Communities of Practice and Management Hierarchy: Can it work?

In this blog post from 2004, I muse on the interaction between the network forms of Communities of Practice and managerial hierarchies (cascades of Real Bosses, not simply organizational charts). I wonder if now, years later, this is still valid. Let me know what you think.

How do Communities of Practice (CoP) interact with the self-organizing principles that Elliott Jaques and his fellow researchers discovered about Stratified Systems Theory / Requisite Organization? The Knowledge Construction Glossary [http://www.cs.colorado.edu/~ostwald/glossaries/kc-glossary.htm, no longer alive] defines CoPs as:

A group of practioners involved in a common activity, albeit performing different roles. Essential characteristics of communities of practice are: 1) they are not defined by organizational mandate (e.g., the “org chart”), but rather by the ways people actually work together, 2) they involve many different “roles”, as opposed to a flat structure, and 3) they experience a ongoing flux of community members, who enter the community from the periphery and gain status as knowledgeable members through participation in the community of practice.

Functional silos in companies were created to take advantage of this natural organizing principle. Unfortunately, management attention can kill communities of practice. They can be fostered but not managed. You can lead a horse to the bar, but still you can’t make him drink a screwdriver. Today’s team-based organizations have made it more difficult for those who share a common activity to learn from each other. I suspect that this is especially true of “staff” functions such as OD, training, communications and quality control. As these workers get attached to particular teams, they lose track of their fellows.

You can definitely see how this would have some of the emergent qualities described by Jaques. If the community of practice (say a guild of sorts) were allowed to develop its own leaders, they might be seeing higher stratum persons rise to leadership. Of course, that would involve overcoming the qualities of money and inherited power. But you can see how our siloed approach to management accountability hierarchies (MAHs) — to use Elliott Jaques’s term — would fit the inherent needs of the communities of practice. Where organized requisitely, the CoP would provide a mix of best of both worlds.

Richard McDermott, an oft-cited Communities of Practice expert, has an interesting article about the problem of team-based companies (“Learning Across Teams: The Role of Communities of Practice in Team Organizations” [PDF]) that describes how to mix the two in a “double-knit organization”. He notes that CoPs develop organically, even developing the “hot topic” organically through discussions. Topics of conversation can change over time. Lurkers, those who listen in but do not contribute are an important part of the community fabric, as they may become contributors later on.

One of the most irritating things that can happen within a community of practice is having an inexperienced person (or simply an incompetent) contribute rather than lurk. As the proverb says, ”

Even a fool who keeps silent is considered wise;

better to keep your mouth closed than to open it and remove all doubt.

Having a fool on your community can kill it because he or she will drive away the wise. Someone has to protect the community. Harder to do in today’s inclusive environment, but you’ve got to make the person shut up. Like all communities, communities of practice are incredibly fragile, susceptible to any sign of distrust of movement in the wrong direction. Given the history of the world is littered with war and famine and plague, it’s amazing that communities of practice ever existed. No wonder the medieval guilds were so protective, so secretive, so demanding. They had to be to survive over time. America’s relatively sedate existence (the US Civil War is the most recent on our soil) may have allowed for more CoPs to exist, which gave rise to the high amount of associativeness in the US. Of course, Alexis de Tocqueville found this in America as early as the 1830s:

Americans of all ages, all stations of life, and all types of disposition are forever forming associations…In democratic countries knowledge of how to combine is the mother of all other forms of knowledge; on its progress depends that of all the others. [from Democracy in America]

That’s almost 200 years after my ancestors got here, of course. Perhaps the “knowledge of how to combine” arose from 200 years of relative peace. Indian wars and the French-Indian War aside, America lived in a great peace compared to its “fellow Europeans” on the continent and isles.

Hundreds of people a year visit PARC and ask, “How do you manage creativity?” The answer is simple: you can’t. You can manage innovation. But you cannot manage creativity, only foster milieus that promote it. You let the world do more of the work of creativity for you. That means, for one thing, giving people the freedom to fail and then reflect, because we learn far more from failures than from successes. [Brown, John Seely (1999). “Sustaining the Ecology of Knowledge” [PDF]. Leader to Leader, 12 (Spring 1999): 31-36.]

Brown also talks about the affinity between artists and scientists, who both tend to be inwardly focused, and between designers and engineers, who both tend towards outward focus. We often think that artists and designers — because they are both artistic — would gravitate together but it isn’t true, at least not at Xerox PARC.

