Early Childhood Research & Practice is in the process of moving to the early childhood special education program at Loyola University Chicago after 17 years at the University of Illinois at Urbana-Champaign. We are delighted by the opportunity to “pass the torch” to our Loyola early childhood colleagues.

We suggest you visit ECRP’s Facebook page for future updates.

Mirar esta página en español HomeJournal ContentsIssue Contents
Volume 11 Number 2
©The Author(s) 2009

“And Then a Huge, Huge Giant Grabbed Me!” Aggression in Children’s Stories

Chiara Bacigalupa
Sonoma State University

Cheryl Wright
University of Utah


Children’s stories, like children’s play, often contain aggressive elements. This research study identified the themes and ideas that children between the ages of 2 and 6 years old included in 290 dictated stories with aggressive elements. Among the stories that contained aggressive elements, 42% were dictated by girls, and 57% were dictated by boys. Although stories dictated by older children (4 to 4.5 years old) were more likely to contain aggression, age did not seem to be a significant factor in the preschooler’s interest in telling stories with aggressive content. Quantitative and qualitative analyses indicated that children mostly attributed aggressive acts to fantasy characters, and they set aggressive action within fantasy situations. In addition, children did not usually dwell on the details of the aggressive acts they portrayed. Analysis of these stories suggests that rather than expressing an unhealthy preoccupation with aggression or an alarming indifference to the consequences of aggressive behavior, these young children often showed a remarkable ability to keep aggression within appropriate boundaries and a profound lack of interest in the details of the aggressive acts they referenced. Stories provide opportunities for adults to help children develop an understanding of the consequences of aggressive behavior and to reinforce distinctions between appropriate and inappropriate real-life strategies for dealing with interpersonal conflict.


In the past few years, the inclusion of children’s storytelling in preschool curriculum has become widespread. Although children’s storytelling is generally seen as a beneficial activity, adults are sometimes concerned when children include topics such as violence or aggressive behavior. Just as aggressive play sometimes concerns adults (Holland, 2003; Levin & Carlsson-Paige, 2006), children’s dictated stories with belligerent characters, antagonistic conflict, and violent actions also have the potential to worry teachers and parents of young children. Some adults believe that these stories reflect an unhealthy preoccupation with aggression and may lead to increased aggressive behavior.

In our analysis of 858 stories dictated by 201 children between the ages of 2 and 6, we found that 34% of the stories contained verbs that represented aggressive actions. Our analysis of these stories provides insight into children’s interest in developing narratives with aggressive characters and actions. Rather than expressing an unhealthy preoccupation with aggression or an alarming indifference to the consequences of aggressive behavior, these young children often show a remarkable ability to keep aggression within appropriate boundaries and a profound lack of interest in the details of the aggressive acts they reference. Stories may actually offer valuable opportunities for children to explore aggression and negative emotions in the early childhood classroom.

Theoretical and Research Framework

Paley is perhaps the best-known researcher on children’s stories (see, for example, Paley, 1981, 1990). While working as a teacher, Paley collected and had children dramatize the stories that children wrote as part of their everyday curriculum. Paley used the children’s words to show how play—including spontaneous storytelling—influences children’s cognitive, emotional, and social development.

Nicolopoulou (1997) argues that children use stories as a vehicle for making sense of the world: “[Children] use their stories as a way of expressing and working over certain emotionally important themes that preoccupy them and of symbolically managing or resolving these underlying themes” (p. 208). Nicolopoulou also reminds us that the meaning that children express in their stories is the result of a complex interplay between children and their social environments, with the result that children’s narratives have the potential to tell us much about how the sociocultural context influences children’s understandings of specific concepts.

Like Nicolopoulou (1997), Engel (1999) argues that children’s stories enable children to order and examine their thoughts. She proposes that children use these narratives to make sense of their worlds, to solve problems, to participate in culture, to make and keep friends, to aid in construction of the self, and to invent new ideas. Engel categorized the kinds of stories that children tell into three groups: stories of personal experience, anecdotes about shared experiences, and stories of fantasy and fiction. Nicolopoulou and Engel provide our study with a theoretical understanding of why children tell stories and an analytical framework by which to categorize their content.

A few researchers have attempted to understand some of the more specific themes and cross-cultural trends associated with children’s stories. Some studies have focused on how young children structure and incorporate narrative forms into their stories (Groth & Darling, 2001; Marjanovič-Umek, Kranjc, & Fekonja, 2002). Others have looked at the themes that children include in their stories, including themes of power, good versus evil, and attack and defense (Broström, 1998).

A few studies have investigated what children’s stories might tell us about their conceptions of themselves and others. Stories written by Polish 8-year-olds, for example, provided no evidence that children used stories to portray self-image (Uszyńska-Jarmoc, 2004). A study by Richner and Nicolopoulou (2001), however, showed that children did construct character types, and those types depended on the gender of the author. Girls constructed characters who were defined by their relationships, while boys were more likely to depict characters who were defined by their actions.

