Educating Student Teachers to Teach in a Constructivist Way - Can it all be done?

Our challenge as teacher educators and researchers was to design a teacher education program module that centered on an ambitious constructivist teaching model. How could such a program be designed that stirred vision, motivation, and inquiry on classroom, self, and the aims of education, that furnished considerable disciplinary and design knowledge and management skills, and that hatched professional community? The project experimented with three different versions over three years. In the first year, the program generated a great deal of inspired pioneering; but technical skill and keen observation was submerged at times in ideological commitment, and understanding of the model was truncated. In the second year, the program placed great emphasis on the mastery of the model aiming at clinical tryouts. Unfortunately, this format sapped the novices' inspiration by over-burdening them with abstract theory and fixed pedagogical forms, thus disconnecting the model from the philosophical and moral reasons of teaching it. In the third year, the program concentrated on practical inquiry and careful bottom-up reflection to develop classroom community. Novices maintained their vision and motivation for the constructivist model, left the project with "reflective prompts," but missed fundamental design competencies. Thus, none of the program iterations stands out as a shining example of success, but together they demonstrate the indispensability of all the components.

In his 1987 article "Knowledge and Teaching: Foundations of the New Reform," Lee Shulman advocates a broad conception of the expert teacher knowledge base. Exceeding the generic behaviors and skills that teacher effectiveness literature has identified (Brophy and Good, 1986), expert knowledge is grounded in acts of pedagogic reasoning. This has important implications for teacher education. Shulman writes:

The goal of teacher education is not to indoctrinate or train teachers to behave in prescribed ways, but to educate teachers to reason soundly about their teaching as well as to perform skillfully. . . . Teaching is both effective and normative; it is concerned with both means and ends. Processes of reasoning underlie both. The knowledge base must therefore deal with the purposes of education as well as the methods and strategies of educating, (p. 13)

Postulating both skill and reason, Shulman wonders "how the extensive knowledge base of teaching can be learned at all during the brief period allotted to teacher preparation" (p. 7). While Shulman's 1987 article went on to explore the first part of his pronouncement, that is, the knowledge base of teaching, ten years later, "Fostering Communities of Teachers as Learners" (FCTL) (Shulman, 1997), a research-cum-reform project under his directorship, tackled the second part of his pronouncement—the question of how it can all be done. This article reports on the .project's repeated attempts to find an answer to this question.

It was a main task of the project to help student teachers understand and practice constructivist pedagogy. Our efforts revolved around one specific constructivist model, "Fostering Community of Learners" (FCL), a complex and ambitious instructional and learning model developed by Brown and Campione (1994, 1996).

THE MODEL OF FCL AND ITS CHALLENGES FOR TEACHER EDUCATION

"Fostering Communities of Learners" originates from the work of cognitive psychologists Brown and Campione (1994, 1996), but embodies many principles of democratic, student-centered, and inquiry-based instruction in the tradition of Dewey, Vygotsky, and Bruner. Underlying the model is a philosophy of learning that places emphasis on generativity, active discovery, reflectivity and meta-cognition, cooperation, and community. Brown and Campione's model mirrors processes of cooperative scientific research. The proto-typical FCL teaching unit consists of research-share-perform cycles. During the research phase, students generate their own questions and hypotheses about a topic based on prior experiences and knowledge and formulate a main question for inquiry. Research is carried out in expert groups that specialize in specific sub-topics. Subsequently, students come together and share their specialized knowledge in jigsaw groups (Aronson, 1978). Students integrate and synthesize their specialized and shared knowledge in a consequential task consisting of presentations to the community public. Group-work based sequences (mainly of the expert-jigsaw type) alternate with whole-class, teacher-directed benchmarks and student-induced "crosstalk" that facilitate conceptual generalization and dialogue. Thus, knowledge is generated constructively in a dialogue among students and teachers; knowledge is evaluated publicly by the community. FCL rests on a set of learning principles and instructional forms that make up, as Brown and Campione point out, a new constructivist system of classroom instruction that diverges radically from more traditional instructional systems (Doyle, 1986). In many of its principles the model shares features with other constructivist approaches (Schifter and Fosnot, 1993; Lampert, 1990), but its systemic character of interrelated task and participant structures is a benchmark of instructional complexity for the project.

Not unlike other ways of teaching in a constructivist manner, (Ball, 1990; Cohen, 1990; Lampert, 1990; Schifter and Fosnot, 1993), FCL extensively engages a teacher's thinking, feeling, and acting.1 We hypothesized that (student) teachers who successfully incorporate ambitious constructivist models, such as FCL, into their teaching repertoires were individuals with a vision compatible with constructivism that could inspire and sustain them through experiences of failure and through the adversities of the institution. Motivation to engage in FCL, we hypothesized, might be sustained by this vision, the experience of success with the pedagogy or the model, and the support received from a group of colleagues.

Collegial support was especially germane in the case of FCL since teachers were to re-invent, not merely replicate, the model that heretofore had been developed in a small number of classrooms by expert teachers working in close contact with the "founders." Re-invention also highlighted the importance of reflection in and on practice (Schon, 1983). FCTL project participants were the first group of teachers who experimented with FCL under more average conditions. Numerous studies (McLaughlin and Tal-bert, 1993; Little, 1982) have shown that innovative practices that go "against the grain" (Cochran-Smith, 1991) of the institution are not sustained by teachers unless the latter are involved in collegial relationships that nurture their engagement and effort and involve them in learning through reflection. We assumed, therefore, that the project needed to facilitate a collegial group setting that would deepen student teachers' inquiry into contextual and institutional constraints and potentialities, lend support, and strengthen student teachers' resolve and understanding in their pedagogical experiments.

Only knowledgeable student teachers will be able to translate their fantasies and communally generated norms into practice. Constructivist teaching of the scope of FCL commonly calls for a greatly expanded knowledge base of teaching (Shulman, 1987; Mosenthal and Ball, 1992). In the dimension of foundational knowledge, teachers need to grasp the model's formal structures and Philosophical and psychological rationales. In the dimension of content knowledge, they need to recognize the generative ideas of the discipline that underlie the constructivist curriculum. In the dimension of pedagogical content knowledge, they need to be able to design units in which "adult" concepts are transformed into age appropriate representations (Black and Ammon, 1992), content matter is subdivided into "jigsawable" chunks that are meaningful for students' specializations and interdependent sharing, and specializations are synthesized in a culminating project.

The systemic character of FGL (Brown, 1992) poses a particular challenge to teacher educators since the full-fledged FCL model requires student teachers to design fairly comprehensive teaching units that create research-share-perform cycles through the interplay of various task formats and participant structures that are laid out in the model. Furthermore, as the founders of the model point out (Brown and Campione, 1990), "fostering community of learners" in the classroom cannot operate in a cultural vacuum. Students need to interact with each other in a climate of community. Thus, student teachers need to gain knowledge and skill in creating classroom culture characterized by mutual respect, dialogue, and shared expertise, and in directing and monitoring work flow in the complex research and group work cycle of the FCL system.

A SUITABLE PROGRAM

A suitable program for FCL had to address teachers' vison and motivation, their understanding and skills, and their dispositions to be reflected in a professional community. The all-encompassing nature of such a program preempted an approach based on the notion of teaching as craft and student teaching as apprenticeship, to follow Feiman-Nemser's (1990) typology. Given that constructivist teaching in general and FCL in particular could be expected to fall outside of the norm of teaching and learning in prospective student teachers' classrooms (Goodlad, 1984; Cuban, 1991), student teachers would not be likely to encounter teachers whom they could emulate as models. Rather we conceived student teachers' role in practicing FCL pedagogy as transformative. Thus our program had to build in safeguards against the conservative pull of student teaching as apprenticeship (Lortie, 1975; Feiman-Nemser and Buchmann, 1985). Likewise, a notion of student teaching as learning to apply generic principles of effective teaching (Berliner, 1986; Brophy and Good, 1986; Goodman, 1986) was too narrow. Although FCL is a "model of teaching" in some respects (Joyce and Showers, 1980), it was first a model in its infant (that is, laboratory-like) stage of development, and second, a model so profound in its personal and cultural implications that student teachers presumably could not learn it as a skill, but had to reinvent and appropriate it as part of their own personal style (that is, learn it constructively themselves).

A suitable program for FCL would therefore be grounded in the tradition of reflective and inquiry-based teacher education (Dewey, 1904; Zeichner and Liston, 1987; Zeichner & Tabachnik, 1991; Feiman-Nemser, 1990; Calderhead, 1988, 1989) that centers on student teachers' understanding of their situation. Situated and practical knowledge (Elbaz, 1983; Schon, 1983; Elliott, 1993) develops through reflection on personal goals (Combs, 1982), biases (Elliott, 1993), and moral dilemmas (Berlak and Berlak, 1981). Technologies of teaching are subsumed under educational, moral, and political considerations. Conventional models of clinical supervision with their emphasis on lessons and lesson structures are avoided in favor of participant structures and group rituals that foster broader inquiry (Cochran-Smith, 1991). Case interpretation and writing in a group (Shulman, 1988; Richert, 1991; Shulman and Colbert, 1988), the use of narratives and stories (White, 1991), life history accounts (Knowles, 1993), student teachers' personal theories (Tann, 1993), and classroom artifacts, such as video clips, are to facilitate situated inquiry and fusion of theory and practice, competence and interpretation.

