Writing and Publishing as Conversation

La Escritura y la Publicación como Conversación

A Escrita e a Publicação como Conversação

GAEA LEINHARDT
University of Pittsburgh, Pittsburgh, PA, United States of America

Correspondence concerning this article should be addressed to Gaea Leinhardt, e-mail: gaea@pitt.edu.
720 Learning Research and Development Center, 3939 O'Hara Street, Pittsburgh, PA 15260.

ARTÍCULO DE REFLEXIÓN
RECIBIDO: 22 de diciembre de 2011 - ACEPTADO: 14 de marzo de 2012

Abstract

This article asserts that the publishing process will be better understood as a conversational task that connects empirical precedents and theoretical debates in a given domain to produce new claims. It proposes that the lack of a conversational focus on the current academic environment and an excessive emphasis on impact indexes have created a citational inflation, in which authors and journals try to artificially increase the impact indexes without contributing to disciplinary progress. Based on current literature in psychology and conversational analysis, the article suggests that conversations are collaborative efforts that, in the case of publishing, must respond to two principles: quantity and quality. Quantity refers to making contributions as informative as possible, and quality refers to providing strong support for every claim. To explain this perspective, the author presents two case studies regarding the elaboration of seminal papers on educational psychology and education.

Keyword: publishing, conversation, bibliometrics, impact indexes.

Resumen

El artículo argumenta que el proceso de publicación puede entenderse como una conversación que conecta antecedentes empíricos y debates teóricos en un campo dado con el fin de producir nuevas propuestas. Sugiere que la falta de un enfoque conversacional en el ambiente académico actual y el excesivo énfasis que se otorga a los índices de impacto han producido una inflación de citación, en la que autores y revistas intentan aumentar sus índices de impacto sin contribuir al progreso académico. Con base en la literatura reciente en el campo de la psicología y en el análisis de la conversación, el artículo sugiere que las conversaciones son esfuerzos colaborativos que, en el caso de la publicación, deben responder a los principios de cantidad y calidad. La cantidad se refiere a que las contribuciones sean lo más informativas posible, mientras que la calidad a que sustenten coherentemente cada posición. Para explicar esta perspectiva, la autora presenta dos estudios de caso relativos a la elaboración de trabajos influyentes sobre psicología educativa y educación.

Palabras clave: publicación, conversación, biblio- metría, índices de impacto.

Resumo

O artigo argumenta que o processo de publicação pode se entender como uma conversação que conecta antecedentes empíricos e debates teóricos em um campo dado com o objetivo de produzir novas propostas. Sugere, ainda, que a falta de um enfoque conversacional no ambiente acadêmico atual e a excessiva ênfase que se outorga aos índices de impacto vêm produzindo uma inflamação de citação, na qual autores e revistas tentam aumentar seus índices de impacto sem contribuir para o progresso acadêmico. Com base na literatura recente no campo da Psicologia e da análise da conversação, o artigo sugere que as conversações são esforços colaborativos que, no caso da publicação, devem responder aos princípios de quantidade e qualidade. Quantidade se refere a que as contribuições sejam informativas, enquanto a qualidade se refere a que sustentem coerentemente cada posição. Para explicar esta perspectiva, a autora apresenta dois estudos de caso relativos à elaboração de trabalhos influentes sobre psicologia educativa e educação.

Palavras-chave: publicação, conversação, bibliometria, índices de impacto.

The research reported in this paper emerged
from a dialogue between the two authors.
The dialogue itself shared two goals ...