Other studies have identified further gender-related differences in stories written or dictated by boys and girls. Libby and Aries (1989) found that 3-year-old boys were more likely to introduce aggressive elements into their stories, while their female peers were more likely to introduce friendly, helpful characters. Peirce and Edwards (1988) found that boys were more likely to use violence to resolve conflicts in their stories, while girls were more likely to use reasoning and analysis. Nicolopoulou, Scales, and Weintraub (1994) found that the stories of boys and girls differed in the degree to which they represented order and disorder. Stories dictated by girls tended to feature stable settings and stable relationships—especially family relationships. When girls introduced disorder into their stories, they usually restored order before the story ended. Boys, in contrast, introduced characters and events that disrupted the scene, without tidy resolutions. These last four studies suggest distinct gender differences in how children construct characters and character behavior.

Although the work reviewed here has made a start at analyzing specific themes and ideas in the stories children tell, there is much more to be learned about both the stories and the children who tell them. We initiated this research study to identify the kinds of themes and ideas that children include in their narratives. Once general themes were identified, we planned to conduct a content analysis of specific stories to better understand the ideas children express through storytelling.

One of the most common themes we identified in the stories was aggression—a large proportion of the stories told by both boys and girls included aggressive characters and events. Although the teachers and researchers who worked with the children in this study did not view the aggressive stories as particularly problematic, we realized that other adults—parents in the program and colleagues who reviewed our early manuscripts—were sometimes disturbed by the aggressive characteristics of the stories. Adults were worried about the content of the dictated stories, as well as how that content would be enacted during story dramatizations. This article focuses only on the aggressive content of the dictated stories.

Specific adult concerns included (1) the apprehension that the stories reflected an unhealthy focus on aggression, (2) the concern that the stories might reinforce the view that aggression is a positive or effective approach to solving real-life problems, and (3) the worry that fictional portrayals of aggression might preclude children from adequately appreciating the negative consequences of aggressive behavior. We wondered whether an analysis of the stories might address these concerns. Using as a guide Engel’s (1999) and Nicolopoulou’s (1997) arguments that children use stories to make sense of the world and to represent that understanding, we decided to take a closer look at the stories in our collection that contained aggressive elements in order to answer the following questions: What percentages of the aggressive stories in our sample were told by boys and by girls, and what were their ages? How do children represent aggression in stories? What do those representations tell us about their understandings of aggression?


Setting and Participants

Our collection of stories comes from children enrolled in a child development center associated with a human development department of a large university in the western United States. The center is a part-time program serving primarily families from the local community. The center runs six classes (20 children in each class on average) during the week.

To strive for balance, the program selects children for admission based on age (2 to 5) and gender. Ethnic, racial, and economic diversity are goals of the program, and enrollment policies give priority to children from single-parent, low-income families, and from families whose racial and ethnic makeup is not European American.

All the classrooms were involved at various levels of participation in the storytelling project. The only demographic data available for this research were child age and gender.

We collected stories from the children over a period of 2 years, from September 2005 through May 2007. We collected 858 stories from 201 children between the ages of 2 and 6. Most of the stories collected were from 4-year-olds (N = 455; 53%). Most of the children contributed one (N = 51; 25%), two (N = 42); 21%, three (N = 23; 11%), or four (N = 24, 12%) stories to our sample. Two hundred and ninety of the 858 stories contained aggressive elements; these 290 stories are the focus of this article.

Data Collection

The methods we used to collect stories are similar to the methods described by Paley (1990). Children in the child development center told stories as part of their regular curriculum. For this study, three research assistants made themselves available to transcribe the stories in mixed-age preschool classes at least once a week (the toddler and kindergarten classes participated less frequently). During the scheduled time for center activities, the research assistants asked individual children, “Would you like to tell a story?” This question was the only prompt. Those children who elected to tell a story dictated it to the researcher. The researcher transcribed the child’s words verbatim, using carbon paper to make a copy. When the child was finished dictating, the researcher read the story back to the child to check for accuracy. Later, each child author had the opportunity to choose fellow actors from among his or her classmates to dramatize the story during class time. Many of the dramatizations were also videotaped for future use. The present project does not make use of the videotapes for data analysis.

The researchers kept the original copy of the story and sent the carbon copy home for the child to share with adults in the home. All of the children’s stories were entered into an electronic database. The database included information about the author of each story (child’s first name, gender, age, and class), and the date of the storytelling. Of the 858 stories collected, 46% were told by males (N = 398), and 54% were told by females (N = 461). Most of the stories (N = 598; 70%) were told by children between the ages of 4 (N = 455; 53%) and 5 (N = 143; 17%) years old.


Once the database was created, two researchers and two research assistants analyzed the content of the stories. All four researchers originally read a subset of approximately 300 stories in order to create coding categories that reflected emerging patterns. The major classifications that we developed from the subset of stories were then used to analyze the content of all 858 stories.

The major categories that we created included types of characters (such as superheroes, witches, wizards, monsters, and dinosaurs), the presence of aggressive characters or acts, the kinds of conflicts and resolutions that were portrayed, whether the content was based in reality or in fantasy, themes of good versus evil, inclusion of death, types of relationships portrayed, and whether stories seemed to be heavily influenced by media. The categories of most interest in the present analysis are aggression and fantasy/reality.