“The challenge,” writes Brown, “is to devise dynamic structures that bridge these worlds. That is a leader’s own creative contribution.”

Peter Block talks about this divide between inward and outward focus in his discussions on the difference between the Economist/Engineer and the Artist. He argues for a new role, one of social architect. I’ve talked about this before, so I won’t go back into it.

Communities of practice are a powerful social phenomenon. We all long for a place of like-minded people, a place where we can explore that which we hold important. Work, so central to our lives, holds a powerful place for in our lives, but not the only place. “Communities of practice” are simply communities, groups of people tied not by a hierarchical command but interest and affinity.

Of course, they don’t really get away from a hierarchy. It simply doesn’t exist on the paper descriptions. These groups still create informal hierarchies where deference to some occurs.

Interestingly, Communities of Practice seem to be an informal forms much like the more formal medieval guild. In order to contribute, you must first have a time of apprenticeship. Breaking the apprenticeship rules (by the fools mentioned above) creates dissonance. Masters should have the right to discipline or expel unruly apprentices. Write John Seely Brown and Estee Solomon Gray, “The practice and knowledge is embedded in the community that created it. The only way to learn the practice is to become a member. The best way to access the knowledge is to interact with the community.” (“The People Are the Company“, Fast Company, 1(11/1995):78)

Fortune had an article a few years ago (“The Invisible Key to Success“, 1996 Aug 5) that better described the joy of belonging to good communities of practice:

Watch a bunch of scientists at a convention: They swap secrets like street vendors opening their jackets to flash contraband Rolexes.

Whether it’s woodworking, wordworking or changing the world through the clever application of stratified systems theory (SST), finding a community of like-minded folks goes straight to the soul.

Brown and Gray (in “The People Are the Company”) say “[Communities of Practice] seldom grow beyond 50 members — that’s about as big as they can get before they lose the intense collaborations needed to build shared commitment.” That number, or close to it, has been bandied about in the online discussions of the Dunbar Number. Dunbar proposed that the ratio of neocortex size to social group size in primates meant that the upper bound of average social groups in humans would be 150.

Because the time social grooming and gossiping takes, it is likely that the effective limit in diverse civilizations would be lower, although Dunbar posits that the development of language made social grooming more effective, thereby decreasing the amount of time that had to be spent to keep a cohesive social group. Of course, modern humans live in civilisations where they live in several communities at once.

For example, I have a work community, a professional community, my residence community and my religious community. Each takes some time to maintain, thereby reducing the total number of people that I can effectively maintain a social bond with within each group.

The real interesting thing about Communities of Practice, as an idea, is how they seemed so revolutionary even though they simply talked about what the ancients already knew.

The problem? “[T]here may situations where the community of practice is weak or exhibits power relationships that seriously inhibit entry and participation”, Mark Smith (1999) points out, and the ancients would agree. The medieval guilds hoarded knowledge that would have benefited from being free. Even today, many of the techniques of the guilds are lost. The methods that artists used to capture in paintings those elusive moments (such as Johannes Vermeer’s “Girl With Pearl Earring”, where he captures the moment of the girl glancing over her shoulder) are now heavily debated. The artistic guilds, jealous of their precious knowledge, did not write anything down. As a result, when the advent of photography made realism available to all, the technique for painting realistically disappeared. Or you can think of the mystic knowledge purported to many groups.

Communities of Practice have gone under because they prevented ethnic minorities from entering their ranks or even protected immoral or illegal practices by their members. Unions today can easily fall into the same traps. A powerful CoP works wonderfully for its members but can also create dictatorial powers for its leaders.

Perhaps this ability to form CoPs based solely on skill and practice — this “associativeness” of Tocqueville — is what has helped make America so resilient. Unlike many other nations, America espoused the belief that it was who you showed yourself to be, rather than who you were born as, that was important. It made the places where we rejected that idea for racist policies all the more painful. But our ability to associate freely with each other, across many boundaries that would have prevented men of other nations, allowed more robust communities of practice to form and thrive. I’d bet that there’s something in the ideas of social capital (see references below) and CoPs that I’m only seeing obliquely.

