Against Best Practice: Uncertainty, Outliers, and Local Studies in Educational Research

Introduction

Everywhere one turns, there is talk of “best practice” in education and teacher education, a concept taken as an unqualified good – and always and everywhere an appropriate aim. It is also a source of much mischief, as will be noted. This chapter explores aspects of this ambition, situating it in an admittedly sweeping history and in relationship to a few select insights drawn from the writings of John Dewey, among others. Along the way, I wish to question not just the possibility but also the desirability of the aspiration, while arguing for a more modest ambition – “better practice.” Better practice is born of research especially attentive to outliers and to local contexts understood not merely as being research venues but as representing a moral stance. This stance is characterized by humility in facing the complexity of education, a complexity that is not yet fully or adequately appreciated (Mason, 2008), and of profound respect for those whose work researchers seek to understand.

The Roots of an Ambition: The Royal Society and Political Arithmetic

To set the context for discussion of best practice, a good place to begin is 1667, when Thomas Sprat published a rather remarkable book, The History of the Royal Society. Although in some respects beginning here is arbitrary, the founding of the Royal Society set a clear ambition that played out across the centuries, culminating in a culture-wide belief in, and seductive pursuit of, what we now call in the West “best practice”: single, best responses or solutions to complex human problems. The full title of Sprat’s book is The History of the Royal-Society of London, for the Improving of Natural Knowledge. Described by Wood (1980) as an apology for the work of the Society, then only a few years old, the history gives a glimpse into the values of at least some prominent Society members and into the issues that concerned them as they tried to make the case for the worth of the then emerging forms of inquiry loosely comprising what we would now call science. The history assures potential Society sponsors, most notably the Crown, that its work will be a force for maintaining social stability and order, will strengthen the church by shedding light on God’s creations, but will do nothing to challenge the place of the inherited knowledge of the ancients, while creating economically useful knowledge and new technologies, which will increase national wealth. The promise was that Society members would speak the truth both as good and humble Christians and as rational natural scientists.

Seeking to avoid the twin dangers of skepticism and enthusiasm, such knowledge would result from the firm facts of experimentation, presented dispassionately and simply in short declarative sentences devoid of all dogmatism or any hint of speculation. As Sprat wrote:

There is one thing about which the Society has been most solicitous; and that is, the manner of their Discourse: which, unless they have been very watchful to keep in due temper, the whole spirit and vigour of their Design, had been soon eaten out, by the luxury and redundance of Speech. The ill effects of this superfluity of talking, have already overwhelmed most other Arts and Professions; insomuch, that when I consider the means of happy living, and the causes of their corruption, I can hardly forbear concluding, that eloquence ought to be banished out of all civil societies, as a thing fatal to Peace and good Manners. (Sprat, 1667, 1958, p. 111)

Sprat argued for forbearance when making truth claims. It was, he thought, the Society’s responsibility to make judgments of fact, and once set, the facts would speak for themselves. Setting aside partisan debate, the “redundance of speech,” facts were to be verified systematically through “repetition of the whole course of the Experiment […] never giving over till the whole Company has been fully satisfied of the certainty and constancy; or, on the other side, of the absolute impossibility of the effect” (Sprat, 1667, 1958, p. 99). Consensus was the aim. Note in the above, three phrases: a “mind too changeable” (to be avoided), “right practice” (to be embraced as means for assuring), and “certainty and constancy.” Later, I will have a few words to say about “repetition of the Experiment.”

As Sprat was writing the history, other developments of interest were unfolding, especially on the continent, that would have a direct bearing on the Royal Society’s early concerns. The 1660s, as Porter (1986) has shown, witnessed the emergence of what William Petty came to call “political arithmetic.” The purpose of political arithmetic “when not confined to the calculation of insurance or annuity rates, was the promotion of sound, well-informed state policy” (p. 18). Political arithmetic, as Petty wrote, was to bring “puzzling and perplext Matters to Terms and Number, Weight and Measure” so that official policy might be grounded in an understanding of the land and its inhabitants (p. 19). William the Conqueror’s Doomsday Book of 1086 anticipated this aim, and ever since and through various means, governments have actively sought to know as much as possible about those they govern for various administrative, social engineering, and social and political purposes.