Stories were labeled as fantasy if they contained characters that do not exist in the current real world (such as witches, unicorns, and dinosaurs) or if they contained actions that cannot really happen (such as acts of magic or talking animals). Dinosaurs were included in fantasy themes because children do not encounter dinosaurs in their everyday lives. Stories were categorized as realistic if they contained real people (such as parents, relatives, or teachers) participating in events that children might reasonably have seen or experienced (such as cooking, going to school, and camping). If the stories featured realistic people who performed unrealistic actions (such as a child flying), the stories were categorized as fantasy but not as realistic. If the stories contained both realistic and fantasy actions, they were included in both the realistic and fantasy categories. 

Although young children are typically believed to be unable to distinguish between reality and fantasy, we believe this coding is nonetheless valid. Recent evidence suggests that children may be more adept at distinguishing between reality and fantasy than previously thought. Sharon and Woolley (2004) found that many 4-year-old children are able to correctly identify characters as either real or pretend. Recall that the majority of aggressive stories in our study were told by 4-year-olds (59%) and older children (5- and 6-year-olds; 16%).

When children in the Sharon and Woolley (2004) study were unable to identify a fantasy character, they often said they were “not sure,” which the authors argue is very different from wrongly saying that the character was real. In addition, the fantasy characters that were most consistently misidentified as real were Santa Claus and the Easter Bunny—a belief that is widely supported by adults and others in the child’s social environment. In their analysis of children’s stories, Pitcher and Prelinger (1973) also code story characters as fantasy and realistic characters, using criteria similar to ours.

Stories were categorized as aggressive if they contained an action that appeared to be (1) intentional and (2) was likely to result in harm to another character. We looked for verbs that denoted aggressive actions, such as “kick” or “kill.” Stories that used only a descriptive adjective for a character (e.g., mean witch, evil wizard) but did not contain an aggressive act were not classified as aggressive stories.

Thirty four percent (N = 290) of the original 858 stories were coded as aggressive. We compared statistics for the aggressive stories with our statistical data on the larger collection of stories in order to see if some themes and content were more likely to appear in aggressive stories. We also read and re-read the entire collection of aggressive stories to refine the codes that we had used with the larger collection. For example, one researcher read all 290 stories and classified the verbs that children used to denote aggressive actions, such as “fight” or “attack.” Another researcher read all 290 stories and noted the kinds of characters who perpetrated the aggressive acts. For example, if the verb “fight” appeared, the researcher noted who did the fighting and whether that character was realistic or fantasy, animal or person.

We also conducted an interpretive analysis of the stories with the goal of exploring what children’s representations of aggression might tell us about their understandings of aggression. Following Nicolopoulou (1997), we embarked on a study of how children “use narrative as a tool to grasp reality and to confer meaning on experience” (p. 194). Our guides in this interpretive analysis were the trends that we had already identified (e.g., fantasy characters as perpetrators and lack of graphic representation of violence) and the concerns that other adults had voiced about the stories (e.g., stories about violence might normalize violence as a problem-solving method). For the interpretive analysis, we read each aggressive story at least twice more, this time focusing on the meaning children seemed to attach to the aggressive elements of the stories.


For this study, multiple units of analysis were explored. We wanted to use stories as the unit of analysis, simply because we knew that we would be describing the content of stories in our interpretive analysis. Thus, we began by coding every story, using the story as the unit of analysis. Once we had coded the stories, we were able to run simple statistical tests. For example, we compared the number of stories with superheroes written by boys with the number written by girls.

To account for the fact that some children told multiple stories, two other methods were used in our simple statistical analyses. First, the characteristics of each child’s stories were averaged, taking an aggregate story for each child. For example, if a child told two stories, and included two aggressive acts in one and four aggressive acts in another, the final aggregate number of aggressive acts would be three. Second, statistical analyses were completed on a sample of one random story from each child. These two methods each use the child as the unit of analysis. We ran the same statistical tests using all three methods, with no significant difference in the results among the different methods. For convenience, we list here only the results that were obtained when we used the story as the unit of analysis.

Percentages of Stories by Gender and Age

Of the 858 stories in our original sample, 27% of stories dictated by girls contained an aggressive act, while 42% of stories dictated by boys contained an aggressive act. A Pearson chi square test of this difference indicates that stories dictated by boys are significantly more likely to contain aggression than are stories dictated by girls (p < .0001). However, among the subset of stories that contained aggressive elements, many were authored by girls (girls dictated 42% of aggressive stories, while 57% were dictated by boys). Although this difference is also significant, it is nonetheless noteworthy that girls authored a high proportion of the aggressive stories. This finding suggests that girls may not be quite as fascinated with aggression as boys are, but it would be inaccurate to suggest that they show little interest.

When we categorized the children into half-year intervals (2.5 to 3.0 years, 3.0 to 3.5 years, etc.), the highest percentages of aggressive stories were from 4- to 4.5-year-old children (39% and 36%, respectively). However, these numbers were not significantly different from the percentages of children in the other age groups who tell aggressive stories. Thus, although stories dictated by older children in our study were more likely to contain aggression, age did not seem to be a significant factor in the preschooler’s interest in telling stories with aggressive content. 

