Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

  • Father: And we had to go up at Christmas time for Grandpa’s funeral, didn’t we?

  • Adolescent: mmmhmm

  • Father: mmmhmm, and, what do you remember about that?

  • Adolescent: It was very sad and…

  • Father: (simultaneous with adolescent) …It was sad…

  • Adolescent: …scary

  • Father: Especially that it was right at Christmas time and Grandma had just passed away about a year before that, which made it really hard for everyone. And I had to get out of work early and Mom had to get out of work and you had to get out of school and we all had to go up to Michigan and it was sad cause it was around the holidays

  • Mother: Yeah, I thought it was especially sad for, um Grandpa, you- your Father.

  • Father: Mmmhmm

  • Mother: You know having to deal with that

  • Father: Yeah, it was sad for me cause it was my last grandparent too

  • Mother: Yeah

  • Father: How did it effect you Rachel?

  • Adolescent: I thought it was sad even though I didn’t spend a whole lot of time with Grandpa, our Great Grandpa, but it was still sad you know a member of the family had passed away

  • Father: mmmhmmm

In this excerpt, a family with a 12-year-old adolescent co-narrates a shared sad experience, the death of the adolescent’s great-grandfather. As this example points to, family narratives are a window into how families construct a shared sense of history, understand and validate each others’ emotions and create a sense of who they are as a family, and as individuals, in the present.

Many theorists have argued that narratives are the way in which we make sense of our experiences and our selves (Bruner, 1987, Chafe, 1990; Fivush, 2008; McAdams, 2003; Ricoeur, 1991), and this is, at core, a social process, in which meaning emerges from the telling and sharing of the stories of our lives with others (Fivush, Haden, & Reese, 1996; McLean, 2005; Pasupathi, 2006). Narratives move beyond simple descriptions of what occurred, to include information about causal explanatory frameworks, about how and why the event happened as it did, and about reflective interpretation that evaluates the event from a personal perspective that provides a subjective stance on what occurred (Fivush & Haden, 2005; Fivush, 2001; Labov, 1982; Linde, 1993; Peterson & McCabe, 1982). Essentially, personal narratives provide the fabric of our self-understanding, weaving together our experiences with our evaluations to produce a story of our life in relation to others.

Recent research has begun to examine the emergence of the life narrative in adolescence, a time when cognitive and socioemotional skills mature, allowing children to begin to reflect on their own and others’ perspectives, values, and goals and to question previously accepted interpretations and meanings (Erikson, 1959/1980; Kroger, 2000). With the developing ability to construct more complex and temporally extended sequences (Friedman, 2000, 2004), linking individual events into larger life motifs (Habermas & Bluck, 2000; McAdams, 1992), and to provide more nuanced and reflective interpretation of self and other (Harter, 1999; Collins, Gleason, & Sesma, 1997), adolescents begin to construct the story of a life that defines who they are and who they want to be (see Habermas & Bluck, 2000, for a full theoretical discussion).

Narratives and Identity

Narratives are both the process by which identity is created and a reflection of that identity. More specifically, the ways in which individuals make sense of their experiences, as expressed in the structure and content of their personal narratives, provide the foundation for understanding of self, and this emerging understanding of self is reciprocally expressed in personal narratives. For example, McLean and Pratt (2006) have shown that adolescents who engage in less meaning-making in their personal narratives show higher levels of diffusion and foreclosure on identity measures, and, critically, adolescent meaning-making in narratives predicts subsequent generativity. Similarly, Pasupathi, Mansour, and Brubaker (2007) have shown that the way in which self-event connections are created within narratives helps to create a stable sense of self over time. Thus self and narrative evolve in an ongoing dialectic (McLean, Pasupathi, & Pals, 2007).

Moreover, personal narratives and identity are co-created in social interaction; through sharing the stories of our lives with others, these stories come to take on new meanings, new interpretations, and new evaluations (Fivush, 2008; McLean, 2005; Pasupathi, 2006). We argue from a sociocultural perspective on narrative and identity, stemming from Vygotsky’s (1978) developmental theories, that personal narratives are embedded in the sociocultural world in which individuals create meaning through daily social interactions that provide the structure and the value of specific activities (Gauvain, 2001; Rogoff, 1990). In terms of autobiographical memory, telling and sharing one’s personal past is a culturally mediated activity that is more or less valued by particular cultures and particular members within a culture (see Fivush, Bohanek, & Duke, 2008, Fivush & Haden, 2003, for an overview). In Western culture, to have and tell one’s autobiography is highly valued (McAdams, 2003; Pillemer, 1998) and is incorporated into the everyday activities in which children and adults are expected to engage. Through participating in adult-structured reminiscing, children learn the forms and functions of personal narratives (Fivush & Nelson, 2004; Nelson & Fivush, 2004).