The ambitions of modern statistics have been rooted in political arithmetic. The aim was to gain a clear picture of a nation’s lands and its peoples for the purpose of exercising ever more effective control over both. As Porter wrote, the “great merit of statistics was that it eliminated perturbations by ignoring individuals and letting their unpredictable activities average out” (Porter, 1986, p. 152), thus facilitating consistent social planning and policy. A foundational assumption of political arithmetic was the existence of what Porter described as a “common personhood”: “statistics tended to equalize subjects. It makes no sense to count people if their common personhood is not seen as somehow more significant than their differences” (p. 25). Hence, the dominating interest of statistics was in averages, understood as offering a human or national type, which now includes even something known as a “proficient fourth-grader” or an “on grade level first-grader.” Concern for human variability and diversity came late, very late indeed. Averages (and the human types that presumably portrayed them) came to be linked tightly to conceptions of normalcy, a linkage that has proven both enduring and deeply troubling, as Hacking (1990) argued:

Words have profound memories that oil our shrill and squeaky rhetoric. The normal stands indifferently for what is typical, the unenthusiastic objective average, but it also stands for what has been, good health, and for what shall be, our chosen destiny. That is why the benign and sterile-sounding word “normal” has become one of the most powerful [of] ideological tools […] (p. 169)

Holding a “cautious optimism for improvement,” yet desirous of buttressing social order, by the 1830s statisticians hoped that “the confusion of politics could be replaced by an orderly reign of facts” (Porter, 1986, p. 27). The ambition was utopian: the normalization of individuals for the good of the state and of Society. Underpinning both Sprat’s dream for the Royal Society of establishing natural facts and the history of statistics, with its driving concern for social order, was a lively and often desperate quest for certainty.

John Dewey, Uncertainty, and a Science of Education

At a much later period of great uncertainty, 1929, Sprat’s ambitions were widely thought to be within reach, but not by every thoughtful commentator. In that year, John Dewey published The Quest for Certainty: A Study of the Relation of Knowledge and Action (1929a), his Gifford lectures. Speaking to his time (but echoing in ours), Dewey well understood what is felt when life is precarious and outcomes seem arbitrary: “The quest for certainty is a quest for a peace which is assured, an object which is unqualified by risk and the shadow of fear which action casts […] Perfect certainty is what man wants. (Dewey, 1929a, pp. 8, 21). For Dewey, certainty in a perplexing, contradictory, and dangerous world was the worst sort of illusion, ultimately requiring disengaging from the world and encouraging a passive acceptance of life’s offerings. In contrast, he called for courage in the face of uncertainty, along with engagement, openness, responsibility, moral action and, perhaps ironically, for humility, a recognition that there are many things in life we cannot control, some things we should not even seek to control, and many questions that strongly and perhaps always will resist resolution.

In education, most of the important issues come in the form of dilemmas to be managed, not problems to be solved. On this point, Dewey stated, “intellectual progress usually occurs through sheer abandonment of questions altogether. We do not solve them: we get over them” (Dewey, 1910, p. 19). Ultimately, his call was to move away from relying on others to set life’s aims and means: for example, the Royal Society deciding what would count as the facts and the statisticians determining what data would be legitimate for aspiring social engineers and policy-makers. He embraced a wide-ranging experimentalism, taking life as an adventure requiring of each person their full engagement and most honest and skilled thinking and practice. This sort of thinking results in a kind of functional “stability rather than certainty” (Baez & Boyles, 2009, p. 63), a flowing but precarious unity grounded in an acute situational sensitivity that facilitates and also requires consistent and moral action (see Gale, 2010).

The subtitle of Dewey’s book is worth noting: “A study of the relation of knowledge and action.” His concern was for the sort of knowledge that enables purposeful and effective action, even though in the important matters of life, certainly when confronting the problems of education, that knowledge is always partial, never really fully adequate. Also in 1929, Dewey’s Kappa Delta Pi lectures, The Sources of a Science of Education (1929b), were published. Connections between his Gifford lectures and the Kappa Delta Pi lectures, which were directed specifically toward educators, are readily apparent. In an uncertain, and in some senses crazy world, what can educators do and where should they look for guidance as they seek to better nurture and educate the young? What, then, are the sources of a science of education?

