Toward a More Pragmatic Knowledge Management: Toyota’s Experiences in Advancing Innovation
Central Connecticut State University, USA
Managers often conceive knowledge management processes in ways that unduly limit its potential. Toyota has avoided falling into this narrow paradigm trap by creating its own version of knowledge management that is well suited to its culture. They have woven their knowledge management strategy together with process improvement and innovation methods. Toyota’s knowledge management system is a theoretically sound, yet practical, business approach built on a set of scientific principles based on a philosophy known as Pragmatism. This chapter examines how Pragmatic principles used by Toyota can achieve superior innovation results. The chapter concludes by explaining why the Pragmatic approach delivers superior performance at lower cost than conventional knowledge management methods.
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Some researchers see organizational innovation as a product of individual skills, such as creativity and imagination. Conventional wisdom regards it as being an art, not a science. Still others, such
DOI: 10.4018/978-1-61350-165-8.ch018 as many management theorists, regard innovation as being the organizational outcome of a properly designed business strategy. In their eyes, strategy becomes the activator ofpredictable organizational processes where ‘B’ follows ‘A’ - as ifa clockwork. They envision a tightly controlled system capable of driving innovative ideas through a pipeline
flowing through various phases of development toward an outcome - releasing a product or service to market. For example, Bacon and Butler (1998:11) define innovation as a commercially successful use of invention, and invention as being a “solution to a problem (unmet needs)”. Similarly, Davila et al (2006), propose that innovation IS a management process - one requiring specific tools, rules, and discipline.
This chapter’s purpose is to explore an alternate paradigm for innovation. It suggests a strategy based in the logic of scientific reasoning and experimentation. Specifically, this alternative spurs innovation by radically improving the quality of knowledge held by an organization’s members. Its theoretical basis is a system designed to improve the quality of scientific discoveries known as Pragmatism. Pragmatism’s founder, scientist Charles Sanders Peirce, set forth a number of principles for being pragmatic in improving the effectiveness of one’s action. Many management theories, such as Total Quality Management, incorporate pragmatic principles. Peirce’s (1958:293) approach to scientific discovery begins with the pragmatic maxim that holds we should always,
“Consider what effects, that might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is the whole of our conception of the object.”
This maxim is the basis for an innovation strategy that focuses on conducting frequent mini-experiments throughout an organization. For example, Spear (2009:215) cites an example of how plant-floor workers at Aisin-a first-tier supplier for Toyota - routinely use such an experimental approach to improving the quality of their common knowledge.
“Problem solving is done in a disciplinedfashion. Assumptions about cause and effect are made explicit and stated clearly, then they are tested in rigorousfashion so improvement efforts both make processes better and deepen process knowledge.”
To Peirce, the key to creating high quality knowledge is through rigorous experimentation and the integrated application of three types of inferential reasoning, namely deduction, induction, and abduction. Abduction produces radical insights of the sort that companies seek for gaining competitive advantage. The chapter proposes an innovation strategy based on pragmatic principles designed to leverage the power of abductive reasoning (Wiener, 1958). Toyota is one of the leading companies in both applying pragmatic principles for increasing knowledge quality, experimentation, and innovation. This chapter examines Toyota’s methods for increasing knowledge quality and innovation and their origins in pragmatic philosophy.
What is innovation? Research by Baregheh (2009:1334) found the term innovation describes, “the multi-stage process whereby organizations transform ideas into new/ improved products, service or processes, in order to advance, compete and differentiate themselves successfully in their marketplace.” By contrast, such flow-oriented definitions oppose those that focus on identifying the sources of knowledge necessary to produce innovations. For example, scholars, such as Peirce and noted economist Joseph Schumpeter (1950), view knowledge as being the primary force behind innovation. Peirce’s main interest in studying knowledge was in discovering how it drives processes of scientific discovery, whereas Schumpeter’s primary interest was in how knowledge influences the entrepreneurial economic potential of a firm. Peirce studied how scientists conduct research and how their methods lead to breakthrough innovations. He concluded that scientific innovations often happen suddenly, but
Table 1. Approaches and forms of scientific reasoning
usually occur after prolonged periods of analysis and many experiences. A surprising find was that these scientists often employed a special type of reasoning called abduction to achieve these radical innovations. The process of abductive inferential reasoning builds on conclusions gained from extensive analysis followed by a period of rearranging perceived facts in the minds of the scientist. Burks (1946) interprets Peirce’s claims about abduction’s power to drive innovation as depending on two key elements. First, abduction relies on inferences taken from a body of data used to explain prior hypotheses, and second, in contrast to inductive methods oftesting hypotheses, abduction provides a way of discovering new, broader, hypotheses. Martin (2009) interprets abduction as being a synthetic way of thinking that integrates many of the lessons of experimentation into new ideas ofwhat might be possible. He concludes that synthetic thinking can fuel innovations in firms, such as new products like Research in Motion’s Blackberry communications device.
Peirce’s approach to innovation is very systematic. It views innovations as resulting from a combination of extensive analysis and experience, followed by abduction, synthesis and breakthrough. He proposes, “abductive suggestion comes to us like a flash. It is an act of insight, although extremely fallible insight. It is true that the different elements ofthe hypothesis were in our minds before; but it is the idea of putting together what we had never before dreamed of putting together which flashes the new suggestion before our contemplation.” (Peirce, 1988:227) In other words, innovation is the result of a kind of mental synthesis that rearranges previously diverse ideas now joined together. While Pragmatic analysis alone does not stimulate radical innovation, it sets the stage for the abductive process where breakthrough discoveries are often the result. Today, many organizational processes such as Total Quality Management, strategic management, KM and innovation, include pragmatic principles in their designs. Peirce wrote over forty volumes dedicated to presenting his core principle that innovation was mainly a scientific process resulting from integrating three different types of scientific reasoning that conclude in abduction. These three types of logic include (1) Deduction; (2) Induction; and (3) Abduction. Each of these types of logic enable decision-makers to draw inferences from evidence gained observation and experience. Management processes employ one of more of these types of reasoning. Ideally, these forms of reasoning become part of an integrated whole system and used together. (Table 1) Each of the first three methods listed focus on one type of inferential reasoning, whereas 5-Point Dynamic Mapping (Cavaleri & Seivert, 2005) is an integrative strategic management process designed to improve the quality of shared knowledge in an organization.
