The Ambiguities of Experience

James G. March

• Publisher: Cornell University Press
• Year: 2011
• Print ISBN: 9780801448775
• eBook ISBN: 9780801457777
• Topic: Business
• Subtopic: Organisational Behaviour
• Index: Business

"The first component of intelligence involves effective adaptation to an environment. In order to adapt effectively, organizations require resources, capabilities at using them, knowledge about the worlds in which they exist, good fortune, and good decisions. They typically face competition for resources and uncertainties about the future. Many, but possibly not all, of the factors determining their fates are outside their control. Populations of organizations and individual organizations survive, in part, presumably because they possess adaptive intelligence; but survival is by no means assured. The second component of intelligence involves the elegance of interpretations of the experiences of life. Such interpretations encompass both theories of history and philosophies of meaning, but they go beyond such things to comprehend the grubby details of daily existence. Interpretations decorate human existence. They make a claim to significance that is independent of their contribution to effective action. Such intelligence glories in the contemplation, comprehension, and appreciation of life, not just the control of it."—from The Ambiguities of Experience

In The Ambiguities of Experience, James G. March asks a deceptively simple question: What is, or should be, the role of experience in creating intelligence, particularly in organizations? Folk wisdom both trumpets the significance of experience and warns of its inadequacies. On one hand, experience is described as the best teacher. On the other hand, experience is described as the teacher of fools, of those unable or unwilling to learn from accumulated knowledge or the teaching of experts. The disagreement between those folk aphorisms reflects profound questions about the human pursuit of intelligence through learning from experience that have long confronted philosophers and social scientists. This book considers the unexpected problems organizations (and the individuals in them) face when they rely on experience to adapt, improve, and survive.

While acknowledging the power of learning from experience and the extensive use of experience as a basis for adaptation and for constructing stories and models of history, this book examines the problems with such learning. March argues that although individuals and organizations are eager to derive intelligence from experience, the inferences stemming from that eagerness are often misguided. The problems lie partly in errors in how people think, but even more so in properties of experience that confound learning from it. "Experience," March concludes, "may possibly be the best teacher, but it is not a particularly good teacher."

1
THE PURSUIT
OF INTELLIGENCE

Organizations pursue intelligence. It is not a trivial goal. Its realization is imperfect, and the pursuit is endless. Every day there are failures to temper any successes. Nevertheless, the pursuit is often exhilarating. It exalts the subtle textures of life and elevates coping with ordinary tasks to the artistry of history. The present book considers one aspect of the pursuit of intelligence—the effort to extract lessons from the unfolding episodes of life. Organizations and the individuals in them try to improve by contemplating and reacting to their experiences.

Folk wisdom both trumpets the significance of experience and warns of its inadequacies. On the one hand, experience is described as the best teacher. On the other hand, experience is described as the teacher of fools, of those unable or unwilling to learn from accumulated knowledge or the teaching of experts. The disagreement between the folk aphorisms reflects profound questions about the human pursuit of in telligence through learning from experience that have long confronted philosophers and social scientists.

Despite extensive enthusiasm for it, the prima facie evidence for organizational improvement through extracting lessons from experience is mixed. Contemporary organizations certainly engage in practices, follow procedures, and exhibit forms that are notably different from organizations of a century ago. By most measures of productivity, contemporary organizations are more efficient than their predecessors. At the same time, experience is often ambiguous and the inferences to be drawn from it are unclear; and the contribution of experiential learning to long-term improvements in organizations is difficult to establish. The ambiguity of history makes the matching of beliefs and actions to experience both complicated and prone to misdirection (March and Olsen 1975; 1995, chap. 6).

There are well-documented cases of the apparent failure of organizations to adapt to their environments. Failures of business firms to copy successful practices from other firms are a familiar refrain of organizations research. Wars are filled with instances of what appears, with the clear vision of after-the-fact perspicacity, to be organizational blindness on the part of armies. The hugely successful American steel and automobile industries of the first half of the twentieth century withered in the last half. The American public school system changed from being the pride of the country and the envy of other countries to being an embarrassment. The American political system found it difficult to adapt to declines in American prosperity and power as the twenty-first century began.