Another impulse for a suitable program comes from an orientation of teacher education termed "academic" by Feiman-Nemser (1980). This orientation recognizes the importance of teachers' subject matter conceptions for classroom instruction (Wilson and Wineburg, 1988; Stodolsky and Grossman, 1995; Stodolsky, 1988). Academic or subject-matter concerns seemed particularly pertinent for a constructivist model such as FCL with its special ways of making the world intelligible to children and with its requirement for deep understanding of content (Mosenthal and Ball, 1992). As Feiman-Nemser (1980) points out, subject-matter concerns often fall out of view of teacher educators since disciplinary preparation is usually not part of education programs. For matters of FCL, however, they had to be included. We, therefore, placed great value on student teachers wrestling with content during the planning of curriculum.

A third impulse for a suitable program, besides the inquiry and academic impulses, was a communal orientation. Not surprisingly, given the development of the field, the category of "communal" does not appear in Feiman-Nemser's (1980) typology. But in recent years, as insights from the literature on educational reform (Fullan and Hargreaves, 1992), learning organizations (Fullan, 1993), professional culture (McLaughlin and Talbert, 1993; Louis and Kruse, 1995) and the neo-Vygotskyan strand of cognitive psychology (Brown and Campione, 1996) have intersected, a communal orientation has made forays into programs of teacher education. In some of these communally oriented programs, action research and collaborative learning on the part of teachers (Comeaux, 1991) are to model and mirror desired classroom cultures.

How could a program be designed that stirred vision, motivation, and inquiry on classroom, self, and the aims of education, that furnished considerable disciplinary and design knowledge and management skills, and that hatched professional community? How could such a program integrate skill, reason, and community; and how would it "fit" its wide-ranging and complex goals with the learning needs of student teachers within the one-year time limit of professional preparation? Thus, how could it all be done?

The literature on reflective and inquiry-based teacher education programs presents approaches that answer this question in a variety of ways. One such approach suggests that student teachers are initially so preoccupied with classroom and instructional routines (Russell, 1988) and with discipline and classroom order (Veenman, 1984) that programs should revolve around the mastery of those issues before more complex tasks and deeper reflection can be tackled. Being mindful of these putatively primary learning needs, programs often limit themselves to inquiry on micro teaching/ learning processes (Valli, 1993). In a staged model, described by Mclntyre (1993) this limitation is avoided by dividing the student teaching year into three phases that follow Van Manen's (1977) typology of technical, practical, and critical reflection. In the first phase, student teachers concentrate on effective ways of teaching, models of teaching, and so forth; in the second phase they are encouraged to "concentrate on learning to evaluate and to develop their own teaching" (Mclntyre, 1993, 45). In a parallel third phase, students take classes in which theory is provided for critical reflection. While in this model student teachers are exposed to the full depth of reflection, attaining teaching skills and reasoning about those skills apparently fall apart into separate programmatic steps.

A different phasing seems to occur in a program described by Beyer (1984). In order to redirect "the tendencies of utilitarian teaching perspectives, excessive realism, and uncritical replication" (p. 40) commonly exhibited by student teachers, student teachers in this program engage in a thorough inquiry component prior to being exposed to the student teaching experience. Studying and experiencing situations outside one's assigned classroom, for example, through a human service project (Beyer, 1991), are to create distance to the taken-for-grantedness of classroom routines and are to enable the novice to see educational problems in larger social contexts.

The programmatic sketches discussed so far are problematic in that they tend to separate technical skill development and reasoned inquiry, theory and practice. FCL, being both a technical model of teaching and a cultural mode of communicating, requires a closer "linking [of] value and meaning to the technical" (Bullough and Gitlin, 1991, 43). This closer link is forged in a student teacher seminar that Bullough and Gitlin describe as an "educative community." Student teachers who participate in this seminar are members of a cohort and are for the most part placed in a small number of schools together. Parallel to student teaching they begin their work together with the exploration of their own personal histories, visions, and theories of teaching. A central activity of the seminar is the planning and implementing of teaching lessons. However, Bullough and Gitlin do not elaborate on the content and structure of the lessons. Implementation of the lessons is observed and analyzed by small groups of student teachers who visit each other's classrooms. Bullough and Gitlin term these peer observations "horizontal evaluation" (p. 48). Work on the teaching lessons is complemented by an action research project through which student teachers analyze problems in their classrooms. Mentioned also as an ingredient in other inquiry-based teacher education programs (Zeichner and Liston, 1987; Cochran-Smith, 1990), the idea of student teacher seminars as precursors of educative communities seemed a good organizational base for our project. However, we placed at the heart of the seminar an ambitious predefined technical model of constructivist teaching (FCL).

While we were in no way aware initially of all the challenges that would unfold in the course of the project, the literature directed us to a number of pitfalls that other constructivist programs had encountered, bolides the problem of phasing and "fitting it all in" discussed above. RiSardson (1997) points to uncertainty about the very nature of what constructivism is. Is it a way of teaching with a specific set of content, tasks, and participant structures, or is it a set of inquiry-oriented principles that can find expression in many pedagogical forms? In the former case, students need to become familiar with the theory and practice of a particular model through direct instruction (p. 10); in the latter case, broad reflection and a slow build-up of insights about constructivist pedagogy in student teachers' classrooms is indicated. Though we settled on FCL as a theoretical and practical model, inquiry orientation and community building remained centerpieces of the project's thinking as well. Though /ye knew that we, as teacher educators, needed to model the kind of constructivist practices for student teachers that we wanted them to implement in their classrooms (Meyer-Smith and Mitchell, 1997), we did not know how to convey to them the considerable technical knowledge that the model presupposed and be constructivist at the same time (Winitzky and Kauchak, 1997). Thus we, as teacher educators, were faced with the same tension between formal knowledge and students' own knowledge construction with which teachers are faced in constructivist classrooms. How would we balance the need for theory and content knowledge with the presumed and well documented tendency of student teachers to privilege practical classroom experiences and relationships with their students as the more relevant sources of knowledge construction? The following pages will give a vivid account of how we wrestled with these questions.

THE PLANNED PROGRAM

The following blueprint evolved for the project's operation. FCL project participants would join a student teacher seminar. During an initial phase participants would explore personal theories, pedagogical assumptions, and visions of community in general and in particular classrooms. Adult learning during this phase was to mirror student learning in prospective classrooms. In a next step, participants were to be familiarized with psychological rationale, "first principles," and formal participant structures of the FCL model. The seminar was then subdivided into small curriculum planning and design teams. The design teams would also be responsible for peer analysis and evaluation of the planned units. Findings from FCL teaching experiments would be shared with the whole seminar. Inquiry into the culture of classrooms was to happen parallel to curricular planning. Each design group would be accompanied by a mentor/researcher whose role was initially fairly undefined.

An effort was made to place participants in a small number of schools. In some cases student teachers experienced "critical dissonance" (Cochran-Smith, 1991) between the cons true tivist project and the philosophies and practices of teachers at their home schools. In other cases we were able to create "collaborative resonance" when student teachers worked with veterans who were involved in the project's professional development segment or when they were placed in a school affiliated with the Coalition of Essential Schools.

THE CONTEXT OF THE UNIVERSITY

The project was housed within a teacher education program that was a conducive environment for constructivist pedagogy. Furthermore, the material conditions of the project allowed for intense contact with each individual participant. In many respects, the university environment in which we carried out the research was atypical for the great majority of teacher education programs in the United States (Goodlad, 1990). The University Teacher Education Program offers a fifth year combined master's and certification program for a secondary school credential. A cohort of 50 to 70 students begins the program each year in the summer with a practicum. Candidates teach in their student teaching assignments for a whole school year—usually two periods a day, five days a week. The school of education is prestigious and selective in its admissions. Students are exceptionally well educated by comparison to the average teacher education candidate (Good-lad, 1990) and often come from solidly middle class backgrounds. As interviews with participants reveal, for many teaching was one of a variety of career options, consciously chosen, often in the face of expectations and pressures by parents and peers to pursue other more lucrative and prestigious professions.

During the three years of the project's duration, the constructivist message was reinforced in other classes within the program. The mandatory Foundations class, taught by one of the principal investigators of FGTL, gave ample space to FCL pedagogy in its syllabus and utilized many of the techniques described above under the rubric of inquiry-oriented teacher education. In subject-matter based methods courses, students were encouraged to concentrate on essential ideas and questions of the disciplines for their curriculum and to reflect on the aims and knowledge structures of the subject domain. Supervisors, either graduate students with teaching experience or retired school teachers, brought a greater variety of approaches into the program. Some project participants worked with FCL mentors as their program supervisor; others worked with supervisors who concentrated their efforts on classroom order and management.

