Daniel Ginsberg | American Anthropological Association (original) (raw)

Ph.D. Thesis by Daniel Ginsberg

The practice of mathematics education is fundamentally multimodal. It incorporates not only talk ... more The practice of mathematics education is fundamentally multimodal. It incorporates not only talk and embodied action, but also technical notation and diagrams, brought into discourse through verbal and gestural reference. As this interplay of semiotic systems arises in interaction, it can be interpreted by analyzing sequences of talk, writing, and gesture, but a better understanding requires an ethnographic perspective to contextualize interaction with reference to its physical surroundings, institutional setting, and enduring relationships within the community. Thus, classroom interaction is best understood as a multimodal ecology in which micro-level discursive practices, the history of a community, and the biographies of its members mutually influence and determine one another.

Beginning from the ecological perspective on classroom interaction, this dissertation presents an analysis of observations and video recordings collected during a semester’s multi-site ethnographic fieldwork with both a middle school math class for English learners and a quasi-remedial college calculus section. To model how any perceptible feature of the environment may be taken as meaningful, I draw on the semiotic theory of C. S. Peirce, not only in interpreting observational data, but as an organizing principle, as the analysis moves from qualities, to particular instances, to recurring patterns. First, I consider the ontological status of mathematical notation as a quality of interaction, investigating its capacity for representing continuous phenomena. Second, I take up actually occurring sequences of interaction, showing how students’ participation in conversational repair offers insights into ideologies of classroom authority and mathematical knowledge. Third, I address students’ and teachers’ identities in the classroom as social perceptions that are constructed and recognized through patterns of interaction. Each area of inquiry is then reconsidered to identify semiotic affordances that are made available in classroom interaction, and to explicate problems of practice that prevent students from seeing themselves as successful mathematics learners. The problems I identify are similar to those addressed by mathematics education researchers in the learning sciences, so I conclude by proposing a future research trajectory combining linguistic anthropology, the learning sciences, and classroom teaching practice.

Papers by Daniel Ginsberg

Educational linguistics can have little impact on student learning unless it is effectively conve... more Educational linguistics can have little impact on student learning unless it is effectively conveyed to and applied by teachers in actual classrooms. This chapter represents teachers’ perspectives on linguistics in the classroom, from three authors with many years of experience teaching linguistically and culturally diverse learners at the high school level. Currently, Kimberly is a teacher educator, Daniel is an educational linguist and anthropologist, and Iris is a practicing classroom teacher. The vision of linguistically-informed education cast by Fillmore and Snow captures many of the ideas about language instruction that we have tried to embody in our classrooms over the years. However, we are aware that these approaches to teaching are not practiced in all classrooms, and we recognize the daily barriers faced by teachers and students in a world that does not share our perspective. In this chapter, we describe our approaches to linguistically-informed teaching, the barriers we have encountered, and our recommendations for teacher education, professional development, and future research.

... I struggled to understand how the students who had produced such sophisticated research paper... more ... I struggled to understand how the students who had produced such sophisticated research papers, who had led such engaged class discussions, had come to feel so poorly served by me and so demoralized by the course. I studied their comments closely, trying to square the experience they described with my own memories of the semester...

Students and teachers alike commonly view mathematics as an objective discipline to be learned th... more Students and teachers alike commonly view mathematics as an objective discipline to be learned through rote memorization. This view of the field is rooted in a traditional ideology of classroom roles in which any question has exactly one answer and the teacher is positioned as the ultimate authority. The case for reform has been made to deepen students’ engagement with mathematics content, but this paper identifies another justification: to provide opportunities for learners to develop proficiency in academic English. I analyze video-recordings of a middle school mathematics class for English learners in which the teacher knows mathematics and language pedagogy, integrates language and content instruction, and incorporates accepted best practices for teaching English through mathematics, such as explicitly teaching technical vocabulary and “unpacking” word problems. Considering how these techniques are instantiated through talk-in-interaction, however, shows that students are given a limited degree of semiotic agency over meaning-making resources such as classroom discourse; even when students evaluate one another’s work, their evaluations become tools for reinforcing teacher authority. Socializing students to academic language requires changes in ideology, not simply technique; only by decentering their own authority will teachers make space for students to engage in complex mathematics discourse.