In this small volume, Dewey had a lot to say about improving educational practice, offering several wise warnings directed to those who primarily taught and those who primarily conducted research on teaching that speak directly to the aspiration for best practices and sure outcomes. Throughout his argument, Dewey assumed the impossibility of certainty in human affairs, suggesting that education is most of all an art, “either a mechanical art or a fine art” that “progressively incorporates more and more of science into itself” (Dewey, 1929b, p. 13). His warnings began with a discussion of the dangers that follow the reduction in educational practice to rules, the deontological and pharisaic stuff of Sprat’s right practice:

When in education, the psychologist or observer and experimentalist in any field reduces his findings to a rule which is to be uniformly adopted, then, only, is there a result which is objectionable and destructive of the free play of education as an art. (p. 14)

It is very easy for [the results of research] to be regarded as a guarantee that goes with the sale of goods rather than as a light to the eyes and a lamp to the feet […] prized for its prestige value rather than as an organ of personal illumination and liberation. (p. 15)

While Dewey’s specific topic in these lectures was the “sources of a science” of education, his argument grew out of a very generous view of science, methodologically more ecumenical than liturgical, of research and of data. “All thinking is research, and all research is native, original, with him who carries it on” (Dewey, 1916, p. 174). Dewey further warned of the dangers of borrowing research methods from other fields wholesale, noting, remarkably, that “quantity is not even the fundamental idea of mathematics” (Dewey, 1929b, p. 27). Methods must follow problems and respond sensitively to intentions; aims and means are ineluctably linked.

When means and ends are viewed as if they were separate, and to be dealt with by different persons who are concerned with independent provinces, there is imminent danger of two bad results. Ends […] become empty, verbal; too remote and isolated to have more than an emotional content. Means are taken to signify means already at hand, means accepted because they are already in common use [and the task is to perfect] the existing mechanism of school operations. (Dewey, 1929b, p. 59)

What was (and is) overlooked, Dewey argued, was a “fundamental issue,” of “How far do the existing ends, the actual consequences of current practices go, even when perfected? The important problem is devising new means in contradistinction to improved use of means already given” (p. 60). Representing frozen ideals, in a dynamic and ever-changing culture, all best practices must eventually give way, some to even better practices, others, perhaps unfortunately, to more politically popular or expedient practices.

Expanding on his argument, Dewey issued an oft-quoted warning against over-reliance on quantitative methods of research for seeking guidance.

That which can be measured is the specific, and that which is specific is that which can be isolated […] How far is education a matter of forming specific skills and acquiring special bodies of information which are capable of isolated treatment? It is no answer to say that a human being is always occupied in acquiring a special skill or a special body of facts […] [The] educational issue is what other things in the way of desires, tastes, aversions, abilities and disabilities he is learning along with his specific acquisitions. (Dewey, 1929b, pp. 64–65)

Problems of Verification and “Best Practice”

What, then, of Sprat’s and the Royal Society’s claim about repetition of experiment to verify the facts? Confidently, Sprat asserted that consensus among Society members achieved through experimental repetition promised natural knowledge, an agreement on the facts of nature. From such agreement about the facts, it was but a short step to rules of practice, a hope and aspiration that have flowed across the centuries and poured into the social sciences. While Dewey did not consider education a social science, he thought that as an art it drew on a range of sciences for insight and support. Notwithstanding Dewey’s conclusion that laws and facts do not yield rules of practice, we have witnessed over the past few decades (especially in the US and England) determined government-sponsored effort to forge a science of education due to the belief that specific rules of practice would follow. Randomized sample designs, the gold standard widely held for research, promise reliability, but reliability depends on the existence of three conditions, each of which is at best suspect: (1) a common personhood and comparable life stations, (2) fundamentally simple and similar social systems, and (3) universally accepted educational ends simple enough to be measured.

Another art, roughly of the same sort as education, has been the model for much educational thinking in the US since the 1980s (Holmes Group, 1986): medicine. Considering both programs and practices, educators and teacher educators have looked to medicine for inspiration, often enviously, as a high-status profession with prestige directly linked to funded research (usually randomized controlled trials) and to generation of ever more impressive forms of technology. Randomized controlled trials promise the natural facts that Sprat so admired. Surely, no one would question the power of medical research or its success in reshaping human life, although questioning the similarity of the two practices is certainly wise. Given its many successes and the place of medicine as a model for thinking about education, it is a very good place to test the Royal Society’s and now the National Research Council’s general research ambitions (see Chapter 3).

Recognizing several highly publicized ethical lapses in published medical research, increasing scrutiny is being directed toward medicine’s research-supported claims. In a provocative article, Freedman (2010) described the work of John Ioannidis, described as “one of the world’s foremost experts on the credibility of medical research” (p. 78). Ioannidis’ research raises serious questions of interest here.