While acknowledging the power of learning from experience and the extensive use of experience as a basis for adaptation and for constructing stories and models of history, the chapters in this book examine particularly the problems with such learning. The essays argue that although individuals and organizations are eager to derive intelligence from experience, the inferences stemming from that eagerness are often misguided. The problems lie partly in correctable errors of human inference forming, but they lie even more in properties of experience that confound learning from it (March 2008, chap. 5). As a result, the book is somewhat more conservative about the possibilities for experiential learning than is characteristic of some writings about it (Kolb 1984; Sternberg and Wagner 1986; Kayes 2002). It identifies some endemic ambiguities and mistakes of experience.

PROLEGOMENON

The ideas pursued in this book reflect three rather different traditions of scholarship. The first is the tradition of scholarship on organizations. That tradition draws particularly from economics, psychology, management science, sociology, political science, public administration, and artificial intelligence. For the most part, scholars within the tradition are Cartesian, scientific, and analytical. They emphasize the formal analysis of data and testing of hypotheses, exercising models, and proving theorems. They are prone to deductive cleverness in the form of relatively limited but provocative ideas, as reflected in enthusiasms for such things as game theory, cognitive dissonance, structural equivalence, absorptive capacity, and garbage-can theories. They practice a style of analysis that focuses on relatively simple relations among discrete variables. They speak, for the most part, from the perspective of the social and behavioral science establishments. And they presume a history that moves haltingly toward greater knowledge and intellectual progress.

The second tradition is that of scholarship on storytelling, narrative, and myth. These humanistic traditions of scholarship draw particularly from literature, criticism, history, anthropology, linguistics, legal studies, and religion. They emphasize language, metaphor, and the elaboration of meaning. Like Dario Fo, who emulates the jesters of the Middle Ages, attacking authority and glorifying the dignity of the downtrodden, many of them place themselves in opposition to the establishment. They are prone to expansive speculation in the form of sweeping generalizations in the manner of Freud, Marx, or Foucault but also revel in close investigation of the human estate in its fine detail. They practice a style of scholarship that exploits the nuances of language to evoke meanings. And they generally eschew notions of intellectual progress in favor of ideas of intellectual and historical embeddedness, social construction, and cultural consciousness.

The third tradition is that of scholarship on adaptive processes. Although the work is intertwined with empirical studies of change, it is heavily theoretical. It emphasizes properties of the mechanisms of adaptation in organisms, species, technologies, organizations, industries, and societies. The traditions of research on adaptation draw heavily from evolutionary biology and psychological ideas about human learning. They attempt to further understanding by identifying adaptive processes and exploring their consequences in complex ecologies of multiple adapting agents. They pay particular attention to the capabilities of adaptive processes to realize outcomes that are, in some meaningful sense, improvements—even optimal. They generally presume a historical process that is heavily history-dependent with multiple unstable equilibria, but an intellectual process that allows some elements of nonrandom forward movement.

The chapters in this book draw copiously and inadequately from all three traditions, but they focus on a relatively narrow topic: When and how do organizations learn intelligently from their experience? What are the possibilities and problems? The chapters address such issues, but they hardly resolve them. They have all the limitations and perversities of brevity.

Among other limitations, strictly organizational factors in organizational learning from experience are subordinated in these chapters to the ambiguous features of experience, the structural features of adaptive processes, and the intellective features of human adaptation. Organizational factors are discussed from time to time, but there is no attempt to be exhaustive with respect to them. The subordination is one of convenience, not a symptom of unimportance. Organizations shape the way experience unfolds and the way in which it is observed and comprehended. They define the networks through which information flows and reinforce or reduce the cleavages of conflict. They create and maintain the goals and expectations that calibrate aspirations. They are systems of rules, routines, capabilities, and identities that both resist and record the lessons of experience. Such factors can be ignored only as a temporary expediency, and even so at some risk.