University students are peripheral participants in various institutional communities, including t... more University students are peripheral participants in various institutional communities, including the university itself as well as the professional communities for which they are preparing. As students are socialized to these communities, their processes of identity navigation become visible in their narrative practices. Using interview and classroom observational data, this paper examines undergraduate, law, and business students’ self-positioning relative to institutions and individuals in their academic and professional lives. We conclude that students envision types of successful members of imagined professional communities, and that their academic motivation and success depend on their positioning their future selves as tokens of that type.

Mathematicians understand their discipline to be a process of logical reasoning about abstract ma... more Mathematicians understand their discipline to be a process of logical reasoning about abstract mathematical objects such as numbers, sets, and relations between them, through which discoveries may be made with potential implications for our understanding of the physical world. Students, on the other hand, often conceive of mathematics as a collection of algorithmic techniques for manipulating formal notation, disregarding the concepts that those symbols might represent; as a result, the study of math is seen to be nothing more than a sequence of progressively more sophisticated manipulations that enable the student to answer more complex questions posed by the teacher or textbook. While this aspect of mathematics is the most readily apparent to students, practitioners view it as the least interesting part, talking derisively about "plugging and chugging," the mechanistic computation of solutions to equations, as well as "drilling and killing," a style of teaching that presents students with long sets of plug-and-chug problems in order to increase their fluency with the notation.

A. Baczkowska & R. Zimny (eds), Impoliteness in Media Discourse, 2014

Among internet users, online discussions about controversial topics are known for quickly turning... more Among internet users, online discussions about controversial topics are known for quickly turning acrimonious, but as Locher (2010) points out, this aspect of computer-mediated communication has received little scholarly attention. To investigate impoliteness online, the present study follows Eelen (2001) in proposing that there can be no meaningful objective definition of impolite language, focusing instead ‘on the production of (im)politeness evaluations’ (Eelen 2001: 249). A corpus of over 20,000 words was collected, comprising comments on articles from the Washington Post local news blog dealing with the coexistence of cyclists and drivers on city streets. Each post was tagged for alignment and politeness/impoliteness, reflecting the judgments of the researcher; judgments from other readers were collected as well, to gauge the extent to which generalizations can be made. Word and bigram frequency analyses were performed along each dimension (cyclist vs. driver; impolite vs. polite) as well as between cells (impolite cyclist vs. impolite driver, etc.) to identify words and phrases that show a statistically significant difference of frequency across categories. These results were then analyzed qualitatively to investigate the underlying ideology of each position. The final analysis suggests that impoliteness is interpreted as a resource for claiming group membership or affiliation.

Language and Linguistics Compass, May 2011

This paper reports on a pilot project that explored the potential of linguistic inquiry in a high... more This paper reports on a pilot project that explored the potential of linguistic inquiry in a high school English as a Second Language (ESL) class. In class meetings across the school year, students worked collaboratively to investigate noun phrase pluralization, language acquisition, writing systems, and translation in their own and other languages. Classroom observations and students’ oral and written work provide evidence that:•Examining the structures of the spoken and written languages represented in the ESL classroom captures students’ interest and engages them in critical inquiry about the nature of linguistic knowledge and their beliefs about language.•The cross-linguistic analysis of students’ home languages validates their languages in the school context, defining them as a rich resource worthy of study, rather than as a hindrance to education.Examining the structures of the spoken and written languages represented in the ESL classroom captures students’ interest and engages them in critical inquiry about the nature of linguistic knowledge and their beliefs about language.The cross-linguistic analysis of students’ home languages validates their languages in the school context, defining them as a rich resource worthy of study, rather than as a hindrance to education.These findings are of particular significance in this time of English-only education in the United States, when students’ home languages are often rejected in schools.

Conference Presentations by Daniel Ginsberg

Traditionally, mathematics is understood to be a field in which any question has a single answer ... more Traditionally, mathematics is understood to be a field in which any question has a single answer that is objectively correct—even outside of mathematics discourse, “two plus two equals four” is a prototypical example of a simple, obvious truth—and teachers are treated as absolutely authoritative sources of mathematical knowledge. This traditional ideology leads students to be positioned in the passive role of “received knowers” (Boaler 2002) whose task is not to think creatively or debate mathematical ideas, but simply to replicate the calculation procedures modeled by their instructor. This paper investigates the way the traditional ideology is produced and reproduced in classroom interaction. I consider a situation in which students seem to be positioned in agentive roles: college calculus students attempting to correct their professor. Theories of epistemic status (Heritage 2013) and representational agency (Kockelman 2007) are used to demonstrate that even in these situations, students’ agency with respect to mathematics discourse is highly constrained. At best, students are able to make claims about the material, but the instructor is always the one to bring mathematical objects into discourse and to validate processes of mathematical reasoning. The extent to which students accept this non-agentive positioning contributes to their sense of alignment with the discipline of mathematics, and ultimately to their level of success in mathematics classes.