[He] zoomed in on 49 of the most highly regarded research findings in medicine over the previous 13 years, as judged by the science community’s two standard measures: the papers had appeared in the journals most widely cited in research articles, and the 49 articles themselves were the most widely cited articles in these journals. (Freedman, 2010, p. 80)

Ultimately, the problem is with the aspiration to produce a or the best practice or set of best practices – practices warranted to produce specified outcomes – when the best that can be hoped for is a much more modest, yet more useful, honest, and responsible better practice for some, not all, teachers and students. Here, again, I return to Dewey. Rather than encouraging hubris, research should always lead to humility: an expression of an appropriate awe for those who perform well the work of teaching and of a deep respect for the complexity of the challenge.

Consistent with his generous view of the nature of research, a reason for Dewey’s optimism about the future of education was the possibility of radically expanding who should be considered a researcher.

It seems to me that the contributions that might come from class-room teachers are a comparatively neglected field. It is to be hoped that the movement [toward educators becoming involved in research] will not cease until all active class-room teachers, of whatever grade, are drawn in. (Dewey, 1929b, pp. 46–47)

Remarkably, a similar conclusion follows from studies of failed modern business practices, from researchers who are coming to appreciate a robust conclusion arising from complexity theory: “Complex systems are, in some special sense, ‘individuals,’ whether or not they are also members of some species” (Lemke & Sabelli, 2008, p. 121). Reporting on a study of the attempt to impose “best practices” in training courses for Xerox technical representatives (reps) who serviced and repaired copiers, Brown and Duguid (2000) wrote,

tasks were [not] so straightforward, and machines, despite their elegant circuit diagrams and diagnostic procedures, exhibited quite incoherent behaviors. Consequently, the information and training provided to the reps was inadequate for all but the most routine of the tasks they faced. (p. 100)

Conclusions like these, supported by Dewey’s arguments, point toward the importance of local studies to quality schooling and teaching, the third topic I shall address, where the aim is a progressively more intelligent and responsive practice directed toward achieving ever evolving but valued ends within specific work contexts: better not best practice. Before discussing local studies directly, a few words are in order about the place of outliers in educational research, those “non-normal” behaviors and experiences that Porter (1986) suggested have usually been set aside as “perturbations,” distortions of “common personhood.”

On Outliers, Educational Research, and Better Practice

Speaking out of what he described as a “skeptical pragmatism,” Toulmin (2001) deepened Dewey’s analysis of uncertainty. Noting the complexity of all social and natural systems, Toulmin argued that it is unreasonable to ask of the social sciences “accurate forecasts of people’s actions”: rather, “the virtue of the social sciences is that they sometimes help us understand just why, and under what special conditions, our expectations of people’s behavior – either as individuals or as institutions – can reasonably be relied on” (pp. 208–209, emphasis added). As Toulmin suggested, one result of recognizing the inevitable limitations of their ability to forecast is that the “claims of contemporary sciences, both natural and human, are a good deal more modest [than in the past], seeking neither to deny nor to explain away the contingency of things” (pp. 209–210). Unfortunately, a similar appreciation and modesty are generally lacking among education policy-makers and politicians who have happily embraced and generously funded the quest for best practice, including value-added research efforts designed to support the ranking, rewarding, or punishing of individual teachers and their schools (Ballou, Sanders, & Wright, 2004). Conceptual and philosophical issues aside, such efforts face virtually impossible technical impediments to realizing accurate, let alone fair and equitable, judgments of teacher quality or value (Berliner, 2014, 2018; Ford, Van Sickle, Clark, Fazio-Brunson, & Schween, 2017; Martineau, 2006; Papay, 2011). Think again of the Xerox machines. Moreover, questions of teacher quality or value are themselves highly contentious, as they should be.

Given the human condition, Toulmin suggested that the appropriate course of action is to “map the range of possible futures open to us – either as individuals or as political and social collaborators – and do our best to create conditions that will help us move in better instead of worse directions” (2001, p. 211 emphasis added). Like Dewey, he pleaded for better (not best) practices: actions framed by reflection, driven by dreams and emerging ideals, but disciplined by conflicting interests, differing life conditions, and prevailing stubbornness of human and institutional histories. On this view, what counts as effective practice is a type of intelligent, forward-looking, well-informed and opportunistic tinkering, and a sensitive and responsive feeling of one’s way along with others toward evolving but worthy and maturing goals.