Participation in the FCTL project was an elective, to which student teachers came voluntarily with initial motivation and commitment, and from which they could withdraw at any time. This meant that we had to be sensitive to student teachers' needs, in order to keep them in the FCL seminar, but it also forced us to down-scale our demands and expectations with respect to participants' workload. As is the case in many research projects of this kind, FCTL had more material and human resources at its disposal than regular programs of teacher education could ever hope to marshal. Each project facilitator worked with no more than three student teachers; at times, intense one-on-one work was possible. The FCL seminar took place for the duration of three quarters each year; meetings were scheduled about once a week for one to three hours. In sum, due to the character of the university setting and the project, conditions for this teacher education experiment were highly favorable. Clearly, given these atypical conditions we could not aim at experimenting with programmatic design features that then might be replicable under more average conditions; rather, we regard the project as an exceptional laboratory in which we draw insights from a number of unique cases which might shed light on the successes and problems more commonly encountered in efforts of learning how to teach in a constructivist manner.

THE CASES

In the following section, I will present selections from case narratives that highlight how the project unfolded, I rely on case studies of individual students and groups that were compiled by project personnel over the duration of the project. These case studies, naturally laced with the interpretations of the writer, who often functioned as mentor and chronicler of events, will be cross-referenced with primary material stemming from project participants themselves: evaluations and reflective cases they wrote about their involvement with FCL, reflective interviews, observations, and unit plans. Material from individual cases will be woven into the story of the group in which the individuals learned about FCL.

The project lasted for three years, from 1994 to 1997. Each year, a new group of student teachers enrolled in the FCL seminar. During the first two years, about fifteen student teachers completed the university-based seminar (a larger number enrolled initially, but did not continue with the program); in the third year, two groups of four student teachers at two school sites were involved in the project.2 Each year, the action plan of the project was revised based on the implementation results of the previous year and the student teachers' responses to the program. Though overall personnel continuity was maintained, in the second and third years some project members left and new personnel influenced the programmatic course of the project. Material from three different instantiations of the program will be presented. In each instantiation, the project coped with the problem of integrating skill, reasoning, and community within a one-year program. The cases will show what was gained and what was lost in each iteration. My report will be short on answers with regard to most effective practices; the project did not succeed in getting it all done in the time allotted. But in the tradition of case study research, the cases will give a critical account of the practical dilemmas that participants and researchers faced in their endeavor of teaching and learning an ambitious constructivist pedagogical model.

YEAR I—INSPIRED PIONEERS

The first year of the project was the pilot year. Project staff had to learn how to balance their roles as learners of the FCL model, as teacher educators, and action researchers. A sense of enthusiasm and productive naivete about the prospects of Learning Communities pervaded the project. The spirit of optimism touched the student teachers as well. By the time students were introduced to the FCL model in the project seminar, they had already explored broader implications of community and had already had an intense experience of each others' company in a previous summer course on the topic of community that was taught by one of the project directors.

The Model

At the time the project seminar was taught, FCL was still in its laboratory phase. Due to the stage of development of FCL, presenting FCL as a pedagogical model-in-action was a challenge. A video of FCL teaching samples and presentations by some of the original creators was available. But on the whole neither a solid clinical base (for example, teaching materials, video tapes) nor the option of observing FCL in vivo was available to seminar participants. Instead, participants relied on the publications of the "founders," which were framed in the technical language of cognitive psychology. Hence both student teachers and project facilitators had to "feel" their way through the literature and pioneer ways to transform science into art, to borrow James' famous phrase.

A first step of re-invention of the model occurred when Lee Shulman (1997) translated the "first principles" of FCL from the psychological language of the founders into the more engaging language of educators. In his reformulation, "Communities of Learners" for both adults and children incorporated a vision of learning as generative, active, reflective, collaborative, passionate, and communal. In the absence of a live technical model, the project rallied around these inspiring, but necessarily diffuse principles as guiding posts and relied more heavily than desirable on exposition and written texts for the explanation of the model's formal structures. The tension between the inspiring principles of FCL and the model's lifeless structures remained a recurring theme throughout the duration of the project. The absence of a live model-in-action underscored the exploratory and experimental nature of FCL reinvention.

A facilitator recounts:

During Year I, B. and I were truly learning about FCL with the teachers. The fact was that neither B. nor I had ever laid eyes on "true" FCL classrooms which we could hold up as models for the teachers; what further centered the idea that we could be "teachers as learners" was the fact that we could not draw directly from our prior experience as classroom practitioners. In other words, as we were teaching . . . our on-going learning was quite authentic.3

Although they recognized this state of staff expertise, Year I participants appreciated the vagueness of the project:

I liked that feeling, too, that we were all collaborating, both teachers and students, at the same level. There is something to be said for that. It would have benefited me some to have a bit more clear direction, but. . . sometimes it's one or the other. I wouldn't want to give up the feel of what we had for more assertive top-down direction of where we were going. (Tim)

I think [we had] a different kind of learning because we were all equals, none of us really understood where we were going with this, so we started off on an equal standing. And I think any time that there's community, like community is food, community is conversation, community is environment, so I think that made the learning really different . . . . What else was different about the learning was that, while we were expected to reflect the whole year, I mean, that was like the big word was reflection, is that it was coming from us and not from a mandate from somebody else. So it was really exciting. (Melissa)

The seminar apparently could not fulfill the utilitarian needs that are typically ascribed to student teachers. What counted more than a need for clear direction, recipes, or practical expertise, and what made student teachers tolerate confusion and insecurity, was a sense of discovery and self-actualization.

Participants, as expressed in the following group interview, professed to have been quite confused about grasping the model with its dense psychological jargon:

• —I mean I continued to be confused throughout the entire year as to exactly what . . . was FCL, what did it look like.

• —I felt like I'd been missing some essential component of exactly what FCL was or was not, or anything. (Exit Interviews STEP Plus)

But unfazed by this, students voiced excitement when they recalled the spontaneous and emphatic way with which they groped for understanding:

There were four or five of us sitting at a table. It was just the . . . students, and we, somebody or all of us together figured out what zones of proximal development was and we did this ... all of a sudden there were five of us doing this little ballet thing and saying, "Oh my God I get it now!" ... we are at a round table, our arms are making these round shapes ... it was like the circle is expanding. (Case Study—Melissa)

Rather than being a specific teaching tool, FCL at this stage functioned as the language of practice furnishing bold practical arguments (Fenster-macher, 1986) to the novices for pre-conceived pedagogical visions of student-centeredness:

- "The thing about FCL is that it fits my vision so I can make a lot of structures work."

- "I would probably still do a lot of the same things, I just wouldn't be sure they were right."

-"[Through FCL,] I can find my own vision, enough to figure out how it's going to have to play out in class."

Community

Sharing of stories happened quite frequently and consisted of relating incidents or commenting on artifacts that in the mind of presenters had some personal or broader relevance for the group. The sharing was experiential, not evaluative. It stressed support, validation, encouragement for experimentation, and interpretation of incidents in light of hopes and ideals. Neither students nor project facilitators held each other accountable for a particular standard of excellence or expertise. Frequently, inspiring moments during which FCL supposedly came to life were related, such as David's revelation about the power of "jigsaw" groups while he taught an abbreviated version of FCL in his mathematics class.

The day of the jigsaw turned out to be one of my greatest teaching days ever! ... The mathematics discourse in the classroom was greatly improved by the distributed expertise of the jigsaw. There was 95% on-task behavior, and the students did not want to stop teaching when the class period ended.

Other novices reported similar incidences which the group relished. Common problems of group work were also aired, such as the recurring case of the student who refuses to cooperate with his group.

Community, not mastery of the model or considerations of implementation, was the main entry point into the designing of FCL units for Year I participants. The seminar split up into three smaller design teams. Two of the three groups chose to plan interdisciplinary units. I will concentrate on one of the groups, the "equality group," which brought Social Studies, English, and Mathematics teachers together. Although interdisciplinarity added complexity to the already complex task of designing in the FCL mode: "We [the equality group] decided that we wanted to design an interdisciplinary FCL unit. We had such an amazing experience designing the unit and talking about its implementation" (Exit Interview STEP Plus). The group generated a common unit with multiple materials, yet actual thematic emphasis, design, and implementation differed widely among participants. In all units, expert groups were based on the disciplinary "lenses" of English, history, law, and mathematics. The main effort of group planning, remembered as the highlight by participants, was the integration of the different disciplinary angles into the unit.