How Aggression Was Represented

The most common aggressive actions in the stories, in order of frequency, were “fight,” “died,” “ate,” and “killed.” Some actions may have appeared one or more times in the same story, but most actions appeared only once in what typically were very short stories. (We recorded stories on wide-ruled notebook paper, and stories were typically half a page or 6-10 lines. Because the stories were to be dramatized, we preferred to keep the stories short. Children were encouraged to end their stories or mark them as “to be continued” when they reached the bottom of a page. The children were rarely inclined to exceed the one-page limit.) The most common aggressive action was fighting (see Figure 1). The verb “fight” occurred in 145, or 50%, of aggressive stories. The following story is a very typical rendition of a fight in the stories we collected:

My story is about a robot. The robot finds the clue. And the robot finds the bad guys and he fights them. And he found them and fights them. The End.

Notice that the child author of this story conveys the entire aggressive conflict with the single verb “fight.” It was typical for children to use a single verb to denote all of the aggression in a story.

The verb “died” also appeared often—in 70, or 24%, of the 290 aggressive stories. If a story said that a character “died” as the result of an action performed by another character in the story, the story was coded as containing aggression. In some rare instances, characters died as a result of illness or other indirect causes, and these cases were not coded as aggressive. For example, one father died after going to the dentist, and a “grandmother died” in the middle of one child’s story.

“Ate” was only coded as aggressive if it resulted in harm to another character. Consider the following story:

Devin was writing in the castle on the table where they were eating. The giant came there too and the giant found and ate her. 

The first instance of “eating” would not be considered an aggressive act in this story, but when the giant eats Devin, that is considered aggressive because it is an unwelcome action that would likely result in harm. The verb “ate” was used aggressively in 61, or 21%, of the 290 aggressive stories.  

The verb “killed” appeared in 58, or 20%, of the 290 aggressive stories. The following story fragment is an example of a typical use of the verb “killed”:

…[Be]cause one night there was a giant. Then the dragon killed the mermaid. Then the witch killed the prince. The end.

Figure 2

Figure 1. Percent of aggressive stories containing each of the four most popular aggressive verbs.

Other common aggressive actions included hit (28 instances), destroy (21 instances), and shoot (21 instances). 

A wide range of characters populated these aggressive stories (see Figure 2). Animals were the most popular character in aggressive stories, appearing in 46% of the stories (however, they did not always appear as the aggressor). Dinosaurs (15% of stories), dragons (12%), and monsters (20%) were not classified as animals, but they also appeared fairly frequently in children’s aggressive stories. 

The second most popular characters were superheroes, appearing in 27% of stories containing aggression. Popular superheroes were Superman, Batman, Power Rangers, Ninja Turtles, and Transformers. Superheroes were more likely to appear in aggressive stories told by boys (22% of boys’ stories versus 7% of girls’ stories). Royalty (kings, queens, princes, and princesses) appeared more often in girls’ aggressive stories (43% of girls’ stories versus 8% of boys’ stories). Queens and princesses appeared in 23% of stories containing aggression, while kings/princes appeared in 20% of stories containing aggressive elements. Witches/wizards (11%) also made regular appearances in aggressive stories.

Figure 2

Figure 2. Number of aggressive stories containing each of the eight most popular characters.

Common Elements or Trends in Stories

We identified two trends in the aggressive stories as being particularly noteworthy. First, the preferred characters in these stories, whether taken from fairy tales, superhero cartoons, or some other source, were firmly rooted in the world of fantasy. This finding is related to the fact that the aggressive actions in the stories that we collected were most likely to occur in stories categorized as fantasy (N = 237; 82% of aggressive stories) or as both fantasy and reality (N = 50; 17% of aggressive stories). In fact, only 26 aggressive stories (7%) had no fantasy elements at all. Half of the stories without fantasy elements involved an animal as an aggressor (N = 13), including crocodiles, alligators, sharks, tigers, elephants, jaguars, snakes, frogs, and dogs. Two stories contained insects that were aggressive. The following story captures typical aggressive acts involving animals:

We had a pet crocodile. And he’s biting.

Of the entire set of aggressive stories, only nine (3%) depicted real people as perpetrators of gratuitous aggression and violence. Two were stories about Native Americans and white settlers, one involved rescue workers, another cowboys, and two others depicted “bad guys” and “someone” in aggressive roles. The remaining story depicted a moderately aggressive act—a sister pinching. Only the pinching sister had descriptive features that made the aggressive character identifiable as a real person known to the author. All other characters who initiated aggressive acts—the majority of aggressors in these stories—were fantasy figures or animals (see Table 1). 

Table 1
Stories Coded as Fantasy or Realistic
Story Category Number and Percentage of Aggressive Stories
Total number of stories containing aggression N = 290
Stories coded as fantasy N = 238 (82%)
Stories coded as fantasy and realistic N = 49 (17%)
Stories coded as realistic only N = 20 (7%)
Stories containing realistic characters who initiate aggression N = 9 (3%)

In stories that contained both realistic elements and fantastical characters, events, or settings, the aggression was always closely tied to fantasy. The following story is an example.

Once upon a time, me and my cute little baby—we were going to the park, and I was talking to her. Then we were going to the park. And then we were going home to take a nap. And then a huge, huge giant grabbed me and took my baby to a castle.