In this chapter, we present research on the Family Narratives Project, in which we examine the ways in which families reminisce together and how this process is related to children’s developing sense of self and well-being as they undergo transition into adolescence. We focus on family co-narration of a negative event, a time that was stressful for the family because stressful events are critically related to well-being. Stressful events create a problem to be solved as well as arousing aversive affect which must be regulated. There is growing evidence that the ability to create more coherent and emotionally integrated narratives of stressful events is related to better physical and psychological well-being (Pennebaker, 1997). For example, using an expressive writing paradigm, in which adults are asked to write about the stressful events of their lives for 10–20 minutes per day for several days, those adults who are able to create more coherent explanatory narratives subsequently show better outcome (see Frattaroli, 2006, for a meta-analysis).

However, children and young adolescents are still in the process of developing the cognitive and socioemotional skills needed to create coherent narratives (Habermas & Bluck, 2000). For them, narrating stressful events may raise anxiety without the concomitant skills to regulate the affect or cognitively restructure the event (Fivush, Marin, Crawford, Reynolds, & Brewin, 2007). Thus adolescents may still be relying on more cognitively and emotionally competent adults to help them structure their narratives in ways that allow processing. In fact, in examining mother–child co-constructed narratives about highly stressful events, Sales (Fivush & Sales, 2006; Sales & Fivush, 2005) found that mothers who were more emotionally expressive and explanatory had children with better coping skills and fewer behavior problems. Thus we were particularly interested in examining how families as a whole co-narrated the stressful events of their lives and how the structure and content of these narratives might be related to adolescent well-being.

Based on a family system’s perspective (e.g., Kreppner, 2002), in which it is assumed that the dynamics of family interaction is a complex process in which each family member influences, complements, and contrasts with each other, we chose to study the family co-constructing narratives of their past. To foreshadow, we first present research examining the family as the unit of analysis, describing the process of narrative co-construction and the emotional content of family narratives in relation to child well-being. We then turn to an analysis of how mothers and fathers differentially scaffold these co-narrations within the family, and argue that reminiscing is a gendered activity (Fivush, 1998; Fivush & Buckner, 2003), with mothers and fathers having differential effects on child outcome. Throughout, we argue that narratives and self are dialectically co-constructed. It is in the process of sharing the stories of our lives with others that we create a personal past, and this evolving life story is reflected in and emerges from everyday co-constructed reminiscing. Individuals who are better able to create elaborated, coherent narratives of their lives have higher levels of self and emotional understanding. Because this research is informed by earlier research examining the emergence of autobiographical narratives during the preschool years in parent–child dyadic reminiscing, we first briefly describe this research and the theoretical framework to place the family narrative research in perspective.

Early Parent–Child Reminiscing

Beginning in the first days of life, parents are already telling their infants the stories of their lives, well before infants are able to understand or participate in this activity (Fiese, Hooker, Kotary, Scwagler, & Rimmer, 1995). The telling of these stories signals that this is a valued activity, one that brings this new life into this family and this culture, and sets the expectations that this new life will have stories of her own to tell (Fivush, 2008; Pratt & Fiese, 2004). Children are encouraged to participate in telling the stories of their lives virtually as soon as they begin to talk (Eisenberg, 1985, Hudson, 1990). Even quite young children are guided to tell daddy what they did during the day, to tell grandma where the new toy came from, and to tell sister what happened at daycare. These early promptings to share one’s life are embedded in cultural values in which it is expected and encouraged to share the stories of one’s day and one’s life with others as a ways of connecting with others and creating emotional bonds.