By using scientific reasoning in an integrative manner, it progressively reveals the mechanisms by which actions produce expected effects under specific conditions. This is not only a scientific way of achieving expected outcomes it is also sets the stage for enabling knowledge to be created and improved.
Pragmatism relies on the scientific method to drive the process of experimentation necessary to improve the quality of knowledge. The founder of the discipline of knowledge management, Karl Wiig (2004:213) sets forth four basic premises as the foundations of modern knowledge management. (1) Knowledge is the primary driver of enterprise performance; (2) Knowledge affects performance through people; (3) Effective Knowledge Management must be people-focused; and (4) Personal knowledge effectiveness. He concludes that in doing knowledge management “we must facilitate and strengthen the knowledge-related processes, activities, and practices that make it possible for people and organizational entities to make effective actions.” Some knowledge management advocates claim that information technologies should be the focus of KM initiatives. Yet, there is little evidence to support the effectiveness of such technologies in the types of business environments where knowledge management is most important - complex, dynamic ones. (Malhotra, 2005) Stewart (2002) also questions the effectiveness of technology-enabled KM approaches, noting they neglect to ask ‘what knowledge should be managed and toward what end?’ Firestone and McElroy (2005) argue that it is moot to seek a precise meaning of the term ‘knowledge management’ and more important seek the answer to question - Has KM has ever actually been done? They argue that unless KM is formulated using clear, non-contradictory ideas, it is impossible to lay any claims to its effectiveness or value. They see KM more simply as being a set of processes designed to enhance an organization’s present pattern of knowledge processing to improve organizational outcomes. Fahey and Prusak (1998) have voiced a similar skepticism about the way KM has been defined expressed as what they term the ‘eleven deadly sins of KM’.
Pragmatic KM’s focus is on managing knowledge processes to improve the quality of knowledge in organizations. Knowledge processes are the task-focused social interactions among team members designed to formulate, evaluate, and validate knowledge claims. (Firestone & McElroy, 2003) Knowledge claims are conjectures, assertions, arguments, or theories proposing which action are most likely to lead to expected states of affairs. Here, knowledge is, reiteratively, improved and refined, over time, through many cycles of experimentation. Experimentation means not only taking new actions, but also looking at the same situation through new eyes, and interpreting performance feedback differently. To the pragmatic manager, knowledge is a storehouse of rules for actions under various specific situations to achieve an expected outcome. Simply, knowledge is a type of contingency plan for action that is based upon on a set of situational cues and lessons learnt from experiences. Due the apparent fallibility of knowledge, it requires continuous improvement and validation to increase its value in generating effective action. Over time, knowledge’s main use is to solve problems and its effects become the basis for conclusions about the value of that knowledge. It becomes iteratively refined through the lessons learnt during these problem-solving episodes. This is a hallmark of Toyota’s approach to KM. Knowledge is continuously improved based on lessons learned from its application in various problem solving initiatives throughout the company.
Firestone and McElroy (2003) propose that knowledge matures over the course of a knowledge life cycle when organizations take systematic steps to improve knowledge quality. Improvements in knowledge quality also owe to the continuous problem solving processes used to drive knowledge processes. Unlike conventional KM approaches used to supply knowledge to workers based on its availability, pragmatic KM employs what McElroy (2003: xxiv) terms supply-side knowledge management. This stands in contrast to demand-side KM which “instead of focusing on the supply of existing knowledge to a workforce, seeks instead to enhance their capacity to produce it.” Firestone and McElroy designed their Knowledge Life Cycle model of knowledge management in a manner that “owes much to Karl Popper’s ideas on problem solving (1972, 1994) and the role of problems: detecting them (while engaged in business processing), arriving at tentative solutions (knowledge claim formulation), and performing error elimination (knowledge claim evaluation) to produce knowledge.” (2003: 197) The Life Cycle KM approach is significant for many reasons, but most importantly, it improves both the quality of knowledge and rate of innovation in organizations. Such developmental knowledge strategies have significant advantages over conventional static KM strategies due to their impetus to continuous improvement in the quality of knowledge that results from its use. The essence ofthe Life Cycle KM approach is the formulation, evaluation, and validation of know ledge claims by subj ecting them to the scrutiny of subject matter experts within the team or organization. Each time a problem emerges, it becomes viewed as being an experiment to select the potentially most effective course of action and study its effects for future codification. At Toyota, this becomes a self-organizing process where potentially interested workers, from around the company, receive invitations to view how a problem is being addressed -- so they may participate in problem solving and learn. In the true spirit of pragmatic knowledge processing, it is not sufficient to solve the problem, the quality of knowledge needs improvement too. Spear (2009:217) cites an example at the Toyota Supplier Support Center where a problem-solving team had solved a problem by getting better-than-expected results, yet failed in terms of learning outcomes. “It was true they had succeeded, but not completely. Yes, they had made the changeover process much better than it had been. Their shortcoming was not that they had failed to reduce the changeover further. It was that they had failed to learn from reducing it further.”
