Building Ba to Enhance Knowledge Creation and Innovation at Large Firms
figures

Ikujiro Nonaka

Professor

Graduate School of International Corporate Strategy, Hitotsubashi University;

Xerox Distinguished Professor in Knowledge, Haas School of Business, UC Berkeley; and

Visiting Dean and Professor, Center for Knowledge and Innovation Research, Helsinki School of Economics

Ryoko Toyama

Associate Professor

Japan Advanced Institute of Science and Technology

Graduate School of Knowledge Science

Tatsunokuhi, Ishikawa, 932-1292, Japan

+81-761-51-1722

rtoyama@jaist.ac.jp

Otto Scharmer

Lecturer

Massachusetts Institute of Technology

Visiting Professor, Center for Knowledge and Innovation Research, Helsinki School of Economics

(+1) 617-258-8132

scharmer@MIT.EDU

First DRAFT, June 2001

Please do not quote without authors ’ written permission

The authors want to thank the "Boston conversation" on leadership and knowledge, including participants Brian Arthur (Santa Fe Institute), Joseph Jaworski (Generon), Michael Jung (McKinsey), Jonathan Day (McKinsey) and Peter Senge (MIT), for inspiration and feedback. The NTT DoCoMo case is based on the case study prepared by Hiroki Nakagiri and Kenji Yoshioka.

"Innovation happens in places." John Kao, the founder of The Idea Factory, looked straight into our eyes as he explained his guiding principle for achieving breakthrough innovation. What he said resonated deeply with what we felt when entering the San Francisco based Idea Factory space, a large factory hall with elements of design, theater, and high tech situated in a former sweatshop of the previous industrial era. Kao continued: "There’s a reason why artists have studios. There’s a reason why in Hollywood the work happens in protected enclaves where you have to go through a guard’s gate and have a pass to get onto a studio lot. Sound stages are often locked, and they have red lights outside that signify when they’re in use. There are all kinds of ceremonial elements to make sure people know that there’s a distinction between ordinary life and the creative environment where the work goes on."

While all artists know that the quality of place is critical for the quality of their performance, it seems obvious that companies also focus on creating configurations of places at large that allow the corporate performance system to continuously evolve, innovate, and thrive. And yet, after about a decade of discussion on knowledge management and organizational learning, we are still short of a deep and truly actionable understanding of what it takes to lead and organize around profound knowledge creation and sustained innovation.

We believe that the single most important factor shaping the quality of knowledge creation is the quality of place. To differentiate the various aspects of place we use the concept of ba, a Japanese word that translates roughly as "place." According to the Japanese philosopher Kitaro Nishida, each ba has a physical, a relational, and a spiritual dimension. Although many approaches to knowledge management have focused on the relational dimension, emphasizing the mental and social conditions of knowledge networks and respective communities (Prusak and Cohen, 2001), there has been relatively less attention paid to the physical and spiritual ends of the spectrum and almost no attention to how all three dimensions of place are interlinked as elements of a larger field or the organization as a whole.

In this perspective, the company is reframed as arising from a dynamic system of places–as an organic configuration of ba. The leadership role is to generate an evolving system of interwoven ba such that it enables its participants to tap into their best individual and collective sources of creativity and innovation. While Hagel and Singer (1999) suggest that companies should unbundle and organize around a single focus such as infrastructure (economies of scale), customer interface (economies of scope), or innovation (economies of speed), we believe that an organic configuration of ba is not about separation but about both differentiating and interweaving these three spheres of value creation. How to better perform the leadership role creating the ba of high-performing companies and organizations is the question that underlies this work. In the following pages we look at two case studies, the Toyota Prius and NTT DoCoMo’s i-mode service, and conclude with a number of principles and practices regarding what leaders can do to build a high-quality ba.

Knowledge is created in situated action (Suchman, 1987). The knowledge-creating process is always context-specific in terms of time, space, and relationships between people. Information becomes knowledge when it is interpreted in context (Schoenhoff 1993). Such a context is interactive (Ueno, 2000).

Ba offers such a context. Based on the concept that was originally proposed by Nishida (1921, 1970) and was further developed by Shimizu (1995, 1999), ba is here defined as a shared context in motion in which knowledge is created, shared, and utilized (Nonaka, Toyama, and Konno, 2000). Ba is the context shared by those who interact with each other, a process through which the context itself evolves through a self-transcending process of knowledge creation. Ba provides the energy, quality, and place for individual knowledge conversions and for moving along the knowledge spiral. Knowledge emerges out of ba.