Importantly, children are encouraged not just to tell what happened, but to reflect on thoughts and emotions about the event. It is not simply that one went to the park, but that one had fun on the swings; it is not just that one’s friend moved away, but that one is sad and upset about it. Events are interpreted and evaluated, and it is in this process that they come to have meaning for the self (Fivush & Baker-Ward, 2005; Fivush & Haden, 2005; Fivush & Nelson, 2006). As children begin to participate, even minimally, in parent–child reminiscing about the shared past, the ways in which parents scaffold, or guide these reminiscing conversations, influence children’s developing autobiographical memory skills. Highly elaborative mothers, who talk frequently and in rich embellished detail about the past, provide a coherent narrative structure for their children, helping them to move beyond the simple reporting of an event to creating a narrative replete with interpretation and evaluation. Children of mothers who are more highly elaborative in their reminiscing style come to tell more detailed and more coherent narratives of their own experience by the end of the preschool years (see Fivush, Haden, & Reese, 2006, Nelson & Fivush, 2004, Reese, 2002, for overviews).

Critically, given the theoretical relations posited between narratives and self and well-being, mothers with a highly elaborative reminiscing style also facilitate the development of self-understanding and emotional understanding. By the end of the preschool years, children of more highly elaborative mothers have a more differentiated and a more coherent sense of self (Bird & Reese, 2006; Welch-Ross, 1997, 2001), a more advanced theory of mind (Reese & Cleveland, 2006; Welch-Ross, 1997), a more sophisticated understanding of their own and others emotions, and show higher levels of emotional well-being (Laible, 2004a, 2004b, 2006) than children of less elaborative mothers.

Research on maternal reminiscing style has thus demonstrated that autobiographical narratives emerge in social interactions and that the form of these interactions is critical. Children of mothers who engage in highly elaborative and evaluative reminiscing come to have richer memories of their own past and more advanced understanding of self and emotions. Reminiscing and elaborating on emotional aspects of the past may be particularly critical in this process (Fivush, 2007). By the time children approach adolescence, and turn to the developmental tasks involved in constructing an adult identity (Erikson, 1959/1980; Harter, 1999), they have a long developmental and familial history of shared reminiscing.

The Family Narratives Project

In the Family Narratives Project, we examine how families continue to reminisce about the shared past as their children grow older and how this might be related to children’s developing understanding of self and emotional well-being as children undergo transition into an adult identity. We studied 40 middle class racially and ethnically diverse two-parent dual-earner families, each with at least 1 preadolescent child between the ages of 9 and 12. Most families had other children as well. Only 3 families had 1 child, whereas 17 families had 2 children, 13 families had 3 children, 5 families had 4 children, 1 family had 5 children, and 1 family had 6 children. Of the 40 families, 29 self-identified as Caucasian, 5 as African American, 1 as Asian, and 5 as interracial. Finally 30 families were considered traditional, 8 families were blended, and 2 families were extended with grandparents living in the household.

Families were visited in their homes when the entire family was present and asked to tell multiple stories together, and family dinnertime conversations were also recorded. Here we focus on family narratives of shared stressful experiences (but see Bohanek et al., in press; Fivush, Bohanek, Robertson, & Duke, 2004; Fivush et al., 2008, for discussion of other aspects of these data). Families sat together in a comfortable place of their choosing, usually the living room, with a tape recorder placed between them. The research assistant left the room, or sat in a corner, and the family selected the events to talk about; no time limits were imposed. About 30% of the families discussed the death of a family member or friend, 22% discussed illness or injury, 20% discussed the death of a pet, 20% discussed accidents or disasters, and 7% discussed a family conflict, a move to a new city, or mishaps during a vacation.

At the end of the visit, we asked the target child to complete measures of self-esteem (the Rosenberg self-esteem inventory, Rosenberg, 1965) and mothers were asked to complete the Child Behavior Checklist (CBCL; Achenbach, 1991) as a measure of their child’s internalizing behaviors (e.g., anxiety, depression, withdrawal) and externalizing behaviors (e.g., aggression, acting out, substance abuse). Two years later, we again visited the families in their homes, although only 24 of the original families were able to participate. At this visit, children were asked to complete The Self-Perception Profile for Children (Harter, 1985) from which we derived measures of self-concept and self-competencies. We focus here on social competence and academic competence because these are rated as the most important competencies by preadolescents and early adolescents (Evangelista, 2001).

We examined the family narratives in several ways to address different questions of interest. We first examined the family as a unit. Stemming from family systems theory (Kreppner, 2002), we reasoned that most spontaneous family narrative interactions are complex co-constructions with multiple family members participating and the ways in which any one family member engages in this task will be a function of the other family members. Thus, examining the family as the unit of analysis allowed us to explore how the family as a system co-constructs the past together. We examined both the process of narrative interaction, that is, how the family co-constructs the narrative together, and the content of narratives, focusing specifically on the emotional content. As already discussed, narratives are a critical mechanism for processing difficult emotions by providing a framework for understanding and regulating aversive affect. Research on early mother–preschool child reminiscing has demonstrated that highly elaborative reminiscing about emotional events is related to children’s developing emotional understanding and regulation (see Fivush, 2007, for a review), and thus we expected that expression and explanations of emotion within family narratives would be related to higher levels of self-understanding and well-being.