One night I spent two hours with Annette and Sarah walking them through the statistics and the mathematical opportunities that existed through the "math lens" of the unit. By the end they both were happy to have learned something valuable, and quite appreciative I might add.... I felt that night quite good about the work I did and the work that we did as a threesome. (David)

Sarah, as well, extols the virtues of this collaboration by relating this particularly moving experience: I could tell you guys about the kid who got excited about statistics, that there were more white people in jail than Hispanics. And you could relate because you knew the context. So we shared vocabulary. I think that our "crosstalk" was great.

And another student teacher summarized, appropriating the vocabulary of FCL:

The fact that we came from such different backgrounds and such different strengths, we were able to expand our own limits of knowledge, zones of proximal development, by sharing information, teaching each other, and watching each other teach.... We began to develop as a community on our own. (Case Study STEP Plus)

Thus, seminar participants hallowed their communal planning and sharing. FCL language became the medium through which group members communicated their common experience and vision. Critical inquiry centered around narratives of problematic or uplifting cases of student conduct, teacher insights, or discrete experiences with FCL structures, such as reciprocal teaching (Palinscar, David, and Brown, 1992) or jigsaw events. Missing from the analysis were questions, however, that systematically inquired about the units' quality with regard to central ideas, the scaffolding of group work, or the integration of distributed expertise. Year I ended for many of the seminar participants with a vague understanding of the model, reflection that stressed anecdotes, and analysis that traversed primarily social, rather than intellectual problematics of constructivism. "It was a new thing and like, wow, we felt, in some sense, like we were apart, separate, special in a way" (Case Study STEP Plus). Many participants were clearly inspired and eager to continue with the project beyond their pre-service program. Perhaps there was even a surfeit of ideological commitment to the detriment of keen observation. But would participants be able to use their communal and reflective potential to reach higher levels of technical competence in constructivist teaching?

YEAR II; SKEPTICAL DESIGNERS

In revising our plan of action for Year II, our goal was to reproduce the excitement of the first year, but also to remedy apparent shortcomings in the sphere of technical competence. While the first year seminar induced vision and motivation around the ideal and quest for community, in the second year the project paid more thorough attention to understanding and practice. Good curricular design and careful scaffolding of group work that avoided the "free rider" phenomenon in group work was to increase the chances for community in classrooms and to alleviate classroom management problems student teachers often encountered when attempting a complex student-centered unit. Thus, in Year II the FCL model was treated as a technical tool.

The seminar was planned for a three-quarter sequence. During the first quarter, the students met bi-weekly, in the subsequent two quarters weekly for two and a half hours. Initially 24 student teachers enrolled. Like the "Community" class in Year I, the seminar in Year II engaged in modeling practices of professional community through sharing and reflection exercises during the first quarter. But this time, the philosophical and experiential grounding in the idea of community was given up in favor of a more thorough introduction to principles and structures of FCL. As in Year I, again, confusion set in, but this time participants were less tolerant of the insecurity this caused. This time around, "community"—as philosophy, promise, or experience—had not developed prior to grappling with the intricacies of the model. By the end of the quarter, the seminar lost half of its participants. Some of those leaving indicated that the class had not been "all that helpful" or "not useful enough." Some of those who remained in the seminar desired a program that would not be "just talk." This brought about the utilitarian shift towards structured planning and more t clinically oriented implementation.

The Model

On the part of the instructors of the seminar, the model of FCL was subjected to another reinterpretation that both simplified and complicated matters at the same time. Rather than aiming at the fully fledged FCL instructional system, the seminar aimed at "starter units." The less elaborate starter units consisted of a brief session of initial cognitive dissonance and hypothesis generation, greatly reducing the student research component, followed by a pre-structured expert group phase based on teacher-prepared material. A jigsaw phase and the culminating project complimented the starter unit. Work on the starter unit revolved around structuring content and scaffolding for the participant structures of expert and jigsaw groups. This emphasis on scaffolding, borne out of concerns for the quality of student-centered teaching, had the unfortunate consequence that questions of student autonomy and knowledge construction were neglected.

A strong subject matter emphasis complicated the planning process. The instructors initially asked participants "to refer to the generative ideas of the subject's discipline" (Personal Notes) when making decisions on the big ideas that underlie the unit. A dialogue sequence from one of the early meetings illustrates the complications:

Student teacher: What do you mean by generative ideas of the discipline? Instructor: These are concepts that structure the knowledge base of the subject, for example "race" or "class" in Social Studies, or "adaptation" or "system" in Science.

[Great confusion]

Student teacher: For me generative ideas generate a lot of questions and ideas from the students.

Instructor: They usually do, but they also go a long way in covering territory in your discipline. They generate a lot of answers. If one assumes that the disciplinary knowledge base is organized in a hierarchical pyramid from abstract to concrete concepts, then they would be high up in the structure of the discipline.

Student teacher: I don't understand. It all depends on your own view. For example, in Social Studies my structure may be very different from yours or that of the kids.

[Participants are confused. The exchange continues for a short while and then breaks off without clarification.]

— Instructor's notebook entry: "From now on 'Big Ideas' will do. More intuitive." (Personal Notes)

The sequence not only reveals the difficulty of the instructor in translating the model's framing within cognitive psychology into the less etched, more intuitive frame of the practitioner, it also shows how for the participants the idea of the "discipline" conflicted with the idea of "constructivism," which views the learner's voice as key in constructing knowledge. This tension was coupled with the student teachers' observable reluctance to delve into subject matter content. Dealing with subject matter, to the degree it was initially emphasized in the seminar, caused frustration and resistance. Participants, on one hand, acknowledged feeling uncomfortable because of a lack of knowledge and, on the other hand, did not see the relevance of work on content. Echoing many evaluative comments from participants| Samantha explained:

Like there was a part of me that was engaged somewhat to the intellectual plane, trying to understand what [the instructor] was talking about and make sense of it, but it just really wasn't addressing what I wanted to address, I guess. . . .It was just taught in a non-developmentally appropriate . . . way.

The biggest problem in the seminar was that conceptual work on content (for example, understanding the "generative ideas") and designing group work tasks was a struggle that preempted other activities. According to plan, parallel to planning the FCL unit, student teachers would experiment with and reflect on less complex forms of cooperative learning in their classrooms that were identified as pre-requisites for creating well-developed "Communities of Learners" (for example, working in pairs or groups, reciprocal teaching, listening skills, etc.). But in reality work on these social pre-requisites in student teachers' classrooms increasingly competed with the technical design process. Eventually, attention to these "small steps" towards classroom community ceased as planning the "big unit" became an ever expanding preoccupation. Thus, swapping stories and reflective inquiry on the culture of a constructivist classroom, so important a feature in Year I, dwindled.

The challenges of the planning process made the small planning or design groups ever more important in the work of the class. Wherever possible, the small design groups were composed of teachers who taught the same subject and grade level. In this way, practical collaboration was being facilitated. In these small groups, student teachers did their hands-on work, and they received close mentoring and support from project staff. Eventually, here they gained a more intuitive understanding of the model's knowledge and participant structures so that by the time planning was completed many participants had a fairly good command of the structures. By comparison, the importance of the whole seminar faded. Spending much time and energy in the small design groups increased the effectiveness of actual planning and underscored the product orientation of the seminar, but this task orientation compressed visionary aspects and moral impetus of "community." It is not surprising, then, that in Year II ties between the project and participants extending beyond the duration of the seminar tended to be relationships between student teachers and design group mentors.

Apart from one design group, the ten teachers who finished the seminar taught a starter unit or a shorter jigsaw experiment and debriefed it in the seminar. As could be expected, the results were mixed. Nina described her implementation experience this way:.

The students completed tasks without major discord, but the classroom lacked an atmosphere in which the students valued each others' knowledge and took responsibility for their own learning. I tried to structure activities that forced students to be interdependent, but what resulted instead was "form without substance"—students going through the motions. . . . Why didn't this well-planned unit succeed?

Usually in groups, the less motivated, less skilled students goof off and copy answers. . . . However, [in the expert group] no one read the same material . . . they had to share what they learned (or so we thought). The result was this: Students read their pieces . .. then each person filled out his or her own section of the chart independently and silently. . . . There was very little group discussion of ideas. . . . Despite all these disappointments, the debate [the consequential task] went well, and the students reported learning quite a bit from this unit. They even found it pretty interesting. Through the debate . . . students were able to express their own ideas and for once listened and learned from each other.4

Cindy, in dealing with the refusal of one student to complete an assignment for fear of reprisals from fellow students, came to question her unit. Although she was confident that students "had gained some important insights" and she was "proud of [her] students' efforts" during the unit, she voiced doubts:

Having decided beforehand what we thought the students should learn, we left no room for FCL's notion of student generated questions as the foundation of research. We controlled the direction of the unit from the outset. My first indication that what we had planned to teach was not what students wanted to learn occurred on the first day of the unit. . . . I casually asked my students what they thought about immigration. Their question back to me was, "Do you mean legal or illegal immigration?" That was what they wanted to learn about. . . . It became clear to me that I needed to see what drove my students' opinions if I hoped to change them.5

And Samantha looked back::

I think . . . most of the stuff we did helped me really understand. . . . I think of it as the formal structure of an FCL unit, but not very much how that is sort of a manifestation of the philosophy behind it. . . . It was a pretty rigid set of pedagogical tools that had to go all together and didn't allow for—it was hard to deal with absences and kids that were at different levels. (Exit Interview)

While Cindy and Nina acknowledged the experience of a well-structured and well-planned unit, and Samantha felt familiar with the formal structures of the model as a result of the seminar, they also sensed the limitations of the seminar's approach. The carefully planned units with their intricate and interdependent tasks and activities suffocated what was at the heart of constructivist communities for participants: norms of collaboration, students talking to each other about their interests, teachers being guided by students' curiosities.

