John Seely Brown and Estee Solomon Gray
The success of a volume like this can be measured by its power to compel us to browse through its pages (thank goodness for paper!), to take excursions into its texts (praised be prose!), to create or extend relationships with its authors (thank G-d for friends!), to sense the shape of the landscape by hovering over its table of contents (some structure is good!), and, finally, to settle back, lengthen our focal point, and take the time to reflect on critical questions (time, oh precious time!). How did we get here? What, if anything, is being said here that couldn’t have been said before? And why are we saying it now?
The contributors to this volume and the diverse participants at a Darden School of Business Administration Colloquium in spring 2002, which set this book in motion, pose these questions even more pointedly: How have we — the practitioners and stakeholders in the art of creating learning cultures — learned what we know? What do we need to learn next? Beyond articulating these essential questions, the contributors to this volume offer some answers.
It Takes 20 Years…
It was in 1990, with Peter Senge’s The Fifth Discipline: The Art and Practice of the Learning Organization, that learning was first catapulted from the peripheral corporate domains of training and development departments to a place much closer to the center of business discourse. E-mail was still a creature of early adopters and large institutions, and PowerPoint (or its aptly named predecessor, Persuasion) was just coming onto desktops and into conference rooms across the world. Because each technology purported to change the way people communicate rather than what they think, neither was considered particularly relevant to learning. In contrast, the five disciplines — personal mastery, mental models, shared vision, team learning, and systems thinking — appeared as tools to change the organization precisely by changing its thinking (and its thinking about thinking) and were easily recognized as valuable management tools for a knowledge-based, competitive era.
For those paying attention, the management conversation about learning had begun almost two decades earlier, when Chris Argyris and Donald Schön published Theory in Practice. They challenged organizations to recognize the limitations of “single-loop learning,” familiar from the quality movement, which fosters the ability to detect and correct errors within the frame of current assumptions and policies, and to aspire instead to “double-loop learning,” the ability to detect, determine, and perhaps even modify the organization’s underlying norms, policies, and objectives.1 The first type of learning implies assimilation, the domain of experience curves, which is relatively straightforward — both for people and for organizations. The second, considerably harder, implies accommodation — altering one’s frame of reference or basic assumptions about the world. Double-loop learning involves changing the kinds of stories we construct to make sense of the world and, using the terms of gestalt therapy, requires a fresh, unbiased hearing of the “other.” It is the ultimate goal of any learning culture. In corporations, double-loop learning is also the domain of strategic shifts. When Senge’s five disciplines showed up on management’s radar screens, they provided instant utility to the many organizations then engaged in strategic efforts to reframe existing markets and envision new business models. Yet Agryris’s Model II learning organizations remain rare to this day.
Meanwhile in the mid-1980s, from a more personal perspective, a community of researchers at and around Xerox PARC (Palo Alto Research Center) resolved to crack the learning problem by coming at it with multifocal conceptual lenses. One result was the founding in 1987 of the independent Institute for Research on Learning (IRL), a multidisciplinary community that undertook research to explore “everyday learning.” Merging the practices of diverse fields — cognitive science, computer science, social linguistics, educational technology, and ethnography — proved painful but instructive. By the early 1990s, IRL began to inject a new, more social constructivist voice into the business conversations cascading from the learning organization work.2 Amplified on one flank by workplace practitioners who worked with companies to enact new products, markets, and business models and on its other flank by educational practitioners who were elaborating new means to teach secondary school physics and mathematics, IRL put forth two fundamental understandings. First, that learning is fundamentally social and second, that learning about is quite different from learning to be, which is a process of enculturation.