We then examined how mothers and fathers differentially contributed to family reminiscing. Research on individual narratives has found that, beginning in childhood and continuing through adulthood, females tell longer, more detailed, more emotional, and more relationally oriented narratives than do males (see Fivush & Buckner, 2003, for a review). These gender differences in personal narratives are related to gender differences in self-concept, with females being more relationally and emotionally oriented and males being more autonomously and agentically oriented (Gilligan, 1982), and underscore the evolving relation between personal narratives and self-understanding. Because females and males differ in their autobiographical narratives, we would assume that mothers and fathers would differ in how they reminisce with their children. Although there has been limited research on differences between maternal and paternal reminiscing, there is some suggestion that mothers are more elaborative and more emotional overall than are fathers when reminiscing with their preschool children (Fivush, Brotman, Buckner, & Goodman, 2000; Kuebli & Fivush, 1992); we wanted to examine possible parental gender differences with older children as well. In addition, previous research has found relations between maternal reminiscing and emotional well-being; thus, we wanted to expand this research to also assess fathers’ role in child outcome.

The Family as a Unit

Family Reminiscing Style

In order to explore the process of family narrative interaction, as well as to be able to directly compare these findings to previous research on parent–child dyadic reminiscing style, we coded the narratives using the standard coding scheme for reminiscing style (Fivush & Fromhoff, 1988; Reese, Haden, & Fivush, 1993), coding each utterance as an elaboration, defined as the inclusion of any new information, an evaluation, defined as confirming, validating, or (rarely) negating anothers’ contributions, or a repetition, defined as repeating what one said. Elaborations and evaluations, which tend to be highly correlated with each other (Bauer & Burch, 2004; Fivush & Vesudeva, 2002), are the hallmarks of an elaborative style. Evaluations serve the function of keeping the child engaged in the conversations, as well as validating and thus valuing the child’s contributions to the co-constructed narrative. Elaborations, at least for preschoolers, are the critical variable for predicting children’s developing autobiographical memory skills (Farrant & Reese, 2000) and also seem to be the variable most related to emotional outcome (Laible, 2004a, 2004b, 2006). Thus a highly elaborative and evaluative reminiscing style is one that both helps the child to create a more detailed, coherent narrative of the past event as well as imparts the positive value of sharing the past with others.

We also divided the narratives into sections focused on the facts of what happened versus the emotional aspects of the event, in an effort to examine the role that elaborative reminiscing plays in child outcome in more detail. Facts were defined as the objectively observable aspects of an event, such as who was there (e.g., “Did Dick go with us?”), what activities were engaged in (e.g., “And then we had to drive all the way to Minnesota for the funeral.”), and descriptions of objects and events (e.g., “And it was really quiet in the room.”). In contrast, emotional aspects of the events focused on emotional states and reactions (e.g., “I was really sad when Grandpa died”) as well as the ensuing discussion of that emotional state or reaction (e.g., “Yes, you cried and cried,” or “It was hard on all of us.”). Whereas emotional aspects of the narratives virtually always began with mention of an emotional state or reaction, the additional utterances coded as emotions may or may not have included a specific emotion word, but had to extend the conversation about emotion. It should be noted that discussion of other internal states, such as cognitions (e.g., “I thought it was strange” or “I don’t remember the trip.”), was actually rare in these transcripts, or else incorporated emotional language (e.g., “I thought it was so sad,” “I guess I was really upset by all that.”), and so was not examined separately.