G. LEINHARDT and M. GREGG (2002)
"Burning buses, burning crosses:
Student teachers see civil rights".

WE, AS scholars, find ourselves in the midst of a surging press to publish. This is by no means a local or even national phenomenon; rather it is close to universal, ranging from China to Europe from South America to North America, from the Pacific rim to the Atlantic. Not only are we pressed to publish but it even matters where because our journals are rated and ranked according to the frequency with which articles in them are cited in other journals, the impact factor. Our own articles are also rated as to their level of citation by still other scholars -we are in some senses in the midst of citational inflation. One reason for this press is the matter of simple convenience- by using things like impact factors and H scores (a modification of a scholar's average level of citation across all publications) we can be deluded that we have found an objective international metric of quality. On the other hand, pressing for publication is a press for the scholarly community to engage in the full range of scholarly activities, not just a few (such as teaching and community activism). The press for publication also sets the stage of extended and internationally based dialogue (for example, of the last 40 articles published in the Review of Educational Research (RER), eight or 20% were by non-US based scholars). The question we need to ask ourselves is how we can make the activity of both writing and publishing more meaningful. The risk is that we might just add to a set of mindless mechanically developed tasks. To counteract this possibility, I argue for a subtle change in stance from seeing publishing as performance to seeing publishing as conversation. In this short essay, I will try to discuss the process of producing publications by dissecting (or backward engineering) two of my own publications in light of a particular theory of conversation. I hope that my efforts are seen in the light of their intention -to open a dialogue about this set of processes and not as an act of self-aggrandizement.

Conversation

This essay makes the following claim: academic writing is a peculiar but definite form of conversational activity rather than a performance action. I assert that academic writing is a form of asynchronous conversation rather than an act to be performed for an audience, an act with a script for how to conduct the performance. I make this assertion based on my own experience as a long time consumer of academic work, as a writer of academic pieces, and as an editor of the Review of Educational Research (2009-2011). Having made this assertion, of course I need to back it up -something I hope to do; but I also need to show how considering writing as conversation shifts or alters the activity toward meaningfulness. Many scholars assume that they need to publish (because there is an institutional press to do so, or because it is a part of their identity as scholars) -as if publishing were an independent act with a set of rules, almost like a game. If the rules are followed then an article will be published. Part of writing for publication requires that some other work be cited and acknowledged, but all too often that action is tacked on to the main one of building the account rather than a serious engagement with a set of ideas that are in constant formation.

I would suggest reversing this way of thinking and instead starting with wide and somewhat promiscuous reading. This wide reading is analogous to joining an ongoing conversation. One generally starts by listening in order to determine how to join in. What are people talking about? What have they already said or assumed? What is the purpose of the discussion? How carefully are they detailing or explaining shared or unshared views? To list just a few of the cues that one listens for. Thus, if a group is discussing the latest economic crisis, I cannot enter the discussion with glowing reports about my grandson. It is not that a grandmother is forbidden such statements, or even that the assembled set of speakers might not be interested in the content, but simply that it is irrelevant to what is actually being discussed no matter how informally. The implication of this anecdote is not simply the notion that one should not submit an empirical research paper to a journal that only publishes review articles nor submit a mathematics study to a reading journal; rather, the idea is that journals build up currents of ideas over time and that worthy publications contribute to those currents. In fact, a worthwhile publication must simultaneously connect to the ongoing discussion and put forward new and interesting ideas.

Systematic analysis of how oral conversations (discourse) work has a long and interesting history (Bales, 1950; Forman, Minick, & Stone, 1993), but for our purposes one set of conversational analyses suffices to guide the discussion of the publication of written work. As a start it is worth remembering that conversations are successful -that is, they have a convergence and coherence when they are about a shared and valued topic- to accomplish that requires reading. There are also shared norms for making contributions, while somewhat superficial, they do need to be listened to and in many journals this means following APA style standards or some other set of known standards. But what of the actual content and its worth?