Community and Curriculum

There was a mismatch between the emphasis of the seminar and the concern of the student teachers. While the seminar dealt primarily with structure and content, student teachers were primarily concerned with students and their role as teachers. For the novices, these relational aspects were the primary meaning of "Community of Learners" pedagogy. Most participants had joined the seminar because of the "Community" label of FCL and expected a seminar on social concerns, while they were less aware and less interested initially in the intellectual challenge of content that is a component of the model.

The tension between community and curriculum concerns came to a head in one design group that veered from the rest of the FCL seminar.6 In the words of the group's mentor, the group had a strong aversion to thinking about curriculum and pedagogy in terms of content knowledge. Rebecca, one member of the group, articulated the concerns of the group by stating that she "was not used to designing curriculum this way." FCL, the group felt, called for a set of social skills; and group members did not see how to create these kinds of interactions in their classes. Under these conditions, the group stalled at curriculum development. By the end of the second quarter frustrations ran so high that two of the three members left the seminar. One of those leaving commented: "If my concerns [about students] are not important to the discussion of FCL, why do I keep coming?"

Rebecca continued with the seminar, now interacting with her mentor on a one-on-one basis. Rather than resuming the ill-fated design work, Rebecca addressed more pressing issues. She had begun asking herself, as she stated, "What's the point?" and she felt that she had "lost the sense of what she was trying to accomplish in her class." She was hoping to find answers by learning more about her students' lives and needs. This, to her, was the true meaning of "community of learners." With the mentor's help and encouragement, she embarked on an ethnographic study of her classroom, the results of which she movingly presented to the seminar as her final presentation. Clearly, her presentation, in which she criticized the curriculum focus of the seminar, energized other participants. Inquiring about the culture of community seemed eminently sensible and perhaps more necessary and promising than wrestling with curricular structures of perhaps dubious value. On the other hand, Rebecca felt that FCL did not have an influence on her teaching.

Summary

The Year II seminar, with its focus on planning and implementing content and participant structures, helped student teachers gain an understanding of FCL as a tool. Both designing and reflecting mainly took place on the level of the "technical," in Van Manen's (1977) typology. Student teachers discovered that beyond good planning and a well thought-out structure, norms of conduct and receptivity to student interest and ideas were essential. These discoveries were made on the foundation of basic design and implementation skills. The units could function as personal benchmarks of achievement from which novices could move beyond the "technical." But would they?

The problem with the Year II approach is that it challenged novices to foster technical competence without nourishing their most pressing concerns, learning interests, and philosophical convictions. These concerns and convictions, we learned from Year I exuberance and Year II resistance, revolved around culture, not structure. They were grounded in Rebecca's question of "what's the point," and thus in reflections on personal purpose and the moral good of teaching, rather than its technical efficiency for student learning. And the discernment of "what's the point" required practical reflection and inquiry, that is, the comparing of personal ideals with reality. The "specialness" of constructivism and community, Year I and Year II both indicate, flowed from the novices' expectation that FCL principles contained their visions of better teaching and personhood for both teachers and students. By emphasizing skill, keen observation, and technical reason, inspiration was compressed. In making the seminar a challenge of mastering an ambitious model of teaching, FCL ceased to symbolize student teachers' hopes and dreams. Since the teaching results were not perceived as convincingly positive on utilitarian grounds, the seminar ended on a skeptical note. In the second year, constructivist visions of teaching were given practical tools of teaching by emphasizing skill and technical reason. But, inspiration was sapped. Would Year III bring integration?

YEAR III: REFINEMENT OF THE DILEMMA

The Year III plan of action revitalized elements of Year I that had been identified as creating the desirable dynamic of inspiration, most notably the experience of professional community and the sharing of stories and artifacts as a way of conversing about the moral and practical reasons for FCL. But the pendulum did not swing back all the way. Year III was not a rerun of Year I.. The "technical innocence" of the model that had accounted for the exuberance with which first-year participants delved into the planning of their units had disappeared as a result of experiences from Year II. In the third year, the project kept up its careful attention to technical aspects of the model but, in an effort to avoid the mistakes from Year II, was prepared to take greater liberties with the model.

In contrast to the two previous years, the Year III seminar was based at two sites. One group, based at Riverdale High School, an innovative school affiliated with the Coalition of Essential Schools, was composed of four student teachers, two first-year teachers who had graduated from the university program the previous year, and three facilitators. Another group of four student teachers and their project mentors was based at Lincoln Middle School.

The Model

At the heart of the seminar was a "conversation about practice." Contrary to the two previous years, FCL was presented as a set of principles that were connected to concrete images and vignettes of classrooms. The "system-ness" of the psychological learning model was deemphasized. For each of the principles introduced, student teachers were encouraged to bring "artifacts" that would concretize what these principles meant to them. The facilitators emphasized "that these [the principles] are not recipes, but these are ideas that inform the ways . . . to create a learning community" (K. Hammerness, Riverdale Field Notes). And, "an FCL classroom might look different for each of us." After a cursory introduction of an FCL model unit, one of the facilitators assured:

She didn't want anyone to worry that [they] expected them to do these units in this entirety . . . that they hoped to try some bits Mid pieces of these ideas together and talk about it. Maybe eventually [students] might design something more ambitious. (Riverdale Field Notes)

Furthermore, the model starter unit presented to the seminar as a practical example was simplified even further than the Year II type. It now consisted of a student inquiry phase, some expository teacher-centered lessons, an expert phase, and a final group presentation, omitting jigsaw groups. Thus overall complexity of the model, the design task, and facilitators' expectations were reduced for third-year participants. Rather than seeing a whole unit in action, student teachers gradually familiarized themselves with "small steps" that would lead to community in their classrooms. It was not until the end of Year III that participants learned of a whole FCL unit when Samantha, the one member who had been part of the Year II seminar, presented her FCL unit that she had planned the previous year and revised this year. By this time, participants had become familiarized with various elements of the model's principles and structures and had discussed what these elements might mean for their ideals and practices of teaching in their own classrooms. But the timing of the seminar was such that once the pieces of the puzzle had come together, the whole could not be explored any further.

As a result, during Year III, none of the four student teachers taught an FCL unit (or a scaled-down version of a whole unit) that was planned in the context of the seminar. One unit was planned by the two first-year teachers and one student teacher, but only implemented by the most expert novice teacher* Samantha, who had been in the project for her second year. Three student teachers completely opted out of design work when the seminar split off into two small groups. like Rebecca in Year II, they instead embarked on a more thorough exploration of what they called "the social way." Thus, in Year III structure and culture, inquiry on knowledge and interaction did not come together—they bifurcated, with student teachers opting out of curricular work.

The Importance of Units

Martha, a student teacher who decided to opt for the design group, but chose not to implement the unit, described her experiences this way:

All of a sudden I have six awesome teachers to work with. . . . I do believe in group work and the success of group work now whereas I was a very hesitant group work user. Even though I said I was into group work, because I know that you are supposed to be, it's like I didn't really believe it because I hadn't been in a lot of successful groups. ... Well I think that my Romeo and Juliet unit that I did in a group, that was probably my best curriculum design project.

This unit was planned as part of her Methods class. Martha then went on to describe how she used reciprocal teaching in her class and how she adjusted it to her liking and needs. About jigsaw groups she said:

I used a whole massive jigsaw for the whole book, so that is directly out of FCL . . . , but we also learned about that in [another class] and we also learned about that in foundations. . . . In terms of practice, . . . I kept trying to do things because I agree with it and I like it and I wanted to. And I would have, even if I stopped working with my FCL group. I would still keep going on. . . . I want to see students talking to students about [issues], it makes sense to use these practices. . . . If I believe that learners construct their own knowledge, then I'm going to want to use [them], so it's based on a philosophy of education. . . . And I guess my suggestion to FCL would be, if you could show me . . . a few benchmark lessons. And I'd love to see one that really shows me, this is not just like . . . a teacher-centered lesson.