We ground the concept of ba in an existentialist framework. The key platform of knowledge creation is the "phenomenal" place. A place of knowledge creation can emerge between individuals, in working groups, project teams, informal circles, temporary meetings, and e-mail groups, or through contact with customers. Accordingly, ba evolves along three different dimensions: the physical space, such as a meeting room; the relational space, such as the patterns of interaction, emotion, and thought; and the deep tacit or spiritual place, such as the sources of one’s creativity, purpose, and energy (see Figure 1).

The role of leadership is to build and utilize ba within and across organizational boundaries so that high-quality knowledge can emerge. A firm is a constantly unfolding organic configuration of ba, and the relationships with other entities such as suppliers and alliance partners can also be understood in this perspective. The following case studies of innovations at Toyota and NTT DoCoMo explain how leaders can manage ba to lead a knowledge-creating process by synthesizing the knowledge assets both inside and outside the firm.

The Prius is the world's first mass-produced hybrid vehicle, which is powered by both a gasoline engine and an electric motor. Introduced to the market in December 1997, the Prius has received numerous awards, including Japan Car of the Year and RJC New Car of the Year, for its innovative product concept and technologies. According to Toyota, the Prius’ s fuel economy and engine efficiency are 100 percent and 80 percent better than that of its conventional gasoline-powered cars, respectively. The Prius emits about half as much carbon dioxide and about one-tenth as much carbon monoxide, hydrocarbon, and nitrous oxide as gasoline-powered cars. As a hybrid car it uses its gasoline engine to charge its electric motor and therefore does not need to stop to recharge its battery, unlike a typical electric car.

The Prius was revolutionary for Toyota in three areas: product, technology, and knowledge creation. First, it is an innovative product that does not fit into any existing product line. Toyota has several product lines in which each product is designed to complement the others so that all Toyota products have a consistent image. The Prius, however, was designed to give Toyota a new image.

Second, the Prius uses many innovative technologies in its engine, motor, battery, and brakes and in combining them into a hybrid system. These technologies are important for both the development of Toyota and the company’s future. Toyota emphasizes the importance of yokoten, which roughly means " horizontal deployment" of technologies. Many new technologies developed for the Prius were later applied to other products.

Third, Toyota used a new method of product development to produce the Prius. It usually takes about four years for Toyota to develop a new model of an existing product, which is one of the shortest cycle times in the automobile industry. The Prius, using a totally new design and new technologies, took only 15 months from design freeze to the start of production. To achieve the speed and innovativeness that the project required, the engineers adopted an unconventional approach. Because the hybrid system was still in the research stage during the development of the Prius, both the research and the development were done at the same time.

To achieve such simultaneous innovation, top and mid-level leaders built, utilized, energized, and connected ba throughout the development of the Prius. The effective use of ba was also important for the development of i-mode service by NTT DoCoMo, a spin-off company of NTT, the Japanese PTT giant. The development of i-mode did not conform to the conventional bureaucratic way of doing things at NTT and DoCoMo.

DoCoMo ’s i-mode service, which allows subscribers to connect to the Internet via their cellular phones, makes it "the most advanced wireless Net access service on the planet," according to Business Week. The number of subscribers to the i-mode service had exceeded 15 million by November 2000, making DoCoMo the largest Internet provider in Japan and the world's largest and most valuable single-country cellular phone company. Using the buttons on their cellular phones, subscribers can check stock prices and bank transactions, read news and their horoscope, and play games.

The following section explains how Toyota and DoCoMo built and utilized the creation of ba to realize their innovations.

Ba can be built intentionally or created spontaneously. Leaders can "find" and utilize spontaneously formed ba, which can also change or disappear very quickly. Hence, leaders have to improvise in terms of how members of the organization interact with each other and with outside environments in order to quickly capture the naturally emerging ba, as well as to form ba effectively.

Leaders can build ba by providing time, space, attention, and opportunities for relationship-building. They can provide physical space such as meeting rooms, cyberspace such as a computer network, or mental space such as common goals to foster interactions. Creating mental space that fosters " love, care, trust, and commitment" among organizational members is important because it forms the foundation of knowledge creation (Krogh, Ichijo, and Nonaka, 2000).

Forming a task force is a typical example of the intentional building of ba. The Prius project started in September 1993 as a small study group called G21, meaning Generation 21^st Century, with the strong backing of top management. At first, G21 was a working group to study " the car for the 21^st century." While still working at their regular jobs, G21 members met once a week to discuss the concept of the 21^st-century car.

To build ba, leaders also have to choose the right mix of people to participate and then promote their interaction. It is important to choose participants with different backgrounds and viewpoints who bring their own experiences and contexts into ba to make the shared context a rich one. Although it is often difficult for a large organization to know exactly what it knows, effective leaders must know where in the company to find those with the appropriate knowledge. In January 1994, Takeshi Uchiyamada was selected as the Prius project leader. This was Uchiyamada’s first chance to lead a new-model development, although for the previous two years he had been in charge of reorganizing Toyota's research and development laboratories. The R&D project gave him a broad knowledge of the technologies Toyota was working on and where to find the people and skills he would need. When he was named to head the Prius project, Uchiyamada spent time learning about those technologies in more detail.