Toyota is one of the most highly effective knowledge management and innovation-based companies in the world. Even though Toyota has often received the MAKEAward as one ofthe best KM companies in the world, it rarely ever uses the term ‘knowledge management’ in its literature. One of the features of Toyota’s strategy is the pragmatic methods it uses for driving knowledge and innovation. This formula for innovation did not occur merely as the result of happenstance. Nearly half a century ago, Toyota’s leaders learned about a scientific approach to operating a company based on pragmatic principles from quality guru W. Edwards Deming. Deming’s system not only emphasized continuous improvement in quality, it also taught the value of knowledge and learning for innovation. Toyota has seamlessly interwoven knowledge processes directly into its business processes and systemic problem-solving initiatives. Designing work this way not only amplifies the value of existing improvement processes, such as TQM, but it also improves the quality of shared knowledge within the company. Why do such knowledge-based processes work so well at Toyota? The core ofthe Toyota’s high-performing knowledge-based system is a highly interactive collective form of scientific experimentation based upon the tenets of Pragmatism. The implications of adopting such a comprehensive organization - wide strategy in business and commerce are indeed profound.
Management scholars, such as Liker (2004) have closely studied Toyota, yet despite this extensive analysis, the theoretical sources of many of the improvement principles were relatively unknown. The founder of the famed Toyota Production System, Taiichi Ohno (1978) attributes many of the core ideas used in the framework to Henry
Ford. On the other hand, various members of the founding Toyoda family have credited the influence of the founder of the Total Quality Management (TQM) approach, W. Edwards Dem - ing, with having a major impact in shaping the company. Today, there is little doubt that many of the pragmatic principles, first advocated by Deming, now play an important role in innovation processes at Toyota. However, less obvious is the influence of Deming’s teachings on the firm’s structured approach to scientific experimentation and leveraging knowledge gained for improving business performance. MIT researcher, Steven Spear, describes Toyota’s approach to scientific experimentation in these terms, consider what the Toyota people are attempting to accomplish. They are saying before you (or you all) do work, make clear what you expect to happen (by specifying the design), each time you do work, see that what you expected has actually occurred (by testing with each use), and when there is a difference between what had actually happened and what was predicted, solve problems while the information is still fresh. (Johnston, 2001)
Near the end of his life, Deming wrote about the importance to companies of developing profound knowledge of how their improvement systems operate. What is far less well known is that Deming’s views were derivatives of Peirce’s Pragmatism. Some scholars, such as Towns, (1997) have argued that Deming himself was a Pragmatist who employed the pragmatic framework as a practical tool for effecting change in organizations. Similarly, Barton (1999) notes “a set oftouchstones exist which allow us to interpret systems and systems thinking in terms of Peirce’s pragmatism. (p. 7). Pragmatism, at its very core, is an approach that employs the scientific method to determine the possible future effects of actions to taken by understanding how past actions have produced particular outcomes. Pragmatism embraces the scientific method of inquiry toward improving the quality of knowledge-in-use in firms. Its methods ask managers to pay careful attention to understanding the mechanisms that enable their actions to produce effects. Over time, such analysis not only improves the quality of knowledge available for use in the organization, but it also prompts innovation.
The Global MAKE (MostAdmired Knowledge Enterprises) Award Hall of Fame has honored Toyota as one of the top twenty-six Knowledge Management firms in the world. Yet, Toyota executives rarely have used the term knowledge management. Toyota’s leaders have avoided using the types of KM techniques that are so popular in the Western countries, in favor of building their own custom knowledge-centric processes. Doing so enables them to interface seamlessly with its internal business and improvement processes. Toyota’s most well known knowledge-based approach is itsyokoten system. Yokoten is Toyota’s way of propagating new ideas and promoting improvement in other past of the company in conjunction with its quality, Lean, and problem solving approach.
Toyota’s approach to KM is essentially a pragmatic one. What makes it pragmatic? At it simplest, yokoten follows many of the same principles of kaizen, such as a mindset based on experimentalism, and the focus on causal analysis found in TQM. Pragmatism employs the scientific method to the question: How do we make actions more effective? Is it coincidental that Toyota’s approach to KM originates in the same philosophical doctrine as its quality improvement process? MIT professor Steven Spear (2004:7) spent many years observing Toyota’s workers and production systems in operation at various plants in Japan and the United States. He discovered at the core of Toyota’s vaunted Production System is a single-minded drive to engage in continuous rigorous scientific experimentation. “At Toyota, the focus is on many quick simple experiments rather than on a few complex lengthy ones. This is precisely that Toyota workers practice process improvement. They cannot “practice” making change, because a change can only be made once.
But, they can practice the process of observing and testing many times.” These are essential ingredients in Toyota’s yokoten system that integrates problem solving with knowledge management.
Toyota’s yokoten system focuses on creating and sharing knowledge that arises during problem solving initiatives. This focus on creating knowledge applied in problem solving initiatives is an example of what McElroy (2003) calls demand-side KM in action. Liker (2008:164) explains yokoten as being not merely a way of communicating or sharing best practices. Rather, it is a way to engage in collective problem solving and use the resulting innovations to propagate even more innovations in other parts of the company. He distinguishes yokoten from conventional ways of sharing best practice s by tracing the origins of yokoten. Toyota takes an organic, non-mechanical, view of it business. “It is not just ‘go, see and then copy’. For Toyota, it is ‘go, see, and improve upon.’” Inside Toyota, the backbone ofthe improvement process is a self-organizing collective problem solving process. Their justification for investing heavily in problem solving initiatives is that there are significant cost savings from eliminating the effects of problems before they become deeper and more widespread in their influence on performance. A key tenet ofthe Toyota system is that everyone has two jobs, first to perform a specific set of tasks, and secondly, to improve the job. The reason for employing this approach is deeply rooted in Toyota’s culture. The company’s focus on kaizen and continuous improvement is well known, but a more fundamental philosophy embedded in the fabric of Toyota’s culture is the belief people can never know in detail what will happen in the future. (Liker 2008:154) Consequently, the chief priority of managers is the training of operational employees in the techniques of problem solving to address unexpected problems that result from unforeseen changes in the organization before they become serious issues. This approach exemplifies Firestone and McElroy’s Life Cycle knowledge-processing methods discussed earlier in this chapter. The intellectual roots of this approach originated in Charles Peirce’s notion of a community of committed inquirers first conceived over a century ago.