---

Other reading:

• CONSORTIUM LEARNING FORUM (2001) “Building and Sustaining Communities of Practice: Continuing Success in Knowledge Management“. APQC International Benchmarking Clearinghouse. [link to executive summary of report] The big problem with the APQC report is that they try to make CoPs something manageable and accountable. CoPs aren’t accountable, even though they can produce great learning. It makes the CoP into a team, and the ones who end up leading are not the most competent but the most aggressive. Almost worthless report.
• Andre, John P. (2004) “Communities of Practice – A Primer“. Decent primer on the concept. Links to other relevant articles.
• Andrews, Peter (2002) “Executive Tek Report: Social network analysis: Tracing relationships” [PDF]. IBM Advanced Business Institute.
• Cross, Rob; Laurence Prusak; and Andrew Parker (2002) “Where Work Happens: The Care and Feeding of Informal Networks in Organizations“. IBM Institute for Knowledge Based Organizations. “This paper describes our work using social network analysis (SNA), a rich set of analytic tools developed by social scientists, to help executives systematically assess the structure of a network as well as the roles of the individuals in the network. An accurate diagnosis of a social network is a critical first step in targeting interventions to foster connectivity in areas where collaboration is critical.
• Cross, Rob; Laurence Prusak, and Andrew Parker (2002b) ““. IBM Institute for Business Value.
• Cross, Rob; Nitin Nohria; and Andrew Parker (2001). “Managing in a Networked World: Assessing and Supporting Collaboration in Social Networks“. IBM Institute for Knowledge Management.
• GuerillaKM.org (2004) GuerillaKM.org’s list of resources on Communities of Practice. A decent list of available web resources.
• Lesser, Eric and Kathryn Everest (2001) “Communities of practice: Making the most of intellectual capital“. IBM Global Services.
• Lesser, Eric and Larry Prusak (1999) “Communities of Practice, Social Capital and Organizational Knowledge” [PDF]. IBM Institute for Knowledge Management. (Accepted for publication in the Information Systems Review).
• Levin, Daniel Z.; Rob Cross; Lisa C. Abrams; and Eric L. Lesser (2002) “Trust and knowledge sharing: A critical combination“. IBM Institute for Knowledge-Based Organizations.
• Millen, David R. and Michael A. Fontaine (2002) “Understanding the Individual, Community and Organizational Benefits of Work-Based Communities”. A Collaborative User Experience Technical Report. IBM Institute for Knowledge-Based Organizations.
• Millen, David R.; Michael A. Fontaine; and Michael J. Muller (2002) “Understanding the Benefits and Costs of Communities of Practice“. Communications of the ACM, 45(4).
• Muller, Michael J. (2002) “Patterns of Participation in Two Communities of Practice: Community of Engagement vs. Community of Reference“. A Collaborative User Experience Technical Report. IBM Institute for Knowledge-Based Organizations.
• Muller, Michael J. and David R. Millen (2000?) “Social Construction of Knowledge and Authority in Business Communities and Organizations“. Lotus Development Corporation / IBM Research Cambridge.
• Muller, Michael J. and David R. Millen (2002?) “Workshop: Creating and Refining Knowledges, Identities, and Understandings in On-Line Communities“. An interesting proposal from some IBM-ers for a workshop that includes a decent bibliography.
• Muller, Michael J. and Kenneth Carey (2002) “Design as a Minority Discipline in a Software Company: Toward Requirements for a Community of Practice“. A Collaborative User Experience Technical Report. IBM Institute for Knowledge-Based Organizations.
• Thomas, J. C.; W. A. Kellogg; and T. Erickson (2001) “The knowledge management puzzle: Human and social factors in knowledge management“. IBM Systems Journal. 40(1) [DOI: 10.1147/sj.404.0863]
• Sharp, John (1997) “Key Hypotheses in Supporting Communities of Practice“. (no longer available)
• Snyder, William M. (1997) “Communities of Practice: Combining Organizational Learning and Strategy Insights to Create a Bridge to the 21st Century“. Organization Development & Change. (28-page research article with lengthy bibliography)
• Smith, M. K. (2003) ‘Communities of practice‘, the encyclopedia of informal education, www.infed.org/biblio/communities_of_pratice.htm. Last updated: 14 February 2004. Outlines the theory and practice of CoPs, and examines some of issues and questions for informal educators
• Smith, M. K. (1999) ‘The social/situational orientation to learning‘, the encyclopedia of informal education, www.infed.org/biblio/learning-social.htm, Last update: February 14, 2004. Goes with Smith’s article on CoPs.
• Other public technical reports published by IBM Watson Research in Cambridge

[Note to self: D. Schoen. The Reflective Practitioner: How Professionals Think in Action. Basic Books, New York, 1983]

Image Credit: “Employees at Mid-Continent Refinery” [ca. 1943 Tulsa, OK]. FSA photography by John Vachon. Via Library of Congress Collection. Public domain.