We divided the narratives between facts and emotions because research with preschoolers has established that mothers who reminisce in more elaborative ways about stressful events have children with higher levels of emotional understanding and regulation (Laible, 2004a, 2004b, Laible & Thompson, 2000). But it is unclear from this research whether elaborating on emotion per se is helpful, perhaps by helping children understand and regulate aversive affect, or if elaborating on the facts of what happened, helping children to construct a more elaborated narrative of the event in general, is helpful. On one hand, elaborating on emotion may help children better understand and regulate aversive affect (Denham, 1998; Gottman, Katz, & Hooven, 1996), but if reminiscing about emotion simply ruminates on the emotional experiences, it is likely not to be helpful (Fivush & Buckner, 2000; Nolen-Hoeksema, 1987). In terms of the factual content, in expressive writing studies, in which adults are asked to write about stressful events for 10–20 minutes a day for several days, adults who construct more explanatory and causally connected narratives of stressful past experiences subsequently show higher levels of physical and psychological well-being (Frattaroli, 2006; Pennebaker, 1997; Smyth, 1998). The focus of this research has been on the use of specific causal and explanatory words, such as “because,” “therefore,” that convey a sense of coherence and explanation to why things happened as they did, as well as more cognitive insight words, such as “realize” and “understand.” Although we did not code for cognitive states, as discussed earlier, we reasoned that families that elaborate on the facts of what happened may be creating a more detailed and perhaps more temporally and causally coherent account of the past event.

Thus we examined family reminiscing style for factual and emotional aspects of the event separately as it relates to adolescent emotional well-being (McWilliams, 2007). We summed across family members and analyzed the number of elaborations, evaluations, and repetitions that families provided about the factual and emotional aspects of events in a repeated measures analysis of variance. Main effects and interactions of type of utterance (elaborations, evaluations, and repetitions) as well as type of content (factual versus emotions) were followed-up with t-tests on each utterance type by content. Not surprisingly, families elaborate and evaluate more about the factual aspects of negative events than the emotional aspects (see Table 3.1 for means and standard deviations). There is simply more factual information to include about past experiences than emotional information. More interesting, families that elaborated and repeated more when reminiscing about the emotional aspects of negative events had preadolescents with concurrently higher levels of internalizing (r = 0.29, p < 0.10 for elaborations and r = 0.34, p < 0.05 for repetitions) and externalizing behaviors (r = 0.26, p < 0.10 for repetitions), whereas families that elaborated and evaluated about the factual aspects of negative events had preadolescents with lower levels of internalizing behaviors (r = −0.29, p < 0.10 for evaluations) and externalizing behaviors (r = −0.30, p < 0.05 for elaborations, r = −0.25, p < 0.10 for repetitions, and r = −0.24, p < 0.10 for evaluations). Thus, it seems that co-constructing a more factually oriented elaborated and evaluative narrative that helps delineate the who, what, where, and when of a stressful event might be beneficial for children, but focusing on emotions may be detrimental, especially if that focus is repetitive.

Table 3.1 Means (and standard deviations) for reminiscing style and emotional content variables for the family, mothers and fathers
Full size table

Family Reminiscing About Emotion

In order to examine relations between reminiscing about emotions and adolescent well-being in more detail, as well as longitudinally, we did a more fine-grained coding of just the emotional content of these narratives (Marin, Bohanek, & Fivush, 2008). We coded all emotional utterances as either an expression of emotion or an explanation of emotion, as displayed in Table 3.1. For example, stating that one was sad would be an expression but stating that one was sad because Grandma died would be an explanation. We also coded whether families referred to specific emotions (e.g., “We were sad.”) or general affective terms (e.g., “That was hard on us.”). Families that expressed and explained more specific negative emotion (but not general affective terms) when reminiscing about negative events had adolescents who, 2 years later, showed higher self-esteem (r = 0.31, p < 0.12 for expressions and r = 0.33, p < 0.12 for explanations) and higher levels of social competence (r = 0.44, p < 0.05 for expressions and r = 0.41, p < 0.05 for explanations) and academic competence (r = 0.35, p < 0.05 for expressions and r = 0.36, p < 0.10 for explanations.

Together, these two sets of analyses suggest that families that create highly elaborative narratives about the factual aspects of shared negative experiences have adolescents with higher levels of self-understanding and well-being, but simply elaborating and repeating on emotions seem to be detrimental, perhaps because this leads to a more ruminative emotional style (Nolen-Hoeksema, 1987). However, integrating specific emotions into an explanatory framework is beneficial for adolescents. Thus, families that co-construct narratives of stressful experiences that provide elaborative detail of what happened and explanation of specific negative emotions seem to help their adolescents understand and regulate aversive emotional experiences, and thus these adolescents show higher levels of well-being.

