himself, if not more so. In Joseph Conrad’s Heart of Darkness , for example, the elusive ivory trader Kurtz is characterized with more detail and nuance than Marlow, the protagonist of the story.

In general, individuation involves three sets of defining characteristics or traits: 4

• Physical : these are the features of the body, such as whether the character is tall or short, slim or fat, blue-eyed or brown-eyed, fair or dark, male or female, etc. Many physical characteristics are external and can be observed with the naked eye (e.g. the shape of the nose or a scar on the forehead), while others might be internal and thus difficult to perceive directly (e.g. diabetes or heartburn)
• Mental : these are the features of personality or psychology, such as whether the character is modest or arrogant, upbeat or depressive, cruel or kind, dreamy or practical, etc. These traits compose what is commonly understood as the character of a person . They might include traits that are perceptual (e.g. powers of observation), emotive (e.g. excitability), volitional (e.g. ambition), and cognitive (e.g. shrewdness), and
• Behavioral : these are the features of habits, such as whether the character is punctual or habitually tardy, whether she shouts or whispers when speaking, laughs easily or never laughs at all, drinks or avoids alcohol, etc. Sometimes it is difficult to distinguish mental and behavioral traits, as they tend to be intimately connected. Behavioral traits may be related to any actions that characters undertake, including communicating and interacting with other characters.

The aim of individuation is to represent characters in such a way that they appear, speak, and act like real individuals. Just as we have verisimilitude in setting , authors often seek individuation in characterization. In the context of psychological or realistic fiction , a fully individuated character should be endowed with a particular set of physical, mental, and behavioral characteristics so as to allow readers to imagine him or her as a person living in the same kind of world in which we all live. William Faulkner in his 1949 Nobel Prize acceptance speech called good writing an exploration of “the human heart in conflict with itself,” which is an apt description for a fully individuated character. Of course, storyworlds might be quite different from our own lifeworld. It is possible, for example, to imagine a storyworld where individuality as we understand it does not exist and all ‘individuals’ are actually clones or genetic replicas of the same organism. In such a context, the notion of individuation would lose most of its sense. This kind of fiction, however, is notably difficult to create, precisely because individuality is such a central assumption in the worldviews of writers and readers.

As long as we stay within the boundaries of mimetic fiction –storyworlds that imitate or are extrapolated from our own world–it makes sense to strive for individuality in characterization. As social animals, we have evolved a set of perceptual and cognitive mechanisms that allow us to identify and distinguish other human individuals from each other. Given the importance of individuality for our own social existence, it is not surprising that our narratives attempt to represent characters as plausible and self-standing individuals, endowing them with a distinctive set of characteristics.

Not all human cultures, however, give the same importance to individuality. We should not forget that modern novels and short stories are largely the products of the individualistic culture that emerged from the European Renaissance (see Chapter 1 ), closely associated with the scientific and industrial revolutions, the expansion of capitalism, and a philosophical conception of the human being as an isolated, autonomous, and self-reflecting individual. In this culture, which has now become globalized, narrative characters that are not fully individuated seem to lack something important , as if not being properly distinguishable from other characters would make them less real. This has not always been the case. In mythical narratives , for example, the characters are not so much individuals as types (e.g. the ‘messenger’) or universals  (e.g. the ‘hero’). Both typical and universal characters are still important in modern fiction, although their nature and function has been somewhat modified by the prevailing individualism of modern culture.

Fig. 4.2 Fan art representing Lord Voldemort and Nagini, from the Harry Potter saga, made with charcoal, acrylics and watercolours. By Mademoiselle Ortie aka Elodie Tihange, CC BY 4.0, https:// fr.wikipedia.org/wiki/Fichier:Lord_ Voldemort.jpg

Typical characters (or simply, types) represent an aspect of humanity or a particular group of humans. For example, characters representing evil in a concentrated and simplified form, like Lord Voldemort (Fig. 4.2), have become common in young adult or popular fiction. While these ‘villains’ might be individuated to a certain extent, they are not so much individuals as types. Many other typical characters, like the ‘mad professor,’ the ‘femme fatale,’ or the ‘wise old man,’ can be found in modern short stories and novels, where they tend to play secondary or supporting roles as stock characters . When types become ingrained in the psychology and culture of a society and start appearing in many different storyworlds, they are said to be archetypes .

In some respects, every character, no matter how well individuated, can be regarded as a type. 5 Even in real life we often perceive other individuals as types (e.g. blue-collar worker, lawyer, businessman, nerd), a simplification that helps us to classify and group people into categories. This is the basis of prejudice and negative bias, but it is also an evolved mechanism to cope with complex social information. Similarly, characters in fiction, even those who have been individuated with care, cannot avoid being cast as types by readers. Emma Bovary, the eponymous protagonist of Gustave Flaubert’s novel Madame Bovary , is one character in realistic fiction who has been characterized with detail and subtlety (see Fig. 4.3). Yet she is often perceived as a typical adulterer, trying to balance the social imperatives of marriage with her romantic longings.

Fig. 4.3 ‘Madame Hessel en robe rouge lisant’ (1905), oil on cardboard. By Édouard Vuillard, Public Domain, https://commons.wikimedia.org/ wiki/File:%C3%89douard_Vuillard_-_ Madame_Hessel_en_robe_rouge_lisant_ (1905).jpg

Fig. 4.4 ‘Don Quixote and Sancho Panza at a crossroad,’ oil on canvas. By Wilhelm Marstrand (1810–1873), CC0 1.0, https://commons.wikimedia.org/wiki/ File:Wilhelm_Marstrand,_Don_Quixote_og_Sancho_Panza_ved_en_skillevej,_ uden_datering_(efter_1847),_0119NMK,_Nivaagaards_Malerisamling.jpg

There are times when fictional characters transcend their individuality and typicality to attain some form of universality. Universal characters represent a general aspect of humanity or the whole human species. For example, Don Quixote and Sancho Panza, the protagonists of Cervantes’ novel, have become a pair of universal characters, representing two fundamental and contrasting attitudes towards life that are generally found in human beings: idealism and materialism (Fig. 4.4). Similarly, in her desperate longing for a more fulfilling and authentic life, Emma Bovary may represent the alienation of all individuals in modern society, torn between reveries of plenitude and the unsatisfactory realities of everyday existence.

4.4 Kinds of Character

Typologies have been proposed to classify and distinguish the kinds of character most often found in fiction. Two of these typologies are still used extensively by critics and writers, even though their psychological assumptions only make them applicable to mimetic, realist fiction, that is, to storyworlds that attempt to imitate or replicate our own lifeworld. 6

The first of these typologies 7 distinguishes characters based on their degree of individuation :

• Flat characters : these characters, which are sometimes equated to what we have called types in the previous section, are constructed around a limited number of traits or characteristics. Of course, there are varying degrees of flatness. At one extreme, we would find characters with a single characteristic or trait, such as a messenger whose only purpose in the story is to deliver a message at a certain point of the plot. Flat characters can be individuated, but their identity, personality, and purpose can often be expressed by a single sentence. They lack depth or complexity and are easily recognizable and remembered by the reader. Because of their limited qualities, however, they also seem artificial, and most readers have a hard time identifying with them or taking them for real human beings. Minor or secondary characters in fiction tend to be flat , even when the main characters in the same story are not. In genres like comedy or adventure , flat characters are quite common. And some writers, like Charles Dickens or H. G. Wells, populate their novels and short stories with flat secondary characters. An example of a flat character from the Harry Potter series is Argus Filch, the caretaker of Hogwarts, characterized almost exclusively by his love for cats and obsession with catching students who break the rules of the school (Fig. 4.5).

Fig. 4.5 Warner Bros. Studio Tour, London: The Making of Harry Potter. Source: Karen Roe, CC BY 2.0, https:// commons.wikimedia.org/wiki/File:The_ Making_of_Harry_Potter_29-05-2012_ (7358054268).jpg

• Round characters : these characters are endowed with many traits or characteristics, some of which might even be contradictory and cause them internal or psychological conflicts . With well-crafted characterization, round characters can appear to be as complex and multifaceted as any human being we might encounter in our world. Major characters in realist prose fiction, such as Emma Bovary, Rodion Raskolnikov, or Anna Karenina, are often round. And there are writers, like Gustave Flaubert or Jane Austen, who tend to characterize even minor characters with such nuance and complexity that they appear to be round, even though they might not have a prominent role in the story. An example of a round character in the Harry Potter novels is Hermione Granger, one of Harry’s closest friends at Hogwarts. While roundness of character is the aim of many realistic and popular stories, in modernist and postmodernist fiction the notion of character has often been questioned. In Robert Musil’s novel The Man Without Qualities , for example, the main character is presented as devoid of any of those stable characteristics, individual or typical, which would allow him to fit comfortably into the preconceived patterns of modern bourgeois society (Fig. 4.6).

Fig. 4.6 ‘Man without Qualities n°2’ (2005), oil and metal on canvas. By Erik Pevernagie, CC BY-SA 4.0, https://commons.wikimedia. org/wiki/File:Man_without_ Qualities_n%C2%B02.jpg

Another typology, also based on a psychological-realist conception of character and often confused with the previous one, distinguishes characters in terms of their ability to change or evolve throughout the plot:

• Static characters : these characters do not experience any profound change or personal evolution from the moment they appear in the plot until they disappear. Most flat characters are also static, although these classifications are based on different variables. It is possible, although relatively unusual, to have a flat character whose limited characteristics undergo a radical transformation in the story. More common is to have round characters who are static, retaining the same personality, identity, or characteristics throughout the whole narrative. In the Harry Potter novels, for example, most major characters, including Harry, Ron, and Hermione, are fairly static, evolving only superficially from their initial appearance until the end of the series, even if the author has tried to add dynamism into their characterization to take account of their growing into adulthood.
• Dynamic characters : these characters undergo profound and significant changes as the story develops, showing some degree of personal evolution or growth which transforms them at the end of the plot. This evolution is not always positive or constructive, and the changes experienced by the character may involve different forms of crisis, physical or psychological degradation, depression, and other negative or destructive changes. Given their complexity and depth, round characters experience this kind of dynamism, although there are many cases where round characters remain static. In short stories, dynamic characters are far less common than in novels , where the length of the narrative provides more opportunities to show character development and evolution. To continue with examples from the Harry Potter novels, Neville Longbottom is one of the few characters who undergoes a significant evolution throughout the series, as he develops a more confident and bold personality.