Building on observations in workplace, school, and craft settings, IRL researchers noted that successful learning happens with and through other people and that what we choose to learn depends on who we are, who we want to become, what we care about, and which communities we wish to join. In this frame, learning is also a matter of changing identity, not just acquiring knowledge. Learning of this nature occurs primarily through the process of gaining membership in a community of practice and is critically enabled by what Jean Lave and Etienne Wenger described as “legitimate peripheral participation” — the essence of classical apprenticeship. By this measure, a marketing manager has learned enough about wireless networking to drive his or her company’s participation in that market when and only when she or he can understand the goings-on at an insider’s wireless conference or have a mutually satisfying conversation with a committed member of the wireless community. Practice is not merely the measure of learning but the medium of it. In communities that arise less through organizational fiat (the authorized infrastructure of work) and more through pursuit of common work by the ecology of crafts, disciplines, and personalities needed to accomplish that work (the emergent infrastructure of work) practice is invented — and learning captured — each step of the way.3 Members in such communities are co-constructing knowledge, which is literally embodied in their practice. Practice is not the stuff in libraries but knowing in action. Words, books, simulations, tool kits, and the like are artifacts deliberately crafted to transfer knowledge by evoking practice in the participant; they are not the knowledge itself.
In 1995, 20 years after Argyris and Schön, five years after The Fifth Discipline, and a year after the extended IRL community’s first corporate client retreat, a pair of former Harvard Business Review editors launched Fast Company, a “handbook of the business revolution” targeted at readers “old enough to make a difference and young enough to be different.” Readers were enjoined to “leap into the loop” by using e-mail to interact with the editors — a novel thought at the time — and to watch for a Web site yet to be constructed. By this time, PowerPoint was fully established as the first-language tool of business. Conference rooms were filled with people engaged in shoulder-to-shoulder knowledge sharing, literally returning to the ancients’ practice of reading and writing knowledge on the walls, although this time with beams of light instead of charcoal, chalk, or pigment.
Learning was so central to the new rules of business that an article by the two of us entitled “The People Are the Company” anchored the core Big Idea section of the magazine’s first issue. “Work Is Personal … Computing Is Social… Knowledge Is Power” blared the cover art. “Learning is about work, work is about learning, and both are social,” we wrote. In one of the most-cited articles in the publication’s history, we asserted that the Community of Practice is the “critical building block of a knowledge-based company,” the place where peers in the execution of real work create and carry the competences of the corporation. Veterans of numerous internal change initiatives, we quietly faced down the tanks of prevailing workplace ideology by proclaiming, “Processes don’t do work, people do.” We pointed out that “the real genius of organizations is the informal, impromptu, often inspired ways that real people solve real problems in ways that formal processes can’t anticipate. When you’re competing on knowledge, the name of the game is improvisation, not rote standardization.” We also took on the sister shibboleths behind the traditional corporate approach to learning and knowledge; namely, that learning means individual mastery and that everything knowable can be made explicit. We did so in the way we knew would work: by telling stories. We told stories about Xerox field reps using radios and an “electronic knowledge refinery” called Eureka, and about how National Semiconductor’s PLL (for “phase locked loop,” a specialized kind of circuit) designers coalesced almost instantly into a powerful, strategic, and ultimately much emulated presence in the company simply by being given the language, the license, and, eventually, the funding to organize. On one hand, these stories about the tacit and collective dimensions of learning and work eased quite naturally into readers’ experiences. On the other hand, partly by design, the words emerge and social seemed to jump off the pages into people’s faces — simple and familiar yet mysterious and somehow uncomfortable.
A Decade Distilled
Internet-time was upon us. The knowledge economy roared in, reshaping mainstream and management culture. It inflated. Burst. Rolled on. Contributors to this volume were deeply engaged in these formative years of the knowledge culture, as individuals and as professionals. As a result, things are being said within these pages that couldn’t have been said before. Here we can begin to comprehend the fruits of the first decade of the knowledge economy.
In light of our early work at Xerox and that nascent whiff of learning culture themes in our Fast Company article, Marcia Conner and Jim Clawson, the editors of this collection, asked us to introduce the burgeoning learning movement and to assist readers traversing through these essays. Here is what we glean from this volume and what we would urge readers to consider.
Reflecting upon the learning trajectory of the last decade captured so well in this volume, the days when learning usually meant training, knowledge meant information, and “content was king” seem to be fading. Community of Practice is now a common term in business language and a sanctioned, funded approach to global knowledge sharing and postmerger competence integration in leading companies. Learning is clearly no longer synonymous with individual mastery. It is now tacitly expressed in practice that not everything knowable can or should be made explicit, that content must be delivered in context to be effective. High-performance workscapes are built less through training and more through creating opportunities for collaboration and continual renewal, usually through teams, communities, networks, or forums. The words “social” and “emergent” no longer crimp business conversations about learning cultures but spark them.