At Toyota, identifying problems is more often the result of continuous reiterative problems solving and learning cycles. Research on Toyota’s problem solving approach by Spear & Bowen (1999:103) finds “.. .all managers are expected to be able to all jobs of everyone they supervise and also teach their workers how to solve problems according to the scientific method.” This is a systematic approach to concurrently both produce products and find problems. This process may seem inefficient to some, but it promises many benefits including reducing costs, and spurring innovation. Toyota’s approach stands in bold contrast to conventional Western manufacturing strategies. Critics wrongly assume its collaborative organic nature is inefficient because the focus on cost-reduction is indirect. In conventional mechanical production systems there is little actual learning, knowledge, or innovation that results among operators from running the system. By contrast, Toyota de signs all of its training, problem solving, and improvement initiatives to interface seamlessly with yokoten processes. This is feasible due to using the same conceptual framework of pragmatic principles. (Fearon and Cavaleri: 2006)
The yokoten process is simple. After identifying a problem, other employees throughout the plant receive a signal that a particular issue holding potential interest for them has surfaced. Then, the problem solvers send a brief description of the problem to all interested parties throughout the company. They are then invited to ‘come, look, and see’ the problem and to participate in the ongoing efforts to solve it. The ‘visiting’ co-workers can contribute their own insights to the problems - solving process as well as they may take away what lessons may be learned from the dialogues with their fellow problem solvers. This type of swarming activity attracts those other organization members who feel they may have something to learn and gain from participating in the problem solving effort because it may benefit them eventually. In accord with pragmatic principles, knowledge exists in a problem-focused context where the chief questions that arise focus on the design of the experiment to take place before the eyes of the workers and how they will measure the quality of their knowledge. This is a quite different approach from standard forms of KM in which various information technology systems send ‘information’ to employees - whether they perceive it may hold any value for them in solving problems they face.
At Toyota, the shared culture supports simultaneous efforts to improve people, processes, and competitiveness as part of a complex system. In a long-term study of numerous Toyota plants, Spear & Bowen (1999: 103) found “All of the organizations we studied are managed according to the Toyota Production System and share an overarching belief that are people are the most significant asset and that investments in knowledge and skills are necessary to build competitiveness.” Further, leading, learning, and problem solving at Toyota are all what Spear (2009) calls ‘high velocity skills’. He notes that Toyota is one of a number of high velocity organizations able to differentiate and gain sustainable performance advantages by nurturing superior improvement, innovation, and invention processes. At Toyota, work is an ongoing experiment that, over time, yields accumulations of knowledge capable of driving innovation. Toyota goes to extreme lengths to design its experiments. It starts by establishing its control variables. They do this by specifying the exact proce sses with little room for variation in performing tasks. They set up rigid standards at the outset introducing a high level of standardization. This standardization often takes the form of following what is termed The Toyota Way (Liker, 2004) of doing things. The purpose of making such extreme prescriptions for work is not as a means to create a lock-step production system. Rather, it is to facilitate the process of experimentation by aiding problem solvers in isolating causes of problems. The logic is simple - the fewer unknown variables, the simpler to isolate the causes of variation. While such a way of doing business may seem impractical, its pragmatic knowledge improvement value is subtle, yet critical. Let us revisit Spear’s example taken from the Toyota Supplier Support Center. The problem-solving team forgot its dual mission to improve both performance and the quality of knowledge. In Spear’s (2009:218) words, “Thus in falling short, they not only missed their target but missed the chance to push further to understand factors they had assumed to be true but that their experience had proved to be false. The process had gotten better, but their understanding of it had not improved as much as much as it might have had they made clear their expectations at the start and the assumptions underpinning them, thereby having something tangible to investigate when those assumptions were proven false.” Toyota provides the most visible instantiation of what a company looks like when it follows Pragmatic knowledge-based principles, and is deserving of further evaluation by scholars and practitioners alike.
It is ironic that business leaders tend to perceive their ways of managing to be pragmatic when in fact they are far from it. This owes to the common misconception of being pragmatic as being goal-oriented rather than being scientific and experimental. Pragmatic business strategies include fundamental elements, such as abduc - tive reasoning, inquiry, knowledge processing, experimentalism, and formal systems to improve the quality of learning and knowledge. At minimum, transforming business strategies toward being more pragmatic provides companies with a relatively simple way to boost innovation by improving the quality of shared knowledge within the firm. Haner (2002) proposes that innovation initiatives can produce varying levels of quality in innovation within a firm - depending on the expertise exercised in managing the effort. All innovations are not equal quality. Similarly, the conventional wisdom holds knowledge as being all of the same quality. It is axiomatic that breakthrough innovations can reliably flow from low quality knowledge. While leading economists, such as Nelson and Winter (2002), acknowledge the relevance and importance of organizational knowledge to the innovation process, they fail to define the specific mechanisms by which knowledge converts to innovation. Without recognizing the relevance of knowledge quality to the innovation process, many companies enthusiastically allocate significant resources to sharing low-quality knowledge. Low-quality knowledge imposes unnecessary limits on firms. Yet, such oversights are probable when firms do not have systems in place to evaluate the quality of knowledge-in-use. These oversights are traceable to the operational definition of the terms knowledge or knowledge management in use today. If managers assume knowledge and information are equal, then there is no need to be concerned about the issue of knowledge quality. This is because information is of singular quality in their minds.