4.5 Representing Characters

Like environments, characters in literary narrative need to be represented through words. They cannot be shown directly to the audience, as in film or drama. There are basically two ways to represent characters in prose fiction: 8

• Indirect characterization : The character is presented by the narrator, who describes his or her physical, mental, or behavioral characteristics. Character descriptions are similar to environmental descriptions. They can be long and detailed or short and cursory. And they often rely on significant details that connect characters to the setting, the plot, or even the reader. In certain cases, some details in a character description might be unnecessary or insignificant, but they can serve to make the character seem more realistic. In indirect characterization, the narrator also tends to use commentary to qualify or evaluate the character, providing a subjective interpretation that goes beyond mere description. Indirect characterization has the advantage of conveying a lot of information about characters in a short time . But, as a form of ‘ telling ’ (see Chapter 5 ), it creates some distance between the reader and the character, making the latter less moving and memorable than when direct methods of characterization are used.
• Direct characterization : The character is revealed through actions, words, looks, thoughts, or effects on other characters. Here, the narrator simply records external or internal events related to the character , including words and thoughts, without undertaking a descriptive summary or evaluation of the character’s traits. Direct characterization is, therefore, a form of ‘ showing ’ (see Chapter 5 ). It leaves the reader to interpret the character based on the information provided in the narrative. This kind of characterization is more vivid and effective than the indirect method. But it also asks more from readers , who are required to participate in the construction of characters through their interpretations.

Both forms of characterization are often used in short stories and novels. But direct characterization is generally preferred in modern works of fiction, as it does not require the mediation of an intrusive narrative voice and allows characters to appear more like real people. There are five methods of direct characterization that are commonly used in narrative: speech, thoughts, effects, actions, and looks. These can be easily remembered with the acronym STEAL .

• Speech : what characters say and how they say it is one of the most important components of direct characterization. Verbal language is the fundamental semiotic system that we humans employ to communicate meanings, emotions, intentions, and so on. When it involves an interaction with other people, we call this dialogue. In prose fiction, speech is a widely used method of characterization, as it can be effective in revealing explicit and implicit information about the characters engaged in dialogue. At the same time, speech can serve to move the plot forward and provide information about events, environments, or other characters in the storyworld
• Thoughts : knowing what the characters think (or desire, want, plan, etc.) can also help to define their characteristics. Of course, in our lifeworld, we have no access to what other people think, except when they tell us about it. This is the reason why some modern writers try to avoid this method of characterization, constraining the narrator to represent what can be observed from the outside, but never entering the characters’ minds. However, in many other works of fiction, both classical and modern, readers are allowed access to the thoughts of at least some of the characters, which are necessarily expressed in words, as some form of interior speech or monologue. Modern writers often use more sophisticated techniques like free indirect speech or interior monologue ( stream of consciousness ) to try to convey the complex and fluid mental processes of characters. 9 Molly Bloom’s interior monologue at the end of James Joyce’s Ulysses is perhaps the most famous example of this technique in modern literature
• Effects : how other characters are affected or react to a character can be used to characterize–not just those characters (using speech, thought, or action) but also the character that causes the effect. For example, if the characters surrounding John laugh every time he says or does something, readers will tend to assume that John is either funny or ridiculous. This is an effective characterization technique because it replicates how we judge character in our own life. As social animals, we are always attentive to the impression people make on other people. For instance, we would tend to see as attractive someone at whom others stare with desire or interest, even without knowing what that person looks like
• Actions : what characters do, their behavior, is perhaps the most important method of direct characterization. In general, actions often involve some kind of physical movement (e.g. gesturing, walking, running, etc.), but they can also be passive states (e.g. sleeping, sitting, etc.), or even internal changes reflected in the face or body of the character (e.g. staring, frowning, etc.). Nonverbal communication , which usually accompanies and supports dialogue, is based on actions. Since the characters in fiction are almost always doing something as part of the plot, every action is an opportunity to characterize them in one way or another in the mind of the reader. This is also how we judge each other in life, not only by our words, but also by our deeds, and
• Looks : how a character looks or appears in the story can also be a useful method of direct characterization. Appearance includes the physical traits of the character’s face and body (e.g. eye color, hair length, height, skin complexion, etc.) and their way of dressing or presenting themselves in front of others. Even a character’s choices in decorating a room or a desk may reflect character. In our lifeworld, appearance provides important cues about a person’s social status, occupation, mental and physical state, intentions and thoughts, etc. In prose fiction, looks are often employed to provide the same kind of information, typically through some form of description. In some cases, it can be difficult to distinguish between descriptions that use indirect characterization (presented from the subjective perspective of the narrator) and those that use direct characterization (without any subjective intervention by the narrator).

4.7 Summary

• At the level of discourse, characters are mere actants with no features other than those defined in the text and no reason for being other than their function in the plot. At the level of story, however, we can regard them as existents of the storyworld .
• In realistic, mimetic prose fiction , characterization aims to individuate characters by ascribing to them physical, mental, and behavioral characteristics or properties that distinguish them as individuals.
• Most characters can be classified according to their degree of individuation ( flat vs. round ) or their degree of personal evolution throughout the plot ( static vs. dynamic ).
• The representation of characters in short stories and novels is generally achieved through indirect or direct methods of characterization. Direct methods involve the representation of characters’ speech, thoughts, effects (on other characters), actions , or looks .
• In prose fiction, dialogue (the representation of speech interactions between characters) is usually an important element of the story, contributing both to emplotment and characterization.

Licenses and Attributions:

CC licensed content, Shared previously:

Ignasi Ribó, Prose Fiction: An Introduction to the Semiotics of Narrative. Cambridge, UK: Open Book Publishers, 2019.  https://doi.org/10.11647/OBP.0187

Version History: Created new verso art. Added quick links. Bolded keywords. Made minor phrasing edits for American audiences. Adopted MLA style for punctuation. Changed paragraphing for PressBooks adaptation. Moved footnotes to endnotes. Added reference to John Updike’s Gertrude and Claudius . Added vocabulary: formal elements of a story, verisimilitude, and young adult fiction. Included “Ben’s Bonus Bits,” with a host of new vocabulary. Included therein references to African American fiction and Rowling’s recent Blog controversy, October, 2021.

Linked bolded keywords to Glossary and improved Alt-image text for accessibility, July, 2022.

• see Paxton, James. “Personification’s Gender.” Rhetorica: A Journal of the History of Rhetoric Vol. 16, No. 2 (Spring 1998), pp. 149-179 (31 pages). While there are many examples of ideas embodied in masculine figures (Death, Father Time, Old Man River), in Western literature the balance is tipped firmly in favor of feminine personification. Some of the earliest literary examples (Anger as ‘Ira’ and Greed as ‘Avaritia’ in Prudentius’ medieval allegory Psychomachia ) still speak to sexist stereotypes today.
• Uri Margolin, ‘Character,’ in The Cambridge Companion to Narrative , ed. by David Herman (Cambridge, UK: Cambridge University Press, 2007), pp. 66–79, https://doi. org/10.1017/ccol0521856965
• Algirdas Julien Greimas and Joseph Courtés, Semiotics and Language: An Analytical Dictionary , trans. by Larry Crist and Daniel Patte (Bloomington, IN: Indiana University Press, 1982), pp. 5–8.
• Uri Margolin, ‘Individuals in Narrative Worlds: An Ontological Perspective,’ Poetics Today , 11:4 (1990), 843–71.
• H. Porter Abbott, The Cambridge Introduction to Narrative (Cambridge, UK: Cambridge University Press, 2008), pp. 129–31, https://doi.org/10.1017/cbo9780511816932
• See Seymour Benjamin Chatman, Story and Discourse: Narrative Structure in Fiction and Film (Ithaca, NY: Cornell University Press, 2000).
• E. M. Forster, Aspects of the Novel (San Diego, CA: Harcourt Brace Jovanovich, 1985).
• Janet Burroway, Writing Fiction: A Guide to Narrative Craft (Chicago, IL: University of Chicago Press, 2019), https://doi.org/10.7208/chicago/9780226616728.001.0001
• Dorrit Cohn, Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction (Princeton, NJ: Princeton University Press, 1988).
• Norman Page, Speech in the English Novel (London, UK: Macmillan, 1988).
• Mikhail M. Bakhtin, Problems of Dostoevsky’s Poetics , ed. by Caryl Emerson (Minneapolis, MN: University of Minnesota Press, 1984), p. 6.
• Mikhail M. Bakhtin, The Dialogic Imagination: Four Essays , trans. by Michael Holquist and Caryl Emerson (Austin, TX: University of Texas Press, 2011).

Abbott, H. Porter. The Cambridge Introduction to Narrative (Cambridge, UK: Cambridge University Press, 2008), https://doi.org/10.1017/cbo9780511816932

Bakhtin, Mikhail M. Problems of Dostoevsky’s Poetics , ed. by Caryl Emerson (Minneapolis, MN: University of Minnesota Press, 1984).

Bakhtin, Mikhail M. The Dialogic Imagination: Four Essays , trans. by Michael Holquist and Caryl Emerson (Austin, TX: University of Texas Press, 2011).

Burroway, Janet. Writing Fiction: A Guide to Narrative Craft (Chicago, IL: University of Chicago Press, 2019), https://doi.org/10.7208/chicago/9780226616728.001.0001

Chatman, Seymour Benjamin. Story and Discourse: Narrative Structure in Fiction and Film (Ithaca, NY: Cornell University Press, 2000).

Cohn, Dorrit. Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction (Princeton, NJ: Princeton University Press, 1988).

Forster, E. M. Aspects of the Novel (San Diego, CA: Harcourt Brace Jovanovich, 1985).

Greimas, Algirdas Julien, and Joseph Courtés. Semiotics and Language: An Analytical Dictionary , trans. by Larry Crist and Daniel Patte (Bloomington, IN: Indiana University Press, 1982).

Margolin, Uri. ‘Character’, in The Cambridge Companion to Narrative , ed. by David Herman (Cambridge, UK: Cambridge University Press, 2007), pp. 66–79, https://doi.org/10.1017/ccol0521856965

Margolin, Uri. ‘Individuals in Narrative Worlds: An Ontological Perspective,’ Poetics Today , 11:4 (1990), 843–71.

Page, Norman. Speech in the English Novel (London, UK: Macmillan, 1988).