Paul Grice, in his Studies in the Way of Words (1989), put forth a small set of conversational implicatures that serve as a useful anchor. Grice started by asserting that conversations are cooperative efforts -the speakers are trying to construct, share, define something together; they are trying to converge on meaning making. This conversational goal can be defined at the beginning (I have agreed to write a piece or discuss the issues surrounding the publication of scholarly work, I want to discuss how we have developed an instructional unit on chemical equilibrium, etc.) and/or the conversation can have its goal emerge and shift as the conversation develops over time (editors and reviewers may point out connections I had failed to make; or the redundancy of some of my assertions -the conversation is molded in a way that sometimes alters the goal). Grice formulates this goal sharing among speakers as the cooperative principle of conversation which requires that speakers in the conversation contribute in a manner that reflects the particular moment and in the general direction of the overall shared goal. The cooperative principle is a stance toward writing (or talking) rather than a formulation of it. One can argue, and it has been, that many conversations do not share a common goal and that in fact is what the conversation is actually about. But for our purposes, I think we can assume that academic writing does share the goal of communicating important new information in a convincing manner and that it needs to be responsive to the particularities of the specific moment. In fact, we can be even more precise and suggest that the cooperative principle is enacted when an important and valuable question is posed, and posed in a manner that permits it to be at least partially answered.

Grice (1989) goes on to introduce four ideas (maxims) that support the cooperative principle: quantity, quality, relation, and manner. Of these the quantity and quality maxims are most relevant for discussions of writing. The quantity maxim requires that the speaker/writer make their contribution as informative as required -neither more than nor less than. In writing this means selecting the right entry level to justify the research or theory and continuously supporting the discussion with information that is complete enough but not overly detailed. The entry level is both a level of detail and a level of significance. Some papers go back to an almost Neolithic beginning, for example, a paper on spelling instruction might begin, "It is in the nature of mankind to want to learn," or "ever since the beginning of writing the notion of spelling has been a challenge" these may be true but they are a long way from spelling. Starting so far away from the target ideas means that much too much information will need to be included. Other papers may make you feel as if you have been dropped in mid-steam, for example, a paper on history instruction might begin, "The Rüsen matrix is of little current use" also possibly true, but if the community is not enmeshed in the discussions of historiography and the tensions between US and German historical philosophy, the reference might be quite obscure. Starting so close to the target ideas leaves the reader unable to understand what the new research is actually about -there is insufficient information.

The quantity maxim when applied to writing a scholarly piece suggests that the authors read deeply in the area of their research and in the journal to which they are submitting. Thus, a general learning journal such as Journal of the Learning Sciences or Cognition and Instruction will need less information about learning theory (the idea is not to suppress all information, but not to provide an entire exegesis), but a bit more about the complexities of learning spelling or history. On the other hand, in The American Statistician even a short article on statistics learning requires a somewhat deeper discussion of learning theory but less explanation of the statistical content (see Larreamendy-Joerns, Leinhardt, & Corredor, 2005). One way to think about selecting the right level of detail is to spend some effort in justifying the value of the topic, that is, trying to invoke the cooperative principle: Why should we care about spelling? Why should we care about history learning? Why should we care about how and what people learn when they visit a museum? The rhetorical move here is to build an argument for value rather than to assume value. So, the main idea to be taken from this maxim of quantity is the notion of connecting at the right level and about something of value.

Grice's (1989) second conversational maxim refers to quality: "Do not say what you believe to be false; do not say that for which you lack adequate evidence" (p. 27). While the first can be assumed as well understood by academic writers, it is the second of these quality maxims that is elusive. Perhaps the most common flaw in early drafts of papers has to do with the broken connection between claims and warrants. This is not to suggest that speculation needs to be eliminated, but rather that it should be clearly marked textually as speculation rather than treated as if the data actually supported the claim. What is more important is that the connection must be actually shown not just asserted as deriving from the data. The quality maxim interacts with the quantity maxim for writing in that the writer needs to have a sense of how much needs to be established and what can be assumed.

A third maxim, relation, asserts the need to be relevant. Grice's fourth maxim, manner, has four sub areas: obscurity, ambiguity, brevity, and order (the first two are to be avoided and the second two to be sought). I would interpret these for the writing activity as requiring a kind of consistency of presentation, where central points are raised and dealt with repeatedly throughout the work starting with the introduction and carrying through to the conclusion.