Creating Learning Cultures: What’s Next?
So, what do practitioners and stakeholders in the art and practice of creating learning cultures need to learn next? Not surprisingly, our response begins with a critique of current practice — individual and collective. For all we have learned and for all that learning cultures have ostensibly changed, there is surely more learning and changing ahead of us.
Very often, sharing knowledge across an enterprise requires leaving the rails of a shared practice and jumping between two different practices (marketing/sales and research, or materials science and production engineering, for example) or organizational cultures. In these cases, we must literally find ways to bridge different practices. Bridging practices is never easy, even (or especially!) when accompanied by process-imposing tools like Lotus Notes or enterprise systems like those from SAP, PeopleSoft or Oracle. Bridging requires nuanced knowledge brokers, people who can span practices and speak multiple languages at the same time. It requires intentional boundary objects — documents, prototypes, phase gates of a process, and the like, around which a negotiation-in-practice can be afforded. It is in reflection upon this negotiation that the second loop of learning occurs — the ability to accommodate, to change underlying models, methods, and our own view of others. Yet few strategies or technologies honor the role of practice — of action on the ground and meaning negotiated in the crucible of work, among people. And too many focus, instead, on the warm friendly notion of communities.
The common corporate goal of sharing best practices is related to but distinct from the challenge of having actionable knowledge jump across distinct communities of practice. In this case, it is crucial to realize that every best practice emerged in a highly situated way; it was grown and honed in a particular context. In order for it to travel, it must first be disassembled from that context and then re-embedded in a new context (that is, in a different part of an organization or in a different organization entirely). The process of re-embedding is highly problematic since the best practice must be viewed as a seed that is allowed to germinate in its new context and sprout in a form that honors the nuances of this new context. It takes time and a willingness to let the people influenced by this new best practice do their part to shape it and grow it, preserving its essence but also modifying it to fit its new circumstances.
Practice does not come in discrete pieces like Legos but in clumps and clusters of yarn like a knitter’s remnant box after a three-year-old child has played in it. To move a strand from one community to another, from one type of product to another, from one country to another means to disentangle, snip, and re-entangle — without consuming the yarn.
Let’s start with a zero-digital-technology example of such a practice that builds directly on knowledge sharing and innovation. Say you want to transfer a new, hard-earned strategic shift from business unit A, where it was hammered out over 18 months, to business unit B, which faces a similar set of strategic issues and, furthermore, sits directly up- or downstream from A. Time is of the essence. There is very little shared practice between A and B, although there is significant hand-off and therefore some history of communication. Bridging A and B, we know, will take nuanced brokering, mediating boundary objects, and time-time to negotiate meaning in practice and time to dis-embed and re-embed key innovations.
The technique is called 2 x 2 x 2 x 2 x 2: Take two people from group A and two from group B, and bring them together for two meetings, each two hours long, two days apart. Ideally, there is a preexisting positive professional relationship between one of the As and the one of Bs. Perhaps they are both current or formers members of a particular engineering network of practice; perhaps they both served on a corporate change-initiative task force related, even tangentially, to the strategic issues on the table; perhaps they have functioned as customer and supplier to one another within the organization’s value chain. Equally important is the relationship between the two members of each unit. Within their dyad, they must be able to reflect on and articulate elements of the practice they share; they must be able to share stories, hash out details, follow each other’s leads, and refine each other’s thoughts. What happens around the table the first day (and it really should be a physical table if possible) is intense. It takes tacit teaming by each side to establish and maintain the conversation — one talking while the other watches body language or searches for the next example. During the two hours, A1 and A2 help B1 and B2 enter into the new way of thinking and doing by describing, showing illustrative artifacts, answering questions, identifying, and if possible addressing objections, and working with B1 and B2 to map the new way into at least two specific situations or practices under way in B. Each of these situations is explored in depth, often primarily in dialogue between the two Bs with by now only intermittent interjection by an A. These situations then become the subject of continued exploration and experimentation in practice by the two Bs over the next two days. Success rests on the fact that with two representatives (the smallest possible representative of a community that is still a community), each side can bring its practice into the room. The second meeting brings all four people back to reflect and continue verbally negotiating meaning. Reframing occurs continuously. Repeat the last two steps as necessary. Unlearning alternates with learning throughout as the threes sets of dyads (A1A2, B1B2; A1B1, A2B2; A1B2, A2B1) argue, test, witness, internalize, challenge, and change.