Language most naturally expresses categorical distinctions among types of phenomena rather than g... more Language most naturally expresses categorical distinctions among types of phenomena rather than gradual variation along a continuum, functioning in a “typological” rather than “topological” fashion (Lemke 1999). Deacon (2003) argues that other universals such as unbounded recursivity are in fact due to semiotic processes underlying symbolic reference itself; if this is the explanation for the categorizing functioning of language, then we should expect to find typological representation not only in all languages, but in all possible symbolic systems. To model continuous phenomena in a way that language cannot accomplish, many technical fields use mathematical formalism as a supplemental semiotic system. Since this formalism demonstrates that symbolic systems are potentially capable of representing continuity, how do we explain the fact that no language does? This paper addresses this question by analyzing video-recorded interaction in mathematics classrooms, where participants reinterpret the formalism as a typological system, for example, understanding a continuous function by calculating specific values, or algorithmically manipulating chunks of notation. This observation leads us to reconsider the relationship between cognition and semiosis: even though algebraic notation encodes a type of meaning that no language can, participants in mathematical discourse may perceive it in linguistic fashion rather than taking full advantage of the system’s unique affordances.

During postsecondary education, students undertake a process of identity navigation, which become... more During postsecondary education, students undertake a process of identity navigation, which becomes visible in their narrative practices. Having interviewed undergraduate, law, and business students, we examine their taking on of various social and enregistered voices to identify their self­positioning relative to other individuals and institutions, and the ideologies so expressed.

This study presents an examination of the structure of teacher talk in a middle school mathematic... more This study presents an examination of the structure of teacher talk in a middle school mathematics lesson in order to describe the features of classroom discourse through which the process of socialization is enacted. Starting from the observation that a lesson is a semantically and interactionally cohesive text of spoken language, I aim to identify the production resources related to academic content and social participation that teachers draw on to create cohesion. Teacher utterances are analyzed using both a discourse analytic and a functional linguistic analytical framework, showing how social and pedagogical discourse functions are realized at the word and clause level through the lexicogrammatical functions of Theme and reference. The analysis shows the complementary roles of forward-reaching talk such as topic initiation, backward-reaching talk such as revoiced student utterances, and outward-reaching talk such as reference to mathematical visual images. The use of outward reach is shown to be particularly important to mathematical discourse because of its specialized function in connecting linguistic and nonlinguistic meaning making resources to create and contextualize mathematical knowledge. These three discourse strategies work in combination to socialize students to both the local classroom discourse community and the broader context of mathematics as an academic discipline.

In this study, a probabilistic computational model is used to identify shifts in conversational s... more In this study, a probabilistic computational model is used to identify shifts in conversational style. “Conversational style” in particular, as distinct from “style” more broadly, is defined as the aspects of discourse that speakers use to mark involvement with their fellow participants in talk, following Tannen’s (2005) classification of “high involvement” versus “high considerateness,” that is, higher and lower levels of explicit marking of involvement in discourse. The computational model allows us to consider style shifting as something that is emergent from and co-constructed in discourse, rather than as individual speakers’ statistical preference or dispreference of a given linguistic variant.

Style is typically considered as intraspeaker variation, that is, the linguistic choices that an individual social actor makes in a given context to express group affiliation or otherwise conduct identity work (Bell 1984, Bell & Johnson 1997, Coupland 2007). In contrast, this study takes an approach inspired by Fred Erickson’s observation that “talking with another person ... is like climbing a tree that climbs back” (cited in Tannen 1989: 13); in other words, participants in discourse are constantly orienting and reorienting themselves to one another. Therefore, no attempt will be made to categorize one or another speaker’s individual stylistic choices. Instead, style is viewed as a property of discourse rather than of speakers, a moving target that is jointly constructed and subject to continuous change.