The first two elements of this framework allow us to consider writing through the lens of conversational activity. Writing is not the conversational action Grice was describing, but considering it a conversational process is an important start to making the real shift in thinking that has to take place among the scholarly community. Conversation requires listening and responding; writing well requires reading and connection, not just describing, declaring, or soap boxing.

Examples

To flesh out these ideas, I will explore them a little more closely by pulling apart two published articles: "Two Texts, Three Readers: Distance and Expertise in Reading History". Leinhardt and Young (1996) (Two Texts), and "Learning from the Birmingham Civil Rights Institute". Gregg and Leinhardt (2002) (BCRI). I chose these two articles to elucidate the ideas of conversation and the idea of significant but answerable question for two reasons: first, I know the back stories and intentionality as well as the negotiation of the conversation with the editors; second, they are both examples of unusual questions and unusual methodologies which in combination might prove interesting but definitely require a depth of explaining and rationale presentation that interact with both the quantity and relevance maxims. The abstract of Two Texts is displayed in Table 1, and the abstract of BCRI is displayed in Table 2.

The back story of Two Texts. During the 1990-1991 academic years, I was a Fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford. As part of my activity there, I engaged in an extended year-long discussion with the nine historians who were also Fellows. The conversations included ideas about explanation, argumentation, the role of theory, and the meaning of being an historian. Evidential traces of my presence can be found in the introductory pages of at least two of the books that have emerged from that year (See D'Emilio, 1992; Mallon, 1995). These conversations led to several small "experiments" that were captured through audio tapes and artifacts.

The conversation reported. Given this setting, what was the research conversation we were planning to enter? And, in the longer time sense, with whom? The Two Texts paper related to some current issues but also to a longer line of discussion first launched by Joseph Jackson Schwab in the 1960's (Schwab, 1962, 1969) and to the discussions that Lee Shulman (2004a, 2004b) voiced a bit later in the 1980's. Both of these scholars had at different times raised the issue of the need to attend to the unique epistemic aspects of learning in different domains, the structures of disciplines. Essentially, they both argued that learning calculus was different from learning poetry or Greek. They were concerned that psychologists had focused too heavily on "learning" as a generic activity and not enough on the details and differences in learning within the disciplines¹.

As the first few sentences of the abstract suggest, there was also an important line of research dealing with reading comprehension. Up to the mid-1990's, reading comprehension discussions had focused on the learner and on first-time reading acts. I felt, however, that many readers read things more than once; and, what is more, they tended to find new things on each subsequent reading. So, in terms of Grice's cooperative principle, I was essentially saying, "I think I can add to the discussion of reasoning in reading by looking at a different condition of reading ... ". I wondered what repeated reading might look like among people who were known to read a lot and to re-read a lot.

The prevailing model of reading, at the time (1990's), was that of Walter Kinstch, who posited the complex reading task as being one of juggling the text model and the situation model (Kintsch & van Dijk, 1978). There was also an exciting relatively new line of work that had been investigating what experienced historians did when they looked (read) at carefully selected documents (Perfetti, Britt, Rouet, Georgi, & Mason 1994; Wineburg 1991). The work on historians was, itself, connected to a larger line of work about the differences between novices and experts (Simon, 1979; as well as-at least in Wineburg's case -connecting to the Schwabian argument.) What I was trying to add to this conversation was the idea of examining expertise -of historian- under rather unique conditions. The two differences that I was adding to the conversation comprised the following: first readers were responding to both unfamiliar and familiar texts; second, that participating expert historians were selecting documents not the experimenter. So, I was bringing to the conversation two new situations against which to think about some core ideas.