Paxton, James. “Personification’s Gender.” Rhetorica: A Journal of the History of Rhetoric Vol. 16, No. 2 (Spring 1998), pp. 149-179.

An entity with agency in a storyworld.

The capacity to act in an environment.

A figure of speech that attributes personal or human characteristics to a nonhuman entity, object, or idea.

Anything that represents something else by virtue of an arbitrary association. In narrative, symbols are existents of the story that become arbitrarily associated with internal or external meanings.

Visually descriptive or figurative language. Despite its association with vision, auditory, tactile, or olfactory imagery also exists but may sometimes be labeled sensory detail.

A figure of speech that establishes a relationship of resemblance between two ideas or things by equating or replacing one with the other.

Related to dehumanization and an inverse to personification, the description of human characters as animals.

A classification of fiction based on moral content. Didactic fiction seeks to teach or enlighten readers.

The main character of a story, the one who struggles to achieve a goal.

A change of state occurring in the storyworld, including actions undertaken by characters and anything that happens to a character or its environment. Also called a “plot point.”

Everything that surrounds the characters in the storyworld.

Semiotic representation of a sequence of events, meaningfully connected by time and cause.

The means through which a narrative is communicated by the implied author to the implied reader.

A complete chronological sequence of interconnected events.

The ascription of mental, physical, or behavioral properties (characteristics) to a character.

A character that represents a particular aspect of humanity or a particular group of humans. (See also Type.)

A character that represents a general aspect of humanity or the whole human species.

Representation of verbal or speech interactions between characters, often accompanied by dialogue or speech tags.

The world of the story, which includes different types of existents (events, environments, and characters).

The figure of discourse that tells the story to a narratee.

The world experienced by writers and readers in their lives.

A character in the story that opposes the protagonist and struggles to frustrate his or her goals.

Features of narrative discourse that attempt to convince readers that the storyworld is a faithful imitation of the ‘real’ world.

The meaningful arrangement or representation of the environments in the story.

Narrative discourse that aims to construct a storyworld that is an accurate reflection of the lifeworld (i.e. the “real” world).

A classification for literature that attempts to mimic the real world. Fiction that seeks verisimilitude.

A character that represents a particular aspect of humanity or a particular group of humans. (See also Typical Character.)

A type that has become part of the psychology and culture of a society and appears in many different storyworlds.

The representation of a story through the mediation of a narrator, who gives an account and often interprets or comments on the events, environments, or characters of the storyworld.

The direct representation of the events, environments, and characters of a story without the intervention (or, in the case of narrative showing, with minimal or limited intervention) of a narrator.

The meaningful arrangement or presentation of the characters of the story.

Narrative indications that often accompany dialogue in prose fiction to provide information about the speakers, the quality and tone of speech, the environment, etc. (See also Speech Tags.)

The inclusion in narrative of a diversity of points of view and voices.

The arrangement of the events of the story into a plot.

Prose Fiction Copyright © by Miranda Rodak and Ben Storey is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License , except where otherwise noted.

Share This Book

Dialogue Definition

What is dialogue? Here’s a quick and simple definition:

Dialogue is the exchange of spoken words between two or more characters in a book, play, or other written work. In prose writing, lines of dialogue are typically identified by the use of quotation marks and a dialogue tag, such as "she said." In plays, lines of dialogue are preceded by the name of the person speaking. Here's a bit of dialogue from Alice's Adventures in Wonderland : "Oh, you can't help that,' said the Cat: 'we're all mad here. I'm mad. You're mad." "How do you know I'm mad?" said Alice. "You must be,' said the Cat, 'or you wouldn't have come here."

Some additional key details about dialogue:

• Dialogue is defined in contrast to monologue , when only one person is speaking.
• Dialogue is often critical for moving the plot of a story forward, and can be a great way of conveying key information about characters and the plot.
• Dialogue is also a specific and ancient genre of writing, which often takes the form of a philosophical investigation carried out by two people in conversation, as in the works of Plato. This entry, however, deals with dialogue as a narrative element, not as a genre.

How to Pronounce Dialogue

Here's how to pronounce dialogue: dye -uh-log

Dialogue in Depth

Dialogue is used in all forms of writing, from novels to news articles to plays—and even in some poetry. It's a useful tool for exposition (i.e., conveying the key details and background information of a story) as well as characterization (i.e., fleshing out characters to make them seem lifelike and unique).

Dialogue as an Expository Tool

Dialogue is often a crucial expository tool for writers—which is just another way of saying that dialogue can help convey important information to the reader about the characters or the plot without requiring the narrator to state the information directly. For instance:

• In a book with a first person narrator, the narrator might identify themselves outright (as in Kazuo Ishiguro's Never Let Me Go , which begins "My name is Kathy H. I am thirty-one years old, and I've been a carer now for over eleven years.").
• Tom Buchanan, who had been hovering restlessly about the room, stopped and rested his hand on my shoulder. "What you doing, Nick?”

The above example is just one scenario in which important information might be conveyed indirectly through dialogue, allowing writers to show rather than tell their readers the most important details of the plot.

Expository Dialogue in Plays and Films

Dialogue is an especially important tool for playwrights and screenwriters, because most plays and films rely primarily on a combination of visual storytelling and dialogue to introduce the world of the story and its characters. In plays especially, the most basic information (like time of day) often needs to be conveyed through dialogue, as in the following exchange from Romeo and Juliet :

BENVOLIO Good-morrow, cousin. ROMEO Is the day so young? BENVOLIO But new struck nine. ROMEO Ay me! sad hours seem long.

Here you can see that what in prose writing might have been conveyed with a simple introductory clause like "Early the next morning..." instead has to be conveyed through dialogue.

Dialogue as a Tool for Characterization

In all forms of writing, dialogue can help writers flesh out their characters to make them more lifelike, and give readers a stronger sense of who each character is and where they come from. This can be achieved using a combination of:

• Colloquialisms and slang: Colloquialism is the use of informal words or phrases in writing or speech. This can be used in dialogue to establish that a character is from a particular time, place, or class background. Similarly, slang can be used to associate a character with a particular social group or age group.
• The form the dialogue takes: for instance, multiple books have now been written in the form of text messages between characters—a form which immediately gives readers some hint as to the demographic of the characters in the "dialogue."
• The subject matter: This is the obvious one. What characters talk about can tell readers more about them than how the characters speak. What characters talk about reveals their fears and desires, their virtues and vices, their strengths and their flaws.

For example, in Pride and Prejudice, Jane Austen's narrator uses dialogue to introduce Mrs. and Mr. Bennet, their relationship, and their differing attitudes towards arranging marriages for their daughters:

"A single man of large fortune; four or five thousand a year. What a fine thing for our girls!” “How so? How can it affect them?” “My dear Mr. Bennet,” replied his wife, “how can you be so tiresome! You must know that I am thinking of his marrying one of them.” “Is that his design in settling here?” “Design! Nonsense, how can you talk so! But it is very likely that he may fall in love with one of them, and therefore you must visit him as soon as he comes.”

This conversation is an example of the use of dialogue as a tool of characterization , showing readers—without explaining it directly—that Mrs. Bennet is preoccupied with arranging marriages for her daughters, and that Mr. Bennet has a deadpan sense of humor and enjoys teasing his wife.

Recognizing Dialogue in Different Types of Writing

It's important to note that how a writer uses dialogue changes depending on the form in which they're writing, so it's useful to have a basic understanding of the form dialogue takes in prose writing (i.e., fiction and nonfiction) versus the form it takes in plays and screenplays—as well as the different functions it can serve in each. We'll cover that in greater depth in the sections that follow.

Dialogue in Prose

In prose writing, which includes fiction and nonfiction, there are certain grammatical and stylistic conventions governing the use of dialogue within a text. We won't cover all of them in detail here (we'll skip over the placement of commas and such), but here are some of the basic rules for organizing dialogue in prose:

• Punctuation : Generally speaking, lines of dialogue are encased in double quotation marks "such as this," but they may also be encased in single quotation marks, 'such as this.' However, single quotation marks are generally reserved for quotations within a quotation, e.g., "Even when I dared him he said 'No way,' so I dropped the subject."
• "Where did you go?" she asked .
• I said , "Leave me alone."
• "Answer my question," said Monica , "or I'm leaving."
• Line breaks : Lines of dialogue spoken by different speakers are generally separated by line breaks. This is helpful for determining who is speaking when dialogue tags have been omitted.

Of course, some writers ignore these conventions entirely, choosing instead to italicize lines of dialogue, for example, or not to use quotation marks, leaving lines of dialogue undifferentiated from other text except for the occasional use of a dialogue tag. Writers that use nonstandard ways of conveying dialogue, however, usually do so in a consistent way, so it's not hard to figure out when someone is speaking, even if it doesn't look like normal dialogue.

Indirect vs. Direct Dialogue

In prose, there are two main ways for writers to convey the content of a conversation between two characters: directly, and indirectly. Here's an overview of the difference between direct and indirect dialogue:

• This type of dialogue can often help lend credibility or verisimilitude to dialogue in a story narrated in the first-person, since it's unlikely that a real person would remember every line of dialogue that they had overheard or spoken.
• Direct Dialogue: This is what most people are referring to when they talk about dialogue. In contrast to indirect dialogue, direct dialogue is when two people are speaking and their words are in quotations.

Of these two types of dialogue, direct dialogue is the only one that counts as dialogue strictly speaking. Indirect dialogue, by contrast, is technically considered to be part of a story's narration.

A Note on Dialogue Tags and "Said Bookisms"

It is pretty common for writers to use verbs other than "said" and "asked"  to attribute a line of dialogue to a speaker in a text. For instance, it's perfectly acceptable for someone to write:

• Robert was beginning to get worried. "Hurry!" he shouted.
• "I am hurrying," Nick replied.

However, depending on how it's done, substituting different verbs for "said" can be quite distracting, since it shifts the reader's attention away from the dialogue and onto the dialogue tag itself. Here's an example where the use of  non-standard dialogue tags begins to feel a bit clumsy:

• Helen was thrilled. "Nice to meet you," she beamed .
• "Nice to meet you, too," Wendy chimed .

Dialogue tags that use verbs other than the standard set (which is generally thought to include "said," "asked," "replied," and "shouted") are known as "said bookisms," and are generally ill-advised. But these "bookisms" can be easily avoided by using adverbs or simple descriptions in conjunction with one of the more standard dialogue tags, as in:

• Helen was thrilled. "Nice to meet you," she said, beaming.
• "Nice to meet you, too," Wendy replied brightly.