A hidden Markov model was chosen as the computational representation of this style-shifting hypothesis. To implement a Markov model, a number of discrete states are identified ― in this case, the high-considerateness and high-involvement conversational styles ― together with a transition probability distribution that specifies how likely the model is to move from any state into any other state, or to stay in the same state, as the process moves forward. In a hidden Markov model, the state is not directly observable; rather, each state could give a number of different possible outputs, which are predicted according to an emission probability distribution. In this study, the length, turn construction type, and speaker identity are modeled as outputs of the Markov process. Based on observations of the outputs, the model allows us to infer the underlying state that is most likely to have generated each turn in the transcript. These results can then be interpreted using both quantitative (linear regression) and qualitative (discourse analysis) methodologies.

This model creates an opportunity for a quantitative study of conversational style that cannot be achieved using traditional quantitative sociolinguistic methods. Rather than taking either turn length or turn construction type as a proxy for conversational style, it represents style through a complex probabilistic model that accounts for the ways speakers construct turns at talk throughout the discourse. Preliminary results suggest that the computational model can identify coherent stylistic episodes in discourse that feature the expected linguistic markers of high and low involvement. More broadly, this result suggests that computational methods can be successfully applied to the study of talk in interaction.

Talks by Daniel Ginsberg

You've all read my problem statement so I'm not going to spend time rehashing ittoday I'm going t... more You've all read my problem statement so I'm not going to spend time rehashing ittoday I'm going to look at this idea of "multimodal semiotics of mathematics teaching and learning" and look at two different directions the research is going at the moment. So I'm going to talk for five minutes about multimodal semiotics, and then five minutes about teaching and learning.

Teaching Documents by Daniel Ginsberg

Whether they know it or not, everyone is interested in sociolinguistics. Comments such as "People... more Whether they know it or not, everyone is interested in sociolinguistics. Comments such as "People from Philadelphia talk funny," "I didn't understand the point of that story," or "How was I supposed to know you were kidding?" are the kind of casual remarks that come up in day-to-day conversation, but behind each of them is an implicit theory about the way language and culture are connected -that a way of talking is connected with a geographical location, for example, or that a good story should have an identifiable "point." As participants in the social world, each of us does this kind of theorizing all day, every day, but social scientists have developed philosophies and methodologies that allow us to conduct more principled investigations into questions such as "What does it mean when people from a given social group use a certain pronunciation?" "What counts as a 'good' story in this social setting?" and "What kind of cues do people give to signal that they're not intending to be taken literally?"

Book Reviews by Daniel Ginsberg

Language in Society, 2013

The practice of mathematics education is fundamentally multimodal. It incorporates not only talk ... more The practice of mathematics education is fundamentally multimodal. It incorporates not only talk and embodied action, but also technical notation and diagrams, brought into discourse through verbal and gestural reference. As this interplay of semiotic systems arises in interaction, it can be interpreted by analyzing sequences of talk, writing, and gesture, but a better understanding requires an ethnographic perspective to contextualize interaction with reference to its physical surroundings, institutional setting, and enduring relationships within the community. Thus, classroom interaction is best understood as a multimodal ecology in which micro-level discursive practices, the history of a community, and the biographies of its members mutually influence and determine one another.

Beginning from the ecological perspective on classroom interaction, this dissertation presents an analysis of observations and video recordings collected during a semester’s multi-site ethnographic fieldwork with both a middle school math class for English learners and a quasi-remedial college calculus section. To model how any perceptible feature of the environment may be taken as meaningful, I draw on the semiotic theory of C. S. Peirce, not only in interpreting observational data, but as an organizing principle, as the analysis moves from qualities, to particular instances, to recurring patterns. First, I consider the ontological status of mathematical notation as a quality of interaction, investigating its capacity for representing continuous phenomena. Second, I take up actually occurring sequences of interaction, showing how students’ participation in conversational repair offers insights into ideologies of classroom authority and mathematical knowledge. Third, I address students’ and teachers’ identities in the classroom as social perceptions that are constructed and recognized through patterns of interaction. Each area of inquiry is then reconsidered to identify semiotic affordances that are made available in classroom interaction, and to explicate problems of practice that prevent students from seeing themselves as successful mathematics learners. The problems I identify are similar to those addressed by mathematics education researchers in the learning sciences, so I conclude by proposing a future research trajectory combining linguistic anthropology, the learning sciences, and classroom teaching practice.