If I examine Two Texts with respect to the first of Grice's maxims, quantity, I need to consider what linkages I made to existing working and how much detail I gave about it. What did I, and the editor, believe was enough and what was too much? I explained the breadth of theories of reading comprehension in use at the time but focused on one, Kintsch and van Dijk (1978). I explained exhaustively the set of work on historians as readers but again focused on one, Wineburg (1991). Wineburg's work was very close to the work described in that, he too, had asked historians to read in and out of their areas of knowledge, but the specific documents were unknown to any of the participants and had been selected by him. So, the reader of this piece needed to be made aware of the dual purpose and the basic assumptions that grounded the paper -but did not need to know about the longer discussions of distinctiveness of content or even issues of expertise.

I also took the risky position of making several assertions -within the context of connection to other conversations. These assertions were essentially my own claims with only experiential evidence for their veracity (a very weak claim warrant connection indeed!). Specifically, I asserted that historians were in some senses special readers. I asserted that they read more dialogically than the average reader because they were constantly pushing the information into their own theoretical account, as well as consistently building up an interpretation of the situation. I suggested that historians used documents in several different ways; this was relevant because of the use of the Document Based Question (DBQ) in high school exams.

We also spent considerable time explaining the construction of the Document Based Question in the history exams and of the role of documents in historical research. This explanation bears on the issue of relevance. Knowing what the DBQ was not strictly relevant to the issues of reading or historical expertise, but it did help to justify the importance and significance of using documents. Now, had the paper been targeted for a journal of research in history, this balance would have had to be switched. Thus, we would have had to expand the detail surrounding the theories of reading and shrink the details surrounding the role of documents in historical thinking and reasoning. Implied in the work of Grice is the idea that something new or original is added to a conversation. In this case, what was new was the idea of the "subjects" selecting their own experimental material in order to understand how historians actually reasoned with documents. Also implied in Grice's work is the idea that the conversation is on-going. So, even though we did not cite Schwab, Shulman, or Simon per se, the paper belonged to a tradition of discussion about the unique features or cultures of specific content and pressed against the prevailing view of general psychological processes as primary explanations for learning.

Before turning to the second paper, I want to briefly discuss how the issue of quality, or the claims warrants connection was made in this paper. The claims were the following: two schemas governed the reading (identify and interpret); each schema had several subcomponents. Identify had four sub-schemas (classify, corroborate, source, contextualize); interpret had two (textual read, historical read). In his work, Wineburg (1991) had several of the same schemas, but he did not have the classify nor the details of the historical read. The warrants for these claims came in the form of the analyses of the protocols generated by the historians. All three historians' identification actions were displayed in a chart of direct quotations from the transcripts; the chart helped to show how the historians made use of the schemas and how they differed based on their awareness of the material. A similar chart was constructed for the interpretation actions. The supporting text for the charts quoted appropriate sections from the historians' transcripts and unpacked each of the actions in detail. We also made use of the complaints of the historians about their lack of knowledge or information. We took these complaints as evidence of the presence of an element in the schemas that they felt was missing and was something they would have used had it been available.

To further connect the claims to warrants for them, we coded the historians' protocols according to how much of the text they quoted directly, how much was a direct paraphrase, and how much was an analysis. The point here is that the evidence for claims was made using several converging sets of qualitative data, not just one. In the opening of the paper we had made a broad claim about the general uses of documents by historians and we returned to this idea and showed how it was played out. We did not show that these were the only uses, however. So what was left for the future was to find out if there were other uses of documents that needed to be considered.

At the request of the editor, we ended the paper by extending the conversation to a larger group, namely historical educators. The editor felt strongly that we needed to make suggestions for how the work might alter or expand the activities in the classroom. To be honest, I am still not totally happy with this part of the paper since I view it as quite speculative. However, a very recent paper describes an innovative approach to teaching history using documents. The paper is very empirical and shows that using documents in much the way we had suggested does improve several instructional outcomes (see Reisman, in press). Thus, the conversation continues.