Martha was clearly well versed in the language of FCL; she understood the participant structures, experimented with them, and was motivated to continue experimentation. But it was not the Year III seminar that received the credit for this:

To be really honest, I don't think that it's really affected my pedagogy much. . . . I am talking about specifically my group at Riverdale that I totally value, I mean, it's great... I don't think that, if you control for all the other factors, like say I'm doing stuff, but I'm just not doing my FCL project, then I think you would get the same result.

What was particularly useful in the seminar, and what would have made the class more helpful to Martha?

Seeing, for instance, [Samantha's] unit "What Makes A Good Leader"— that has been influential, I would say, in how to actually structure a unit which is not something I knew before, and it's not something that I learned in another class. . . . I think to actually show an FCL unit earlier on, and like on video. Either do some video or just show the unit on paper. . . . Like I said I never got a binder, I don't know if that would have helped me either, though, but we talked about the mystery that was appearing in the beginning because they [the facilitators] wanted to sort of create it out of the group, but like I said, I just really didn't think that was effective in terms of having it come from the group, because it wasn't coming from the group. . . . They [the facilitators] wanted to do what was good for us, but they were definitely, like they—it wasn't a bottom up organization and that was fine, so you pretend that it was. Do you know what I mean? . . . So that's why I think it would be really good to see, kind of get more information earlier on, in terms of very specific models. I think it would have made everything just a little bit easier, sort of being explicit about what was going on.

In the seminar, Martha did join the design sub-group and therefore could have learned more about the model during the design phase of the unit. She confided that she "lost excitement for this project" 7 early on once she had realized she was not going to implement the unit in her class and it became a planning effort for somebody else's classroom.

Andrea, a student teacher who joined the "social way" group, did not have Martha's opportunity of planning while in the seminar. She viewed her experience in this way:

I look forward to that time of the week [of the seminar] because it's the only time I have where I feel I'm pushed idealistically. Like so much is about survival. . . . So anyway we started with the large group and it started having this philosophical switch, but it was clear it had too many focuses. If anything I would say that was a weakness of the whole program, is that inevitable problem you have when you empower your students, which is you stop having an agenda and you start losing focus. . . . One piece that I feel like I missed is the chance to come out of this class with something concrete, in the sense of like, here is a lesson. . . . I don't have a unit, I don't even have a lesson. What I do have, which is also very helpful is a list that we created in [the Social Way] group about our ideal classroom. And I have found that when I have time, I go back to it sometimes and try not to get overwhelmed with it and take one issue and think about it for a while, sort of a reflective prompt." (Andrea, Exit Interview)

For these participants, the great value of having been in the FCTL seminar was the strong sense of being part of a group of like-minded and similarly affected teachers and the "reflective prompts" they received from the group. They seemed to exit this seminar inspired, reflective, and inquisitive. The seminar attended to their idealistic impulses, their vision, and their quest to understand themselves and their students. Thus, clearly it traversed the terrain of inspiration and reason, yet participants felt, in Andrea's words, "incomplete at this time." The seminar did not satisfy a need for the curricular unit that could have been a useful vehicle to organize knowledge in an extended inquiry and in a set of dialogic participant structures. In Andrea's graphic language: "I would have loved to suck those three [facilitators] dry for all they're worth, because they're worth so much more; so that, not only [would] I have the memories of what they've said in my head, but I'[d] have a model to carry with me" (Exit Interview).

The salience of the unit is corroborated by Abbie's experience as a student teacher based at Lincoln Middle School. As was the case with the Riverdale group, her mentor did not try to teach the FCL model per se, but strongly encouraged her proteges to implement the principles of FCL in their classrooms. These principles conflicted with the segmented and drill-and-practice oriented classroom set up by Abbie's cooperating teacher. The tension between a constructivist vision of teaching and the reality of her classroom climaxed in Abbie's quest for a unit. Here is her vivid account:

I think, number one, I was just tired of what I was doing in the classroom, just dissatisfied with how I was teaching, hating how I was teaching, because it was just so much—I just didn't feel it was the right way to teach, and I wasn't feeling happy with that. I remember the day [the supervisor] came and . . . observed, and I was like crying and I just didn't like what I was doing. The kids seemed okay because that's what they are used to, you know just kind of really, really teacher-centered kind of stuff or whatever, or just things that don't build on each other. They're used to that so they didn't feel like it was a bad thing what I was doing, but I was the one who was dissatisfied, you know, just really unhappy about the whole thing. And I was going: this is not how I want to teach, and I really wanted the kids to get involved and just really to interact with one another, to learn from each other. And most importantly for me, to feel satisfied with my teaching, was to see a real build-up. I remember I kept saying to [the supervisor], "I have not created a lesson." My goal is to create a lesson that's complete and that flows and has a coherent beginning, middle, and end, that kind of thing; and that really bothered me that I hadn't created a unit like that yet. And that was one of the breaking points. That day [the supervisor] observed, I was just like I can't, I don't like this at all. . . . And then meeting with [the supervisor], that really just brought stuff together, how to lay out a unit and how to go from beginning, middle, and end; and how to start from the end and go back to see what is the final product, what's the goal of this unit, and then to come back. And just go about implementing it. And seeing the results, too. The results that kids are really grasping something and getting some skills here.8

Thus, Abbie's breaking point was turned productively with the help of her supervisor who understood that she needed the tools to make her vision practical. That tool was the design of a unit:

I think I was heading to the breaking point thing because of what I was learning [in the teacher education program]. . . . Although all that stuff is theoretical and is hard to apply practically, that stuff really shook me up practically. . . . I think the key was like for me the scaffolding, the integrating of everything, the integration of all the steps into the unit, and having that goal.

Yet, it is important to note that in Abbie's case the practical design work could only become fruitful once she had built up to this tension between vision and practice. An earlier exploration of an FCL starter unit had no effect on her. She said:

I know when we met [at first], the focus was . . . mainly on a Social Studies unit, . . . and I personally felt it wasn't really helpful then because it started focusing so much just on that one unit. I was just sitting there listening passively and not really being able to get anything from it.

Abbie, Martha, and Andrea confirm that for novices to engage in con-structivist teaching vision and reasoning about practice are indispensable entry points, but also that skills and particularly design knowledge on the scale of the unit become essential at some point in the process. None of the student teachers in the Riverdale group implemented units that were planned in the context of the seminar; yet, in the end they came to miss not having had exposure to design and implementation of a whole unit.

Samantha, who more than anybody else in the project achieved a synthesis of skill, reasoning, and inspiration during her two years of involvement and her intense co-teaching experience with one of the FGTL facilitators, compared Year II and Year III this way:

Last year, . . . where we actually did an FCL unit, that was one of the most important things for me to understand. . . . So [when I compare] what I had last year which was a lot of the time too theoretical. . . with this year, [during] which we talked about practice a lot, like combining the two a little bit, I think, would be really helpful. . . . I already have the theory . . . but I think for them, for people who are just getting one year, I think that it would be helpful to combine the two a little bit more. But . . . it would, I think, involve a little bit more investment on the part of student teachers, so . . . there would have to be a more tangible reason why this was helping them. (Riverdale Group Interview)

How could this combination be achieved, and is it possible to do it in one year?

DISCUSSION

In our attempt to help student teachers teach FCL in their classrooms, the project tried fusing inspiration, reasoning, skill development, and design work three times in three program iterations. In the first year, the program generated a great deal of inspired pioneering, but technical skill and keen observation was submerged at times in ideological commitment, and understanding of the model was truncated. In the second year, the program placed great emphasis on the mastery of the model aiming at clinical tryouts, but it sapped the novices' inspiration by overburdening them with abstract theory and fixed pedagogical forms that disconnected the model from the vision and the philosophical and moral reasons of teaching it. In the third year, the program concentrated on practical inquiry and careful bottom-up reflection to develop classroom community. Novices maintained their vision and motivation for FCL, left the project with fond memories and "reflective prompts," but missed fundamental design competencies, encapsulated in units. Thus, none of the program iterations stands out as a shining example of success, but together our seesawing endeavor demonstrates that all of the elements that we tried out with different emphasis from one year to the next are indispensable components of a program that aims at constructivist pedagogy in student teachers' classrooms.

Generating vision and motivation sustained by collegial reflection and support, inquiring about social prerequisites of classroom community (that is, the "small steps"), and designing coherent teaching units are all necessary ingredients of a suitable program if the aim is to enable novices to teach ambitious constructivist models such as FCL. When one of these components was neglected, either student teachers themselves lamented its absence, sometimes with the consequence of dropping out of the project altogether, or project facilitators had to content themselves with only limited understanding or implementation of constructivist teaching in the classroom. None of these components turned out to be a lever that could drive the learning process without the others being attended to simultaneously (speaking in the time frame of one year).