Educational linguistics can have little impact on student learning unless it is effectively conve... more Educational linguistics can have little impact on student learning unless it is effectively conveyed to and applied by teachers in actual classrooms. This chapter represents teachers’ perspectives on linguistics in the classroom, from three authors with many years of experience teaching linguistically and culturally diverse learners at the high school level. Currently, Kimberly is a teacher educator, Daniel is an educational linguist and anthropologist, and Iris is a practicing classroom teacher. The vision of linguistically-informed education cast by Fillmore and Snow captures many of the ideas about language instruction that we have tried to embody in our classrooms over the years. However, we are aware that these approaches to teaching are not practiced in all classrooms, and we recognize the daily barriers faced by teachers and students in a world that does not share our perspective. In this chapter, we describe our approaches to linguistically-informed teaching, the barriers we have encountered, and our recommendations for teacher education, professional development, and future research.

... I struggled to understand how the students who had produced such sophisticated research paper... more ... I struggled to understand how the students who had produced such sophisticated research papers, who had led such engaged class discussions, had come to feel so poorly served by me and so demoralized by the course. I studied their comments closely, trying to square the experience they described with my own memories of the semester...

Students and teachers alike commonly view mathematics as an objective discipline to be learned th... more Students and teachers alike commonly view mathematics as an objective discipline to be learned through rote memorization. This view of the field is rooted in a traditional ideology of classroom roles in which any question has exactly one answer and the teacher is positioned as the ultimate authority. The case for reform has been made to deepen students’ engagement with mathematics content, but this paper identifies another justification: to provide opportunities for learners to develop proficiency in academic English. I analyze video-recordings of a middle school mathematics class for English learners in which the teacher knows mathematics and language pedagogy, integrates language and content instruction, and incorporates accepted best practices for teaching English through mathematics, such as explicitly teaching technical vocabulary and “unpacking” word problems. Considering how these techniques are instantiated through talk-in-interaction, however, shows that students are given a limited degree of semiotic agency over meaning-making resources such as classroom discourse; even when students evaluate one another’s work, their evaluations become tools for reinforcing teacher authority. Socializing students to academic language requires changes in ideology, not simply technique; only by decentering their own authority will teachers make space for students to engage in complex mathematics discourse.

University students are peripheral participants in various institutional communities, including t... more University students are peripheral participants in various institutional communities, including the university itself as well as the professional communities for which they are preparing. As students are socialized to these communities, their processes of identity navigation become visible in their narrative practices. Using interview and classroom observational data, this paper examines undergraduate, law, and business students’ self-positioning relative to institutions and individuals in their academic and professional lives. We conclude that students envision types of successful members of imagined professional communities, and that their academic motivation and success depend on their positioning their future selves as tokens of that type.

Mathematicians understand their discipline to be a process of logical reasoning about abstract ma... more Mathematicians understand their discipline to be a process of logical reasoning about abstract mathematical objects such as numbers, sets, and relations between them, through which discoveries may be made with potential implications for our understanding of the physical world. Students, on the other hand, often conceive of mathematics as a collection of algorithmic techniques for manipulating formal notation, disregarding the concepts that those symbols might represent; as a result, the study of math is seen to be nothing more than a sequence of progressively more sophisticated manipulations that enable the student to answer more complex questions posed by the teacher or textbook. While this aspect of mathematics is the most readily apparent to students, practitioners view it as the least interesting part, talking derisively about "plugging and chugging," the mechanistic computation of solutions to equations, as well as "drilling and killing," a style of teaching that presents students with long sets of plug-and-chug problems in order to increase their fluency with the notation.

A. Baczkowska & R. Zimny (eds), Impoliteness in Media Discourse, 2014

Among internet users, online discussions about controversial topics are known for quickly turning... more Among internet users, online discussions about controversial topics are known for quickly turning acrimonious, but as Locher (2010) points out, this aspect of computer-mediated communication has received little scholarly attention. To investigate impoliteness online, the present study follows Eelen (2001) in proposing that there can be no meaningful objective definition of impolite language, focusing instead ‘on the production of (im)politeness evaluations’ (Eelen 2001: 249). A corpus of over 20,000 words was collected, comprising comments on articles from the Washington Post local news blog dealing with the coexistence of cyclists and drivers on city streets. Each post was tagged for alignment and politeness/impoliteness, reflecting the judgments of the researcher; judgments from other readers were collected as well, to gauge the extent to which generalizations can be made. Word and bigram frequency analyses were performed along each dimension (cyclist vs. driver; impolite vs. polite) as well as between cells (impolite cyclist vs. impolite driver, etc.) to identify words and phrases that show a statistically significant difference of frequency across categories. These results were then analyzed qualitatively to investigate the underlying ideology of each position. The final analysis suggests that impoliteness is interpreted as a resource for claiming group membership or affiliation.