Back story of BCRI. My colleague, Sister Madeleine Gregg, and I had been having a multi-year conversation about what could and could not be learned in informal settings. Dr. Gregg, as a teacher educator, wanted to develop a guide for teachers about ways to make use of informal environments such as museums. I was in the midst of a five-year project on learning in a wide variety of museum contexts. By the second year of our conversations, we had formulated a plan for a study of what student teachers might learn in a museum that challenged their initial beliefs and understandings. Gregg is a university professor in the American South; she is, in fact, only a few miles away from where much of the drama of the Civil Rights struggle was played out. But as she wrote so clearly in this article, issues of race were not really up for discussion among the young student teachers from that area at that time.

The conversation reported. So what was the larger conversation being entered into here? There are two near term ones: pre-service education of teachers and out-of-school learning. These, like the previous paper, both have histories that were not reflected in the discussion, but make them conversationally familiar. One historical line of teacher learning and role goes back to discussions of Lortie (1975) and general professional learning; the other has roots in Dewey (1938) and the discussions that surround the real world versus the abstract or theory-based world. I decided on discussing this paper for two reasons. First there was a traditional and "easy" way to have presented the paper. Namely a two by two ANOVA layout: two groups of teachers and a pre-post design. But that would have ignored the troubling undercurrent of the material. We were teaching these teachers about the Civil Rights struggle and about the contentious and difficult ideas of race. The "conversation" we both felt had to be much broader and more nuanced than the contrast of groups and contrast of conditions would allow. We needed to actually show the changes.

The second reason for selecting this paper for discussion is that there is in fact a second version using different data and a very different set of claims (Leinhardt & Gregg 2002). In the second paper, we considered the out-of-school experience the focus. It is important to know that, as close colleagues and friends, we simply could not settle on the main conversation we wanted to contribute to. We had enough data to enter both, but this is rare; the discussion here focuses on the Gregg and Leinhardt piece. The duality of purpose connects to the Gricean idea of relevance -deep discussions of informal learning would be needed if that was what the reader might expect while discussions of pre-service education were more relevant if the article was trying to enter into conversation with that community.

The entrance discussion of the BCRI paper deals with how pre-service teachers learn and how to help them learn things that they might not have expected to learn -such as how to deal with the topic of race. We offered a very brief discussion of how new teachers learn with the slightest of nods to extant research that was followed by a somewhat deeper discussion of the way race is or, most commonly, is not discussed in teacher education programs. The opening discussion problematizes the idea of learning about race and begins to offer a solution of considering out-of-classroom experiences -in this case visiting the Birmingham Civil Rights Institute. Since this particular study could have been described or talked about by starting with the role of informal education in learning, there is a marker in the early discussion to a second paper that did that (Leinhardt & Gregg, 2002). However, in this paper we were inviting those who wished to discuss both teacher learning and ideas of race into the conversational arena.

So the quantity maxim of Grice in this case deals with four briefly presented ideas: student teachers have expectations that support learning; some expectations may be embedded in their personal experiences, for example, race; the topic of race cannot be avoided if teachers are going to teach about the Civil Rights Era in United States history; perhaps one way to deal with this is indirectly in an informal setting, BCRI. There is no question that the paper does not say too much as it approaches its topic, but it may say too little. Considerable background knowledge is assumed; I can imagine, and in fact have experienced, someone being concerned that I had not clearly enough defined the differences in attitudes toward race in different parts of the United States among new teachers.

In a manner reminiscent of the approach in Two Texts of describing the nature of history documents and their uses, there is an unusually expansive discussion of the purpose and sense of the BCRI as a museum setting. Because it was this place and its setting that was significant and because we imagined that many of our conversational partners had not seen the place, we were elaborate in that part of the discussion. If this had been a strictly psychological comparison then the description would have been far briefer.

The tasks we had the teachers engage in were also somewhat unusual: draw webs representing their ideas of Civil Rights, design field trip experiences, and discuss race and their expectations of the museum and what it might have in it. We also discussed differing approaches to educating pre-service teachers. One approach was based on the idea of developing a community of learners who supported each other through meaningful conversations about the substance of teaching and learning and who shared experiences in the location and the activity in which the ideas were to be used -thus, adapting a decidedly situative framework. The other approach emphasized the deep mastery of content which was then to be applied to specific settings. The thought was that the former group might benefit more from the experience of the museum because they were familiar with the unfamiliar.