In this story, the child mixed common experiences such as going to the park and eating lunch with a make-believe event—a monster grabbing a baby and taking it away to a castle. It is noteworthy that the “bad guy” in this story is a fantasy character—a giant. We found that in stories that contained both realistic and fantastical elements, realistic characters nearly always played the heroes and heroines, while the “bad guys” were based on people and animals that children do not typically encounter, such as monsters, dinosaurs, wild animals, and superheroes borrowed from both print and electronic media. The following story illustrates this point:

Once there were three unicorns, and the prince locked the princess in the cage. And a blanket comed. And then it was plain, and all of them had tea. Then they goed to school, and before they had breakfast. And the mom picked them up. And the witch locked everyone in the cage. And a teacher came and tried to get them out, but couldn’t, and they didn’t know how to. And the three unicorns came back. And whoever did that was in trouble. The End.

This story contains a mix of realistic and not-so-realistic characters. The nonaggressive characters are both realistic and fantastical, but the aggressive characters (the prince and the witch) are both fictional. 

In fact, realistic characters usually limited their aggressive acts to self-defense against fantastical characters who were the initial aggressors. In the above story, for example, it is a realistic character—a teacher—who tries to save characters who are the victims of a fantasy aggressor—a witch.

A second trend in the stories that we collected was that children provided few details about how violent acts were carried out. Rather than calling upon an arsenal of weapons or bloody tactics, children simply used the verb “fight” to convey the idea that a realistic character had physically fought off an aggressor. For example, the following lines describe all of the action involving this heroic author: “I was a brave knight, and then I fight a monster and he was dead. And I saved the princess.” This nondescriptive way of conveying that a fight occurred was characteristic of most of the stories that we collected. Children used verbs such as “fight,” “kill,” and “die” to indicate both the conflict and the resolution in their stories. They rarely specified weapons, detailed strategies, described wounds, or in any way dwelt on the harm that characters inflicted upon one another.

The following story is one of the stories in our collection that includes a gun:

There was a princess. Her stepsister killed her with a gun. Then she went to heaven. Then she went to the hospital and came right back home. Then her mom took care of her and she was all better.

This was the most worrisome story that we encountered in our collection—the only one to spark concern among some of the adults working in the preschool. Not only does it contain a real weapon, but it also portrays an explicit act of violence. Despite the disturbing elements in this story, we offer three reasons why there is little cause for concern. First, the author has relegated the violence to two characters who regularly appear in fairy tales—a princess and her stepsister. Although the characters are technically members of a family, stepsisters often appear as villains in fairy tales. Second, all of the violence is portrayed in the phrase “killed her with a gun.” The brevity and vagueness of this phrase make the violence seem less immediate and tangible. Finally, the author negates the harmful action of the gun by reversing the stepsister’s death through the introduction of a healing character—a mother who cares for the victim and provides reassurance that everything will be okay. We believe that this author took specific steps to neutralize the harmful elements of the story, thus showing an understanding that such violence is wrong. It is also noteworthy that none of the children who participated in or watched the dramatization of this story were visibly disturbed by it (Wright, Bacigalupa, Black, & Burton, 2008). The ideas raised by our analysis of this story will be considered further in the next section.


Some parents in the center and some reviewers of our early manuscripts expressed concern about the aggressive content of the stories. A common adult concern is that allowing children to dictate and act out such ideas will reinforce negative impressions about how to resolve conflicts and/or lead to increased violence in the classroom. Such concerns fall into the category of what Levin and Carlsson-Paige (2006), writing about aggressive play, identify as the sociopolitical view. The sociopolitical view predicts that aggressive play or storytelling will reinforce aggressive tendencies.

Competing with the sociopolitical view is the developmental view, which portrays aggressive play as a necessary vehicle for exploring, mastering, and diffusing the violence and aggression children see in the media (books, television, movies, and video games) or in their own lives (Levin & Carlsson-Paige, 2006). Many authors who study children’s play support the developmental view, arguing that children need such play to work through their fears, explore conflicting ideas, and experiment with solutions (e.g., Edmiston, 2007; Jones, 2002; Katch, 2001). Prominent authors on storytelling also tend to adopt the developmental view, arguing that the inclusion of aggressive characters and actions in stories empowers children by giving them an alternate way to confront and defeat potential and real adversaries (Cooper, 1993; Engel, 2005; Paley, Cooper, McNamee, Mathes, Capo, & Floyd, 2003).

Our research provides evidence to support a developmental response to children’s stories that contain aggression. The aggressive stories told by the children in our sample tended to be fantasy based, which is consistent with the idea that storytelling may provide children with a safe space to play with the ideas that concern or frighten them. We cannot ignore, however, the reminder from the sociopolitical view that children are developing their ideas about appropriate ways to interact with others and to solve problems. In the following sections, we use the results of our research to address two concerns raised by adherents to the sociopolitical view. One concern is that children will not develop a good understanding of the consequences of aggression when they are allowed to include it in their stories. Closely related to this concern is the fear that children will view stories as models for using antisocial methods to resolve their own interpersonal conflicts.

First, we note that in many cases the primary purpose of aggression may simply be to provide excitement and a plot—a practice that has a strong precedent in the stories that children hear and view in books and other media. Pitcher and Prelinger (1973), for example, note that when children’s stories do not include aggressive content or misfortune, they “seem markedly to lack the urgency” (p. 164) of those that do. Perhaps children are, at some level, aware of this distinction. It is understandable that children would want to make their stories exciting, and aggression is an effective means of accomplishing that end, particularly if the stories are to be dramatized. Adults might be more pleased if a child wrote a story in which the protagonists “use their words” to conquer monsters and save princesses, but we can understand why such a story may not be as satisfying to a child. Thus, some children’s stories may contain aggression simply because it is exciting.