It is here where robust attention to outliers, not just to central tendencies, finds a place, for attending carefully to outliers can be helpful in anticipating what is coming and in envisioning and planning for different and possible futures. Extending the point, Toulmin (2001) drew on an insight from Eudora Welty to argue that often “the eccentric can be used to explain the central, rather than the other way around!” (p. 30). Indeed, outliers reveal and problematize what counts as normal. Writing about normality in medicine, Stigler (1999) made a parallel point: that knowledge may be best advanced by “looking at the more extreme variations” (p. 424). Everyone, Stigler suggested, is in some sense sick, as illness is, in fact, normal. In medicine, the connection between noting the extremes in illness and anticipating what is coming is a connection relatively easily made. In education, such links may prove more elusive, yet they are abundant: whether signaling, say, the first signs of change following a large demographic shift or the technological and cultural developments that manifest themselves in differing forms and levels of feeling for school among students. They reveal variations in interest, ability, and life condition from rising levels of diagnosed autism to increasing familial instability following a severe economic downturn like the Great Recession. Thus, yesterday’s outliers frequently become tomorrow’s norm.

Of even greater consequence for educators, however, is an insight offered by Paul Feyerabend (1994) in his provocative book Against Method. Suggesting the importance of dialectical thinking, Feyerabend wrote,

The first step in our criticism of commonly-used concepts is to create a measure of criticism, something with which these concepts can be compared. Of course, we shall later want to know a little more about the measuring-stick itself; for example, we shall want to know whether it is better than, or perhaps not as good as, the material examined. But in order for this examination to start there must be a measuring-stick in the first place. (p. 52)

Here too the careful and consistent attentiveness to outliers in practice and in research is important. Outliers challenge not only methods of education (with children who do not respond as anticipated to one or another preferred method or intervention, what might be termed methodological outliers), but also the aims of education (with those for whom the established purposes of schooling need to be questioned). It is in part because of this second category of outliers for whom the favored and now sharply narrowing purposes of schooling prove troubling that we are witnessing, for example, the rapid growth in both charter schools and home schooling in the US. Since commonly in education the ends and means are thought of as separate and separable, it might appear that there is no issue here. The domain of means is widely assumed to be the teacher’s purview, while others set the ends of education. However, as suggested previously, such views deny what should be obvious: Whosoever sets the aims of education also determines the means; in turn, available means, including the relative generosity or scarcity of resources, set boundaries around achievement.

Local Studies and Validity

“Educational practices provide the data, the subject-matter, which form the problems of inquiry […] These educational practices are also the final test of the value of the conclusion of all researches” (Dewey, 1929b, p. 33).

By local studies, I mean formal inquiries into questions that arise directly out of specific communities of practice (Wenger, 1998). Such studies usually do not seek generalizable results; rather, they represent, as Gallimore, Ermeling, Saunders, and Goldenberg (2009, p. 550) suggested, “setting-focused interventions” that aim at improving, enriching, and sometimes redirecting a specific practice or set of practices. Thus, local studies are deeply interested studies, drawing on insiders and depending on insider knowledge when framing questions and interpreting, not just implementing, results.

This characteristic of local studies cuts two ways. On the one hand, heavy reliance on insiders makes certain that the questions addressed speak to genuine issues and concerns, makes available a broad range of data, including and especially related to outliers, and enables sensitive and deeply informed interpretations of those data to possess potentially high validity. On the other hand, depending on insiders may lead to avoiding some issues for fear of disrupting relationships or offending a friend, also to insularity and cultural blindness, when data interpretations are bounded by the range of understandings and insights readily available within a given context. Depth of desire, as Jensen (2007) suggested, is crucial to overcoming the first issue. The second requires reaching beyond the common sense of a community of practice to consider alternative interpretative possibilities, ways of thinking, making sense, and acting. Perspective may be gained, as noted previously, by engaging in systematic comparison (Feyerabend, 1994); by reading and discussing published research, the stuff of public theory, related to the current question; and by involving selected outsiders, persons who can knowledgeably speak with the insider from the outside. Lacking outside perspectives, the danger is that what is seen is what is already believed (Rabinow, 2008).

