1. Introduction
Psychologists have usually studied sex differences as qualities inherent in the individual rather than as a social or contextual issue. Many see gender differences as related to biological sex, pre-determined or a part of our genetic make-up. Others agree that there are social forces at work to influence people in the direction of gender-appropriate behavior, but still see these forces as creating stable structures internal to the individual.
The goal of this study is to examine how boys and girls use discourse markers in conversation. Discourse markers can index the conversational activity context in which boys and girls are engaged, and allow us to compare gendered speech within that context. This provides an opportunity to consider the role of context rather than internal forces in gender differences observed in conversation.
Much research has been dedicated to the differences between men and women's styles of speech. Male and female speech patterns have been portrayed as if men and women speak two different languages and even come from two different planets.
It has been argued that the difference between men and women's speech patterns is a cultural difference. Young children segregate into gender-separate play groups, and may be given quite different responsibilities. Because there is little contact between boys and girls after this separation, they are socialized quite differently, developing different values and norms. These cultures become the crucible from which emerge differences in conversational, social and intellectual style. The study of these differences dominates much research on gender.
2. Cultural training vs. situational context
There is another viewpoint, however, which frames gender as a social construction and argues that the situations in which conversation occurs may constrain speech behavior, as opposed to seeing gender differences as dispositional. Gender is about group identity in a given context and there are times when it is relevant and times when it is not. When gender becomes relevant, then boys' and girls' speech will be different. At the times when it is not, other aspects of the social situation tend to dominate, and boys and girls will be more similar in their speech. Thorne also suggests that the separate worlds of boys and girls may not be so separate after all. Although sex-segregation in elementary school children is pervasive, there are still settings where boys and girls are together, providing contexts that encourage same-gender behavior. The social context can provide social spaces for the performance of gender, such as in interactive song games. These song games create a shared structure in which gender and sexuality are negotiated and co-constructed by the participants.
Thorne observed that children often interact in contexts where gender is organized dichotomously, with boys and girls defined as opposing sides. In these contexts, gender becomes highly salient. In contexts where there are cues for socialized sex-roles, children will also be constrained into gender-appropriate play scenarios. Yet there are times when gender may be low in salience, like in a sandbox, in cooperative activities where boys and girls work together, or even when they run in circles and scream. At these times, boys' and girls' behavior is much more similar than different. For these reasons, we must take context into consideration when interpreting girls and boys' conversational patterns.
Sheldon found that in disputes, girls are much more assertive than previously thought, although they masked their assertiveness through mitigation. In her study of preschool age girls, she found that they used "double-voiced discourse," as a way to be confrontational, while softening the social risk of conflict through using indirect speech. Goodwin observed that within a septic context (jump rope), gender was not as important as was expertise for predicting who would use direct speech and tell others what to do. In the use of directives, she found that both boys and girls directly told others what to do if they were more skilled in the activity at hand. Aronsson and Thorell also found that both boys and girls used similar strategies in threat-tell sequences even though they demonstrated knowledge of expected gender roles in disputes. Although children saw the male figures in family disputes as acting more directly than female figures, children of both genders engaged in aggravation and mitigation.
Kyratzis and Guo also challenged the assumption of the universality of gender differences in speech style. In conflict situations, they found that Mandarin-speaking Chinese girls were quite assertive in an argument, and the boys quite compliant. American boys and girls had opposite patterns and were more extreme. Assertiveness in girls has also been observed among Latino girls in cooperative learning contexts and in the bilingual code-switching conversation of Swedish children, in which both boys and girls did not hesitate to engage as adversaries. Thorne presents a challenge to researchers in the field of gender by saying that the "why" question about gender differences is less productive than the "how" question. How do these gender differences come about? How do children express common goals when there is not a gender difference, and when there is?
3. Discourse markers
3.1. Pragmatic form and social index
Discourse markers organize a conversation and index social and propositional moves within the spoken discourse. Schiffrin defines discourse markers as "elements that signal relations between units of talk by virtue of their sequential relations as initial or terminal brackets demarcating discourse units." Discourse markers (DMs) organize discourse at multiple levels and index relationships among people, turns, topics, activities, ideas, and more. The meaning they convey is a relational one, marking the shifts and changes of a dynamic exchange. Discourse markers are not like the other words in an utterance: they are vocabulary without reference and do not have propositional meaning. Instead, DMs signal changes that happen in the flow of discourse; in that way, they have pragmatic meaning, but do not derive their meaning from reference. For the purpose of this study, we will define DMsalong the lines of Schiffrin and use the following criteria:
1. A discourse marker is not grammatically obligatory. You can remove the form from the utterance and preserve both correct syntax and the propositional content.