By co-constructing more elaborative narratives, families may be helping their adolescents to create more detailed and possibly more coherent narratives of these stressful events. Although we could not code for overall coherence in these co-constructed narratives due to their length and complexity, we would argue that through elaboration, families are constructing richer and more detailed narratives that help the adolescents to place the event in larger life contexts and explanatory frameworks. This may be similar to what is found in the expressive writing literature, in which individuals who are able to provide more temporal, causal, and explanatory words subsequently show better outcome. The use of temporal, causal, and explanatory words suggests a more elaborated description of who, what, where, and when, similar to what we see in elaborated family narratives.

It is also intriguing that it is the use of specific negative-emotion words that is related to adolescent well-being and not the use of more general or vague emotion terms. Previous research has not addressed emotion specificity, but our results suggest that families who discuss specific emotion may be modeling a more nuanced and differentiated understanding of emotion than families that discuss more general emotional states. It may be that families that are able to focus on specific emotional states and reactions, and provide a framework for understanding and regulating these specific emotions, are helping their adolescents to learn specific emotional coping and regulation skills. Discussing more general or vague emotional states and reactions may be a way of distancing from emotion that ultimately would not be helpful in coping and regulation.

Family Reminiscing as a Gendered Activity

Whereas examining the family as a unit allowed us to explore the dynamic co-construction of shared events, we were also interested in how mothers and fathers might differentially contribute to these co-narrations as well as to adolescent well-being. As mentioned earlier, gender differences in personal narratives emerge in childhood and are maintained throughout adulthood, with females telling longer, more detailed, more emotional, and more relationally oriented narratives than males (Bauer, Stennes, & Haight, 2003; Buckner, 2000; Friedman & Pines, 1991; MacDonald, Uesiliana, & Hayne, 2000; Shaw & Edwards, 1997), and there is some suggestion that, when comparing mother–child and father–child dyadic reminiscing, mothers are more elaborative and more emotional than fathers (Bohanek et al., in press; Fivush et al., 2000). Examining the ways in which mothers and fathers individually reminisce within the family as a whole allowed us to gain a better perspective on the role that mothers and fathers play within the family in creating shared narratives of the past. Thus we examined both reminiscing style, in terms of elaborations, evaluations, and repetitions, and emotional content, in terms of expressions and explanations, for mothers and fathers separately. Note that this set of analyses differs from previous research on parent–child gender differences in reminiscing because in previous research mothers and fathers were compared across independent narrative conversations with their children, whereas in these analyses, the whole family was reminiscing together, and the differential contributions of mothers and fathers to these family narratives were examined.

Parental Reminiscing Style

We coded all maternal and paternal utterances as elaborations, evaluations, and repetitions for both the factual and emotional aspects of events (Fivush, Marin, McWilliams & Bohanek, 2009). Repeated measures analyses of variance were conducted examining the number of elaborations, evaluations, and repetitions that mothers and fathers provided for the factual and the emotional aspects of the narratives. Main effects and interactions of parental gender, type of utterance, and content of narrative were followed-up with comparisons of mothers and fathers on each utterance type by content. As shown in Table 3.1, mothers were more elaborative than fathers, but mothers and fathers were equally evaluative. That mothers are more elaborative confirms previous findings examining parent–child dyadic reminiscing with preschoolers and suggests that mothers are playing more of a role in helping to construct the shared past, although it must be noted that fathers are involved in creating a validating narrative interaction. High levels of maternal elaborations further suggest that mothers are playing the role of “kin keepers” in the family, creating and maintaining the family stories and family history in order to preserve a family identity over time (Sherman, 1990; Wamboldt & Reiss, 1989).

Intriguingly, higher levels of maternal elaborations and evaluations about the facts of the negative events were concurrently related to lower levels of adolescent internalizing (r = 0.30, p < 0.10 for elaborations) and externalizing behaviors (r = 0.30, p < 0.10 for evaluations), but maternal reminiscing about the emotional aspects of negative events was not related to adolescent well-being. Fathers, in contrast, who elaborated and evaluated more about the emotional aspects of negative events had adolescents with higher levels of internalizing (r = 0.29, p < 0.10 for elaborations) and externalizing behaviors (r = 0.41, p < 0.05 for evaluations). These patterns suggest that the beneficial effects we saw for family reminiscing on adolescent well-being are being carried by maternal reminiscing; mothers who help their adolescents to create more coherent, detailed narratives of who, where, what, and when about stressful events are beneficial for adolescents. However, the finding that fathers who elaborate and evaluate on the emotional aspects of events was related to lower well-being was puzzling. Note that because we examined mothers and fathers within the same narrative conversations, this cannot be explained by differences in what these narratives were about; both mothers and fathers were contributing to the same co-constructed family narratives on the same topic. However, mothers and fathers could be contributing different kinds of information and in different ways to the ongoing co-constructed narrative. Therefore we turned to a closer examination of what mothers and fathers might be contributing to the emotional content, as well as examining these relations longitudinally.

