The national systemic school reform effort has assumed that sending clear and consistent signals to teachers and students about what is important to teach and learn is an essential element of school improvement (Knapp, 1997). As a consequence, the focus for the past decade has been on creating and aligning policy instruments such as curriculum frameworks, standards, and assessments (Cohen & Spillane, 1994; Fuhrman, 1993; Goertz, Floden, & O'Day, 1996). Although those promoting systemic reform "seek much more coherent and powerful state guidance for instruction" (Cohen, 1995, p. 11), the experience of policy alignment in at least one state suggests that this strategy has yet to provide significant assistance for practitioners. Cohen said that

	while systemic reform brought a broad drift toward intellectually more ambitious instruction at the state level [California] for about a decade, thus far it has not brought more coherence to state guidance for instruction...The guidance for instruction that many local central offices offer to schools has begun to shift in the direction of reform, but that shift has so far not been accompanied by greater local coherence...Reforms that seek more coherence in instructional policy have helped create more variety and less coherence...State guidance added messages, but so did local agencies. Nothing was subtracted. (p. 12)

Hargreaves (1994) described similar paradoxes that create challenges for teachers working in an increasingly complex world. Situating the work of teaching in the wider social context, Hargreaves argued that teachers are being asked to do more, but with less time and support to learn how to meet the new demands. It is worth hearing Hargreaves' argument in full.

	First... the teacher's role expands to take on new problems and mandates - though little of the old role is cast aside to make room for these changes. Second, innovations multiply as change accelerates, creating senses of overload among teachers and principals or head teachers responsible for implementing them. More and more changes are imposed and the timelines for their implementation are truncated. Third, with the collapse of moral certainties, old missions and purposes begin to crumble, but there are few obvious substitutes to take their place. Fourth, the methods and strategies teachers use, along with the knowledge base which justifies them, are constantly criticized - even among educators themselves - as scientific certainties lose their credibility. If the knowledge base of teaching has no scientific foundation, educators ask, "on what can our justifications for practice be based?" What teachers do seems to be patently and dangerously without foundation. (p. 4)

Recent reform efforts have had but minimal impact on classroom progress. Teachers are making instructional decisions in a more fluid context that includes new policies, new ideas about learning, instruction, and assessment, and many programs that claim to reflect these new ideas. The multiplicity and diversity of messages about improving classroom practice only confounds the decision-making process for teachers. Teachers interpret these messages in very different ways depending on their experiences, beliefs, students, and school culture. Thus, the way a particular reform program is implemented will vary greatly from teacher to teacher and may be quite different from the expectations of the reformers (Jennings, 1996; Grant, Peterson, & Shojgreen-Downer, 1996; Peterson, McCarthey, & Elmore, 1996). From their study of teachers and a mathematics curriculum reform, Grant et al. (1996) concluded that teachers are not adequately supported in their efforts to make connections between new ideas presented as reform and the enactment of these ideas into practice.

Some educators have cautioned that school improvement will only be achieved when there is greater clarity and coherence in the minds of the majority of teachers (Fullan, 1996), and that "coherence in policy is not the same thing as coherence in practice" (Cohen, 1995, p. 16). From this perspective, educational practice will change only when teachers have the support they need to make sense of new ideas and directives, bring them together in a meaningful way, and construct a coherent practice.

In the new reform climate, it appears teachers have little time and less guidance to learn - or rethink and relearn - how to make the best decisions about what and how children learn. Many teachers, therefore, make instructional decisions based on their immediate needs to comply, survive, conform, or meet a time constraint (Hargreaves, 1994). It is easier for them to rely on external sources of authority, such as curricular documents, assessments, textbooks, and teachers' guides, to provide the guiding vision for their instruction than to rethink and reform that practice. Reliance on these materials, which are designed for use across a large number of classrooms by a diverse group of teachers with some typical student, can promote teaching that is routine and unthinking. Yet, as Coldron and Smith (1995) contend, "teaching which is routine and unthinking sells pupils and teachers short [italics added]; learning to teach and sustaining professional development require reflection which is closely linked to action" (p. 1).

In a similar vein, Elmore (1996) argued that changing the structures of schooling will have little impact on how and what students learn unless there are also changes in the "core" of educational practice (i.e., how teachers understand knowledge and learning and how they operationalize their understandings). Therefore, what Cohen (1995) calls "coherence in practice" depends more on how teachers understand, interpret, and internalize the reform messages for their own practice than on the alignment of those messages at any policy level.

The success of school improvement thus rests squarely on teachers, and, by association, on those responsible for supporting their professional growth. Darling-Hammond (1996) said that "betting on teaching as a key strategy for reform means investing in stronger preparation and professional development while granting teachers greater autonomy...we must put greater knowledge directly in the hands of teachers" (p. 5, 6). The challenges are to create the time for teachers to engage in their own learning, place value in that learning, and develop useful and effective strategies to help teachers learn how to make the best decisions in their teaching practices. Researchers have found that teachers who have opportunities to dialogue, study, and reflect on teaching and learning with their colleagues seem better able to improve their practices (Lieberman, 1995; Wilson, Peterson, Ball, & Cohen, 1996). As a result, there has been a recent shift from professional development models emphasizing "acquisition of discrete skills and behaviors" to more complex models emphasizing teachers learning in professional communities (Little, 1997, p. 2) and enabling teachers to become more reflective practitioners.