In addition, three critical properties of organizational adaptation that are vital to a thorough understanding are not considered significantly here. First, organizations are coalitions of conflicting interests (March 1988, chap. 5; Hoffman 1999; Rao, Morrill, and Zald 2000). Conflicts of interest make theories of conflict-free adaptation notoriously incomplete, complicating particularly the association of success or failure with outcomes and the pooling of information (Cyert and March 1963; Augier and March 2001). Conflicts of interest are conspicuous factors in all organizations and influence not only the pursuit of intelligence but also its definition (Greenwood, Suddaby, and Hinings 2002; Olsen 2009).

Second, organizational adaptation involves the simultaneous, interacting adaptation of several nested levels (March 1994, chaps. 2, 6; Friedland and Alford 1991). Populations of organizations evolve at the same time as the individual organizations within the populations, and organizations evolve at the same time as individuals within the organizations. These nested systems of adaptation affect each other, with adaptation at one level sometimes serving as a substitute for adaptation at another and sometimes interfering with it.

Third, the environment of organizations consists, in part, in other adapting organizations; and the elements of coevolution produced by their simultaneous adjustments are an essential feature of the adaptive story (Hannan and Freeman 1989; Kauffman and Johnsen 1992; Levinthal and Myatt 1994). The treatment of the environment as exogenous, as is common in the literature on organizational learning and, to a substantial extent, in these chapters, is a significant simplification.

TWO COMPONENTS
OF INTELLIGENCE

Intelligence normally entails two interrelated but somewhat different components. The first involves effective adaptation to an environment. In order to adapt effectively, organizations require resources, capabilities at using them, knowledge about the worlds in which they exist, good fortune, and good decisions. They typically face competition for resources and uncertainties about the future. Many, but possibly not all, of the factors determining their fates are outside their control. Populations of organizations and individual organizations survive, in part, presumably because they possess adaptive intelligence; but survival is by no means assured. Although a few organizations, most notably the Roman Catholic Church and older European universities, have survived for many years, the vast majority of organizations endure for only a relatively short time. By that criterion, at least, organizational adaptive intelligence is not guaranteed. It is not even typical.

The second component of intelligence involves the elegance of interpretations of the experiences of life. Such interpretations encompass both theories of history and philosophies of meaning, but they go beyond such things to comprehend the grubby details of daily existence. The desire of human beings to make sense of their experiences permeates much of scholarship. It also permeates much of life. Interpretations of experience are ornaments of casual conversations and of theories of psychological, economic, political, cultural, and social systems. Stylishness of interpretation is a certification of human status and a basis for the social ranking of individuals and institutions. Interpretations decorate human existence. They make a claim to significance that is independent of their contribution to effective action. Raymond Fischesser, a former director of L’École des mines de Paris, defined intelligence as “la préoccupation efficace de l’essentiel” (Riveline 2008, 7). Such intelligence glories in the contemplation, comprehension, and appreciation of life, not just the control of it.

EXPERIENTIAL LEARNING

The tools for achieving intelligence reflect the knowledge technologies of the time and place. At some not-too-distant times and places, intelligence seeking involved extensive use of magic potions, enchantments, and incantations, as well as the anticipations and ambiguities of oracles and other keepers of extrahuman capabilities. No self-respecting manager of early Chinese, Egyptian, Greek, or Roman enterprise would have willingly confronted the uncertainties of life without a suitable pipeline to the gods who manipulated the universe within which their organizations operated.

Though various forms of godlike revelation continue to secure adherents (Eisenstadt 2006), efforts to access the mysteries of the gods have been largely replaced in modern life by efforts to uncover more mundane secrets of knowledge. In that respect, few ideas are as sacrosanct in contemporary sensibilities as the notion that human beings achieve mastery over their lives through learning from experience. Individuals and organizations try to improve their lots by observing and reacting to their experiences, partly by elementary efforts to reproduce actions associated with success, partly by more elaborate efforts to fit the events of their histories into acceptable causal frames. Experience is venerated; experience is sought; experience is interpreted.

Learning from experience is, of course, by no means the only mechanism of human learning. Indeed, most of what is known by individuals and organizations is not discovered in lessons extracted from the ordinary course of life and work. It is generated by systematic observation and analysis by experts and transmitted by authorities (e.g., in books, through web browsers, by teachers), and accepted (or rejected) without direct experiential confirmation. It reflects, in a general way, academic knowledge rather than experiential knowledge (March 2004).