On the other hand, philosophers, such as Plato, Socrates, Dewey, Peirce, Popper, and Quine argue that all knowledge is fallible. If the knowledge were infallible then it could is considered a being truth rather than knowledge. Popper’s (1945) theory of fallibilism outlines an epistemology proposing that all knowledge is capable of containing errors. One means of identifying errors and improving knowledge quality is by using the Life Cycle Model of Firestone and McElroy, 2005. Here, innovation is the natural outcome of ongoing efforts to improve the quality of knowledge where it is both a cause of performance improvement efforts, and its effect. This is a continuous, incremental, knowledge-based approach where innovation breakthroughs result from the accumulation of scientific knowledge acquired over time. While this approach is compatible with the general economic theories laid out by notable economists, such as Schumpeter (1950) and Nelson and Winter (2002), it runs contrary to many of the popular management-oriented theories of innovation proposed by thought leaders, such as Burgelman (1983), Christensen, (2003), Von Hippel (1998), and Utterback (1996). These management-driven theories of innovation often ignore the critical role played by knowledge and knowledge processing in achieving breakthrough innovations. By contrast, Pragmatic theories of performance improvement advocate for ongoing experimentation and learning from experience as a means to heighten the organizations capacity for adaptation. How do Pragmatic knowledge management and innovation approaches actually work? Martin (2009) argues against overreliance on traditional measures ofbusiness success, and for the value of using abductive logic in developing new products and companies. He notes,
While Motorola was projecting future sales volumes of “feature phones,” Mike Lazaridis, founder of Research in Motion, was imagining what executive life would be like if you could receive your emails on a handheld device. How compelling would an ordinary phone be if you could have a BlackBerry attached to your belt? He couldn’t “prove” that this would be a good idea. There was no data on the demand patterns for smartphones, because smartphones existed only in his imagination. But a mere 11 years after the launch of the product of his imagination, RIM leads Motorola by an ever-accelerating margin in sales, market share and profitability. Long ago, Peirce coined a term for the thinking that Lazaridis used to create the BlackBerry: abductive logic. Martin (2010)
Research by Sterman (1994) suggests that people’s capabilities of learning through experience diminish as the result ofa variety of cognitive and perceptual limitations that exist within complex systems. These factors often render human problem solving capabilities ineffective. In a world defined by high levels of dynamic complexity, even the smartest leaders are incapable of accurately predicting the effects on performance of the strategies and policies they formulate. The forces of bounded rationality are often too strong to simply permit effective management by relying on conventional methods ofmanaging (Simon, 1991).Yet, despite the limits on the cognitive processing abilities imposed on all executives by these forces, there is a general reluctance to even consider the possibility that limits exist. Executives often become skillfully unaware oftheir own inability to simulate solutions to complex problems in their minds. They prefer to try solving the wicked problems they face by simplifying them into ever-smaller components through a process of rational analysis. This strategy of reducing complex problems to their elements to find causes tends to produce more unintended consequences than the number of problems it resolves. Sterman (1989) studied the effects on managers and MBA students asked to play a computerized business simulation that incorporated delayed feedback of the effects produced by decisions the players made. The lengths of the feedback delays were comparable to those found in a low-moderate complexity business environment. The length of the feedback delay significantly reduced the game player’s performance in the simulation. In real life situations, unexpected ambiguities amplifying the complexities of learning from experience. Interpreting the exact meaning of feedback and in predicting its implications of for future performance becomes confounding for workers. In sum, the effects of delayed, confounding, and ambiguous feedback comingle to create a dangerous soup of misperceptions with great potential to cause managers to misfire in making decisions -- even in less complex dynamic systems. When the limits of bounded rationality on decision making by managers are well understood it becomes clear that even the smartest of managers struggle to discern optimal strategies in a dynamically complex environment. Given the ubiquitous effects of bounded rationality and dynamic complexity, the importance of creating high quality knowledge, learning rich lessons from experience, and experimenting in systematic way is greater than ever.
Designing knowledge management systems around the use of pragmatic principles promotes reflection, inquiry, and ways to improve the quality of knowledge in an organization. Alternately, an organizational culture rooted in a shared understanding of the value of experimentation driven by continuous learning, discovery, and innovation is likely to pay dividends for many organizations. Pragmatism provides managers with a scientific conceptual framework to guide their efforts to create and improve knowledge for innovation. One ofthe core tenets of Pragmatism is that the quality of knowledge improves directly in proportion to the level of insight gained from understanding the cause-effect mechanisms produce outcomes from actions. Pragmatic analysis studies how gaps between expectations and results trigger the inquiry process. Inquiry is a search for new solutions to improve performance and seeks to settle doubt over how future actions can generate desired results. Inquiry also opens the possibility of reconsidering the validity of gained through prior experiences. This sort of analytical process uses unexpected performance to evaluate the reliability and trustworthiness of the knowledge - in-use at any point in time. This is one element of the double-loop learning process described by Pragmatic theorists, such as Argyris (1990) and Schon (1983). Double-loop learning is a process designed to reevaluate the validity of operative knowledge by contrasting past performance with expected results. This type of action learning ultimately depends on replacing or improving current knowledge. Ideally, the goal is to align such knowledge more closely with one’s expectations.