Almost every important new point of view or piece of technology, we argue, imposes a burden of unlearning on would-be adopters, often swamping or preventing the better known learning demands it makes. No more dramatic examples exist today than “naturalized” Internet citizens literally looking at internet-native genres like MMPOG (massive multiplayer online games). A fundamental act of reframing — learning to swap the periphery for the center — is necessary, we’ve learned, before one can begin to see the game. This is not an easy shift, unless you have a good guide plus an inclination to see.
In John’s case, he realized early on how difficult it was to understand the culture being created by kids who grew up digital. Fortuitously, he met young author J.C. Herz,5 who offered to be John’s “reverse mentor.” Over a year’s time, J.C. structured a set of experiences that would give John a way in to the practice of this emerging digital culture, help him unlearn certain biases, and slowly construct a new set of conceptual lenses through which he could see, hear, and make sense of the massively multiplayer game world. For John, being reverse mentored also presented an opportunity to hone his ability to listen with humility and through engagement. What unfolded over the year was a slow realization that until then, John, like most adult game novices, had focused on the actual playing of the game — at the center of the game screen, if you will — while remaining moderately oblivious to the rich social activities transpiring around the edge of the game. There, at the edge, a rich constructivist ecology was evolving — the sharing of tricks and heuristics, the bartering of magical swords, avatars, and other objects of play , the general swapping of stories, and more. Suddenly, he realized that what he thought of as the center was in fact the periphery and that what he initially considered to be periphery (or context) was in fact the center (or content) of the game. The real game, he saw, is deeply social. The real action, he understood, lies in the new kind of nonlinear, multiauthored narrative being constructed collectively by the players.
In Estee’s case, the guides are her 15- and 11-year-old Internet native sons. For them, summer vacation begins when — and only when — they are allowed to devote entire days in succession to their favorite MMPOG, which this year happens to be the Korean Ragnarok Online. Being Mom, Estee worries about eyestrain, their relative lack of fresh air, sunshine, and exercise, and their willingness to forgo physically apprenticing with their father as he constructs an addition to our house. But, armed with a deeply internalized appreciation for the social and situated aspects of learning and prodded periodically by John to follow their experience closely, she does not worry about wasted time, social isolation, or (lack of) future memories of joyful togetherness. The boys prefer to play on adjacent computers in what they call the “downstairs computer room,” where they are in constant verbal connection with each other. Occasionally, they or a friend are forced to use a third machine upstairs, which means they tie up two phone lines in order to keep up the conversation. Add to this the roughly 200 people with whom each interacts on a good Ragnarok day, in passing, as a close fellow traveler in their current party, as member of their latest guild, as famous personality players, and as buyer or seller of various items. Their ability to multitask is, well, awesome. To them, systems thinking seems natural. Later in the summer, letters from the younger to the older at overnight camp principally feature updates on what John has called a “new kind of nonlinear, multiauthored narrative.” As John learned to see, the narrative is not about kills or game places visited or instances of deploying weapons, spells, or other skills — none of the foreground flora and fauna that capture the adult’s eye when faced with the game. Rather, it’s about how the game is evolving, what particular players are up to, the latest tidbit from one of the three or four user sites they graze, how the strategies they’ve been exploring are working out, what stupid or cool thing Gravity (the company that makes the game and runs the main servers) has done lately. “You know what I learned today, Mom?” starts the daily report. And as the 11-year-old talks, all the cyber-age shifts we talk about are manifest. He freely discovers, links, lurks, tries, asks, borrows, and navigates a complex n-dimensional space while his mom internally fights her need to know-before-acting and wishes for a place to start deducing what to do next. (She’s wondering, is there a document, a set of base rules, something?) His digital world is social and constructivist from the get-go. Moreover, he is constantly shifting center and periphery — at will.