Nevertheless, in the contemporary literature on organizations, experiential learning continues to be seen as one of the more important sources of adaptation in human action, a mechanism for improving the fit of actions by individuals or organizations to the environments they face (Argyris and Schön 1978; Levitt and March 1988; Huber 1991; Payne, Bettman, and Johnson 1993; Cohen and Sproull 1996; Argote 1999; Nooteboom 2000; Starbuck and Hedberg 2001; Greve 2003). Admirable organizations are described as “learning organizations”; advisers on ways to augment and refine learning abound (Argyris and Schön 1978; Senge 1990); and recent proposals for improving organizations have often emphasized learning from experience (Senge 1990; Olsen and Peters 1996; Dierkes et al. 2001; Zollo and Winter 2002).

Learning from experience is seen as relevant to theories of rational action (Arrow 1972; Coleman 1990; Milgrom and Roberts 1992) as found in game theory (von Neumann and Morgenstern 1944; Luce and Raiffa 1957; Kreps 1990b) and in decision theory (Raiffa 1968; Machina 1987; Anand 1993; Augier and March 2002). It is also seen as relevant to theories of rule-based action (March and Olsen 1989, chap. 2; March 1994, chap. 2) as found in ideas of individual identity, institutions, and social roles. In rational theories, the presumed fundamental basis of intelligent action is a logic of consequences (March and Olsen 1989, chap. 1; March 1994, chap. 1); in rule following, action follows from systems of rules, identities, and roles (Scott and Meyer 1983; Ashforth and Mael 1989; North 1990; Becker 2004; Brandstätter, Gigerenzer, and Hertwig 2006); and the presumed fundamental basis is a logic of appropriateness (Günther 1993; March and Olsen 2006b). Within both perspectives, experiential knowledge is seen as a necessary component of human claims to intelligence. In rational action, organizations are seen as looking forward by looking backward (Gavetti and Levinthal 2000; Zollo and Winter 2002; Gibbons and Roberts 2008). In rule-based action, organizational rules are seen as evolving through experience (Alchian 1950; Nelson and Winter 1982; March, Schulz, and Zhou 2000; Akerlof and Kranton 2005; March and Olsen 2006a).

Thus, considerable effort has been devoted to augmenting the effectiveness of intelligence in organizations by drawing on the evidence of history. Sophisticated instruments of estimation, modeling, and strategic planning have been created and implemented to facilitate marketing, financial, production, and human resource decision making in business firms and to facilitate the efficient and effective delivery of social services in governmental agencies. Elaborate systems of accounts have been developed to facilitate control over organizations by various stakeholders. Very large industries of consultants have arisen to diffuse practices and theories deemed useful. Schools of management have become conspicuous components of modern universities, providing education to large numbers of would-be future managers of both public and private organizations. Management education provides a menu of “good practices,” as well as a kit of theories of markets, politics, operations, and firms.

In order to improve their fit to their environments, organizations frequently undertake changes in organizational practices, procedures, and forms. To be sure, the alternative possibility that the fit of an organization to its environment can be improved more easily by changing the environment than by changing the organization is reflected in strategies common among powerful players. From this perspective, adaptation is for the weak; the strong impose themselves and let others adapt to them. The great empires of history (Rome, China, Ottoman, Spain, Great Britain, the Soviet Union, the United States) have all thrived in part by forcing their environments to adapt to them, rather than wasting energy in trying to adapt to their environments. The same thing is true of the great industrial empires (US Steel, General Motors, Royal Dutch Shell, Unilever, Sony, IBM, Microsoft). It is true also of the human species. The long run disadvantage of improving fit by changing the environment is—as all of the examples illustrate—a decay of the ability to adapt to the environment if necessary, a decay that accelerates decline if a dominant position is lost.

The search for aids to experiential learning in organizations has been paralleled by a substantial empirical and theoretical effort in organization studies to develop theoretical ideas about the realities of organizational adaptation. That effort has been directed toward understanding the ways in which populations of organizations and individual organizations change (not necessarily intentionally) over time in response to adaptive pressures (Aldrich 1979; Cohen and Sproull 1996; Nooteboom 2000; Dierkes et al. 2001; Greve 2003; Aldrich and Ruef 2006; Dosi and Marengo 2007). Studies of the birth and death processes of organizational life and the resulting features of organizational demography have provided important contributions to these efforts (Hannan and Freeman 1989; Carroll and Hannan 2000).