Inspiration which combined visions of good teaching, the quest for self-actualization, and the motivation to act was crucial because constructivist teaching in all its complexity could not be adopted by novices in the program on purely utilitarian grounds. Although discrete incidences of success, such as "the jigsaw that did not end when the bell rang," were celebrated by the novices—in their rarity, one should add—results from FGL experiments were at best mixed and at worst disappointing for the novices. Most often, students turned out to be the strongest "obstacle" of success when they did not behave according to the expectations of the pedagogy. Neither did student teachers have the expertise, nor the persistent influence over their own classrooms, to create a successful learning community in their classes. The second reason was that constructivist teaching required additional investment on the part of the novices, both in terms of preparation time and in terms of teaching "against the grain" of the institutional norms of the school. This project was blessed with students who came in with a vision concurrent with constructivist principles. But even those initially committed participants needed space in the program to replenish their inspiration to try the pedagogy. Time for self-exploration and supportive conversations about students were crucial in this regard

On-going collegial inquiry about practical first steps towards learning communities was essential because it sustained motivation and helped bridge the gap between theory and practice. Student teachers were aware of the tension between high-flying theory and the reality of student behavior in their classrooms and suffered from their limited skills in creating an atmosphere of respect and mutual trust. A gradual path of experimentation was therefore called for. In addition, the conceptualization of the FCL model was often abstract and needed to be appropriated by both student teachers and mentors in more practical terms. Particularly Year I and Year II are replete with examples of failed bridging due to the limited understanding of the model on the part of facilitators (Year I) and due to the lack of grounding in daily classroom experiences (Year II). In Year I, novices and facilitators tended to compensate incomprehension with bold strokes that often went astray, while in Year II, the presentation of the model in its abstractness turned participants off.

Designing curricular units was a backbone of experimentation with con-structivist teaching. Student teachers in all three years stressed their value or deplored their absence. Once prepared, units became legacies. They functioned as recurrent blueprints and benchmarks for further experimentation. They integrated experimentation with both the intellectual and social challenges of FCL by providing the format for a more extended inquiry process in student teachers' classrooms that was structured but also allowed for more student autonomy. But without first generating and then sustaining a critical tension between constructivist vision, intrepid experimentation, and structured teaching, student teachers saw the units either as meaningless or as shackles to an inquiry-oriented classroom culture.

In the three years of its existence, the project worked with about thirty-five student teachers who engaged in some form of design work. None of the student teachers reached a level at which they could handle FCL in all of its complexity. It is safe to say that almost none of the teachers we had the opportunity to observe were able to create the kind of learning community envisioned by the pedagogy. Moreover, project facilitators felt compelled to simplify the model gradually from year to year. The model requires both enormous savvy and craft in the fields of both curriculum and classroom management that seem to eclipse the resources of beginning teachers and often the skills of project facilitators as well. This situation was exacerbated by the developmental stage of the model itself that made it hard for both student teachers and facilitators to figure out what FCL would look like in regular classrooms.

Can it all be done? Apparently, the project could not cover all the essential bases for learning FCL in the time allotted, despite the fact that the philosophy of the teacher education program was very conducive to the project's purpose and despite the fact that FCL's message was reinforced by other classes of the larger program even if the exact model was not taught. As a result, the project could either attenuate its goals or extend its time frame. Given severe time constraints in the already crowded teacher education program, the project chose to simplify the model over the years and to scale down its ambitions.

In helping student teachers to master an ambitious constructivist model, we made adjustments from year to year as we encountered problems. Each new iteration made us look at the problem of constructivist teacher education from a different angle, each iteration showing us what was amiss, what was essential. In the practical implementation of the program we ended up with two distinct phases: during one phase we taught the technical model and worked with student teachers on deep content; during the other phase we engaged students in on-going inquiry about the principles and reality of constructivism and community. This happened despite our avowed determination to integrate the two. In Year I, when the two phases were probably integrated the most, participants produced teaching units with great "technical innocence" and only a hazy understanding of what FCL was all about. In Year II, design work preempted sustained inquiry, whereas in Year III student teachers, for the most part, never reached the stage of design within the time frame of the one-year seminar. Some kind of phasing seems necessary in order to cut the sheer complexity of the task. In addition, the one-year seminar could possibly be followed up by learning phases in subsequent years. Judging from the small evidence of three iterations, the "jumping-in" approach (Year I) or the approach of slowly building up to the model through inquiry and experimentation (fear III) may be more advantageous than the Year II approach of teaching the model and designing units, since "jumping-in" and "slowly building up" more strongly sustained motivation for further learning among participants than the design approach.

We could not delimit our definition of constructivism: it was always both a model of teaching with a specific set of content, task, and participant structures and a set of inquiry-orienting principles for which student teachers found many pedagogical forms. When we swayed towards teaching the model and engaging student teachers in structured design work on disciplinary knowledge, we were apt to lose participants along the way for whom the idea of learning communities meant the construction of knowledge through a communal reflection on individual experiences, both in the university seminar and in their own classrooms. When we were truly "con-structivist," that is, when we involved student teachers in "constructivist talk" (MacKinnon and Scarff-Seatter, 1997, p. 51) on their own terms, eschewing the model in theory, student teachers' understandings and experimentations with complex forms of constructivist teaching tapered off while their calls for structure intensified. We were truly caught in a dilemma that seemed unsolvable within the format of the seminar.

Our constructivist teacher education experiment is short on clear answers, but a vivid case of unfolding practical dilemmas. It is possible that our travails were due to the limited experience of project staff with FCL or constructivist teacher education. While the project overall was directed and run by experienced teacher educators and classroom teachers, each year the composi-tiori of the team changed, thus preventing continuity of our debates within the project and contributing to the course corrections from one year to the next. As Year I staffers indicated in the interviews (see Year I: Inspired Pioneers), due to their limited exposure to FCL they were "truly teachers as learners." In Year II, the director of the seminar had prior experience in designing cooperative learning units of similar complexity and form as FCL, but the Year I experience was only accessible to him through data; he advocated an emphasis on design work to increase the quality of units that were planned and implemented in Year I. In Year III, the team was strongly influenced by staffers who had seen the previous two iterations and felt that they could offer a course that might finally get it right, by emphasizing previously successful elements and avoiding pitfalls. Our project was a success in that student teacher participants were for the most part appreciative of the seminar experience and felt they had grown in many ways through their participation. It was a success in that we as teacher educators and researchers learned from it to tailor realistic expectations and to embrace the multiple facets, tensions, and dilemmas that the construction of knowledge entails for teachers and teacher educators. It is up to other teacher educators to learn from our mistakes and find a better way—perhaps, to get it all done.

This article is dedicated to the memory of Ann Brown who with her husband Joe Campione developed "Fostering Community of Learners." Their bold ambition set close to fifty teachers and teacher educators on a course of re-invention. I thank the "founders" for the lessons we learned. I also want to thank the teachers for their valuable time and insights as well as the Andrew W. Mellon Foundation for their generous funding. This article would have been impossible without the work of my collaborators in the project "Fostering Communities of Teachers as Learners." They are Betty Achinstein, Mercy Carbonnell, Karen Hammemess, Robert Jordan, Heather Kirkpatrick, Carolyn Krohn, Edie Mendez, Tom Meyer, Kay Moffett, Miriam Gamoran Sherin, Joel Westheimer, and Jennifer Whitcomb. Special thanks to Lee and Judith Shulman for their guidance and leadership.

REFERENCES

Aronson, E. (1978). The jigsaw classroom. Beverly Hills: Sage.

Ball, D. (1990). Reflection and deflections of policy: The case of Carol Turner. Educational Evaluation and Policy Analysis, 72(3), 263-275.

Berlak, A., &: Berlak, H. (1981). Dilemmas of schooling: Teaching and social change. London: Methuen.

Berliner, D. (1986). In pursuit of the expert pedagogue. Educational Researcher, 15(7), 5-13.

Beyer, L. (1984). Field experience, ideology, and the development of critical ref lexivity. Journal of Teacher Education, 25(3), 36-41.

Beyer, L. (1991). Teacher education, reflective inquiry and moral action. In R. Tabachnik & K. Zeichner (Eds.), Issues and practices in inquiry-oriented teacher education (pp. 113-129). London: Falmer.

Black, A., & Ammon, P. (1992), A developmental-constructivist approach to teacher education. Journal of Teacher Education, 43(5), 323—335.

Brophy, J., & Good, T. (1986). Teacher behavior and student achievement. In M. Wittrock (Ed.), Handbook of research on teaching (pp. 328-375). New York: Macmillan.

Brown, A. (1992). Design experiments: Theoretical and methodological challenges in creating complex interventions in classroom settings. Journal of the Learning Sciences, 2(2), 4-12.

Brown, A., & Campione, J. (1990). Communities of learning and thinking, or a context by any other name. Contributions to Human Development, 21, 108-126.

Brown, A., & Campione, J. (1994). Guided discovery in a community of learners. In K. McGilly (Ed.), Classroom lessons: Integrating cognitive theory and classroom practice (pp. 229-270). Cambridge: MIT Press.

Brown, A., & Campione, J. (1996). Psychological learning theory and the design of innovative environments: On procedures, principles, and systems. In L. Schauble & R. Glaser (Eds.), Contributions of instructional innovation to understanding learning. Hillsdale: Erlbaum.