Uchiyamada decided to build a self-sufficient team with all of the necessary capabilities to develop a car. In the case of usual product development, the product development team just has to make improvements on an existing model and choose necessary components. The development team submits an " order" to related departments, such as the engine department, to develop necessary components. In such cases, the engine is likely to be what the development team wants instead of the engine that fits the car best.

When it began, the G21 project did not yet have a product concept. What it did know was that it needed the best technologies to fit the car and that the car should be developed quickly. Uchiyamada decided to form a team representing all the necessary technologies. Eventually, ten members were selected with expertise in eight areas, such as body, chassis, engine, drive system, and production technology. They were all in their early thirties, old enough to have experience but young enough to be flexible.

The first thing the team did was to acquire a physical space: it "procured" a room to be their base camp, where two computer-assisted design (CAD) systems and personal computers were brought in. Here, team members worked closely with each other. It was the first time Toyota had had a project team working in one room to develop a product. The team also utilized virtual space as ba. Especially effective for the prompt dissemination of information was the electronic mailing list, through which the members exchanged necessary information. Eventually, 300 people were on this mailing list.

In addition to selecting the ba participants, Enoki set the boundaries of ba. Boundaries must be set within which a meaningful shared context can emerge. This is sometimes called "cocooning," the practice of building a unique world or context (de Monthoux, 1996). Leaders should set boundaries and protect ba when it is necessary. Enoki worked as an interface between the project and other departments in DoCoMo to prevent conflicts. But because the " misfits" were familiar with contexts that were very different from DoCoMo’ s, some conflicts were inevitable. "Without Enoki to protect us, we wouldn’ t be able to do anything," said Natsuno.

Still, a ba is an open place where participants with their own contexts can come and go and the shared context can evolve. Ba lets participants share time and space, and yet it transcends time and space. Ba also lets participants transcend self. To participate in a ba means to transcend one's own limited perspective or boundaries. Participants can be both insiders and outsiders at the same time. By providing a shared context in motion, a ba sets binding conditions for the participants by limiting how the participants view the world from the inside. And yet it provides participants with viewpoints other than their own in order to see things from the outside. It is necessary for leaders to keep the boundary of ba permeable, so that such openness can be retained. In the Prius project, team members were required to transcend their own boundaries. Uchiyamada established the guidelines for the project, which he put in the first page of his notebook and always brought with him. It included: "The technology should be evaluated by everyone, regardless of his specialty"; "One should think of what is best for the product, instead of representing one’s own department’s interest"; and "When discussing technologies, one should not care about one’s age or rank." Working closely together, team members learned from each other. At DoCoMo, a room was created for the i-mode team to have conversations with each other and experts in various fields (some are not directly related to telecommunication business). Casually referred as "Club Mari," the room was well furnished and drinks were stocked so that everyone can relax and communicate freely. At Club Mari, participants exchange their ideas and viewpoints, and created new ideas.

But merely building ba is not enough for a firm to manage the dynamic knowledge-creating process. Ba needs to be " energized" to give energy and quality to the knowledge-creating process. Ba needs to be a self-organizing place with its own purpose, direction, interest, and mission. Without purpose (which can be given or can eventually emerge), the energy in ba cannot be directed effectively, and only chaos rules the ba.

From the beginning, the Prius project had a clear goal of creating a car for the 21^st century and introducing it to the market. Another mission for the project was to create a new method of product development. Because the project was initiated by Toyota’s executive vice president, Yoshio Kanahara, and had strong backing from the company’s president, Shoichiro Toyota, its mission was disseminated throughout the company. It was also said that Eiji Toyota, the chairman, commented, " This is a project I would want to do, if I were a bit younger."

The top management decided that Toyota would take the lead among carmakers in dealing with environmental issues. The Prius, the world’ s first mass-produced hybrid car, was treated as the project that would determine Toyota’ s future. Toyota first began thinking about a hybrid vehicle around 1991, when its Electric Vehicle Development Department started to study the hybrid system that uses both an engine and a motor. But after developing several prototypes, Toyota realized that it did not have technologies to make the main components of the hybrid system, such as the battery, motor, converter, and inverter. At that time, Toyota would have been a mere assembler of components supplied by outside companies. Toyota, which prides itself as an automobile maker, took its shortcomings as a serious threat. Realizing that it lacked specific knowledge assets, the top management established initiatives to research, develop, and produce the hybrid system internally. In short, Toyota created the ba in which to study the hybrid system because it realized that the technology held the key to Toyota’ s future.