In the earlier version, the irregular verbs (or "said bookisms") draw attention to themselves, distracting the reader from the dialogue. By comparison, this second version reads much more smoothly.

Dialogue in Plays

Dialogue in plays (and screenplays) is easy to identify because, aside from the stage directions, dialogue is the only thing a play is made of. Here's a quick rundown of the basic rules governing dialogue in plays:

• Names: Every line of dialogue is preceded by the name of the person speaking.
• Mama (outraged)  : What kind of way is that to talk about your brother?
• Line breaks: Each time someone new begins speaking, just as in prose, the new line of dialogue is separated from the previous one by a line break.

Rolling all that together, here's an example of what dialogue looks like in plays, from Edward Albee's Zoo Story:

JERRY: And what is that cross street there; that one, to the right? PETER: That? Oh, that's Seventy-fourth Street. JERRY: And the zoo is around Sixty-5th Street; so, I've been walking north. PETER: [anxious to get back to his reading] Yes; it would seem so. JERRY: Good old north. PETER: [lightly, by reflex] Ha, ha.

Dialogue Examples

The following examples are taken from all types of literature, from ancient philosophical texts to contemporary novels, showing that dialogue has always been an integral feature of many different types of writing.

Dialogue in Shakespeare's Othello

In this scene from Othello , the dialogue serves an expository purpose, as the messenger enters to deliver news about the unfolding military campaign by the Ottomites against the city of Rhodes.

First Officer Here is more news. Enter a Messenger Messenger The Ottomites, reverend and gracious, Steering with due course towards the isle of Rhodes, Have there injointed them with an after fleet. First Senator Ay, so I thought. How many, as you guess? Messenger Of thirty sail: and now they do restem Their backward course, bearing with frank appearance Their purposes toward Cyprus. Signior Montano, Your trusty and most valiant servitor, With his free duty recommends you thus, And prays you to believe him.

Dialogue in Madeleine L'Engel's A Wrinkle in Time

From the classic children's book  A Wrinkle in Time , here's a good example of dialogue that uses a description of a character's tone of voice instead of using unconventional verbiage to tag the line of dialogue. In other words, L'Engel doesn't follow Calvin's line of dialogue with a distracting tag like "Calvin barked." Rather, she simply states that his voice was unnaturally loud.

"I'm different, and I like being different." Calvin's voice was unnaturally loud. "Maybe I don't like being different," Meg said, "but I don't want to be like everybody else, either."

It's also worth noting that this dialogue helps characterize Calvin as a misfit who embraces his difference from others, and Meg as someone who is concerned with fitting in.

Dialogue in A Visit From the Good Squad

This passage from Jennifer Egan's A Visit From the Good Squad doesn't use dialogue tags at all. In this exchange between Alex and the unnamed woman, it's always clear who's speaking even though most of the lines of dialogue are not explicitly attributed to a speaker using tags like "he said."

Alex turns to the woman. “Where did this happen?” “In the ladies’ room. I think.” “Who else was there?” “No one.” “It was empty?” “There might have been someone, but I didn’t see her.” Alex swung around to Sasha. “You were just in the bathroom,” he said. “Did you see anyone?”

Elsewhere in the book, Egan peppers her dialogue with colloquialisms and slang to help with characterization . Here, the washed-up, alcoholic rock star Bosco says:

"I want interviews, features, you name it," Bosco went on. "Fill up my life with that shit. Let's document every fucking humiliation. This is reality, right? You don't look good anymore twenty years later, especially when you've had half your guts removed. Time's a goon, right? Isn't that the expression?"

In this passage, Bosco's speech is littered with colloquialisms, including profanity and his use of the word "guts" to describe his liver, establishing him as a character with a unique way of speaking.

Dialogue in Plato's Meno

The following passage is excerpted from a dialogue by Plato titled Meno.  This text is one of the more well-known Socratic dialogues. The two characters speaking are Socrates (abbreviated, "Soc.") and Meno (abbreviated, "Men."). They're exploring the subject of virtue together.

Soc. Now, if there be any sort-of good which is distinct from knowledge, virtue may be that good; but if knowledge embraces all good, then we shall be right in think in that virtue is knowledge? Men. True. Soc. And virtue makes us good? Men. Yes. Soc. And if we are good, then we are profitable; for all good things are profitable? Men. Yes. Soc. Then virtue is profitable? Men. That is the only inference.

Indirect Dialogue in Tim O'Brien's The Things They Carried

This passage from O'Brien's The Things They Carried exemplifies the use of indirect dialogue to summarize a conversation. Here, the third-person narrator tells how Kiowa recounts the death of a soldier named Ted Lavender. Notice how the summary of the dialogue is interwoven with the rest of the narrative.

They marched until dusk, then dug their holes, and that night Kiowa kept explaining how you had to be there, how fast it was, how the poor guy just dropped like so much concrete. Boom-down, he said. Like cement.

O'Brien takes liberties in his use of quotation marks and dialogue tags, making it difficult at times to distinguish between the voices of different speakers and the voice of the narrator. In the following passage, for instance, it's unclear who is the speaker of the final sentence:

The cheekbone was gone. Oh shit, Rat Kiley said, the guy's dead. The guy's dead, he kept saying, which seemed profound—the guy's dead. I mean really.

Why Do Writers Use Dialogue in Literature?

Most writers use dialogue simply because there is more than one character in their story, and dialogue is a major part of how the plot progresses and characters interact. But in addition to the fact that dialogue is virtually a necessary component of fiction, theater, and film, writers use dialogue in their work because:

• It aids in characterization , helping to flesh out the various characters and make them feel lifelike and individual.
• It is a useful tool of exposition , since it can help convey key information abut the world of the story and its characters.
• It moves the plot along. Whether it takes the form of an argument, an admission of love, or the delivery of an important piece of news, the information conveyed through dialogue is often essential not only to readers' understanding of what's going on, but to generating the action that furthers the story's plot line.

Other Helpful Dialogue Resources

• The Wikipedia Page on Dialogue: A bare-bones explanation of dialogue in writing, with one or two examples.
• The Dictionary Definition of Dialogue: A basic definition, with a bit on the etymology of the word (it comes from the Greek meaning "through discourse."
• Cinefix's video with their take on the 14 best dialogues of all time : A smart overview of what dialogue can accomplish in film.

• PDFs for all 136 Lit Terms we cover
• Downloads of 1924 LitCharts Lit Guides
• Teacher Editions for every Lit Guide
• Explanations and citation info for 40,556 quotes across 1924 books
• Downloadable (PDF) line-by-line translations of every Shakespeare play
• Characterization
• Colloquialism
• Anthropomorphism
• Tragic Hero
• Connotation
• Common Meter
• Static Character
• Pathetic Fallacy
• Anadiplosis
• Parallelism
• Figure of Speech
• Antimetabole

Directness in Speech and Writing

Glossary of Grammatical and Rhetorical Terms

• An Introduction to Punctuation
• Ph.D., Rhetoric and English, University of Georgia
• M.A., Modern English and American Literature, University of Leicester
• B.A., English, State University of New York

In speech and writing , directness is the quality of being straightforward and concise : stating a main point early and clearly without embellishments or digressions . Directness contrasts with circumlocution , verbosity , and indirectness .

There are different degrees of directness, which are determined in part by social and cultural conventions. In order to communicate  effectively with a particular audience , a speaker or writer needs to maintain a balance between directness and politeness . 

Examples and Observations

• "The whole world will tell you, if you care to ask, that your words should be simple & direct . Everybody likes the other fellow's prose plain . It has even been said that we should write as we speak. That is absurd. ... Most speaking is not plain or direct, but vague, clumsy, confused, and wordy. ... What is meant by the advice to write as we speak is to write as we might speak if we spoke extremely well. This means that good writing should not sound stuffy, pompous, highfalutin, totally unlike ourselves, but rather, well—'simple & direct.' "Now, the simple words in the language tend to be the short ones that we assume all speakers know; and if familiar, they are likely to be direct. I say 'tend to be' and 'likely' because there are exceptions. ... "Prefer the short word to the long; the concrete to the abstract; and the familiar to the unfamiliar. But: "Modify these guidelines in the light of the occasion, the full situation, which includes the likely audience for your words." (Jacques Barzun, Simple & Direct: A Rhetoric for Writers , 4th ed. Harper Perennial, 2001)
• Revising for Directness "Academic audiences value directness and intensity. They do not want to struggle through overly wordy phrases and jumbled sentences. ... Examine your draft . Focus specifically on the following issues: 1. Delete the obvious: Consider statements or passages that argue for or detail what you and your peers already assume. ... 2. Intensify the least obvious: Think about your essay as a declaration of new ideas. What is the most uncommon or fresh idea? Even if it's a description of the problem or a slightly different take on solving it, develop it further. Draw more attention to it." (John Mauk and John Metz,  The Composition of Everyday Life: A Guide to Writing , 5th ed. Cengage, 2015)
• Degrees of Directness "Statements may be strong and direct or they may be softer and less direct. For example, consider the range of sentences that might be used to direct a person to take out the garbage: Take out the garbage! Can you take out the garbage? Would you mind taking out the garbage? Let's take out the garbage. The garbage sure is piling up. Garbage day is tomorrow. "Each of these sentences may be used to accomplish the goal of getting the person to take out the garbage. However, the sentences show varying degrees of directness, ranging from the direct command at the top of the list to the indirect statement regarding the reason the activity needs to be undertaken at the bottom of the list. The sentences also differ in terms of relative politeness and situational appropriateness. ... "In matters of directness vs. indirectness, gender differences may play a more important role than factors such as ethnicity, social class, or region, although all these factors tend to intersect, often in quite complex ways, in the determination of the 'appropriate' degree of directness or indirectness for any given speech act ." (Walt Wolfram and Natalie Schilling-Estes, American English: Dialects and Variation . Wiley-Blackwell, 2006)
• Directness and Gender "While some of us will think that without the skills of 'good' writing a student cannot truly be empowered, we must be equally aware that the qualities of 'good' writing as they are advocated in textbooks and rhetoric books —  directness , assertiveness and persuasiveness , precision and vigor—collide with what social conventions dictate proper femininity to be. Even should a woman succeed at being a 'good' writer she will have to contend with either being considered too masculine because she does not speak 'like a Lady,' or, paradoxically, too feminine and hysterical because she is, after all, a woman. The belief that the qualities that make good writing are somehow 'neutral' conceals the fact their meaning and evaluation changes depending on whether the writer is a man or woman." (Elisabeth Daumer and Sandra Runzo, "Transforming the Composition Classroom."  Teaching Writing: Pedagogy, Gender, and Equity , ed. by Cynthia L. Caywood and Gillian R. Overing. State University of New York Press, 1987)
• Directness and Cultural Differences "The U.S. style of directness and forcefulness would be perceived as rude or unfair in, say, Japan, China, Malaysia, or Korea. A hard-sell letter to an Asian reader would be a sign of arrogance, and arrogance suggests inequality for the reader." (Philip C. Kolin, Successful Writing at Work . Cengage, 2009)