Bullough, R., & Gitlin, A. (1991). Educative communities and the development of the reflective practitioner. In R. Tabachnik & K. Zeichner (Eds.), Issues and practices in inquiry-oriented teacher education (pp. 113-129). London: Falmer.

Calderhead, J. (Ed.). (1988). Teachers' professional learning. London: Falmer.

Calderhead, J. (1989). Reflective teaching and teacher education. Teaching & Teacher Education, 5(1), 43-51.

Cochran-Smith, M. (1990). Reinventing student teaching. Journal of Teacher Education, 42(2), 104-118.

Cochran-Smith, M. (1991). Learning to teach against the grain. Harvard Educational Review, (51(3), 279-310.

Cohen, D. K. (1990). A revolution in one classroom: The case of Mrs. Oublier. Educational Evaluation and Policy Analysis, 23(2), 311-330.

Combs, A. A personal approach to teaching. Boston: Allyn and Bacon.

Comeaux, M. (1991) But is it "teaching"? The use of collaborative learning in teacher education. In R. Tabachnik &: K. Zeichner (Eds.), Issues and practices in inquiry-oriented teacher education (pp. 226-252). London: Falmer.

Cuban, L. (1991). Stability and change in curriculum. In P.Jackson (Ed.), Handbook of research in curriculum. New York: Macmillan.

Dewey, J. (1904). The relation of theory to practice in education. In C. Murray (Ed.), The relation of theory to practice in the education of teachers (pp. 1-30). Chicago: University of Chicago Press.

Doyle, W. (1986). Classroom organization and management. In M. Wittrock (Ed.), Handbook of research on teaching (pp. 392-431). New York: Macmillan.

Educational Researcher, 23(4), 4-23, October 1994, Exchange.

Elbaz, F. (1983). Teacher thinking: A study of practical knowledge. New York: Nichols.

Elliott, J. (1993). Professional education and the idea of a practical educational science. In J. Elliott (Ed.), Reconstructing teacher education (pp. 65-85). London: Falmer.

Feiman-Nemser, S. (1990). Teacher preparation: Structural and conceptual alternatives. In W. Houston (Ed.), Handbook of research on teacher education (pp. 212-233). New York: Macmillan.

Feiman-Nemser, S., & Buchmann, M. (1985). Pitfalls of experience in teacher education. Teachers College Record, 87(1), 53-65.

Fenstermacher, G. (1986). Philosophy of research on teaching: Three aspects. In M. Wittrock (Ed.), Handbook of research on teaching (pp. 37-49). New York: Macmillan.

Fullan, M. (1993). Change farces. London: Falmer.

Fullan, M., & Hargreaves, A. (Eds.). (1992). Teacher development and educational change. London: Falmer.

Goodlad, J. (1984). A place called school. New York: McGraw Hill.

Goodlad,J. (1990). Teachers for our nation's schools. San Francisco: Jossey-Bass.

Goodman, J. (1986). University education courses and the professional preparation of teachers: A descriptive analysis. Teaching and Teacher Education, 2(4), 341-353.

Joyce, B., & Showers, B. (1980). Improving inservice training: The message of research. Educational Leadership, 37, 379-385.

Knowles, G. (1993). Life-history accounts as mirrors: A practical avenue for the conceptualization of reflection in teacher education. InJ. Calderhead & P. Gates (Eds.), Conceptualizing reflection in teacher development (pp. 70—92). London: Falmer.

Lampert, M. (1990). When the problem is not the question and the solution is not the answer: Mathematical knowing and teaching. American Educational Research Journal, 27(1), 29-63.

Little, J. (1982). Norms of collegiality and experimentation: Workplace conditions of school success. American Educational Research Journal, 1P(3), 325-340.

Lortie, D. (1975). Schoolteacher. A sociological study. Chicago: University of Chicago Press.

Louis, K., & Kruse, D. (1995). Professionalism and community: Perspectives on reforming urban schools. Thousand Oaks: Corwin Press.

MacKinnon, A., & Scarff-Seatter, C. (1997). Constructivism: Contradictions and confusions in teacher education. In V. Richardson (Ed.), Constructivist teacher ducation: Building a world of new understandings (pp. 38-55). London: Falmer.

Mclntyre, D. (1993). Theory, theorizing and reflection in initial teacher education. In J. Calderhead & P. Gates (Eds.), Conceptualizing reflection in teacher development (pp. 39—52). London: Falmer.

McLaughlin, M., &: Talbert, J. (1993). Contexts that matter for teaching and learning. Stanford, GA: Center for Research on the Context of Secondary School Teaching.

Meyer-Smith, J., 8c Mitchell, I. (1997). Teaching about constructivism: Using approaches informed by constructivism. In V. Richardson (Ed.), Constructivist teacher education: Building a world of new understandings (pp. 129-153). London: Falmer.

Mosenthal, J., & Ball, D. (1992). Constructing new forms of teaching: Subject matter knowledge in in-service teacher education. Journal of Teacher Education, 43(5), 347—356.

Palinscar, A., David, Y, & Brown, A. (1992). Using reciprocal teaching in the classroom: A guide for teachers. Unpublished manuscript.

Richardson, V. (1997). Constructivist teaching and teacher education. In V. Richardson (Ed.), Constructivist teacher education: Building a world of new understandings (pp. 1—14). London: Falmer.

Richert, A. (1991). Case methods and teacher education: Using cases to teach teacher reflection. In R. Tabachnik & K. Zeichner (Eds.), Issues and practices in inquiry-oriented teacher education (pp. 130-150). London: Falmer.

Russell, T. (1988). From pre-service teacher education to first year of teaching: A study of theory and practice. In J. Calderhead (Ed.), Teachers' professional learning (pp. 13-34). London: Falmer.

Schifter, D., & Fosnot, C. (1993). Reconstructing mathematics education: Stories of teachers meeting the challenge of reform. New York: Teachers College Press.

Schon, D. (1983). The reflective practitioner. San Francisco: Jossey-Bass.

Shulman, J., & Colbert, J. (1988). The mentor teacher case book.S&n Francisco: Far West Laboratory.

Shulman, L. (1987). Knowledge and teaching: Foundations of the new reform. Harvard Educational Review, 57(1), 1-22.

Shulman, L. (1988). The dangers of dichotomous thinking in education. In P. Grimmet & G. Erickson (Eds.), Reflection in teacher education (pp. 31-38). New York: Teachers College Press.

Shulman, L. (1997). Communities of learners and communities of teachers. Jerusalem: Mandel Institute. Stodolsky, S. (1988). The subject matters. Classroom activity in math and social studies. Chicago: University of Chicago Press.

Stodolsky, S., & Grossman, P. (1995). The impact of subject matter on curricular activity: An analysis of five academic subjects. American Educational Research Journal, 32(2), 227-249.

Tann, S. (1993). Eliciting student teachers' personal theories. In J. Calderhead and P. Gates (Eds.), Conceptualizing reflection in teacher development (pp. 70—92). London: Falmer.

Van Manen, M. (1977). Linking ways of knowing with ways of being practical. Curriculum Inquiry, 6(3), 205-228.

Valli, L. (1993). Reflective teacher education programs: An analysis of case studies. In J. Calderhead & P. Gates (Eds.), Conceptualizing reflection in teacher development (pp. 11-22). London: Falmer.

Veenman, S. (1984). Perceived problems of beginning teachers. Review of Educational Research, 54(2), 143-78.

White, J. (1991). War stories: Invitations to reflect on practice. In R. Tabachnik & K. Zeichner (Eds.), Issues and practices in inquiry-oriented teacher education (pp. 226-252). London: Falmer.

Wilson, S., & Wineburg, S. (1988). Peering at history through different lenses: The role of disciplinary perspectives in teaching history. Teachers College Record, 89(4), 525-539.

Winitzky, N., &: Kauchak, D. (1997). Constructivism in teacher education: Applying cognitive theory to teacher learning. In V. Richardson (Ed.), Constructivist teacher education: Building a world of new understandings (pp. 59—83). London: Falmer.

Zeichner, K., & Listen, D. (1987). Teaching student teachers to reflect. Harvard Educational Review, 57(1), 23-48.

Zeichner, K., & Tabachnik, R. (1991). Reflections on reflective teaching. In R. Tabachnik & K. Zeichner (Ed.), Issues and practices in inquiry-oriented teacher education (pp. 1-21). London: Falmer.

RICK MINTROP is a former teacher and teacher educator. He is assistant professor at the University of California at Los Angeles, specializing in the relationship between institutional reform and educational practice. He is author of "Changing Core Beliefs and Practices Through Systemic Reform: The Case of Germany After the Fall of Socialism" (Educational Evaluation and Policy Analysis, Fall 1999). Currently he is researching the effect of school accountability on school practices in U.S. schools.