In the case of DoCoMo, the knowledge vision "From volume to value" was put forth by Oboshi, then the CEO, who felt that the firm needed to expand into data communication. It was expected that the voice communication market would be saturated in the near future and that DoCoMo would need another business to sustain its growth. Based on this vision, Enoki built a ba to develop the i-mode service.

The energy of ba is provided by its self-organizing nature. To make a ba a self-organizing place, leaders need to provide conditions that allow for autonomy, creative chaos, redundancy, and variety and that foster love, care, trust, and commitment. The Prius project was a typical self-organizing team. Further, Uchiyamada created several task forces inside the project team to work on particular issues; the leaders of the task forces were given complete autonomy in their work.

The Prius project used creative chaos effectively. Leaders introduced creative chaos to the Prius project by setting very challenging goals. One goal was to find a new way to develop a car, and to do so, Uchiyamada demanded that his team members question every established norm related to new car development. For example, the team planned to make the car 50 percent more fuel-efficient than its conventional cars. But Wada, the executive vice president, established the even more challenging goal of 100 percent improvement. This change threw the team into turmoil, and they eventually discarded their original plan to use the direct-injection engine in favor of the hybrid system.

Further, the top management put the team under great time pressure. The Prius was originally planned to be released in December 1998. It was a tough deadline to start with, considering that developing a new model for an existing line took an average of four years. However, Okuda, the president, pushed the deadline up one year, to December 1997. The team took many unconventional approaches to cutting the cycle time, such as extensive simultaneous engineering.

All of these goals created the opportunity for leaders to put a good ba in place. A good ba can be expressed with the metaphor of a sphere. Because all team members should have equal access to the "center," every participant in a ba is at the same distance from the center. However, the center here is not a fixed point. In a ba, anyone has the potential to be the center, and because ba is a shared context in motion, the center can change as the context evolves. Rather than being a stable entity, ba is a constantly moving sphere. Because a sphere has maximum surface area and minimal volume, it provides maximum external interface and variety.

" Equal access to information" was one of Uchiyamada’s action guidelines for the G21 project. As noted previously, e-mail was used effectively for speedy information exchange. In an ordinary product development project at Toyota, when an employee found a problem he reported it to his boss. And if the problem couldn’t be solved it would be reported to the chief engineer, who would notify other engineers who could be affected by the problem. This was a time-consuming process. In the Prius project, by contrast, the engineers could post e-mail to the mailing list immediately after discovering a problem. Anyone who read the e-mail and had a solution could immediately post the necessary information.

In a good ba, participants cannot be mere onlookers. They must be actively involved in the goals and activities of the ba. The commitment of the participants is what gives the energy to action and interaction in ba and hence determines the quality of knowledge that is created in ba, since commitment underlies the human knowledge-creating activity (Polanyi 1958). Leaders have to foster such commitment among the participants of ba.

In a good ba, dialogue among participants with different viewpoints and backgrounds is essential. Questions such as ﾒWhat is the essence of this thing/event? ﾓ and ﾒWhy do we do this? ﾓ let participants see things and themselves from viewpoints deeply rooted in their own beliefs and values, and from others ﾕ perspectives at the same time.

These dialogues, when they explore deeper questions like "Where does my commitment come from?" and "What is it that wants to emerge here," open the space to the deep tacit, that is, the more spiritual foundations of ba. In this context, a spiritual exploration means inquiring into one’s deeper sources of creativity, commitment, and energy (Loehr and Schwartz, 2001). It is important for a leader to " coach" participants in exploring all three dimensions of ba. A leader can also act as a " third person" in the dialogue, offering yet a different perspective to help participants and groups gain access to their self-transcending sources of knowing (Scharmer, 2001). In the case of DoCoMo, Enoki often offered the perspective of a coordinator, Nastuno and Matsunaga worked together, and the i-mode team worked with outside people such as the sales force.

In knowledge-creating processes, just one ba is not enough. Knowledge creation needs many ba, which exist on multiple levels and are connected to each other organically to form a greater ba (see Figure 2). An organization is an organic configuration of ba, where various ba form a fractal, a whole that is part of and embedded in larger wholes. How ba are connected and interact with each other determines the quality of knowledge created.

Hence, leaders have to facilitate the differentiation and integration among various ba and synthesize the knowledge created there. Connecting and interweaving ba is, in other words, the recontextualization of ba. In many instances, the relationships among ba are not predetermined. Which ba should be connected in which way is often unclear. It is different from the " network economy," in which each part is treated as a module and intersections among modules are standardized. Therefore, leaders have to read the situation to connect various ba and improvise as the relationships among them unfold.