Pronunciation: de-REK-ness

• The Top 20 Figures of Speech
• Brevity in Speech and Writing
• 5 Tips on How to Write a Speech Essay
• Fluency in Language
• What Is Clarity in Composition?
• What Is Attribution in Writing?
• French Grammar: Direct and Indirect Speech
• How to Use Indirect Quotations in Writing for Complete Clarity
• Plain Style in Prose
• The Power of Indirectness in Speaking and Writing
• What Is Tone In Writing?
• Definition and Examples of Plain English
• Definition, Examples, and Observations on Writing
• Ellipsis: Definition and Examples in Grammar
• Pause (Speech and Writing)
• Figures of Speech: The Apostrophe as a Literary Device

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

The Influence of Direct and Indirect Speech on Mental Representations

* E-mail: [email protected]

Affiliation Department of Psychology, Open University, Heerlen, The Netherlands

Affiliation Department of Psychology, Erasmus University, Rotterdam, The Netherlands

• Anita Eerland, 
• Jan A. A. Engelen, 
• Rolf A. Zwaan

• Published: June 12, 2013
• https://doi.org/10.1371/journal.pone.0065480
• Reader Comments

Language can be viewed as a set of cues that modulate the comprehender’s thought processes. It is a very subtle instrument. For example, the literature suggests that people perceive direct speech (e.g., Joanne said: ‘I went out for dinner last night’ ) as more vivid and perceptually engaging than indirect speech (e.g., Joanne said that she went out for dinner last night ). But how is this alleged vividness evident in comprehenders’ mental representations? We sought to address this question in a series of experiments. Our results do not support the idea that, compared to indirect speech, direct speech enhances the accessibility of information from the communicative or the referential situation during comprehension. Neither do our results support the idea that the hypothesized more vivid experience of direct speech is caused by a switch from the visual to the auditory modality. However, our results do show that direct speech leads to a stronger mental representation of the exact wording of a sentence than does indirect speech. These results show that language has a more subtle influence on memory representations than was previously suggested.

Citation: Eerland A, Engelen JAA, Zwaan RA (2013) The Influence of Direct and Indirect Speech on Mental Representations. PLoS ONE 8(6): e65480. https://doi.org/10.1371/journal.pone.0065480

Editor: Krish Sathian, Emory University, United States of America

Received: February 1, 2013; Accepted: April 25, 2013; Published: June 12, 2013

Copyright: © 2013 Eerland et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The authors received no external funding for the studies described in the manuscript. The authors used research money from the Erasmus University Rotterdam to pay the participants. The university played no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Suppose you are reading a story that contains the sentence Joanne said: ‘I went out for dinner last night’ . Would it have made a difference if you had read the very similar Joanne said that she went out for dinner last night instead? Would it have made a difference, in other words, if the writer had used indirect speech rather than direct speech? The fact that the two different forms exist suggests that they serve different functions in linguistic communication. But what are these functions?

Indirect speech (e.g., Joanne said that she went out for dinner last night ) is thought to be description-like, whereas direct speech (e.g., Joanne said: ‘I went out for dinner last night’) is considered to be more depiction-like [1] . We might construe this to mean that indirect speech focuses on what is said (the gist of a particular message) whereas direct speech focuses on creating a mental representation of the described situation. In terms of the Van Dijk and Kintsch [2] levels of representation, direct speech focuses on the surface structure whereas indirect speech focuses on the situation model. This distinction might be responsible for the fact that people perceive direct speech as more vivid and perceptually engaging than indirect speech [3] , [4] , [5] . Little is known about the effects of direct and indirect speech on the nature of mental representations that are formed during reading but research on this topic is emerging [3] , [4] , [6] .

There is a great deal of evidence that people form mental representations of the described situation during language processing (e.g., [2] , [7] , [8] , [9] ). These representations are known as mental models or situation models. Subtly different linguistic constructions can have different effects on situation models. For example, various studies have examined the effects of grammatical aspect (e.g., [10] , [11] ) and negation [12] on the construction of situation models. What are the effects of using direct vs. indirect speech?

Recent studies are supportive of the idea that direct speech is more engaging than indirect speech. In one study, participants read short stories containing a direct or an indirect speech quotation. Context was manipulated so that either a fast or a slow speaking protagonist was implied. Reading times for direct speech were influenced by how fast the speaker spoke but reading times for indirect speech were not [5] . In an attempt to extend this finding, a recent study [6] explored whether not only speech rate but also the speed of the character’s movement influences reading times for direct and indirect speech quotations. People spent less time reading direct speech quotations when these utterances were described as being made quickly than as being made slowly. There was no effect of indirect speech quotations on reading times. There also was no effect for speed of movement on reading times. It thus seems that the use of direct speech causes the speaker’s voice to be more activated in the reader’s mind than the use of indirect speech. What we do not know is whether this more engaging experience influences our mental representations of described situations.

Given that direct speech is apparently perceived as more vivid than indirect speech, it seems likely that there are (subtle) differences in the mental representation of a given situation depending on whether this situation was described in direct or indirect speech. For example, objects that are present in the referential situation (i.e., the situation that is talked about) might be more accessible when they are talked about in direct speech than in indirect speech. This hypothesis is consistent with recent findings [5] , [6] that readers are more likely to engage in perceptual simulations of a situation related in direct speech than in indirect speech. On the other hand, if, as Clark and Gerrig [1] suggest, indirect speech is more descriptive than direct speech, then we might expect situational information to be more strongly represented in indirect than in direct speech. We investigated this idea in Experiment 1.

In all of the experiments described in this paper, we used the same participant-recruitment and participant-exclusion plan, which is very similar to that of Zwaan and Pecher [13] . Criteria were set after we conducted the first experiment. For every experiment, except for Experiment 1a, we recruited 200 participants online through Amazon’s Mechanical Turk ( http://www.mturk.com ). All experiments were presented online in the Qualtrics survey research suite ( http://www.qualtrics.com ). Because we were interested in running native speakers of English only, we excluded participants who indicated at the end of the experiment to be no native speaker of English. We also excluded data from participants with low accuracy scores (<.75 in Experiment 2, <.80 in all other experiments). As these exclusion procedures often left us with unequal number of participants per counterbalancing list, we excluded data from the last-run participants of the longer list to create equal-length lists. After each experiment we asked participants 1) to guess what the purpose of the study was, 2) in what kind of environment they performed the experiment (regarding the amount of distraction and level of noise; on a 9-point scale), 3) what type of monitor participants used to perform the task, and 4) some demographical questions (age, gender, level of education, native language).

For all experiments, response times <300 ms and >10000 ms were removed, as they indicate extremely fast or slow responses. The remaining data were analyzed. Because standard significance testing might lead to false positives in large samples [14] , [15] , [16] , we also calculated the posterior probability favoring the alternative hypothesis using the JZS Bayes Factor ( BF 01 , calculated with Rouder's web based application at http://pcl.missouri.edu/bayesfactor ), which provides the odds ratio for the null/alternative hypotheses given the data. A Bayes Factor of 1 means that they are equally likely, larger values (>3) indicate more evidence for the null hypothesis, and smaller values (<.33) indicate more evidence for the alternative hypothesis. Item analyses for Experiments 1–4 are reported in Appendix S1 .

Ethics Statement

The participants in all experiments were recruited online and voluntarily subscribed for participation in the described experiments. We did not obtain written consent. We did consult with the Ethics Committee of Psychology (ECP) at the Erasmus University Rotterdam, the Netherlands and receive a formal written waiver because the experiment was noninvasive and the results were analyzed anonymously.

Experiment 1a

In this first experiment, we investigated the accessibility of information regarding the referential situation that was either in direct or indirect speech. We used a probe recognition task to do so. Probe recognition tasks are commonly used to probe the strength of situational dimensions such as space [17] , time [18] , character goals [19] , [20] , and combinations thereof [21] . In a probe-recognition task, words are presented after a sentence. The participants’ task is to indicate as quickly as possible whether the word has occurred in the sentence they just read. Responses are usually very accurate but differences in response speed are thought to reflect differences in the strength with which situational information is active in the reader’s working memory [9] . For example, responses are faster when the probe word refers to an event that is still ongoing in the described situation than when the word refers to a past event [18] . Responses are also faster when the probe word refers to an object that is present in the described situation than when it refers to an absent object [12] .

If direct speech is indeed perceived as more vivid than indirect speech, one might hypothesize, based on the findings of Yao and colleagues [3] , that information that was presented in direct speech is more accessible than information presented in indirect speech. If, on the other hand, indirect speech is perceived as more descriptive than direct speech, then one might predict the opposite pattern. Our hypothesis was based on the idea that direct and indirect speech differ in terms of vividness and therefore we predicted that people should respond faster to probe words that were mentioned in direct speech than to probe words that were mentioned in indirect speech.

Participants.

One hundred and eighty participants were recruited online of which 179 completed the experiment. The sample had a mean age of 34 (range = 18–75, 108 females). All participants were residents of the USA and received $1 for their participation, which required approximately 26 minutes. Ten participants did report another language than English as their native language. With the exclusion of these participants, our sample included 169 native speakers of English.

Materials and procedure.

Participants first performed a lexical decision task in which they were randomly presented with eight strings of letters (one at a time). They had to indicate as fast as possible whether a given string was a word (m-key) or not (c-key). Four words and four non-words were included in this task. The lexical decision task was added to the actual experiment to familiarize participants with the task of making speeded responses to visual stimuli.