Parental Emotional Content

For those parts of the narratives that focused on emotion, we counted all expressions and explanations for mothers and fathers separately (Bohanek, Marin, & Fivush, 2008). We conducted repeated measures analyses of variance on parent gender, type of utterance, and type of content, and main effects and interactions were followed-up with analyses examining parental gender by each utterance type by type of content. As shown in Table 3.1, mothers expressed and explained more emotion than did fathers. Again this confirms previous findings that females generally tell more emotionally laden personal narratives than do males, as well as extending previous findings that mothers reminisce more about emotions with their preschoolers than do fathers (Kuebli & Fivush, 1992). Most intriguing, longitudinal relations between parental narrative style and adolescent well-being were clearly related to gender. (Note that because we only had 12 girls and 12 boys at the 2-year follow-up, we focused on the effect sizes of the correlations [see Kline, 2004; Vacha-Haase, Nilsson, Reetz, Lance, & Thompson, 2000; Vacha-Haase & Thompson, 2004, for discussions of this issue]. Because of the exploratory nature of the data and the small sample size, both small effect sizes (r ≥ 0.31) as well as moderate effect sizes (r ≥ 0.55) are considered meaningful.)

Mothers who were more emotionally expressive and explanatory when reminiscing about negative events had daughters with higher self-esteem (r = 0.33 for expressions) and sons with higher self-esteem (r = 0.62 for explanations) and lower internalizing behaviors (r = −0.35 for explanations). Similarly, fathers who were more expressive and explanatory when reminiscing about negative experiences had sons with higher self-esteem (r = 0.64 for expressions) and fewer internalizing behaviors (r = −0.49 for explanations). In contrast, fathers who were more emotionally expressive when reminiscing about negative events had daughters with higher levels of internalizing (r = 0.44 for expressions) and externalizing behaviors (r = 0.38 for expressions).

One issue in this type of research, of course, is causal influence. As we discussed in our introductory remarks, we frame this research dialectically, arguing that narratives and identity are reciprocal. Similarly, we argue that by examining the family as a whole, we are examining the ways in which individual family members mutually and reciprocally influence each other in the co-construction of the narrative. This, of course, raises the question of whether parents are facilitating identity and well-being in their adolescents or if adolescent identity and well-being are soliciting parental style. We argue that this, too, is a dynamic system that is reciprocally interactive. Thus it may be that daughters who, for whatever reasons, may be having more difficulties with identity issues and emotional regulation solicit more emotional expressions and explanation from their fathers. We do not rule out this possibility, but argue that the direction of effect is likely stronger from parent to child than from child to parent for two reasons. First, longitudinal research on parental reminiscing style in relation to child narrative skills and emotional well-being in early childhood has shown that the direction of influence is substantially greater from parent to child than from child to parent (see Fivush et al., 2007, for a review and discussion). Second, many of our correlations are longitudinal, suggesting relations between earlier measures of parental narrative co-construction and later measures of adolescent well-being. Still, it is important to recognize that children are active agents in their own socialization and that theoretical and empirical models that posit more dynamic reciprocal influences are needed.

Summary of Family Narratives

Overall, the patterns suggest that reminiscing is a gendered activity (see Bohanek et al., 2008, for a full discussion); mothers elaborate more than do fathers and mothers express and explain more emotions than do fathers. That mothers do the majority of the emotion talk in the context of family narratives is not surprising. Mothers generally do the majority of “emotion work” in the household (Hochschild, 1979), and family narratives are a critical context for creating emotional meaning and a shared history (Fivush, Haden, & Reese, 2006). Similarly, mothers are responsible for being the “family historian,” for keeping track of family events and milestones, and to place these events in an ongoing narrative of family life (McDaniel, 1999; Sherman, 1990; Wamboldt & Reiss, 1989). Thus, doing more of the elaborations and emotion work within family narratives may be part of the larger gender role that mothers adopt in terms of creating a sense of emotional integration and stability for the family.