This history of scholarship forms a background for the explorations in the present chapters. The chapters consider the complications involved in learning from experience. Some of these complications reflect well-known human informationprocessing habits—for example, biases toward conserving belief, simple causalities, historical certainty, and attributions of significance to human intentions. Those characteristics of human behavior are important, but the focus here is primarily on a separate set of the complications that reflect interactions between experience and the processes of learning. As the following chapters will argue, these problems lie primarily not in the learners but in the nature of experience.

2
LEARNING THROUGH
REPLICATING SUCCESS

Extracting lessons from natural experience presumes a learning cycle that begins with observing associations between actions and outcomes, the rudiments of finding order in history. Learning takes place when the observation of associations produces changes in actions or rules for actions. Learning serves intelligence when those changes improve the actions or the rules. Changes from learning occur often and relatively easily, but their contribution to intelligence is more problematic. Experiential learning makes many mistakes.

TWO MODES OF INTELLIGENT
ADAPTATION

It is useful to distinguish two different modes of achieving intelligence through experience. The two modes reflect different kinds of processes and encounter different kinds of complications, so the distinction may be useful as long as it is recognized that actual occasions of learning involve mixtures of the two (Zollo and Winter 2002; Winter, Cattani, and Dorsch 2007; Starbuck, Barnett, and Baumard 2008; Winter 2009). The first mode—which might be called “lowintellect” learning—is one in which actions associated with success are replicated with little or no effort at causal understanding. The second mode—which might be called “highintellect” learning—is one in which explicit efforts are made to understand the causal structure of the events of experience and to derive action implications from that understanding. The distinction is closely related to that made by Starbuck, Barnett, and Baumard (2008) between noncognitive and cognitive learning and by Gul and Pesendorfer (2008) and Camerer (2008) between mindless and mindful economics, though the different terms contain somewhat different nuances. Low intellect and high intellect are not differentiated by merit. Each has its place; each has its limitations.

Low-intellect mechanisms of learning from experience are built on the replication of success and are common among both humans and other animal species. They often generate rules or heuristics of behavior that are surprisingly effective (Hutchinson and Gigerenzer 2005). High-intellect mechanisms, on the other hand, appear to be less prominent in the learning of other animals than they are among human beings. The entire academic apparatus for developing, recording, and dispensing knowledge derived from theories about the underlying causes of outcomes of experienced history is a distinctively human construction. That apparatus depends on the power of written language and instruments of symbolic manipulation that not only are uniquely human but have existed among humans only for the past few thousand years.

THE REPLICATION OF SUCCESS

The basic idea of replication of success is inherent in a familiar set of ideas in studies of organizations. Organizations are often seen as comparing performance with an aspiration for performance and searching for changes primarily when performance falls below aspiration (Cyert and March 1963; Greve 2003; Baum and Dahlin 2007). Thus, actions associated with success are more likely to be reproduced than are actions associated with failure.

In its barest form, the replication of success is elegant in its simplicity:

(1) Act by choosing an alternative from among those available.

(2) Record a result and evaluate it in terms of its success.

(3) Replicate the choice of alternatives associated with successes more often than the choice of alternatives associated with failure.

Practices, forms, and rules that are associated with good outcomes survive longer and reproduce more than practices, forms, and rules that are associated with poor outcomes.

It should be observed that describing the replication of success as low-intellect (or “noncognitive”) is potentially misleading. Such a description ignores cognitive complications involved in the adaptive process. For example, in a classic binary learning task, a learner chooses repeatedly between two mutually exclusive and exhaustive alternatives. When c...

Table of contents

• PREFACE
• 1 THE PURSUIT OF INTELLIGENCE
• 2 LEARNING THROUGH REPLICATING SUCCESS
• 3 LEARNING THROUGH STORIES AND MODELS
• 4 GENERATING NOVELTY
• 5 THE LESSONS OF EXPERIENCE
• REFERENCES