In complex situations, where cause and effect are difficult to discern, the capacity to improve knowledge, beliefs, and assumptions becomes the fuel that drives adaptation, innovation, and other forms of intelligent action. Schon (1983:50) views simple technical rationality as being insufficient for solving the sort of complex problems faced today by professionals. “Once we put aside the model of Technical Rationality, which leads us to think of intelligent practice as an application of knowledge to instrumental decisions, there is nothing strange about the idea that a kind of knowing is inherent in intelligent action.” Above all, Pragmatism is a science for driving innovation by increasing the level of intelligence embedded in actions - thus increasing effectiveness. There is no intelligent action without reflection. At Toyota, the process of reflection receives much more attention than in most companies. Toyota’s 14th and final principle is to become a learning organization through relentless reflection (hansei) and continuous improvement (kaizen). Liker, (2004:251) studied Toyota extensively and notes that at Toyota reflection and learning drive innovation. “I believe Toyota is the best learning organization. The reason is that it sees standardization and innovation as two sides of the same coin, melding them in a way that creates great continuity.” Ultimately, the Pragmatic perspective defines knowledge as being the product of action, experimentation, prediction and reflection. When the polar forces of action and reflection comes into correct balance it opens the door to learning deeper lessons from experience. Technical rationality is incapable of adequately addressing the challenges posed by complexity or driving the innovation process forward in ways that may confer sustainable competitive advantages. One of the most fundamental assumptions that tend to derail knowledge management-driven innovation is the belief that organizations are machines and innovation is a predictable, deterministic process. The belief that organizations are machines leads to the presumption that innovation is a mechanical process that flows through a company via controlled, simple, and sequential channels.
The validity of such assumptions diametrically opposes much of the relevant academic literature that describes the basic nature organizations as being complex adaptive systems (Stacey, 1996). Forrester (1991:15) has an even harsher view of reductionist strategies designed to deal with the problems posed by complexity. “Complex systems defy intuitive solutions. Even a third order, linear differential equation is unsolvable by inspection. Important situations in management, economics, medicine, and social behavior usually lose reality if simplified to less than fifth-order nonlinear dynamic systems. Often the model representation must be twentieth order or higher.” Management strategies rarely are sufficiently sophisticated to adequately deal with the problems posed by complexity. Rothwell (2005) has taken a historical view ofthe innovation process and identified a series of progressively more complex theories to explain the innovation process that have evolved. He has identified five generations of innovation models ranging from linear models using technology-push systems to more flexible, customized, continuous innovation models. Similarly, Tidd (2006) claims that one of the most important challenges in managing innovation is to make clear sense of complex, uncertain and highly risky sets of phenomena. Yet, such dynamically complex domains tend to render traditional management strategies ineffective. All knowledge encompasses theories that predict some future state or expected outcome. Over time, accrued experiences can provide evidence to verify that theories-in-use and practices match well with reality. The next section of this chapter will explore this type of action-based view of knowledge and the tools that can help to raise the level of quality of this knowledge - including, inquiry, semiotics, and Pragmatic logic.
From the Pragmatic perspective, knowledge is always contingent on situations. In particular, its basis is a triad of interconnected elements that define every problem situation. Collectively, these triads are known knowledge acts. Knowledge acts consist of the following three elements:
1. Case - A perceived problematic situation or ideal future state
2. Rule - Algorithms that serve as guides to action in various situations
3. Result - Expected outcomes follow consequent to prior actions.
A typical case defines a problem situation or an unmet purpose. For example, it might be one where sales of a computer software product are increasing fast, calls to the company’s toll-free technical support hotline are increasing dramatically, and the rate of employee turnover among technical is increasing at steep rate causing the quality of customer to support to decline markedly.
A rule for action is a prescription of a series of steps or processes to follow in a given situation. Rules may range from being particular sets of algorithms to more general heuristics. For the manager ofthe technical support center is to follow the habit or routine for handling such situations in the past, such as to invest more money in recruiting new technical support personnel.
An expected result is a predicted outcome or ideal state of affairs. Whenever actions aim toward achieving a specific purpose, their design targets an imagined outcome or pattern of behavior. The expected result is that hiring new technical support personnel will offset the turnover problem and return the quality of customer to an acceptable range.
Here, a pragmatic manager envisions each possible problem-solving scenario as being an experiment. Experimentation fulfills dual purposes. It operates in the same way that managers at Toyota seek both to improve performance as well as the quality of knowledge about a situation. When a problem is recognized, then actions are taken in accord with specific rules -- based in knowledge - to attain a hoped for outcome. Such experiments seek to determine whether particular problem diagnoses and specific rules-for-action will produce expected outcomes. If results meet expectations -- then, the knowledge used to guide that action was valid for that particular case. Similarly, through a process of logical induction we may reinforce the rules for action used as being appropriate for use in this situation. Over time, as rules are repeatedly used, in becomes possible to ascertain their trustworthiness. In this scientific experiment, if expected results fail, then we may want to question the following:
1. How we interpreted a problem and what meaning we give to it.
2. Whether the rule for action we followed is appropriate to the situation.
3. The extent to which an expected future state is consistent with knowledge acts.
In Pragmatism, the science of semiotics governs the formation of meaning given to a perceived symbol in a problem situation. For example, a decline in market share may symbolize many things ranging from dissatisfaction among customers to aggressive action taken by competitors. The problem we ultimately defined and the desired result will largely govern the type of rules for action we are likely to invoke to achieve that desired state. Over time, as we go through multiple iterations
of following rules-for-action to achieve desired end states so as to resolve a perceived case -- we accrue knowledge of which actions work and which fail. Unlike simple theories of learning that address the importance of lessons learned, the Pragmatist uses the knowledge gained as feedback to measure the degree to which one’s beliefs mirror reality. Knowledge acts, not only become a basis for taking future actions, it also provides input for revising one’s belief systems. When expected performance fails, it triggers the irritation of doubt about the validity of our beliefs about how things work.