Example: "I can't have a pet 'cuz my mom is 'lergic." With marker removed: "I can't have a pet. My mom is 'lergic"
2. A discourse marker is commonly used at the beginning or end of a conversational turn. Although a marker may occur in the middle of a turn or between clauses, the form that is used must also be possible as a turn-initial or turn-final marker. The following example includes a turn-initial use of because:
Example: "I don't like tuna"
Response: "Because tuna is yucky."
3. A discourse marker does not add propositional meaning to the utterance: a form must not gain meaning from reference if it is to be defined as a discourse marker. While ideational markers may mark a relationship between propositions, discourse markers may add pragmatic or social meaning to an utterance. In the following example well adds a sense of politeness but not content:
Example: "Are you free on Friday night?"
Response with marker: "well, no, I have plans"
Response without marker: "no, I have plans"
3.2. Functions of discourse markers
DMs can function at many levels of discourse, and sometimes can mark several levels simultaneously. A DM can mark a shift in the dynamics of the interaction, of the function of the conversation, of the phase of the activity that is being currently engaged in. Schiffrin discusses five different levels at which discourse markers function:
1. Exchange: Indexing shifts at the level of the turn.
Discourse markers can index shifts at the level of the exchange. At this level, a DM manages the exchange of conversational turns. Any move to gain the floor, hold the floor, get attention, or otherwise elicit or initiate a conversational move would be marking at the exchange level.
In (1), below, the discourse marker occurs at the beginning of the turn, to get attention and gain the floor. But markers can also occur at the end of a turn in the form of a tag question, to surrender a turn and confirm the participation of the addressee, as in (2).4. Participation: Marking shifts in the roles of participants.
(1) M (4 yrs.): "Oh, I have peanut butter... Look, I have peanut butter"
(2) J (3 yrs.): "We can be power rangers, ok?"
2. Action: Marking moves designed to accomplish an action.
Discourse markers can also mark moves that are meant to accomplish some kind of action. These functions are mainly found to mark control acts, that is, speech acts that are meant to control or influence another's behavior. Discourse markers index the relation between the utterance and the desired behavior on the part of the other person. Examples include such control act moves as marking arguments, requests or bargaining and justifications, as in (3).
(3) D (4 yrs.): "It's mine because I waited my turn"
3. Ideation: Organizing and marking relations between ideas.
Discourse markers can also organize and mark relations between ideas. This level of marking indexes a relation between the meaning of two clauses or ideas, like marking relational cohesion in a narrative. The plane of ideation is fundamentally referential, involving features such as conversational topics, naming, deixis, and propositions. Ideation refers directly to the informational content of the utterance. In the previous example (3), because is not acting as a causal connective but marks a justification for a claim (action). In (4), because is marking a causal relationship between two propositions (ideation).
(4) D (4 yrs.): "I can't have a pet 'cuz my mom is 'lergic"
4. Participation: Marking shifts in the roles of participants.
At this level, discourse markers index a shift in the roles of the participants in the talk, or introduce upcoming changes in episode or phase. This primarily occurs as a boundary-marking function, especially between phases or kinds of activities. It can also mark a shift in the social roles of the speakers, or indicate an adversarial or affiliative position or a register shift in role-play. In (5), a child is using the marker but (along with a shift in voice) to mark a shift in role-play register, doing her best Elvis impersonation:
5. Information Management: Checking on the flow of information.
Some DMs described by Schiffrin mark information state. At this level, the marker indexes the relationship between the ideas of the speaker and the addressee. It is the speaker's way of checking in with the addressee to confirm that the information contained in a statement is being understood. It also alerts the addressee to the expectations of old and new information on the part of the speaker.
8. Conclusion
The gender differences in children's conversations may be more about activity than a difference in actual conversational goals. Both boys and girls bargain, and when they do, they use strikingly similar action level discourse-marking functions. When they tell stories, both boys and girls mark ideation. In using discourse markers to index the conversational goals of these children, it appears that for this form, activity accounts for gender differences to a great degree.
Discourse markers, however, are limited in the scope to which they can tell us about the interaction itself. This study suggests that gender differences may actually reflect the contextual demands placed on the child, and may be more descriptive of situated behavior, rather than indexing essential gender differences. The study of more complex conversational interactions may give a better perspective into the relationship of gender and language. For instance, the observed gender difference in activity participation presents additional questions. Although the "why" questions may seem the most compelling, Thorne's challenge to ask the "how" questions may prove to be more fruitful. How do children select their conversational activity? The answer to that question may give us not only a better understanding of what role gender plays in creating social contexts, but also an understanding of what role social context plays in the construction of gendered behavior.