That excitement, however, could look very appealing to a child and lead to the mistaken idea that aggression is appropriate in real-life situations. This concern is understandable, since children are often exposed to stories that suggest that aggression is necessary and effective. However, we believe that children show more discernment in the stories that they create. The consistent coupling of aggression and fantasy in the stories that we collected suggests that the children in this study (most of whom have not been exposed to real-life violence) are aware that aggressive acts do not typically take place in day-to-day interactions between people. Instead, the child authors consistently relegated aggression to the province of animals and “bad guys.”

Proponents of the sociopolitical view might not find this argument convincing if they believe that children do not understand that their fantasy creations are not real. Yet older preschoolers do seem to distinguish between reality and fantasy. First, as discussed previously, Sharon and Woolley (2004) provide evidence that 4-year-old children are capable of making a distinction between fantasy and reality. Second, our results show that the children in our study consistently associated aggression with the fantastical elements in their stories, which does seem to indicate that they are able to distinguish between reality and fantasy. If they were making no distinction at all, we would expect that, as a matter of chance, aggression would be linked to realistic characters and events more frequently. Young children may not yet be conscious of the distinction they are making, but our findings offer compelling evidence that they are nonetheless making the distinction. In fact, it is possible that by dictating such stories, children are actively reinforcing a distinction between real-life aggression that is harmful to real people and make-believe aggression that appears in the fairy tales and other fictional stories that they hear and view. 

A related concern often expressed by adults is that fictional depictions of aggression may prevent children from developing a healthy appreciation for the real consequences of aggressive acts. Cartoon characters spring right back from debilitating physical blows, good guys dodge bullets with ease, and the evil nemesis who was eliminated by our favorite superhero last week returns unharmed in this week’s episode. Children see stories such as these in books and electronic media, and they re-create them in their own narratives. As we discussed in the previous sections, children glossed over the details of the violence that they depicted. They did not talk about the harm that comes to characters, and they sometimes pointedly ignored the possibility of such harm by having characters who are killed come back to life. We view this lack of detail as a positive indication that children are not focused on the aggression for its own sake, and we believe the lack of detail keeps the aggression psychologically distant. 

However, one might also argue that this lack of specificity shows a disturbing lack of understanding of the real-life consequences of aggressive acts, thus leading children to think that aggression might be an appropriate way to resolve real-life problems. We agree that children’s stories usually do not reflect a complete understanding of the real-life consequences of aggressive behavior, and we can imagine instances in which their stories might become very realistic, graphic, or otherwise worrisome. But it does not follow that children therefore do not possess a good understanding or that creating aggressive stories renders them incapable of developing such an understanding. A measured approach to such stories may be to allow children the freedom to express such stories so that an adult can then make an appropriate response to them. In other words, the stories that children dictate generate valuable opportunities for finding out what children do know and for helping them to develop better understandings of aggression and real-life problem solving. 

In research on violent play, for example, Dunn and Hughes (2001) report the disturbing finding that children who engaged more frequently in violent play showed poor executive control and language ability, more often experienced conflict with their friends, and were less likely to show prosocial behavior in their peer interactions. Rather than concluding that violent play should be banned, however, Berk, Mann, and Ogan (2006) propose an alternative response. These authors argue that play is exactly the outlet that these children need in order to develop the self-regulation they seem to lack. In their view, these children should be allowed to initiate their own make-believe play—even when it contains violent themes—and the adult’s role should be to intervene when the play becomes antisocial. 

Levin and Carlsson-Paige (2006) recommend that adults respond to children’s aggressive play by reinforcing the idea that play is pretend, encouraging children to think of and try out alternate behaviors and resolutions in their play, helping children to humanize the enemy, and helping children understand the actual effects of violence and aggression. We believe that these approaches can also be appropriate responses to children’s stories. Depending on the context, a teacher might be comfortable enacting the story of the princess who shoots her stepsister, as it is dictated, with only a few remarks intended to remind children that the story is just pretend and/or emphasizing the healing aspects of the story. Another teacher, one who works with children who may have experienced the consequences of gun violence, might decide on a very different response. The point is that the story can be used as a starting place for discussions about violence and aggression that emphasize the concepts that concern a particular group of children. By allowing the stories to be told, teachers allow children to explore the ideas that intrigue and concern them, while also capitalizing on opportunities to guide children’s thinking about aggression. Rather than preventing children from realizing the real-life consequences of aggression, stories with aggressive elements provide a safe space for adults and children to confront and learn about aggression.

In addition to using children’s stories as a springboard for more sophisticated discussions about aggression and its consequences, we would encourage adults to look closely at each aggressive story before assuming that it contains harmful ideas. We have shown that children’s stories often embed aggression in fantasy, thus dissipating much of the negative power of the aggression. 