Next, participants read 48 short stories (24 experimental, 24 filler; adapted from [3] (see Appendix S2 ) online, sentence by sentence. Each story consisted of three sentences with the last sentence always being a direct or indirect speech quotation (see example story below). Two versions were created that differed regarding the last sentence of the experimental stories. Whenever the last sentence contained a direct speech quotation in one version, the sentence contained an indirect speech quotation in the other version. In both versions, half of the quotations were direct. This was true for both experimental and filler stories. All stories were presented in a random order.

Example Story ( probe )

It was 5.30 p.m. and everybody was ready to leave the office.

At one desk, Elaine was having a quick chat with Steven about her work.

Direct: She said: “The amount of paperwork is killing me at the moment. I feel completely exhausted.”

Indirect: She said that the amount of paperwork was killing her at the moment, and that she felt completely exhausted.

Participants performed a probe-recognition task directly after each story to test the accessibility of text information regarding the referential situation. Crucially, for the experimental stories the probe was always a noun that was mentioned in direct or indirect speech, so each experimental story required a ‘yes’ response. The probes that followed the filler stories were also nouns but they were never mentioned previously and thus required a ‘no’ response. All probes we used were never mentioned in one of the other stories (in case of the experimental stories, probes were only mentioned once). Response times to the probes were measured. To make sure participants read all stories properly, comprehension questions followed after 25% of the stories. The right answer to these questions was ‘yes’ 50% of the time. Three practice trials were included before the actual experiment started.

Each trial started with the first sentence of a story. Participants pressed the space bar when they had read a sentence to make the next sentence appear. Whenever participants pressed the space bar after the third sentence of the story, a fixation cross appeared in the middle of the screen for 1000 ms, followed immediately by the probe. Participants had to indicate as fast as possible whether this probe was mentioned in the story they just read (m-key) or not (c-key).

We excluded data from participants with an accuracy <80% on the probes (eight participants) and data from one participant due to problems with the recording of response times. Finally, we excluded data from six last-run participants on one of the lists to make both lists equal regarding the number of participants. Data from the remaining 154 participants were analyzed. Unfortunately, there was a counterbalancing error involving one of the stories (it appeared in the same condition twice), so we excluded this item for all participants.

Mean response times to the probes are displayed in Table 1 . In contrast to what we expected, there was no effect of speech, t (153) = 1.45, p  = .15, BF 01  = 5.55. Accuracy levels were high (.96 for direct and.95 for indirect speech) and did not differ between conditions, | t | <1.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0065480.t001

Because we did not determine all exclusion criteria before collecting the data, this study must be considered exploratory in nature. In Experiment 1b we tried to replicate our findings using the exact same settings as in Experiment 1a. Therefore, the study described as Experiment 1b is confirmatory rather than exploratory [22] . We followed this procedure for all experiments (see also [13] ).

Experiment 1b

Given that many psychology studies are underpowered [23] we started this experiment by conducting a power analysis with the program G*Power [24] to estimate the sample size needed to detect an effect of speech on accessibility of text information regarding the referential situation. According to this power analysis, at least 174 participants were needed to obtain statistical power at the recommended.80 level [25] (An anonymous reviewer suggested to use an ANOVA with repeated measures to estimate the sample size rather than a t -test. The F-test takes into account the real correlation between measures (.73, based upon the results of Experiment 1a) rather than an estimated correlation (.5). According to this alternative power analysis, at least 154 participants were needed to obtain statistical power at.80 level. The number of measures per condition was 12. The effect size was.10 based on the results of Experiment 1a). Because we anticipated that the sample would include non-native speakers of English, we recruited 216 participants online of which 209 completed the experiment. The sample had a mean age of 34 (range = 18–70, 117 females). All participants were residents of the USA and received $1 for their participation, which required approximately 28 minutes. We excluded the data from six participants because they reported another language than English as their native language. With the exclusion of these participants, our sample included 203 native speakers of English.

Except for the fact that we repaired the counterbalancing error of one of the experimental stories, the materials and procedure for this experiment were exactly the same as in Experiment 1a.

Results and Discussion.

We excluded data from participants that had accuracy scores <.80 (11 participants). Data from one participant were excluded because he or she also participated in Experiment 1a and data from seven participants were removed to equalize both lists regarding the number of participants. Data from the remaining 184 participants were analyzed.

Mean response times to the probes are displayed in Table 1 . Again, we found no effect of speech, t (183) = 0.09, p  = .92, BF 01  = 17.02. Accuracy levels were somewhat lower than in Experiment 1a (.95 for direct and.94 for indirect speech) and there was a significant difference between conditions, t (183) = 2.18, p  = .03, BF 01  = 1.66. So people responded slightly less accurately to probe words mentioned in indirect than in direct speech.

The results of Experiments 1a and 1b are similar and show reliable effects. In both studies, text information regarding a referential situation is not more accessible when this information was presented in indirect as compared to direct speech. Although this is not what we expected, Bayesian analyses indicated that the combined data of both experiments provide strong evidence for this null effect, BF 01  = 12.79.

Our results do not support our hypothesis that information regarding the referential situation is more accessible when this information was mentioned in direct as compared to indirect speech. But perhaps direct speech does not focus attention on the referential situation but rather on the communicative situation itself (i.e., the situation in which a conversation takes place). Evidence for this idea comes from a recent study by Stites and colleagues [6] . They found that people tend to read direct speech quotations faster whenever the talker speaks quickly (i.e., when someone was in a hurry) compared to when he was talking slowly. Conversely, they found no effect of talking speed on reading times on indirect speech quotations. It thus seems that the manner of speaking is more important in direct than in indirect speech. If this is true, then information about the manner of speaking should be more available after direct than after indirect speech. We investigated this idea in Experiment 2.

Experiment 2a

Two hundred participants were recruited online of which 188 completed the experiment. The sample had a mean age of 32 (range = 18–66, 116 females). All participants were residents of the USA and received $0.75 for their participation, which required approximately 20 minutes. Eight participants reported another language than English as their native language. With the exclusion of these participants, our sample included 180 native speakers of English.

Participants first performed a lexical decision task (see Experiment 1a). Next, they read 48 sentences online (24 experimental sentences that were adapted from [6] ; 24 fillers that we created ourselves; see Appendix S3 ). Each sentence consisted of a direct or an indirect speech quotation. Critically, an adverb was included in all sentences to provide information about the way of speaking. In the study by Stites and colleagues [6] only speed of speaking was manipulated. We decided to also use other kinds of adverbs (e.g., repeatedly, rudely, respectfully) so that testing the communicative situation was not limited to talking speed. We created two versions of the experiment that differed regarding the quotation in the sentence. Whenever the quotation was in direct speech in one version, it was in indirect speech in the other version. In both versions, half of the quotations were direct, whereas the speech quotation was indirect for the other half of the sentences. This was true for both experimental and filler items. All sentences were presented in a random order.

Participants performed a probe-recognition task directly after each sentence to test the accessibility of text information regarding the communicative situation. This time, for the experimental sentences the probe was always an adverb related to the way of speaking of the agent. As in our previous experiments, each experimental sentence required a ‘yes’ response. The probes that followed the filler stories were also adverbs but were never mentioned previously and thus required a ‘no’ response. Response times to the probes were measured. To make sure participants read all stories properly, comprehension questions followed after 25% of the stories. The right answer to these questions was ‘yes’ 50% of the time. Five practice trials were included before the actual experiment started.

Each trial started with the appearance of a sentence. Participants pressed the space bar whenever they had read a sentence to make the next one appear. After the third sentence, a fixation cross appeared in the middle of the screen for 1000 ms, followed immediately by the probe. Participants had to indicate as fast as possible whether this probe was mentioned in the sentence they had just read (m-key) or not (c-key).

We excluded data from two participants for whom timing data somehow were not recorded and from participants with accuracy scores <75% (eight participants). The removal of these ten participants yielded unequal numbers of participants across lists. Data from the last-run participants of the longest list were removed so that both list were equal regarding the number of participants. Our analysis included data from the remaining 168 participants.

Mean response times to the probes are displayed in Table 1 . We found a small but reliable effect. As in Experiments 1a and 1b, people were faster to respond to probes after reading an indirect than a direct speech quotation, t (167) = 3.51, p  = .0006, BF 01  = .03. Accuracy levels were high (.93 for direct and.94 for indirect speech) and did not differ between conditions, | t |<1.

Experiment 2b

Two hundred participants were recruited online and all completed the experiment. The sample had a mean age of 34 (range = 19–69, 115 females). All participants were residents of the USA and received $0.75 for their participation, which required approximately 20 minutes. Ten participants did not report English as their native language. With the exclusion of these participants, our sample included 190 native speakers of English.

The materials and procedure for this experiment were exactly the same as in Experiment 2a.

We excluded data from participants that had accuracy scores <.75 (nine participants). Data from five participants were removed to equalize both lists with respect to the number of participants. The remaining data (176 participants) were analyzed.

Mean response times to the probes are displayed in Table 1 . We found a small effect showing that people respond faster to probes regarding the communicative situation after indirect than after direct speech quotations, t (175) = 2.20, p  = .03, BF 01  = 1.56. Bayesian analysis shows that the evidence in favor of the alternative hypothesis must be considered ambiguous. Accuracy levels were high (.91 for direct and.93 for indirect speech) and differed between conditions, t (175) = 3.09, p =  .002, BF 01  = .16.

The results of Experiment 2a show that text information regarding a communicative situation is more accessible when this information was presented in indirect as compared to direct speech. The results of Experiment 2b are ambiguous concerning the influence of speech on accessibility of information in respect of the communicative situation. However, Bayesian analysis indicates that the combined data of both experiments provide strong evidence for the conclusion that information regarding a communicative situation is more accessible after indirect than direct speech, BF 01  = .01. The alternative hypothesis (faster responses after indirect than direct speech) is a hundred times more likely, based on these data, than the null hypothesis.

We expected direct speech to make readers focus more on the communicative situation (i.e., the way of speaking) as opposed to the referential situation (i.e., the content of the speech) than indirect speech. However, our results do not support this hypothesis. In fact, we found that text information regarding the communicative situation is more accessible in indirect than in direct speech.

Experiment 3a

In this experiment we wanted to examine an alternative explanation for our finding that text information regarding the communicative situation is more accessible in indirect than in direct speech. Perhaps direct speech is so engaging that it is more difficult to switch from the comprehension task to the probe recognition task after direct than indirect speech. To test this idea, we added a sentence that did not convey speech to each of our stimulus texts, such that the probe word was not presented immediately after the direct/indirect speech manipulation but after an intervening sentence.