Inquiry plays a critical function within the Pragmatic knowledge processes. It leads not only to improving the quality of existing knowledge, but to creating new knowledge as well. Inquiry often takes the form of exploring new ways of thinking about past knowledge and experience to settle one’s mind about why actions taken may not have produced the expected results in the past. The inquiry process is set in motion when actual performance fails to align with goals (expectations). When performance fails to meet expectations it causes doubts to arise about the validity of the hypothesized causal relationship between case, rule, and result. In other words, there might be doubts about whether the right rules-for - action are used or whether the way the diagnosis of the case provided a meaningful description ofthe situation or both. Figure 1 illustrates the dynamic relationship between performance, doubt, and inquiry.
Whenever results fail to match expectations this precipitates what Peirce terms the ‘irritation of doubt’ to emerge. Figure 2 depicts how the unexpected declines in performance cause greater doubt in inverse proportion to the unmet expectations, which cause inquiry to increase. So here, the process of inquiry commences a search
for new answers that typically might involve redefining the case or selecting new/different knowledge acts to follow or both. Once performance improves to expected levels then doubt will decline as will the amount of effort devoted to inquiry. Figure 2 depicts the three basic elements that compose knowledge as envisioned in the Pragmatic tradition.
According to the Pragmatists, most explications of truth are defined, not only by how things work in reality, e. g. it is true that when a lead ball is dropped in the air it falls predictably toward the ground due to the effects of gravity, but there is also social component to defining what is true. Outside of the undeniable physical laws of the universe, in fields such as medical science, there are many interpretations as to the causes of disease based on the shared common beliefs of communi-
Figure 2. Situational Knowledge Act
ties of committed practitioners and inquirers. Figure 3 demonstrates the reciprocal relationship between performance and inquiry as mediated by doubt.
When performance declines and doubt grows, it causes our efforts to seek alternative explanations to expand through a process of inquiry. Inquiry is the driving force behind the science of discovery. Discovery can take many forms in organizations including invention and innovation
Figure 3. The balance of the Performance-Doubt-Inquiry Triad
Historically, Charles Sanders Peirce developed the principles of Pragmatism to provide a general roadmap to increasing the capabilities of scientists to make important discoveries. Peirce’s approach to promoting discovery was to employ the synergistic use of all three forms of reasoning - deduction, induction, and abduction to build knowledge and use it to inform effective action. For example, the logical process of deduction enables practitioners to use the knowledge extracted from prior experiences and now embedded in rules-for - action to guide future conduct. The effectiveness ofthese actions in meeting expectations provides feedback on the validity of the knowledge in use. Similarly, induction draws upon the lessons learned from prior experience to further innovate and revise rules for action thereby increasing their effectiveness. Finally, abduction synthesizes conclusions developed from prior deductions and inductions to create new hypotheses about how key governing relationships might exert their influence on some new initiative, such as releasing a new product or predicting changes in consumer buying habits. This logical process of recursively moving through continuous cycles of prediction and action is also the substrata of many process improvement systems. Most importantly, lessons learnt from revising rules-for-action, set the stage for creating improved knowledge for innovation. Knowledge gained from experience becomes the basis for enriching models of practice. Senge (1990: 176) argues, “The problems with mental models arise when the models are tacit - when they exist below the level of awareness.” Models based on false premises cannot produce highly effective outcomes reliably and in a sustainable manner. Quality experts, Box & Draper (1987:424) observed, “Essentially, all models are wrong, but some are useful.” Pragmatism enables managers to create causal models that will enable them to improve their knowledge, innovation, and performance.
Quality guru, W. Edwards Deming created what has become became the centerpiece of virtually every TQM initiative -- the Plan-Do-Check-Act cycle. (Figure 4) Among the greatest influences on Deming’s thinking were ideas he learned from his mentor, Walter Shewhart. Shewhart was an avid reader of the writings of Harvard professor of pragmatic philosophy C. I. Lewis (Towns, 1997, Barton, 1999) and began to envision the implications of applying their principles to performance improvement and statistical process control. Deming proceeded to marry together quality, performance improvement, and knowledge into the rubric known as Total Quality Management. He also wrote of the importance profound knowledge in quality improvement initiatives. Deming (1993:102) identified four components of profound knowledge including a theory of knowledge. He argued management is a form of prediction and requires having a theory of knowledge about how things work causally so we can anticipate the future consequences of our actions. Deming’s theory of knowledge teaches that any statement, if it conveys knowledge, then predicts future outcomes. Here, Deming was speaking ofthe abductive qualities of knowledge. As the quality of knowledge improves -- through reiterative problem solving initiatives -- it provides opportunities for workers to find errors in their collective thinking, reach deeper insights into how cause produces effect, and enriches mental models with new possibilities. By rearranging facts and impressions to form new knowledge acts, it also enables abductive hypotheses to form. Such new speculations are the seed of innovation because they produce a type of design synthesis that heretofore failed to exist. In effect, a rational series of conclusions and inferences indirectly leads to creative insights that would be otherwise unlikely to form.
The PDCA cycle does not explicitly acknowledge how the quality of knowledge-in-use changes over time as a product of learning and experimentation. Incorporating the aforementioned pragmatic principles into the PDCA cycle there is a greater potential seeing more clearly how learning, reasoning, reflection, and experimentation, when integrated together, all contribute to improvements in the quality of knowledge. The four Pragmatic tools described earlier in this chapter directly govern the extent to which employees can fully realize the potential ofthe PDCA model to increase levels of organizational performance, adaptation, and innovation. Knowledge is the product of using these tools and is the second most important determinant of effective action - behind holding valid beliefs about how things work in practice. Figure 5 provides an integrative framework that unites knowledge, learning, and reasoning into a single perspective.