The analysis became a part of the "conversation" because it needed to be explained and it was the device for connecting claims to warrants, so the reader, in some senses, had to buy into the evidence before it was presented. The differences between webs produced by the teachers were fairly intuitive (more nodes, deeper links). The conversational analysis needed a bit more justification. We focused on three ideas: the connectedness of the discussion (supported or disagreed with another; continued or contrasted another; interrupted to finish another); relation to museum content in the discussions; focus on central versus peripheral ideas. Codes for the field trip manuals were fairly straightforward, focusing on feasibility and depth of use of the museum setting. But the depth of discussion of each of these coding approaches was necessary because it was unusual; it was not like a total number correct on a spelling test. One way of thinking about this paper in the context of conversation is that the conversational depth was continued throughout the paper. We tried to anticipate where our "listeners" might want further justification and detail.

We tried to validate the claims made in the paper by both presenting the quantitative descriptors and rich sets of examples. This is especially clear in the before and after node-link diagrams and between the two different instructional groups. We also showed how individuals added information to their diagrams. We gave samples of the differing conversations that were conducted in the classrooms so that the readers could appreciate the nuances we were trying to capture with the numerical summaries. Finally, we presented the analyses of the field trip manuals to show how much more sensitized to the unique setting one group was when contrasted to another.

The point of summarizing all of these analyses is to show how much effort was put into the quality part of the conversation. The paper(s) was a dialogical argument. We were trying to make credible and important the claims we were making -important in the sense of value- and to show in multiple ways different forms of evidence to back up those claims. This goes further than a p value or a percentage and I believe it is what is needed whether or not one uses quantitative or qualitative techniques.

Conclusions

In this short essay, I suggest that it is fruitful to consider academic writing in terms of conversation -albeit a rather formal conversation, not a coffee chat. In the retelling of the conversational aspects of two papers, I have tended to display the imagined "other" through the authors' lens. But there are actually real conversers. Papers often go through various stages of review: first to one's immediate colleagues, then to a more formal editorial review. Often either the journal reviewers or the editor has specific suggestions that relate to the conversational aspects of the piece. As an editor myself, the two places where the press for conversational connection is made are first through the detailed extension of explanations (coding, topic selection, expanded references), and second through the addition of "expected" material. In the BCRI paper, the editor was concerned that readers might need a more expanded description of the institute. In early conference presentations I learned that important members of the academic community from the North East of the United States did not believe that the student teachers would know so little about the Civil Rights Era because it is so extensively covered in the curriculum in that part of the United States. Both of these concerns led to greater expansion so that the conversation might be more inclusive and meaningful.

There is a serious risk that in discussing writing as a conversational act I might be suggesting that it is casual or informal. When I first started this essay, I had a rather long list of do's and don'ts that included, among other things, the need to use the most rigorous and up-to-date methodology available for answering the core questions posed by the research. This means avoiding research that is basically anecdotal in frame. The conversation that I am pointing to is long-lived, asynchronous, and coherent. It is a thoughtful exchange of ideas over time that builds up the understanding of the broad intellectual community about an idea or a finding. It is honed and clarified through the understandings of editors and appropriate peer reviewers. It is not a free association of detached ideas spewing forth. If you look back at the two abstracts you can see that there is as much left out as there is put in. These conversations are selective and pointed toward a community. A community that holds a particular understanding.

I have put forth a suggestion that considering scholarly written activity as a form of conversation is a better way of proceeding, in that the outcome is better and the activity has more value than considering it a performance of some kind. It is a shift in stance rather than a shift in techniques. One tries to imagine an audience that is actually engaging in the issues with you as author; one tries to attach the details of the conversation at the right level rei -neither too broad nor too fine-grained; one tries to share the interest in the question itself and to find a defensible way to actually answer the question; and, finally, one tries to move the conversation ahead to addressing ever more important and meaningful issues rather than simply re-doing what has already been done. I have tried to show how that was done in two personal examples. As the quotation at the beginning of this paper suggests, some work is in fact the outcome of nested conversations: some between the authors, some between current researchers and the authors, and some between the authors and their intellectual ancestors. I await the replies.