If the probe-response advantage of indirect of over direct speech persists, then we can rule out that this is due to a larger task-switching effect in the direct speech condition. Moreover, by measuring reading times on the added sentence, we could examine whether switching from direct speech to non-speech incurs processing costs. If this is not the case, then this would provide supportive evidence that the probe-response disadvantage for direct speech found in Experiment 2 is not due to task switching.

Two hundred participants were recruited online of which 185 completed the experiment. The sample had a mean age of 34 (range = 18–69, 117 females). All participants were residents of the USA and received $1 for their participation, which required approximately 25 minutes. There were seven participants that did not report English as their native language. With the exclusion of these participants, our sample included 178 native speakers of English.

In this experiment, we added a last sentence to the sentences that were used in Experiment 2 after which the probe appeared. This sentence never contained direct or indirect speech (see Appendix S3 ). The procedure was the same as in Experiment 2, only this time we were also interested in reading times for all last sentences.

Because removal of the data from nonnative speakers of English yielded unequal numbers of participants across lists, we removed the data from six last-run participants of the longest list. Data from the remaining 172 participants were analyzed.

Mean reading times for the last sentences and mean response times to the probes are displayed in Table 1 . We found no effect of speech (direct vs. indirect) on reading times, t (171) = 1.83, p  = .069, BF 01  = 3.18. There was also no effect of speech on response times to the probes, t (171) = 0.10, p  = .92, BF 01  = 16.45. Accuracy levels were lower than in all previous experiments (.79 for direct and.85 for indirect speech). Importantly, however, we found a significant difference between conditions regarding accuracy scores, t (171) = 5.13, p<0.000001, BF 01  = .00009.

Experiment 3b

Two hundred participants were recruited online of which 183 completed the experiment. The sample had a mean age of 33 (range = 18–66, 112 females). All participants were residents of the USA and received $1 for their participation, which required approximately 26 minutes. There were four participants that did not report English as their native language. With the exclusion of these participants, our sample included 179 native speakers of English.

The materials and procedure for this experiment were exactly the same as in Experiment 3a.

Because removal of the data from participants who were no native speaker of English yielded unequal numbers of participants across lists, we removed the data from five last-run participants of the longest list. Data from the remaining 174 participants were analyzed.

Mean reading times for the last sentences and mean response times to the probes are displayed in Table 1 . As is Experiment 3a, we found no effect of speech (direct vs. indirect) on reading times, t (173) = 0.84, p  = .401, BF 01  = 11.71. Also, the analysis regarding the response times to the probes yielded the same results as in Experiment 3a. There was no effect of speech on response times, t (173) = 1.73, p  = .086, BF 01  = 3.80. Accuracy levels were comparable to those in Experiment 3a (.82 for direct and.86 for indirect speech) and differed again between conditions, t (173) = 3.96, p  = .0001, BF 01  = .01.

In both experiments we found no effect of speech on reading times or response times. Moreover, the Bayesian analysis of the combined data provided very strong evidence for the null hypothesis regarding reading times ( BF 01 s  = 14) and response times to the probes ( BF 01 s  = 8). If it were more difficult to switch to a situation with no speech (e.g., the probe recognition task or a sentence that does not contain any speech) from direct speech than from indirect speech, one would expect differences in reading times for the last sentences. Given that we did not find such a difference, it seems unlikely that the results of Experiments 1 and 2 can be explained by more difficulty in switching to the probe recognition task after direct than after indirect speech.

Accuracy levels in Experiments 3a and 3b were lower than in our previous experiments. This finding can be explained by the fact that participants read another sentence before responding to the probe. In our previous experiments the probe immediately followed the sentence in which the probe was mentioned. This lower accuracy level may also explain why we did not find effects on probe-response times. There were fewer correct responses that could be entered into the analysis and participants may have emphasized accuracy over speed. This is why it is important that we found significant differences in probe accuracy between conditions. Participants were more accurate in responding to probes in the indirect than in the direct speech condition. Bayesian analysis of the combined data shows that the evidence is very strong for this conclusion ( BF 01 <.001). This is in line with the results from Experiment 2, which suggest that indirect speech favors the communicative situation.

Experiment 4a

So far, we have found no advantage (in terms of the accessibility of information during language processing) for direct speech over to indirect speech. It is possible that the ‘more vivid’ experience of direct speech does not necessarily influence information processing but prompts a switch from the visual modality (reading) to the auditory modality. In one recent study [3] , participants read some short stories including a direct or indirect speech quotation while their brain activity was recorded. Participants showed more brain activity in the auditory cortex while reading direct as compared to indirect speech. This is consistent with the idea that silent readers are more likely to mentally simulate a character’s voice while reading to direct speech. Thus, if voice areas are more activated while reading direct as compared to indirect speech, then people should be primed to respond faster to auditory stimuli after reading direct speech than indirect speech. This idea is consistent with the modality switching effect (e.g., [26] , [27] , [28] ). It also explains why direct-speech responses to visual probes were slower than expected in our previous experiments; participants had mentally shifted away from the visual modality.

To test this idea, we presented participants with spoken probe words rather than written ones (as in Experiments 1–3). Because of the just-described neuroimaging findings [3] , we expected participants to respond faster to the probe after direct than after indirect speech because reading direct speech activates the auditory cortex more strongly than indirect speech.

Two hundred participants were recruited online of which 193 completed the experiment. The sample had a mean age of 35 (range = 18–67, 125 females). All participants were residents of the USA and received $1 for their participation, which required approximately 28 minutes. There were six participants that did not report English as their native language. With the exclusion of these participants, our sample included 187 native speakers of English.

Instead of the lexical decision task, we had participants perform a categorization task first. They were auditorily presented with four fruits ( grape , lemon , strawberry, mango) and four animals (horse, tiger , turtle , rabbit; words in Italic were pronounced by a male). Participants had to decide as fast as possible whether the word they heard was a fruit (m-key) or an animal (c-key). Words were presented in random order. We included this task to familiarize participants with the task of making speeded responses to auditory stimuli. They were also instructed to use this task to set the volume of their computer to the right level.

Next, participants read the same 48 three-sentence stories that we used in Experiment 1 and performed a probe recognition task. However, this time, the probes were presented auditorily instead of visually. The pronounced words were collected from http://www.merriam-webster.com/ . Some stories were slightly changed to make sure that whenever the probe was pronounced by a male it was also the case that a male spoke in the story (and not a female). We did so because we know that people encode features of speakers’ utterances, like gender [29] , and we wanted to prevent mismatch effects. After each last sentence of a story, a fixation cross appeared on the screen for 1000 ms. Then participants heard an auditory probe and indicated as fast as possible whether the word they heard was present in the story they just read (m-key) or not (c-key).

To make sure participants read all stories properly, comprehension questions followed after 50% of the stories. The right answer to these questions was ‘yes’ 50% of the time. Three practice trials were included before the actual experiment started.

We excluded data from participants with an accuracy <80% (38 participants). Furthermore, we excluded data from nine last-run participants on one of the lists to make both lists equal regarding the number of participants. The remaining data (140 participants) were analyzed.

Mean response times to the probes are displayed in Table 1 . Although we expected people to respond faster to an auditory probe after reading direct as compared to indirect speech, we found no effect of speech on response times to the probes, t (139) = 1.08, p  = .28, BF 01  = 8.56. Accuracy levels were high (.95 for direct and.94 for indirect speech) and did not differ between conditions, | t| <1.

Experiment 4b

Two hundred participants were recruited online of which 189 completed the experiment. The sample had a mean age of 32 (range = 18–65, 116 females). All participants were residents of the USA and received $1 for their participation, which required approximately 30 minutes. There were eight participants that did not report English as their native language. With the exclusion of these participants, our sample included 181 native speakers of English.

The materials and procedure for this experiment were exactly the same as in Experiment 4a.

We excluded data from participants with an accuracy <80% (31 participants) and from six last-run participants on one of the lists to make both lists equal regarding the number of participants. The remaining data (144 participants) were analyzed.

Mean response times to the probes are displayed in Table 1 . We found a very small effect. Although we expected people to respond faster to an auditory probe after reading direct as compared to indirect speech, we found an effect of speech on response times to the probes that was the opposite of this, t (143) = 2.28, p  = .02, BF 01  = 1.2. Accuracy levels were high (.92 and.93) and did not differ between conditions, | t| <1.

The results of Experiment 4a show that there was no effect of speech on response times to auditorily presented probes, while the results of Experiment 4b show a very small effect favoring indirect speech. We thus ended up with mixed effects. Moreover, Bayesian analysis of the combined data provides no clear evidence for the null or the alternative hypothesis regarding response times ( BF 01 s  = 1.63).

The results of Experiment 4 do not support the idea that the more vivid experience of direct speech is caused by a switch from the visual to the auditory modality. We also tested the idea of auditory priming by direct speech in four other experiments (two exploratory and two confirmatory ones). In the first of these experiments, participants read the same 48 stories that we used in Experiments 1 and 4. However, after each last sentence, participants were presented with either a high (650 Hz) or a low (450 Hz) tone. They were instructed to decide as fast as possible whether the tone they heard was either high (650 Hz, always presented after the experimental items) or low (450 Hz, always presented after fillers). In another study, we replaced the tones by the spoken words ‘right’ and ‘left’. Participants decided as fast as possible whether the word they heard was either ‘right’ (experimental items) or ‘left’ (fillers). In none of these experiments we found an effect of speech on response times.

Our findings do not seem to be consistent with the literature. However, an experiment by Kurby, Magliano, and Rapp [30] on auditory imagery experiences (AIEs) during silent reading of direct speech yielded results similar to ours. In this study, participants first listened to dialogues between two characters. Then they read several texts, some of which they heard before, while others were new. While participants read those texts they performed a probe recognition task. Probes were auditorily presented and were either in the voice of the character that originally pronounced that word (match condition) or in the voice of the other character (mismatch condition). Participants were faster in the match than in the mismatch condition but this was only true for familiar scripts. In other words, people only had AIEs during silent reading of direct speech when they had previously experienced the same voice in the same situation. In our experiment, participants had prior experience with the voices that pronounced the probe words but not with the particular context in which they appeared. The fact that we did not find a priming effect of direct speech on auditory probes is therefore consistent with the results of Kurby and colleagues [30] .