Based on our analysis of children’s aggressive stories, we agree with Cooper (1993) and Engel (2005) that preschool children often use stories as a vehicle for exploring, mastering, and diffusing powerful ideas and events they encounter in books, television shows, movies, and video games. Within those explorations, some children seem to be very adept at using fantasy and clever plot lines to keep aggression manageable and safe. We hope that these examples will encourage adults to take time to respond carefully and thoughtfully to the stories that children dictate. Although the topics that children choose may sometimes seem scary and inappropriate for classroom use, a close look often reveals that children are capable of handling aggressive themes in very complex and appropriate ways. Furthermore, stories present excellent opportunities for adults to help children develop a good understanding of the consequences of aggressive behavior and to reinforce distinctions between appropriate and inappropriate real-life strategies for dealing with interpersonal conflict.


Several limitations restrict the scope of the conclusions that can be drawn from this study. These limitations include characteristics of the children who authored the stories, the way stories were collected, the coding categories chosen for this preliminary study, and the number of stories collected per child.

The children in this study were members of predominantly white, middle-class families. Most of these children had not experienced real-life violence or seen its effects firsthand. Thus, the stories discussed here do not reflect the ideas of children who have had to cope with real-life aggression. For example, Haight, Black, Ostler, and Sheridan (2006) found that children who had experienced trauma sometimes engaged in realistic enactments of adult aggression. It is thus possible that teachers who work with children who have experienced real-life aggression and violence may find that the stories their children dictate are very different from those found in our sample. In such cases, children may need additional support or intervention to help them process the disturbing events that they have experienced.

In the classrooms where we collected stories, children were always given a choice of whether or not to dictate stories. It is possible that the children who chose to dictate stories differ from those who chose not to dictate stories. For example, it is possible that the children who behaved more aggressively in the classroom were less willing to dictate stories. Consequently, the stories discussed here might not reflect the ideas of children most likely to engage in real-life aggressive actions with their peers. 

A third limitation of the study has to do with the way we coded the children’s stories. We chose to code stories as aggressive if a perpetrator’s acts were likely to cause harm to someone in the story—we did not code whether or not harm actually occurred. In addition, we did not distinguish between aggressive acts that caused more harm or less harm. In this preliminary analysis of our data, we sought to be overly conclusive on the assumption that children may see acts that look to an adult to be only potentially or only mildly aggressive as actually quite frightening (and thus psychologically harmful). The data could be analyzed so that aggressive acts are defined at different levels of harm and so that only acts specifically identified as actually causing harm are included. Such an analysis might result in fewer stories being identified as aggressive. 

In addition, more nuanced coding would add depth to our understanding of how children use aggression in stories. For example, additional coding would allow a closer look at the kinds of aggression children include and which aggressors they pair with more or less aggressive acts.

A final limitation is the fact that we collected relatively few stories (usually four or fewer) per child. It is likely that the length and level of detail in children’s stories increase over time. If so, it is possible that a child’s aggressive stories might include more details about the aggressive elements of the stories and thus over time be cause for more concern than the stories presented here.

Although we have identified some trends in the stories that we collected, it is important to remember that all stories dictated by children are unique. Each story should be evaluated and responded to with consideration for the author and the author’s context.


Stories seem to provide an important outlet for preschool children to explore the aggression and violence to which they are sometimes exposed, but these stories can be alarming to the adults who care for young children. We encourage educators to create safe spaces where children can express their curiosity and ideas about aggression. A close look at the stories dictated by the children in this study suggests that many children do handle aggressive themes in appropriate ways in their stories. These narratives offer reassuring evidence that children’s stories often really are “just pretend.”


Berk, Laura E.; Mann, Trisha D.; & Ogan, Amy T. (2006). Make-believe play: Wellspring for development of self-regulation. In Dorothy G. Singer, Roberta Michnick Golinkoff, & Kathy Hirsh-Pasek (Eds.), Play = learning: How play motivates and enhances children’s cognitive and social-emotional growth (pp. 74-100). New York: Oxford University Press.

Broström, Stig. (1998, September). Children’s stories and play: Storyride—a children’s culture project. Paper presented at the Eighth Symposium for the European Conference on Quality in Early Childhood Settings, Santiago de Compostela, Spain.

Cooper, Patsy. (1993). When stories come to school: Telling, writing, and performing stories in the early childhood classroom. New York: Teachers and Writers Collaborative.

Dunn, Judy, & Hughes, Claire. (2001). “I got some swords and you’re dead!”: Violent fantasy, antisocial behavior, friendship, and moral sensibility in young children. Child Development, 72(2), 491-505.

Edmiston, Brian. (2007). Forming ethical identities in early childhood play. New York: Routledge.

Engel, Susan. (1999). The stories children tell: Making sense of the narratives of childhood. New York: W.H. Freeman.

Engel, Susan L. (2005). Real kids: Creating meaning in everyday life. Cambridge, MA: Harvard University Press.

Groth, Lois A., & Darling, Lynn Dietrich. (2001). Playing “inside” stories. In Artin Göncü & Elisa L. Klein (Eds.), Children in play, story, and school (pp. 220-237). New York: Guildford Press.

Haight, Wendy; Black, James; Ostler, Teresa; & Sheridan, Kathryn. (2006). Pretend play and emotion: Learning in traumatized mothers and children. In Dorothy G. Singer, Roberta Michnick Golinkoff, & Kathy Hirsh-Pasek (Eds.), Play = learning: How play motivates and enhances children’s cognitive and social-emotional growth (pp. 209-230). New York: Oxford University Press.