Language and Linguistics Compass, May 2011

This paper reports on a pilot project that explored the potential of linguistic inquiry in a high... more This paper reports on a pilot project that explored the potential of linguistic inquiry in a high school English as a Second Language (ESL) class. In class meetings across the school year, students worked collaboratively to investigate noun phrase pluralization, language acquisition, writing systems, and translation in their own and other languages. Classroom observations and students’ oral and written work provide evidence that:•Examining the structures of the spoken and written languages represented in the ESL classroom captures students’ interest and engages them in critical inquiry about the nature of linguistic knowledge and their beliefs about language.•The cross-linguistic analysis of students’ home languages validates their languages in the school context, defining them as a rich resource worthy of study, rather than as a hindrance to education.Examining the structures of the spoken and written languages represented in the ESL classroom captures students’ interest and engages them in critical inquiry about the nature of linguistic knowledge and their beliefs about language.The cross-linguistic analysis of students’ home languages validates their languages in the school context, defining them as a rich resource worthy of study, rather than as a hindrance to education.These findings are of particular significance in this time of English-only education in the United States, when students’ home languages are often rejected in schools.

Traditionally, mathematics is understood to be a field in which any question has a single answer ... more Traditionally, mathematics is understood to be a field in which any question has a single answer that is objectively correct—even outside of mathematics discourse, “two plus two equals four” is a prototypical example of a simple, obvious truth—and teachers are treated as absolutely authoritative sources of mathematical knowledge. This traditional ideology leads students to be positioned in the passive role of “received knowers” (Boaler 2002) whose task is not to think creatively or debate mathematical ideas, but simply to replicate the calculation procedures modeled by their instructor. This paper investigates the way the traditional ideology is produced and reproduced in classroom interaction. I consider a situation in which students seem to be positioned in agentive roles: college calculus students attempting to correct their professor. Theories of epistemic status (Heritage 2013) and representational agency (Kockelman 2007) are used to demonstrate that even in these situations, students’ agency with respect to mathematics discourse is highly constrained. At best, students are able to make claims about the material, but the instructor is always the one to bring mathematical objects into discourse and to validate processes of mathematical reasoning. The extent to which students accept this non-agentive positioning contributes to their sense of alignment with the discipline of mathematics, and ultimately to their level of success in mathematics classes.

Language most naturally expresses categorical distinctions among types of phenomena rather than g... more Language most naturally expresses categorical distinctions among types of phenomena rather than gradual variation along a continuum, functioning in a “typological” rather than “topological” fashion (Lemke 1999). Deacon (2003) argues that other universals such as unbounded recursivity are in fact due to semiotic processes underlying symbolic reference itself; if this is the explanation for the categorizing functioning of language, then we should expect to find typological representation not only in all languages, but in all possible symbolic systems. To model continuous phenomena in a way that language cannot accomplish, many technical fields use mathematical formalism as a supplemental semiotic system. Since this formalism demonstrates that symbolic systems are potentially capable of representing continuity, how do we explain the fact that no language does? This paper addresses this question by analyzing video-recorded interaction in mathematics classrooms, where participants reinterpret the formalism as a typological system, for example, understanding a continuous function by calculating specific values, or algorithmically manipulating chunks of notation. This observation leads us to reconsider the relationship between cognition and semiosis: even though algebraic notation encodes a type of meaning that no language can, participants in mathematical discourse may perceive it in linguistic fashion rather than taking full advantage of the system’s unique affordances.

During postsecondary education, students undertake a process of identity navigation, which become... more During postsecondary education, students undertake a process of identity navigation, which becomes visible in their narrative practices. Having interviewed undergraduate, law, and business students, we examine their taking on of various social and enregistered voices to identify their self­positioning relative to other individuals and institutions, and the ideologies so expressed.