While local studies do not seek generalization of results (i.e., reliability), they do seek high validity; when published, such studies should invite readers into acts of experiential comparison. Readers should be able to understand clearly where their experiences as teachers, as parents, or even as students parallel or divert from what is being described and advanced. As a means for widening and enriching the conversation about teaching and learning and for recognizing work well done, sharing results is critically important, whether by providing newsletters to school patrons and educators working within a school or district or by publishing and distributing an article printed in an international journal for teacher educators. Effective communication of results requires inclusion of sufficient contextual information and detail to allow readers to enter the story and make connections across experiences or locate tensions without undue difficulty. Transparent methods, rich contextual descriptions, clear writing (including consistent use of terms), generous use of data, and some recognition of blind spots or weaknesses are required to establish study relevance, value, and validity. For educators, matters of validity are of paramount importance.

In its various forms, local research, like some forms of action research and of self-study (Loughran, Hamilton, LaBoskey, & Russell, 2004; Pinnegar & Hamilton, 2009), seeks to sharpen practice–theory links in context, where theory is practiced and practice is theorized (Bullough, 1997). The aim is both better theory and better practice within specific programs and settings, for specific purposes and for specific participants. To this end, local research requires attentive observation and careful listening to those being taught or studied and to one’s own experience, or else the noise of external critics will distract and confuse. Sadly, researchers often both serve these critics and invite distraction by requiring even the most able of teachers to deny the authority of their own teaching experience. In addition, local studies respect the rhythm of research, a rhythm imposed by contexts and of work and patterns of learning while working. External researchers are always in a hurry, and in the rush to judgment teacher learning is likely short-changed.

In addition to their promise as powerful forms of professional development, there are abundant reasons for investing in local studies of one kind or another. Perhaps the most obvious is that all change is local; curiosity about the effects and educational possibilities of a new practice is another. Then, there is excitement about a new understanding of a topic or increased interest in a colleague’s program. Despite the generally depressing effects of high-stakes testing on teacher morale (Pearson, 2009), another reason is the need to raise student test scores. It is tempting to dismiss raising test scores as a legitimate reason for inquiry, but to do so would be foolish in the present political climate – a climate that adds a layer or two of complexity to the work of educators. To this end, as Vescio, Ross, and Adams (2008) suggested, work done within professional learning communities that involves groups of teachers sharply focused on studying their practice in relation to student work produces gains in tested learning. Regardless of the driving concern, what is important is that local studies directly address problems of genuine moment to teachers and students in their full living particularity and peculiarity and in full response to their life context and culture. When they do, local studies prove empowering, opening the possibility for educators to become their own “best theorists” (Hunt, 1987).

Local studies that involve what Jensen (2007) described as a “‘back and forth’ looping between theory and practice” (p. 495) open the possibility of revealing and testing implicit or personal theories against public theory and, conversely, of testing public theory, even “best practice,” against personal experience and theory. Rather than cutting them off from their experience, local studies invite teachers deeper into the context of their work and into the lives and experiences of their colleagues and of those they teach. The same is true for teacher educators. That local studies take place in a specific context and with colleagues brings contextual peculiarities and cultural idiosyncrasies to view, potentially leading to identification of unrecognized resources, specific and unique points for action, and opportunities for improvement. No institutional cultures are wholly congruent; every culture both enables and limits meaning, preferring or discouraging some ways of thinking, acting, and being, and also promoting some forms of role enactment over others. When values clash, local studies provide opportunities to confront the limiting functions of culture and to reconsider values and commitments. The focus on outliers holds similar promise. Moreover, such studies offer opportunities to locate strengths that may be extended and built upon. Undoubtedly, some studies may seek to avoid such tensions or minimize their importance, but the tensions are always present to some degree, even if suppressed.

Local studies have revelatory and disciplinary functions; comparing results reveals where central tendencies collapse and where prejudices lie hidden […] They “account for the particular” and encourage (drawing on Garrison, 1997) “outlaw thinking”, normative discourse that enables the raising of questions that reside outside of established methodological parameters and taken-for-granted system imperatives. (Bullough, 2008, p. 11)