Holland, Penny. (2003). We don’t play with guns here. Maidenhead, Berkshire, England: Open University Press.

Jones, Gerard. (2002). Killing monsters: Why children need fantasy, super heroes, and make-believe violence. New York: Basic Books.

Katch, Jane. (2001). Under deadman’s skin: Discovering the meaning of children’s violent play. Boston: Beacon Press.

Levin, Diane E., & Carlsson-Paige, Nancy. (2006). The war play dilemma: What every parent and teacher needs to know (2nd ed.). New York: Teachers College Press.

Libby, Marion N., & Aries, Elizabeth. (1989). Gender differences in preschool children’s narrative fantasy. Psychology of Women Quarterly, 13(3), 293-306.

Marjanovič-Umek, Ljubica; Kranjc, Simona; & Fekonja, Urška. (2002, August). Developmental levels of the child’s storytelling. Paper presented at the Annual Meeting of the European Early Childhood Education Research Association, Lefkosia, Cyprus.

Nicolopoulou, Ageliki. (1997). Children and narratives: Toward an interpretive and sociocultural approach. In Michael Bamberg (Ed.), Narrative development: Six approaches (pp. 179-215). Mahwah, NJ: Lawrence Erlbaum.

Nicolopoulou, Ageliki; Scales, Barbara; & Weintraub, Jeff. (1994). Gender differences and symbolic imagination in the stories of four-year-olds. In Anne Haas Dyson & Celia Genishi (Eds.), The need for story: Cultural diversity in classroom and community (pp. 102-123). Urbana, IL: National Council of Teachers of English.

Paley, Vivian Gussin. (1981). Wally’s stories. Chicago: University of Chicago Press.

Paley, Vivian Gussin. (1990). The boy who would be a helicopter: The uses of storytelling in the classroom. Cambridge, MA: Harvard University Press.

Paley, Vivian G.; Cooper, Patsy; McNamee, Gillian; Mathes, Bernie; Capo, Karen V.; & Floyd, Connie. (2003). Children telling and acting their own stories in the early childhood classroom. Paper presented at the meeting of the National Association for the Education of Young Children, Chicago, IL.

Peirce, Kate, & Edwards, Emily D. (1988). Children’s construction of fantasy stories: Gender differences in conflict resolution strategies. Sex Roles, 18(7/8), 393-404.

Pitcher, Evelyn Goodenough, & Prelinger, Ernst. (1973). Children tell stories: An analysis of fantasy (3rd ed.). New York: International Universities Press.

Richner, Elizabeth S., & Nicolopoulou, Ageliki. (2001). The narrative construction of differing conceptions of the person in the development of young children’s social understandings. Early Education and Development, 12(3), 393-432.

Sharon, Tanya, & Woolley, Jacqueline D. (2004). Do monsters dream? Young children’s understanding of the fantasy/reality distinction. British Journal of Developmental Psychology, 22(2), 293-310.

Uszyńska-Jarmoc, Janina. (2004). The conception of self in children’s narratives. Early Child Development and Care, 174(1), 81-97.

Wright, Cheryl; Bacigalupa, Chiara; Black, Tyler; & Burton, Michael. (2008). Windows into children's thinking: A guide to storytelling and dramatization. Early Childhood Education Journal, 35(4), 363-369.

Author Information

Chiara Bacigalupa is assistant professor of education in the Department of Literacy Studies and Elementary Education at Sonoma State University. She received her Ph.D. in Early Childhood Education (ECE) from the University of Minnesota. She has completed several qualitative studies focused on child care quality, children’s moral and social development, and children’s interactions with various types of media. Currently, Chiara continues her work on understanding children through the stories they tell. She is also collaborating with Head Start teachers on a study of the barriers that ECE professionals face in baccalaureate degree programs.

Chiara Bacigalupa, Ph.D.
Department of Literacy Studies and Elementary Education
Sonoma State University
1801 East Cotati Avenue
Rohnert Park, CA 94928
Email: chiara.bacigalupa@sonoma.edu

Dr. Cheryl Wright is the chair of the Department of Family and Consumer Studies at the University of Utah. Dr. Wright received her Ph.D. from Oregon State University. She oversees the Early Childhood Teacher Certification Program and the Graduate Program for Early Childhood Educators. Her primary areas of research are school readiness and literacy. Her current research focuses on preschool storytelling and literacy. She coordinates a summer early literacy intervention program for low-income kindergartners. In collaboration with the Utah State Office of Education, she developed the Utah Family Literacy Project, which was a series of parent education programs for statewide dissemination regarding preschool literacy. She also participated in the implementation of Reach Out and Read, a pediatric literacy program in an inner-city health care clinic in Salt Lake City. She collaborated on the evaluation and implementation of United Way’s Success by Six early intervention program. She has been actively involved in early childhood education policy at the state and national levels.  

Cheryl Wright
Department of Family and Consumer Studies
University of Utah
225 South 1400 East, Rm. 228
Salt Lake City, UT 84112-0080
Email: Cheryl.wright@fcs.utah.edu