This study presents an examination of the structure of teacher talk in a middle school mathematic... more This study presents an examination of the structure of teacher talk in a middle school mathematics lesson in order to describe the features of classroom discourse through which the process of socialization is enacted. Starting from the observation that a lesson is a semantically and interactionally cohesive text of spoken language, I aim to identify the production resources related to academic content and social participation that teachers draw on to create cohesion. Teacher utterances are analyzed using both a discourse analytic and a functional linguistic analytical framework, showing how social and pedagogical discourse functions are realized at the word and clause level through the lexicogrammatical functions of Theme and reference. The analysis shows the complementary roles of forward-reaching talk such as topic initiation, backward-reaching talk such as revoiced student utterances, and outward-reaching talk such as reference to mathematical visual images. The use of outward reach is shown to be particularly important to mathematical discourse because of its specialized function in connecting linguistic and nonlinguistic meaning making resources to create and contextualize mathematical knowledge. These three discourse strategies work in combination to socialize students to both the local classroom discourse community and the broader context of mathematics as an academic discipline.

In this study, a probabilistic computational model is used to identify shifts in conversational s... more In this study, a probabilistic computational model is used to identify shifts in conversational style. “Conversational style” in particular, as distinct from “style” more broadly, is defined as the aspects of discourse that speakers use to mark involvement with their fellow participants in talk, following Tannen’s (2005) classification of “high involvement” versus “high considerateness,” that is, higher and lower levels of explicit marking of involvement in discourse. The computational model allows us to consider style shifting as something that is emergent from and co-constructed in discourse, rather than as individual speakers’ statistical preference or dispreference of a given linguistic variant.

Style is typically considered as intraspeaker variation, that is, the linguistic choices that an individual social actor makes in a given context to express group affiliation or otherwise conduct identity work (Bell 1984, Bell & Johnson 1997, Coupland 2007). In contrast, this study takes an approach inspired by Fred Erickson’s observation that “talking with another person ... is like climbing a tree that climbs back” (cited in Tannen 1989: 13); in other words, participants in discourse are constantly orienting and reorienting themselves to one another. Therefore, no attempt will be made to categorize one or another speaker’s individual stylistic choices. Instead, style is viewed as a property of discourse rather than of speakers, a moving target that is jointly constructed and subject to continuous change.

A hidden Markov model was chosen as the computational representation of this style-shifting hypothesis. To implement a Markov model, a number of discrete states are identified ― in this case, the high-considerateness and high-involvement conversational styles ― together with a transition probability distribution that specifies how likely the model is to move from any state into any other state, or to stay in the same state, as the process moves forward. In a hidden Markov model, the state is not directly observable; rather, each state could give a number of different possible outputs, which are predicted according to an emission probability distribution. In this study, the length, turn construction type, and speaker identity are modeled as outputs of the Markov process. Based on observations of the outputs, the model allows us to infer the underlying state that is most likely to have generated each turn in the transcript. These results can then be interpreted using both quantitative (linear regression) and qualitative (discourse analysis) methodologies.

This model creates an opportunity for a quantitative study of conversational style that cannot be achieved using traditional quantitative sociolinguistic methods. Rather than taking either turn length or turn construction type as a proxy for conversational style, it represents style through a complex probabilistic model that accounts for the ways speakers construct turns at talk throughout the discourse. Preliminary results suggest that the computational model can identify coherent stylistic episodes in discourse that feature the expected linguistic markers of high and low involvement. More broadly, this result suggests that computational methods can be successfully applied to the study of talk in interaction.

You've all read my problem statement so I'm not going to spend time rehashing ittoday I'm going t... more You've all read my problem statement so I'm not going to spend time rehashing ittoday I'm going to look at this idea of "multimodal semiotics of mathematics teaching and learning" and look at two different directions the research is going at the moment. So I'm going to talk for five minutes about multimodal semiotics, and then five minutes about teaching and learning.

Whether they know it or not, everyone is interested in sociolinguistics. Comments such as "People... more Whether they know it or not, everyone is interested in sociolinguistics. Comments such as "People from Philadelphia talk funny," "I didn't understand the point of that story," or "How was I supposed to know you were kidding?" are the kind of casual remarks that come up in day-to-day conversation, but behind each of them is an implicit theory about the way language and culture are connected -that a way of talking is connected with a geographical location, for example, or that a good story should have an identifiable "point." As participants in the social world, each of us does this kind of theorizing all day, every day, but social scientists have developed philosophies and methodologies that allow us to conduct more principled investigations into questions such as "What does it mean when people from a given social group use a certain pronunciation?" "What counts as a 'good' story in this social setting?" and "What kind of cues do people give to signal that they're not intending to be taken literally?"