¹ In this section, I start the discussion by using the first person singular and then later switch to the first person plural. The work was conducted over some time and many decisions were made by me alone; later decisions and writing were made by both my co-author, Kathleen McCarthy Young, and myself, and I am trying to reflect that distinction.

References

Bales, R. F. (1950). Interaction process analysis. Cambridge: Addison-Wesley Press, Inc.        [ Links ]

D'Emilio, J. (1992). Making trouble: Essays on gay history, politics, and the university. New York: Routledge        [ Links ]

Dewey, J. (1938). Experience and education. New York: Phi Delta Kappa.        [ Links ]

Forman, E. A., Minick, N., & Stone, C. A. (Eds). (1993). Contexts for learning: Sociocultural dynamics in children's development. Oxford: Oxford University Press.        [ Links ]

Gregg, M. & Leinhardt, G. (2002). Learning from the Birmingham civil rights museum: Documenting teacher development. American Educational Research Journal, 39 (2), 553-587.        [ Links ]

Grice, P. (1989). Studies in the way of words. Cambridge: Harvard University Press.        [ Links ]

Kintsch, W. & van Dijk, T. A. (1978). Toward a model of text comprehension and production. Psychological Review, 85, 363-394.        [ Links ]

Larreamendy-Joerns, J., Leinhardt, G., & Corredor, J. A. (2005). Six online statistics courses: Examination and review. The American Statistician, 59 (3) 240-251.        [ Links ]

Leinhardt, G., & Gregg, M. (2002). Burning buses, burning crosses: Student teachers see civil rights. In G. Leinhardt, K. Crowley, & K. Knutson (Eds), Learning conversations in museums (pp. 139 - 166). Mahweh, New Jersey: Lawrence Erlbaum Associates.        [ Links ]

Leinhardt, G. & Young, K. M. (1996). Two texts, three readers: Distance and expertise in reading history. Cognition and Instruction, 14 (4), 441-486.        [ Links ]

Lortie, D. C. (1975). The schoolteacher: A sociological study. Chicago: University of Chicago Press.        [ Links ]

Mallon, F. (1995). Peasant and nation: The making of postcolonial Mexico and Peru. Berkeley: University of California Press.        [ Links ]

Perfetti, C. A., Britt, M. A., Rouet, J. F., Georgi, M. C. & Mason, R. A. (1994). Howstudents use texts to learn and reason about historical uncertainty. In M. Carretero & J. F. Voss (Eds.), Cognitive and instructional processes in history and the social sciences (pp. 257-283). Hillsdale, NJ: Erlbaum.        [ Links ]

Reisman, A. (2012). Reading like a historian: A document based history curriculum intervention in urban high schools. Cognition and Instruction, 30 (1), 86-112.        [ Links ]

Schwab, J. J. (1962). The teaching of science as enquiry. Cambridge: Harvard University Press.        [ Links ]

Schwab, J. J. (1969). College curriculum and student protest. Chicago: University of Chicago Press.        [ Links ]

Shulman, L. S. (2004a). The wisdom of practice: Essays on teaching, learning, and learning to teach. San Fransisco: Jossey-Bass.        [ Links ]

Shulman, L. S. (2004b). Teaching as community property: Essays on higher education. San Fransisco: Jossey-Bass.        [ Links ]

Simon, H. A. (1979). Models of thought. New Haven: Yale University Press.        [ Links ]

Wineburg, S. S. (1991). On the reading of historical texts: Notes on the breach between school and academy. American Educational Research Journal, 28 (3), 495-519.        [ Links ]