The PDCA cycle often acts as a balancing feedback loop to determine the extent to which actual results match expected performance. Ifthey do not match, then it sets off inquiry-driven search processes designed to reduce doubt by closing the performance gap by improving performance. The cycle achieves its learning purpose by enacting specific forms of inferential reasoning to drive each of the four elements in the loop. For example:
Figure 4. Deming’s PDCA Cycle
• The PLAN function seeks to determine the exact parameters of the case;
• The DO function seeks to select the Rule (rules for action) that will drive action to achieve a desired state;
• The CHECK function looks at both the expected result of anticipated action as well as the effects of prior action.
In Pragmatic logic terms, the various steps in the PDCA correspond to the specific types of reasoning necessary for creating new knowledge or improving existing knowledge. The process of deduction drives planning, while checking is primarily a process of induction. Finally, the Act function is largely a matter of abduction. (Reed, 2010) Each type of inferential reasoning builds on the conclusions that result from a logical sequence of considerations among the case-rule-result triad. The Pragmatic system of knowledge quality improvement operates as a rigorously scientific experimental framework designed to ensure that the quality of the knowledge-in-use is continuously improved. It is more likely that knowledge built upon valid premises and using the logic of abduction will be capable of effectively driving the innovation process.
Despite Toyota’s outstanding business performance in terms of quality and innovation, competitors tend to emulate its pragmatic methods far less often than one might expect. Arguably, Toyota’s organizational system is unique and difficult to imitate. It is a complex system composed of a mix of deeply held values, precise standards, relentless learning, scientific experimentation, and personal reflection by its employees. This sophisticated system has evolved over a half-century driven by adherence to the company’s fourteen basic principles. The difficulty other companies experience in imitating Toyota’s knowledge processes provide a competitive advantage to the firm. After all, typically the greatest sources of sustainable
competitive advantage are unique applications of a firm’s intellectual capital reflecting their subtle ways of thinking and perceiving that remain less then apparent to outside observers. However, it is equally arguable that the barriers to imitation of Toyota’s strategy stems even more the extent to which its process are designed on the basis of pragmatic logic that often is nonsensical in the eyes of outsiders. For example, the notion of reducing costs, indirectly, through knowledge and innovation rather than by employing discrete cost - cutting initiatives is purely confounding to many competitors. Further, the pragmatic principles that uphold their yokoten system has little resemblance to the sorts of information technology-based KM approaches that proliferate around the world. Toyota’s KM system is a direct mirror of its own pragmatic principles that subscribe to idea that knowledge, action, feedback, and belief are inseparable. In the original pragmatic writings of Peirce there was a perspective expressed that all of these factors are parts of a larger dynamic system driven by inferential reasoning, inquiry, and reflection. In sum, Toyota’s quality, knowledge, and innovation processes are synergistic and operate seamlessly together. In other words, there is no Toyota ‘innovation system’ or ‘quality improvement system’ per se - rather there is the Toyota pragmatic way of doing business that embraces all ofthe processes that deliver superior performance for the company.
Pragmatic knowledge-based systems stand in stark contrast to conventional knowledge management approaches in several important ways. Conventional KM approaches demand either a high investment in technology and/or employee time. Ironically, many of today’s employees already carry high responsibilities and resent the ‘add-on’ nature of typical KM processes. This is especially so when the direct benefits of doing such types ofKM are not readily apparent to them. The failure of such conventional KM approaches to spur innovation over the past decade has been unsurprising. First-generation KM approaches do not address the sort of knowledge-creation processes that innovation requires. Pragmatic KM innovates by seamlessly integrating knowledge with processes, such as problem-solving and quality improvement. However, today TQM has run out of steam in many companies as managers focus more on its cost-reducing benefits than on its innovation promise. To more effectively leverage the promise of knowledge-based innovation programs individual employees must become capable of changing their beliefs about how things work as the result of ongoing scientific experimentation. They must be taught to be open to evidence that may go contrary to habitual ways of thinking. Typically, such insights are the product of not only experimentation, but also reflection. While the notion of engaging in wide spread reflection among workers in companies has largely been rejected in North America as being inefficient, the need for it remains compelling. In the race among companies to achieve more sustainable forms of competitive advantage, a type of organization known as a high-velocity organization is winning the race (Spear, 2009). They achieve this by consistently investing heavily in developing employee capabilities in four areas: (1) systems design and operation; (2) problem solving and improvement, (3) knowledge sharing; and (4) developing high-velocity skills, such as leadership, in others.
Amidst the hype of promises made about KM, among those many companies that claim to have used it, few of these promises have realized its potential benefits. Yet, there are significant advances by other companies, such as Toyota. Over the past twenty years, Toyota has been developing its own type of knowledge-based system to improve problem solving, quality, and innovation. This system relies on the Pragmatic principles embedded into the foundation of TQM by Deming. The main advantages of adopting a Pragmatic perspective is its focus on operational improvement at relatively low cost. Organizations that have already invested in building their capabilities in organizational processes, such as TQM, Lean, organizational learning, and systems thinking will find the transition to becoming more Pragmatic and innovative to be a relatively smooth journey. Yet, there is no escaping the principles laid out by Schumpeter (1950). He pointed out the importance of knowledge for innovation, and identified the principle that successful innovation results from the accumulated scientific knowledge that develops within a company over time. His use ofthe term scientific knowledge in innovation processes runs parallel to what Peirce described as being necessary for driving abductive reasoning. Innovation may occur in a flash - but it often results from incremental experimentation that has laid a foundation built on high quality knowledge.