Enormous stocks of ink, budget, attention, engineering, and marketing elbow grease have certainly been devoted since then to technologies supporting communities, collaboration, and knowledge management. Few of these have engaged or energized their intended users beyond an early (often enforced) usage spike. For a time, unabashedly transaction-oriented marketplaces and exchanges hijacked both the noun community and the adjective collaborative. Knowledge management is often a synonym for taxonomy-driven content management. So-called collaboration systems are still primarily means for posting, retrieving, and, to a more limited degree, co-producing semistructured content. Even the live-events segment of the collaboration market was sold and purchased largely as a means to broadcast human-delivered presentations or lessons, until demand to replace face-to-face meetings with zero-travel e-meetings skyrocketed after September 11, 2001. New software that honors and activates the emergent has been barely visible.
In the last few months, the term “social software” has arced from the province of bloggers and tech early-adopter conferences to the pages of the Wall Street Journal and the New York Times. In most of those venues, the focus is on weblog creation tools including blogs — an instant personal publishing technology and practice that has enabled hundreds of thousands of people to find their individual voices over the last two years — and wikis, a group voice technology and practice, following on the heels of blogs, but entailing somewhat more structure and shared page ownership.6 Social software also encompasses instant messaging and other emerging forms of presence awareness technology, and hints of tools (still largely academic or researchy in flavor) for tracing, analyzing, and navigating social networks. Some observers include a gaggle of social networking services that interconnect registered individuals (and thereby, theoretically, their social networks) for numerous professional and personal purposes, depending on the service. Whether you believe recent groupware products such as peer-to-peer Groove or contextual collaboration offerings from IBM Lotus also fit the term is left for you, the reader, to decide. Undeniably, most of the Internet-native entries are better classified creatures of the emergent than the authorized. Moreover, the broader software or Web-services market of which they are a fresh part is showing signs of avoiding the tunnel vision that has traditionally excluded social sensibilities from the activities of information technology developers and purchasers. Social software developers and early adopters aspire to a new approach to building adaptive social applications that are easily deployed and can be humanized — not just customized — to support different types of online interaction and different modes of communication. They anticipate a new set of online genres reflecting a tremendous shift in human relationships: from episodic to always-on.7 Many proudly point out the relative simplicity of blog and wiki technology. But the practice around their use is anything but.
Defined most clinically, social software is designed to be used by three or more people. It is much rarer than it sounds. Most interaction supported by technology is narrowcast (one-to-one), such as telephones and simple e-mail, midcast (one-to-small groups), such as e-mail using distribution lists and small ezines, or broadcast (one-to-many), as in standard publishing and large-scale ezines. Clay Shirky of New York University points out, “Prior to the Internet, the last technology that had any real effect on the way people sat down and talked together was the table. Beyond that, there was no technological mediation for group conversations. The closest we got was the conference call, which never really worked right...” We interject that a later midcasting technology, the copier, radically affected how people interacted around that table by giving each a copy of shared and sharable documents but agree with Shirky when he continues: “We’ve had social software for 40 years at most, dated from the Plato Bulletin Board System, and we’ve only had 10 years or so of widespread availability, so we’re just finding out what works.”8
Designing social software that works is important for creating a culture of learning — inside and outside the corporation. It can, and already does in a few places, complement traditional IT systems designed to support the formal business processes and content stores of an organization but ignore the social fabric where learning and knowledge sharing happen. But social computing is hard, since we must now understand the emergent properties of groups of people, down to their social- and psychodynamics — both inside the corporation and in society at large. We must learn to distinguish the natural size and activity classes of various groups, communities, networks, and collections, and handle each appropriately. We learn from repeated online experience that by its very self-organizing nature, a community can quickly degenerate into the tyranny of the masses or be hijacked by weirdos, spammers, and the like. Designing social software is much more like designing a constitution than designing an operating system. The constitution needs to exhibit the right balance between supporting dissenting opinions and guaranteeing that the community’s real work can get done. It must vary with each community; indeed, it must emerge and evolve along with the community. Borrowing Shirky’s language again: “Groups are a run-time effect. You cannot specify in advance what the group will do, and so you can’t substantiate in software everything you expect to have happen.”