In the case of the Prius, Uchiyamada, with strong backing from top management, used his experience reorganizing Toyota’s R&D organization to bring together those with the knowledge necessary to create and connect various ba. When the top management established the new goal of doubling the fuel efficiency, Uchiyamada had no choice but to use the hybrid system. At that point another ba, which had been developing the hybrid technology, was brought in to recategorize and recontextualize the project. Since the hybrid was still in the research phase at that time, Takehisa Yaegashi, head of Toyota's hybrid development project, and his team of 15 engineers worked day and night to coordinate their work with that of the researchers. Meanwhile, Uchiyamada and three others set to work coordinating the overall project. One coordinator dealt with the hybrid system. Another was in charge of body, cost issues, and weight. A third oversaw regulatory issues, production planning, and marketing. The intent was to set parameters early and make key decisions as quickly as possible. Explained Uchiyamada, "You have to keep the numbers down if you want to develop a vehicle quickly."

Once the basic criteria were set, Uchiyamada went to each department–those responsible for brakes, electrical systems, and so on–for the engineers he needed. Again, the fact that senior management was solidly behind the Prius allowed Uchiyamada to obtain the people he needed without running afoul of managers protecting their territories. " Okuda was very concerned that sectionalism would appear," according to Uchiyamada, "and he encouraged everyone in the project to cooperate fully." Eventually, about 1,000 Toyota personnel worked on the car, most of them on a part-time basis. Figure 3 shows various ba for the Prius project and how they were connected to each other.

It was necessary for the engineers who participated in the project to transcend their own boundaries. The boundaries of ba are permeable, and participants from various ba work together to form a greater ba when several ba are connected each other. Because the Prius was developed so quickly, everyone had to understand everyone else’s work in order to carry out their own tasks effectively. Toyota is known for its " simultaneous engineering," but the Prius was an extreme case. For example, Toyota normally utilizes " resident engineers" (REs) who are on-site at the manufacturing plants to solve any problems during the production of a newly developed vehicle. For the Prius, " reverse REs" came from the manufacturing plant to participate in its development. This was to ensure that there would be fewer problems when the Prius got to the manufacturing stage.

Because the Prius utilized a hybrid system, engineers of several components had to work together seamlessly. The engineers in charge of the engine also had to understand the technologies used in the motor, generator, and battery and coordinate with those who were in charge of these technologies. For example, when certain improvements to the engine reduced its fuel economy, engineers of the various components together had to determine which component was causing the problem and design a solution. Again, this required the engineers to cross the boundaries of their specialties.

Ba is not necessarily created within the organization’ s boundary. It is often necessary to share and utilize knowledge that resides outside the firm. In the case of Toyota, a joint venture with Matsushita Battery was created to manufacture the batteries for Prius. Toyota needed the Matsushita’ s knowledge on manufacturing batteries, and Matsushita needed Toyota’ s knowledge on quality control.

Connecting ba was also important for DoCoMo’ s development of the i-mode, particularly across organizational boundaries. I-mode works as a portal, and some of the ﾒ official sites" that can be accessed directly from DoCoMo's i-mode menu charge a fee for their service. DoCoMo collects the fees on behalf of the content providers and takes a 9 percent commission. Subscribers are attracted to i-mode because of the content providers, and the content providers are attracted to i-mode because of the number of subscribers they can reach via i-mode. The result is that both DoCoMo and the content providers make money.

At first, consultants suggested that DoCoMo should buy content from providers. However, Kiyoshi Tokuhiro of NTT DoCoMo said, "Content is knowledge. If you bought the content, that would be the end of the interaction. Because we built a win-win situation in which content providers are part of the process and can make money as well, the interactions [among DoCoMo and providers] won’t end."

I-mode uses a subtext of HTML called compact HTML. The use of compact HTML makes it easy for a content provider to convert an existing Web page to the one for i-mode. There was already an international standard called WAP (wireless application protocol), but Enoki decided against using it. ﾒ In the Wireless world, not using WAP means that DoCoMo is an orphan. However, in the Internet world, WAP is an orphan,ﾓ says Enoki. ﾒ The wireless part can be managed by the money and technology of DoCoMo. But for the content part, we cannot do anything. Content exists on the Internet. So we have to accommodate to the Internet world."

By using compact HTML, i-mode can take advantage of the vast amount of content on the Internet. Apart from the official content providers, nearly 4,000 sites are i-mode accessible today. Content providers interact continuously with DoCoMo to improve their service. Knowledge is created out of such interactions. DoCoMo builds and connects ba with content providers instead of conducting all the knowledge creation within DoCoMo. Natsuno, the Internet entrepreneur, used his experience to cultivate the content providers that became the service’s official sites.

Leaders promote profound innovation and knowledge-creation by creating a good ba. The following ten principles and practices summarize how to build a good ba.