In contrast, researchers mostly concerned with fidelity to best practice seek to shape, control, and direct teachers’ classroom actions. For teachers, research can be seen as distant, something someone else does and experienced as disempowering, as a form of colonialization of the classroom and of the curriculum. There is no doubt that in the pursuit of best practice the control of teachers often becomes a dominating concern manifested through prescribed curricula, tightly monitored and set practices, and aggressive and often punitive evaluation and accountability systems. Clearly, belief in the need for controlling teachers permeates the language of school change, where reforming, reculturing, and restucturing rather than renewing (Goodlad, 1994) dominate discourse. Local studies seek renewal. The difference in intention is evident and large: a difference between seeking to replace or fix something broken, or presumed to be broken, and building to teacher strength and to invigorate imagination. Underpinning these views of school change is a fundamental distrust of teachers and an implicit doubt about their ability to grow on the job and into teaching excellence (Brill, 2011). What is lacking is the kind of trust evident in Dewey’s argument for including teachers in research and teaching as thoughtful practice, always involving some form of research. Excluding teachers from participating in setting the aims of education is a strong manifestation of distrust; in contrast, including them builds trust and confidence.

Remarkably, the dominance of means over ends, of rules over principles, and the acceptance of teaching as primarily concerned with tasks of transmission and content delivery (means narrowly construed) linger even in some of the most sophisticated of current conceptualizations of teaching. For example, Hammerness, Darling-Hammond, and Bransford (2005) offered the metaphor of “teachers as adaptive experts” as a way for thinking about teacher education and teacher development. The argument centers on the importance of teachers being able to teach efficiently and to develop innovative strategies for responding to situations where established routines fail: that is, methodological outliers. Thus, the emphasis on “adaptive” expertise and on being able to “innovate within constraints” (p. 364) is primarily a matter of teachers getting their instructional house in order; ends are set.

In Dewey’s view, research begins and ends with practice. By virtue of the great variability among communities, schools, classrooms, teachers, and students, inevitably the relationships between theory and practice and between ends and means are dynamic; always the proof of the value of any action is found in the resulting consequences, some of which are unanticipated. In this view, the value of even the most robust of rules is only “indirect,” as more or less useful “intellectual instrumentalities” (Dewey, 1929b, p. 28). It is this “more or less” quality of the value of rules to which the concept of “adaptive expertise” speaks, as does Schön’s (1984) concept of reflection in action. There is no doubt that rules offer beginning teachers’ points of departure or of orientation, a place to begin, but only a place to begin framing, making sense of, and initially responding to the problems and dilemmas of teaching and learning. However, over time rules that are externally imposed yet strongly held inevitably limit imagination and cripple innovation.

Conclusion

A few years before the passage of No Child Left Behind in the US, with its aggressive program of accountability, and as Ofsted in the UK was gaining power, John Goodlad (1994) identified what motivates teachers in their daily work:

Good teachers are driven neither by the goals of improving the nation’s economic competitiveness nor that of enhancing the school’s test scores. Instead, they are driven by a desire to teach satisfyingly, to have all their students excited about learning, to have their daily work square with their conception of what this work should be and do. (p. 203)

There is no doubt that Sprat was right to warn members of the Royal Society of the twin dangers of enthusiasm and skepticism. Enthusiasm leads to advocating aggressively, to taking strong ideological positions, and to over-promising – outcomes associated with the quest for best practice. As Toulmin suggested, the best that can be hoped for from the social sciences is a “sometimes” and, undoubtedly, temporary better practice. So it is with an art like education. Blind advocacy and indiscriminate skepticism like Brill’s (2011) undermine any and all efforts at improving practice while encouraging teacher disengagement and assuring displacement of responsibility to others willing and able, even eager, to take charge and dictate directions for change.

Modesty is called for on the part of those of us who see ourselves primarily as researchers, as well as a deep respect of the sort Dewey possessed for the difficulty of teaching and of teaching well. Such respect is prerequisite to producing research worth its salt. A tempering of the ambition to fix things is also needed, replaced by a lively desire to increase understanding and build more far-reaching, inclusive, and generous research communities that do not deny the challenges of uncertainty, but delight in them, especially in outliers from whom much can be learned. As researchers, teacher educators need to be clear about who or what we serve. While our legacy coming from political arithmetic may be that of state service, like teachers, our first and foremost responsibility is to do our work in ways that nurture the young and those who serve them, as well as strengthen our democratic traditions. Ultimately, this is precisely what local studies should seek to do. The last words of this chapter come from two complexity theorists who argue there is a need to move “away from the Enlightenment dream of universal laws, perfect predictability and rational control [in educational research] to a new recognition that all genuinely complex systems are individual, surprising, and not a little perverse. Just like us” (Lemke & Sabelli, 2008, p. 122).