Designing and using social software is therefore like designing and living in ecology; moderators must honor diversity and husband the cross-pollination of opinions and ideas to keep the emergent ever present. There are also business challenges inherent in life on the emergent side of the enterprise in a cost-sensitive era. It can be difficult to garner revenue up front for things that don’t yet exist or provide measurable outcome guarantees. If these design goals and business challenges can be met, social software can act as a true enhancer of our ability to learn from and with each other. We may yet tap the logic of knowledge work and the spirit of community.
Conclusion: A Twenty-First Century Intuition
It appears in big blue letters encased in a cloud-like form floating toward the top of the Darden Colloquium mural: The 21st century mind is a collective mind. We understand why. Learning is the strategic competence for meeting the economic, cultural, and cognitive implications of increased speed and globalization. A new, more social model of the human at work is emerging as biological metaphors, ecological models, and adaptive system approaches predominate. On a daily basis, twenty-first century first-world citizens engage in coproduction — as consumers, as coworkers, at play and in political life. E-mail is the lifeblood of business. Files that end in .PDF and .PPT are the universal currency of knowledge exchange. While corporate learning practitioners are still not taking advantage of the rails of practice, non-Internet natives are still fundamentally confusing the center and the periphery when looking at genres like MMPOG, and industry has yet to deploy software that honors and energizes the emergent, alongside the authorized, as knowledge workers approach their keyboards with expectations beyond the twentieth-century information highway.
The Cartesian worldview of “I think, therefore I am” seems to be finally giving way. A next step, “We participate, therefore we are,” better captures today’s ethos, we think.
That next step is strongly in line with the African proverb, “It takes a village to raise a child.” It takes a community to change a practice. If double-loop learning were a matter of intrapersonal, interpersonal, or even simple intracommunal learning, we would have seen more of it in the last 20 years. But our experience, our theory, and our intuition suggest this goal of all learning cultures, and most certainly twenty-first-century ones, is best achieved as an intercommunal dynamic. That is, it may take one working community pushing another in order to reconsider and recast working knowledge. Each center is the other’s periphery. What ensues is a creative collision of craft, which — if it can take place in a fabric of trust, with appropriate brokering and cultivation practices — can recreate worlds.
The word “intuition” is purposefully chosen here. However the twenty-first century plays out, none of us today knows how to create a twenty-first-century learning culture. In fact, most of us in charge today of the budget and resources for building tomorrow’s learning cultures KNOW we don’t know how to build them for those coming up behind us. It takes courage to breach the barriers of current practice and head knowingly into the unknown. And it takes intuition to navigate there.
Let’s distinguish for the moment between two kinds of intuition. One is the kind of personal intuition that arises from one’s own experiences. The other arises from being embedded in a collective. It incorporates learning in the moment, listening with humility, and being able tap tacitly held beliefs and sensibilities. It is about being able to discern a kind of group resonance. Mystical overtones notwithstanding, some leaders and strategists in the quotidian world already exhibit this ability to make sense at the collective level, but even here it is rarely articulated. As we move forward with the insights in this volume, both types of intuition are necessary.9
“I think, therefore I am” has paled. “We participate, therefore we are” is where we’re heading. Here’s to the next 20 years.
1See C. Argyris and D. Schön, Theory in Practice: Increasing Professional Effectiveness (San Francisco: Jossey-Bass, 1974). For additional information on Argyris, see http://www.infed.org/thinkers/argyris.htm.
6As a point of interest, Eileen Clegg (who created the illuminating images that appear between the sections of this volume) also created visual maps of our dialogs at the Darden Colloquium. The term blogs appeared on these murals; wikis did not.
7Lee Bryant and Livio Hughes, London http://www.headshift.com/moments/archive/sss2.html#_Toc38514168
Shirky in a speech at ETech April, 2003 entitled “A Group Is
Its Own Worst Enemy,” published July 1, 2003 on the Networks,
Economics, and Culture mailing list.
We are indebted to Teddy Zmrhal for their help on this chapter and more generally to Paul Duguid for his continual contributions to our understanding of social practice.