1. The Power of Place: Create places that have their own field and atmosphere, that help people and teams focus on what is most important, and that provide access to their deeper sources of creativity. These places need an open boundary that on the one hand provides a cocoon in which to do creative work and that on the other hand provides openness and a flow of new perspectives, people, and ideas that come with it (see sidebar on The Idea Factory).

2. Prototyping: Create physical embodiments of the real work. A good ba happens not outside but in the midst of the real work. A good ba allows for the real work to move to center stage. In the case of product development teams this often translates into simultaneous engineering and rapid prototyping processes in which the object of co-creation becomes the catalyzing object around which the ba develops and unfolds.

3. Relational Place: Harness the diversity of experience and talent across the whole system. A good ba embodies the diversity of experience and talent throughout the system. Innovation teams often are most successful when they are composed of people who embody the experiences of those on the periphery of the system (younger people, recent hires, and/or people who are distant from the headquarters or even outside the organization), because it is on the periphery where the new shows up first. For example, Enoki, who was not in mainstream of DoCoMo himself, recruited a group of young engineers internally, as well as outsiders with specific expertise, each of whom brought experiences and talents that would challenge the experience base of the internal group.

4. Sphere: Everyone has equal access to the center. Every participant in a good place is at the same distance from the center. However, the center is not a fixed point. In a ba, anyone has the potential to be a center, and the center can change as the context evolves. Ba is a sphere that is constantly moving along an evolving trajectory.

5. Purpose: Uncover a powerful purpose and intention. Both Enoki and Uchiyamada started by focusing on and building a powerful sense of mission and purpose for the overall project. In both cases it was pointed out that the future of the company largely depended on the completion and success of this single project. And both leaders continued to repeat and emphasize the primacy of that underlying purpose.

6. Cocreation: A good ba needs creative chaos, care, and love in order to tap into people’s highest level of energy and commitment. The energy of ba is provided by its self-organizing nature. To make a ba a self-organizing place, leaders need to create conditions that allow for autonomy, chaos, redundancy, and variety and that foster the development love, care, trust, and commitment. Both Toyota’s Prius team and NTT DoCoMo’s i-mode team were typical examples of highly self-organizing teams. Uchiyamada also created several autonomous task forces inside the Prius project team to work on particular issues.

7. Imaginative Seeing: Innovation and inspiration start with seeing differently. The capacity to see differently and deeply is crucial and yet difficult to develop. It involves suspending one’s judgment and allowing the reality to emerge. Both teams consisted of members with different perspectives. They also continually developed their capacity to sense the emerging opportunities that presented themselves throughout their project.

8. Dialogue: Foster high-quality conversations as a key for integrating multiple viewpoints, experiences, and disciplines. A good place enables essential dialogues, which allow participants to share their views and experiences and to see themselves through one another. The quality of people’s conversations is one of the most important measures of the quality of place and the vitality and health of a system. (see sidebar Four Fields of Dialogue).

9. Self-transcendence: Become part of a whole larger than oneself. A good ba is like an instrument. The music doesn’t come from the instrument. The music comes through the instrument when someone plays it skillfully. Such is the ba. The ba is a vehicle that, if "played on" in the right way allows teams to connect to their highest creative possibilities and ‘music.’ Accordingly, ba allows groups to transcend the habitual patterns of the past and to move into the uncharted waters of new interpretations and identities.

10. Synthesizing: Connect different ba to an evolving larger whole. Knowledge creation needs many ba, which exist on multiple levels and are connected to each other organically to form a greater ba. An organization is an organic configuration of ba in motion, where various ba form a fractal. Hence, leaders have to facilitate the differentiation and interweaving among various ba and synthesize the knowledge that emerges from the larger ba. "My role was to get the right timing," said Enoki. When the project needed cooperation from other groups both inside and outside DoCoMo, Enoki and Tokuhiro coordinated and mediated the interaction.

Sidebar: The Idea Factory

At the San Francisco-based Idea Factory the economies of place have been turned into a whole company focused on innovation. In an interview, John Kao, the founder of The Idea Factory, explained how what the Idea Factory does relates to the three dimensions of ba: the physical, relational, and the deep tacit (spiritual) dimension. Says Kao: "We believe that physical place is really important, and we also believe that our physical place should be able to change its purpose at a moment’s notice, depending upon staging or perception or intention."

The Idea Factory uses several concepts from theater and design in order to inspire innovation. Says Kao: "In the theater, you quickly forget about the literal facts of the physical place if the experience is successful. You forget that you’re sitting in a chair that has purple cushions, that the theater has a certain number of rows and a certain kind of architecture, because there’s a suspension of disbelief that changes the mental landscape. You are drawn into action that is occurring on the stage among actors that you have a projective identification with and that leads to a flow state, where you lose track of physical time and space as you are drawn into a story."