Moreover, family narratives are related to adolescent well-being, and this, too, is gendered. Mothers who elaborate on the facts of what happened have adolescents with higher levels of well-being. And, although simply elaborating on emotion is related to lower well-being, perhaps because this facilitates a ruminative style, mothers who express and explain more emotion associated with stressful events have adolescents with higher levels of well-being. Thus, mothers who help their adolescents create more detailed and therefore possibly more coherent narratives of the who, what, where, and when of stressful events, and who help their adolescents to understand the emotional causes and consequences of stressful events, have adolescents, both daughters and sons, with higher levels of well-being.

Father’s role, however, appears to be linked to gender of child. Fathers who express and explain emotions have sons with higher levels of well-being but daughters with lower levels of well-being. Few studies have examined the father–child relationship, particularly in relation to children’s self-esteem and emotional and behavioral adjustment, but there is some evidence that the father-child relationship is much different than the mother–child relationship (Parke, 2004), especially during the adolescent period. Children tend to have a closer and warmer relationship with their mothers than they do with their fathers (LeCroy, 1988; Steinberg & Silk, 2002). The father–adolescent relationship, and especially the father–daughter relationship, is strained. Adolescent girls report more of a distant relationship with their fathers than do adolescent boys (Hill, 1988); sons report being much closer to fathers than daughters do during adolescence (Steinberg & Silk, 2002).

Moreover, children may recognize that their mother and father each have different strengths, which may lead them to turn to a particular parent for help with certain types of problems. Specifically, both sons and daughters report going to their mothers for more emotional and relational issues, whereas they turn to fathers for information and material support (Steinberg & Silk, 2002). Related to this, fathers tend to foster autonomy in their adolescents through discussions of achievements and successes (Fiese et al., 1995). Although family narratives may serve many different goals, in this study, we specifically asked families to discuss highly emotional experiences together, and therefore, in this context, family narratives may serve a more relational or affiliative goal, that of bonding the family together through hard times. In this context, then, it is possible that maternal scaffolding of highly emotional narratives of family relationships may be related to children developing a more emotionally coherent sense of family and self. In contrast, the father’s expression and explanation of emotion in this context may be antithetical to his usual role in the family as fostering independence and problem-solving (see Finley & Schwartz, 2006, for recent empirical support on this issue).

This interpretation may help explain the gender differences in relations between paternal reminiscing and adolescent outcome. Because negative events create problems that need to be resolved, it is possible that fathers may take on a more problem-focused approach in these narratives. Thus one reason we may see consistency in maternal reminiscing and outcome in both daughters and sons is because in reminiscing, mothers are playing their acknowledged role in the family, both as the family historian and as the one who provides emotional scaffolding and support. Fathers, on the other hand, may play very different roles for sons and daughters especially during early adolescence when gendered identity becomes critically linked to romantic partners. For sons, who are striving for autonomy, paternal focus on emotional explanation may provide a healthy model for coping with aversive affect. Daughters, in contrast, may be looking to their fathers to provide strength and protection, and a focus on emotion may undermine daughter’s perceptions of their fathers as strong and unflappable. Obviously, this argument is speculative and clearly tied to developmental stage. Arguing from developmental theory, we would propose that it is only in this transition into an adult gendered identity that daughters may have problems with emotionally expressive fathers. Indeed, we might even argue that, long-term, emotionally expressive fathers may play a positive role as their daughters consolidate a more nuanced and mature gendered identity. The need for additional research is obvious.

Conclusions and Implications

Family reminiscing is a frequent and critical part of family life and has implications for adolescents developing sense of self and well-being. By focusing on the family as a whole, examining the process and content of family narrative interactions, we have demonstrated that families that are able to create coherent, elaborated narratives of stressful family events, and that express and explain specific negative emotions, have adolescents who display higher levels of self-esteem and emotional well-being. However, our research has also highlighted the gendered nature of reminiscing. During family narrative interactions, mothers play a larger role than fathers in scaffolding and supporting more elaborative and emotionally expressive and explanatory family narratives, and mothers who are more elaborative and emotionally expressive and explanatory have adolescents with higher levels of self-esteem and emotional well-being. Fathers’ role in family narrative interaction is more complex; elaborative, emotionally expressive, and explanatory reminiscing is beneficial for sons, but detrimental for daughters. While the interpretations we present in this chapter are clearly speculative, our results point to the critical role of family narratives in adolescent development, and the ways in which family reminiscing is a gendered activity.