So far, we have found no evidence that direct speech enhances the availability of information about the referential and communicative situation relative to indirect speech. If anything, we have found (some) evidence to the contrary. However, so far we have only tested mental representations at the level of situation models (whether these are models of the referential or the communicative situation). It might be the case that the influence of direct speech takes place at another level of mental representation. According to van Dijk and Kintsch’s [2] classic model, linguistic input is represented at three levels: the surface structure (a representation of the exact wording of an utterance), the textbase (a representation of the explicitly stated meaning of an utterance), and the situation model (a representation of the referential situation). It is plausible that direct speech influences mental representations at the level of the surface structure. As we mentioned earlier, direct speech is thought to focus more on the exact words, whereas the gist of a particular message is the focus of indirect speech [1] . A recent study has reported initial evidence for this idea [31] . Participants were presented with a text. Then the text appeared again and participants were instructed to report any difference between the two texts. Speech was manipulated (direct vs. indirect) but also word-change. There could be no change at all between the two texts, there could be a semantically related word-change (flatmate – roommate), or a distantly related word-change (flatmate – brother). Change detection was significantly better in direct than in indirect speech. The authors therefore conclude that the exact wording of what was said by a story protagonist is critical for direct but not for indirect speech.

Based on these results, we expected people to focus more on the exact words in direct speech than in indirect speech. In Experiment 5 we tested this idea.

Experiment 5a

Initially, we recruited 200 participants, but because of a large number of non-native speakers in two of our four lists, we decided to run a few more participants in these lists. In total, 214 participants were recruited online and all completed the experiment. The sample had a mean age of 34 (range = 15–66, 116 females). All participants were residents of the USA and received $0.5 for their participation, which required approximately 18 minutes. There were 15 participants that did not report English as their native language and one participant reported to be 15 years of age. With the exclusion of these participants, our sample included 198 adults who were native speakers of English.

Participants read all 24 experimental stories, sentence by sentence, that we used in Experiment 1. After each last sentence, a fixation cross appeared on the screen for 1000 ms. Then a sentence appeared and participants indicated whether this sentence was exactly the same as one of the sentences of the story they just read (m-key) or not (c-key). For half of the stories, the sentence that appeared after the fixation cross was exactly the same as the last sentence of the story (which was always a sentence in direct or indirect speech). For the other 12 stories, the sentence that appeared after the fixation cross was a paraphrase of the last sentence of the story (see example story below). We created four lists, so that we could manipulate speech (direct vs. indirect) and referential sentence (literally vs. paraphrase) within stories.

Example Story ( Paraphrase )

Direct: She said: “The amount of paperwork is killing me at the moment. I feel completely/ totally exhausted.”

Indirect: She said that the amount of paperwork was killing her at the moment, and that she felt completely/ totally exhausted.

To make sure participants understood that we were looking for subtle differences between sentences, we presented them with three practice trials. They received feedback on their responses during these trials.

We excluded data from ten last-run participants on three of our four lists to make all lists equal regarding the number of participants. Data from the remaining 188 participants were analyzed.

We computed d’ scores [32] . To be able to use d’ , we converted scores of 1 and 0 to.99 and.01 respectively [33] . ‘Yes’ responses to literal statements were considered hits, whereas ‘yes’ responses to paraphrases were counted as false alarms. Mean d’ scores by condition are displayed in Table 1 . The results show a medium effect of speech on the ability to detect subtle changes in surface structure even though the textbase and situation model of the message remained the same, t (167) = 2.76, p =  .006, BF 01  = .42. Participants were better at remembering the exact words that were used in direct than in indirect speech.

This difference cannot be explained by bias. We found no difference between conditions (direct vs. indirect speech) regarding the tendency to respond ‘yes’, | t| <1 (see C-scores in Table 1 ).

Experiment 5b

Two hundred and one participants were recruited online (i.e., most likely due to technical issues, we ended up with data from 51 participants on one of the list) of which 200 completed the experiment. The sample had a mean age of 33 (range = 18–69, 124 females). All participants were residents of the USA and received $0.5 for their participation, which required approximately 18 minutes. There were eight participants that did not report English as their native language. With the exclusion of these participants, our sample included 192 native speakers of English.

The materials and procedure for this experiment were exactly the same as in Experiment 5a.

We removed data from four last-run participants to equal all four lists regarding the number of participants. The remaining data (188 participants) were analyzed.

Again, we computed d’ scores (see Table 1 ) and we found a significant effect of speech on the ability to detect subtle changes in texts even though the gist of a message remained the same, t (187) = 3.14, p  = .002, BF 01  = .15. So, also in this confirmatory experiment, participants remembered the exact words that were used better after direct than indirect speech. This effect is due to differences in sensitivity because we found no differences with respect to the tendency to respond ‘yes’ between conditions, | t| <1 (see C-scores in Table 1 ).

Although the effect we found was stronger for Experiment 5b than for Experiment 5a (due to smaller SDs in the last experiment), the results of both experiments are similar. Participants were better at remembering the exact words that were used, indicating a more prominent surface representation, after direct than after indirect speech. Bayesian analysis of the combined data also showed strong evidence in favor of the alternative hypothesis, BF 01 s  = 0.01.

General Discussion

Language can be viewed as a tool that “allows us to shape events in each other’s brains with exquisite precision” [34] . Ultimately, language comprehension amounts to creating a mental representation of the state of affairs described in an utterance. But how do subtle differences in the form of an utterance have their effect on how its contents are represented? In a series of experiments we sought to answer this question for direct and indirect speech quotations, which make up a major part of everyday communication. Our findings suggest that direct and indirect speech quotations influence mental representations at different levels.

Although direct speech is perceived as more vivid and is thought to be more engaging than indirect speech, we did not find support for the idea that direct speech makes textual information regarding the referential (Experiment 1) or the communicative situation (Experiment 2) more accessible. In fact, we observed no effect of speech in Experiment 1 and an advantage for indirect speech Experiment 2. We were able to rule out that this latter finding was due to greater task-switching costs after direct than after indirect speech (Experiment 3).

At first, these results seem puzzling but they can be explained by the distinction proposed by Clark and Gerrig [1] . According to these authors, indirect speech quotations are a descriptive form of language which means that they are aimed at conveying the gist of an utterance without necessarily drawing attention to its specific realization. Direct speech, on the other hand, is a depictive form of language. It offers the listener a more direct perceptual experience – comparable to looking at a Picasso painting itself, rather than reading a description of that painting. We explored whether this more direct perceptual experience – in this case of a person speaking – involved a switch from the visual to the auditory modality, as suggested by Yao and colleagues [3] . No evidence was found in support of this idea (Experiment 4). A possible explanation for this lack of support might be that the probe recognition task differs from the methods that were used in previous studies on direct and indirect speech and measured sentence processing [3] , [4] , [5] , [6] . However, the absence of a priming effect of direct speech on auditory probes is consistent with the results showing that for auditory priming effects to occur, prior experience with a particular voice in the same context is required [30] . In our experiment, participants did have prior experience with the voices that pronounced the probe words but not with the particular contexts in which they occurred. How is it possible that people perceive direct speech as more vivid and engaging than indirect speech, and yet we found no clue that it makes the mental representation of the referential situation more accessible, or the depicted speech act more perception-like? Taking a cue from a well-known model of mental representations [2] , we hypothesized that direct and indirect speech influence these representations at different levels (just as genre expectations have been found to do [35] ). We found support for this idea. Participants showed superior memory for the exact wording of an utterance when it had the form of a direct speech quotation as compared to an indirect speech quotation (Experiment 5). Apparently, direct speech makes the exact wording of an utterance more salient, enhancing memory for the surface structure of the utterance, whereas indirect speech leads comprehenders to focus more on constructing a situation model.

To summarize, we have systematically addressed several potential consequences of the use of direct versus indirect speech quotations for comprehenders’ mental representations. As it turned out, not all experiments showed an effect in the expected direction or even an effect at all. Nevertheless, these results must be considered informative. Given the large numbers of participants, our experiments had sufficient statistical power to detect possible effects. Moreover, we used Bayesian analysis to determine the posterior probability of the null hypothesis and the alternative hypothesis for each experiment. This approach allows one to combine the results of multiple experiments to compute a single Bayes factor. By doing so with already large samples, we were able to put confidence in our claims regarding the null hypotheses, which would not be possible with the standard procedure of null hypothesis significance testing alone.

Although some of our results seem to be at odds with earlier findings in the literature, they need not be mutually exclusive. For instance, while the effect of implied talking speed on actual reading times may be a pervasive phenomenon, other aspects of the communicative situation [5] , [6] , such as a talker’s voice or manner of speaking, may only be simulated under specific conditions.

Together, our experiments paint a slightly complex, but coherent picture of the effect of direct and indirect speech quotations on comprehenders’ mental representations. While direct speech quotations make the exact wording of an utterance more memorable, this does not necessarily hold for the information it conveys.

Supporting Information

Appendix s1..

Results of the item analyses for experiments 1–4.

https://doi.org/10.1371/journal.pone.0065480.s001

Appendix S2.

Materials that were used in Experiments 1, 4, and 5.

https://doi.org/10.1371/journal.pone.0065480.s002

Appendix S3.

Materials that were used in Experiments 2 and 3.

https://doi.org/10.1371/journal.pone.0065480.s003

Acknowledgments

The authors would like to thank the members of the Language and Cognition lab and an anonymous reviewer for their constructive feedback on these studies.

Author Contributions

Conceived and designed the experiments: AE JAAE RAZ. Performed the experiments: AE JAAE RAZ. Analyzed the data: RAZ. Contributed reagents/materials/analysis tools: AE JAAE RAZ. Wrote the paper: AE JAAE RAZ.

• View Article
• Google Scholar
• 2. Van Dijk TA, Kintsch W (1983) Strategies of discourse comprehension. New York: Academic Press.
• 7. Johnson-Laird PN (1983) Mental models. Cambridge, MA: Harvard University Press.
• 25. Cohen J (1988) Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum.
• 31. Bohan J, Sanford AJ, Cochrane S, Sanford AJS (2008) Direct and indirect speech modulates depth of processing. Paper presented at the 14th Annual Conference on Architectures and Mechanisms for Language Processing (AMLaP), Cambridge, UK.
• 32. Green DM, Swets JA (1966) Signal detection theory and psychophysics: Wiley New York.
• 34. Pinker S (1994) The language instinct: How the mind creates language. Harper Perennial Modern Classics.