Thus, the first level of ba, the physical level, facilitates the emergence of the second, the mental level. "That’s part of what we mean when we emphasize to companies that they need to figure out their story," says Kao. "There’s a big difference between what people physically do in a company and the kind of mental space they’re in, which relates to whether they’re feeling like they’re a part of the corporate story."

And finally, if successful, the second level of ba facilitates the emergence of a third, the spiritual essence of place. "If the story works," adds Kao, "you progress to the third level, which is yet another landscape. The great Zen philosophers and practitioners talk about how, at the moment of enlightenment, space and time have a different meaning and there’s a great mental clarity. That burst of insight or satori– which I think people are seeing increasingly not as one isolated event, but as a quality of experience sustained in one’s spiritual practice–has a different landscape again. It’s yet another shape imposed on the physical and the mental."

Sidebar Four Fields of Dialogue

One of the most critical leadership strategies for promoting a good ba is to enhance the quality of conversation. Without the capacity for dialogue, teams are unable to express their tacit, taken-for-granted assumptions about how reality works.

The model in Figure 4 outlines a process archetype that we have seen in numerous management and organizational settings and developed through many consulting, action research, and community-building experiences (Scharmer 2001). The model is based on four generic stages and field-logics of listening and "languaging" (Figure 4).

Within each of these four different field-logics, people relate to each other in a different mode:

• in field-logic I by talking nice and the primacy of politeness,
• in field-logic II by talking tough and the primacy of debate and advocacy,
• in field-logic III by using reflective dialogue and the primacy of inquiry, and
• in field-logic IV by using the magic of generative dialogue and profound flow.

Rule-repeating (talking nice), rule-revealing (talking tough), rule-intuiting (reflective dialogue), and rule-generating (generative dialogue) speech acts produce different kinds of conversations, each of which allows participants to access and communicate different types and layers of knowledge and knowing.

The leadership challenge is to help teams and institutions get "unstuck" in the first quadrant (talking nice), and increase their capacity to move up across all four quadrants and field-logics of conversational action. Both Enoki and Uchiyamada spent significant energy on creating conditions that allowed their teams to interact in field III and field IV interactions. What sort of interventions can help leaders move the field-logic up?

In shifting from field-logic I to field-logic II, the principal leverage is based on reconnecting what we think with what we say. The work of Argyris (1992) on accessing the "left-hand column" focuses on these kinds of interventions. An example of this kind of intervention would be to create a safe container that allows participants to articulate diverging views and to confront difficult issues. No learning or genuine knowledge creation will ever occur without moving the field-logic from the first to the second quadrant, for field I conversations only reproduce what is already known.

The principal leverage in field III is based on reconnecting what we think and say with what we do. Argyris and Schön (1996), Schein (1992, 1993), Isaacs (1993), and Srivastva and Cooperider (1990) address this issue and focus on "double loop learning" (Argyris and Schön), "taken for granted assumptions" (Schein), "containers of conversation" (Isaacs), and "appreciative inquiry" (Srivastva and Cooperider).

The principal leverage in moving from field III to field IV is based on reconnecting what we think, say, and do with what we see. Examples of this are deeply generative conversations throughout which not only new ideas are generated but people leave such a conversation totally re-energized and with a different and enhanced sense of themselves and their capacity to co-create.

To promote the knowledge-creating process, "seeing things as they are" or the phenomenological method of "letting the reality emerge" often plays a critical role. In organizational knowledge creation, the phenomenological approach is taken by a team of people. Hence, we refer to a "collective phenomenology," which is not just about one person’ s view of the world. The first step of a collective phenomenology is for participants in a ba to suspend their habits of judgment and try to see things from all angles and perspectives without being hindered by preconceived notions. The second step is for participants to reflect on the meaning of these perspectives for their own lives and put that meaning into stories and words. The third step is for participants to tune in to the deeper meaning or essence that wants to emerge through a particular situation and to use this deeper source or inspiration as the point of departure for co-creating the new (Arthur et al., 2001).

In the case of DoCoMo, each of the participants from different backgrounds sensed the emerging cellular phone business in a different way. Enoki sensed it as an evolution of the telecommunications business of DoCoMo. Natsuno, who used to work at an Internet venture, sensed that the new i-mode service was something very different from a Wallet PC. Instead of tiny personal computers, cellular phones would give users the Internet connection. Matsunaga sensed it as something fun. Although the concept that was originally proposed by consultants was for a service that offered proprietary content to elite business people, Matsunaga insisted that the service should be for ordinary people and hence should be easy to use. The content should be something " you can enjoy when you have a bit of time," not just " useful" content such as news and banking services.

Their common ba allowed them to reflect on their own views and share them with each other to achieve a self-transcending coherence. In this process the roles of the first person, the second person, and the third person are important. The first person plays the role of an innovator. S/he is the one who senses the new reality first. The second person plays the role of a coach. S/he attains intersubjectivity by interacting with the first person and brings in his/her own viewpoint. The third person plays the role of manager by seeing the first and the second person from a higher viewpoint. S/he attains transsubjectivity and makes the new reality understandable and touchable for other people. Another important role of the third person is cocooning– that is, protecting the team from outside influence so that the first and the second person can keep and develop their own viewpoints. In the case of DoCoMo, Natsuno and Matsunaga were the first and second persons, while Enoki played the role of the third person. Their process and co-sensing led them to finally co-create and enact a revolutionary new reality and business offering, the i-mode service (Figure 5). This process of co-sensing and co-creating took place between DoCoMo and its content providers as well.

This does not mean that only three persons are on the team. It means that there are three different roles to be played in the collective phenomenology. In the case of Toyota, for example, the gasoline engine team and the electric vehicle team played the role of the first and the second person, and Uchiyamada played the role of the third person to create the new reality of the hybrid system.

To facilitate the knowledge-creating process effectively, leaders need to read the situation to determine where the spiral of knowledge creation is heading and what kind of knowledge is available to be shared and utilized, both inside and outside the organization. Along with this reading and sensing, leaders need to improvise to incorporate necessary changes in the knowledge-creating process. Improvisation is an important factor in dynamic knowledge creation, especially when dealing with tacit knowledge (Weick, 1993). Leaders should be able to improvise and facilitate improvisation by the participants in the knowledge-creating process.

Leaders need to be able to create their own concepts and express them in their own words and thus should be able to use creative and symbolic language effectively. The language used can include tropes (such as metaphor, metonymy, and synecdoche), "grammar" and "context" for knowledge, and nonverbal visual language and artifacts such as design and prototype. Tropes such as metaphor, metonymy, and synecdoche are effective in creating concepts out of vast amounts of tacit knowledge. Therefore, leaders should carefully choose and use language according to the process of knowledge creation.

In the case of DoCoMo, Matsunaga used symbolic language effectively during the project. Her sensitivity toward language based on her experience as a magazine editor was a very valuable asset for the project. Matsunaga used "concierge" as a metaphor for the aspect of i-mode service that could help the user find something quickly. It had to be " concierge," not " secretary" or " agent," because the i-mode service was for ordinary customers. Likewise, leaders also need to use language effectively to foster dialogue about the essence of things.

How can an organization create high-quality knowledge and innovation continuously and efficiently? It is a question that many scholars and practitioners ask in this era of the knowledge economy. "Knowledge management" has become a fad phrase among those who seek the answer to this question. Tools are sold and seminars are held for those who believe that " knowledge management" can give them sustainable competitive advantage. Tools can be useful in managing information and supporting knowledge creation. However, tools do not create knowledge. Humans do. Since knowledge creation is a human process, we cannot really answer the question of how to create high-quality knowledge without understanding human factors.

It is such human factors that make knowledge both unique and so difficult to understand. Knowledge cannot be understood without context, and hence knowledge creation requires shared context. This paper discussed the concept of ba as a shared context in motion. To create high-quality knowledge, leaders have to build and maintain good ba, energize them, and connect them.

An organization can be viewed as an organic configuration of ba, where various ba are created and interweaving with each other. Viewing the organization as an organic configuration of ba in motion means that we have to view an organization as an entity that constantly recontextualizes itself. Management of knowledge and the knowledge-creating organization has to be reexamined from the perspective of how we can promote such a process to constantly create high-quality knowledge.

Further, ba is not limited to the frame of a single organization but often moves beyond the boundary of cooperating companies, or between companies and customers, as was done at Toyota and DoCoMo. Ba can be built as an interactive relationship with customers, a joint venture with a supplier, or an alliance with a competitor. Organizational members transcend the boundary by participating in ba, and further transcend the boundary of ba, as ba become connected to other ba (see Figure 6). In such a case, the legal boundary of a firm is not as important as how it synthesizes various ba, both inside and outside the organization. Some ba need to be built within the company because they will produce the knowledge that gives the firm a competitive advantage, as in the case of Toyota. Especially important for a company is a ba that gives the company the capability to synthesize. Knowledge creation is a dynamic human process that allows managers and workers to grow. Managers become leaders and increase their capability to synthesize by differentiating and interweaving the diverse types of ba. The innovation creating company emerges out of the organic configuration of ba.

Especially important for a company is a ba that gives the company the capability to synthesize. Knowledge creation is a dynamic human process that allows managers and workers to grow. Managers become leaders and increase their capability to synthesize various ba through their participation in ba. Fostering such leaders is the key to become an organization that can create knowledge effectively and continuously.