All Life Is Problem Solving

Lascaux Horse

One of the most popular techniques identified with Knowledge Management is storytelling. Led by Steve Denning, Dave Snowden, Katalina Groh, Larry Prusak, John Seely Brown, and Seth Weaver Kahan, storytelling has become a vibrant movement within KM with a life of its own. Two new books, due out in June 2004, by Denning, and Brown, Denning, Groh, and Prusak, promise to fuel the fire of storytelling and spread it well beyond the disciplinary confines of KM into the general field of Organizational Management.

Storytelling was introduced into Knowledge Management by Steve Denning (See The Springboard: How Storytelling Ignites Action in Knowledge Era Organizations, Woburn, MA: KMCI Press/Butterworth-Heinemann, 2001) as part of a Knowledge Sharing initiative at the World Bank. It was used to present “Springboard” stories to communicate an action-igniting vision of what Knowledge Sharing might mean to the World Bank. But its purpose was not simply to share the idea that Knowledge Sharing might be valuable to the Bank and its clients. It was also to induce listeners to “co-create” their own vision of Knowledge Sharing and its benefits to the bank. The stories used are called springboard stories because they “catalyze” a process in listeners which (a) leads to a new level of understanding about Knowledge Sharing and its significance and (b) leads to action based on this understanding.

Springboard stories are not the only ones that are useful in organizations. In Squirrel, Inc. one of the two new books I referenced earlier, Denning covers stories whose primary function is (1) communicating about who we are, (2) getting people to work together, (3) expressing and transmitting core values, (4) springboarding, (5) taming the organizational grapevine, (6) sharing knowledge, and (7) leading people into the future. Many of these functions of stories transcend knowledge processing, showing that storytelling has applications broader than knowledge processing and KM.

In this blog and a future one, I’ll focus in on the relationship between storytelling and problem solving (or knowledge production) in organizations. In future blogs, I’ll write about storytelling and knowledge integration, and storytelling and knowledge management.

Popper, Storytelling, and Problem Solving

In his later writings, Karl Popper had a good bit to say about stories, storytelling, and problem solving. He viewed stories in an evolutionary framework and in the context of the appearance of language in animals and humans. In animals, primitive languages have two functions: expressive (of inner physiological states) and communicative (signaling). But with the development of human language, an informative/descriptive/explanatory function of language co-evolved, along with human culture and cultural products.

Up to this point Popper’s account is based on a theory of his teacher, Karl Bühler. To this theory of the functions of human language, Popper added (p. 84-91) the critical/argumentative function, which, he believed, was the key to creating objective knowledge. Other functions of human language such as the persuasive, advisory, hortatory, and others also exist. But the first four are the most essential ones and stand in a hierarchical relationship to one another, such that the higher functions cannot occur without the lower ones. Thus, argumentative expression presupposes, description, communication, and emotional/physiological expression. Descriptive/informative/explanatory expression presupposes communication, and emotional/physiological expression, and so on.

The evolution of descriptive human language, according to Popper (1994, p. 89), was accompanied by a number of important biological effects, including:

• “A fuller awareness of time and . . . a more flexible conscious anticipation of future events“;
  
• “The formulation of questions” and “the beginning of objectivization of problems“;
• “The development of imagination” . . . used in myth-making and storytelling“;
• “The development of inventiveness”; and
• The “entrenchment” of newly invented tools, behavior, and social institutions in culture.

Inventiveness (pp.89-90):

“increased tremendously with the invention of storytelling. Its role in the rise of the higher civilizations cannot be exaggerated.
Storytelling is found, as far as we know, in all human communities, however low they may be in their cultural development. Sticks are not found in all human communities, but storytelling is. So I would say that the invention of tools, and the richness of the different tools that men can invent, is connected with storytelling. . .”

Popper connected storytelling with the critical/argumentative function as well as with the descriptive function. He said (p. 90):

“Almost all conversation and even most stories are largely argumentative and critical. Myths are invented as explanatory theories and are, like all explanations, partly argumentative, although often in a primitive way. It is also obvious that the descriptive function cannot fully develop without the critical function: only with the argumentative and critical function can negation and similar things develop, and these, of course, greatly enrich the descriptive and informative function.”

But elsewhere, he contrasted stories and criticism and pointed out that human language (p. 452):

“. . . gives rise to the need to criticize because of storytelling. With the invention of language there also comes the invention of excuses, of false excuses, and of false explanations produced in order to cover up something not quite right that one has done, and so on; and with this arises the need to distinguish between truth and falsity. Thus, with storytelling there arises the need to distinguish between truth and falsity, and this, I think, is how criticism actually arose originally in the development of language . . .”

“. . . What characterizes a descriptive statement is that it can be true or false, and therefore also that it can be used for different purposes: for the purpose of telling the truth – that is to say, for conveying information – or for the purpose of lying; for example, for making certain excuses acceptable, or for covering up failure, and so on. I think storytelling emerges from these descriptive reports, from the telling of lies, or from both. Both descriptive reports and lies fulfill a kind of explanatory function. . . .”

In light of Popper’s work, the connection between language, storytelling, and criticism, on the one hand and problem solving on the other is made by realizing that storytelling, along with other descriptive/explanatory language formulations, addresses real, visualized, or envisioned problems. Stories are one method of expressing knowledge claims whose intent is to describe/inform/or explain a pattern of related events or occurrences. Stories are conjectural in nature. They assume theoretical propositions about cause and effect and about change, even generalizations, in many cases, but they are about specific events or occurrences, and in that way they are different from general theories.

Stories, like other conjectural formulations, fit into Popper’s problem solving schema, and the Knowledge Life Cycle version of it I have described for organizations. Stories, represent tentative solutions to epistemic problems because they explain why things have happened or because they offer predictions about what will happen, or because they prescribe, or because they close an epistemic gap by informing us about a solution. It may help us to get a grip on the meaning and implications of our knowledge claims when we wrap them in stories. And stories may suggest entirely new knowledge claims to us by stimulating us to combine ideas in fresh ways.

But stories, like general theories, and other knowledge claim networks, may be false, or if they prescribe values, may be illegitimate. They may fail to explain, describe or inform about reality, or they may fail to prescribe the right actions. To use them to solve problems, we need, once we’ve formed them, to subject them to criticism and testing through Knowledge Claim Evaluation. Only if our stories survive competition against other stories during knowledge claim evaluation can we conclude that they provide solutions and objective knowledge. Even then, however, we cannot say for certain that they are true. They along with other knowledge claims remain conjectural and subject to further critical evaluation as the need arises.

Storytelling and other areas of Problem Solving
Knowledge Claim Formulation is not the only area of problem solving where storytelling can help. Let’s review the others.

Storytelling and Problem Recognition

Story-telling can help us to recognize and clearly formulate knowledge gaps (i.e. problems) that are relevant for improving organizational business processes. If you’re the storyteller, the act of formulating a good story can be a great aid in increasing your own understanding of the problem. If you’re the listener, a good story can help you to see a knowledge gap clearly and to better appreciate its connection to the practical decisions that cannot be made without closing that gap.

Storytelling and Information Acquisition

Is story-telling useful for acquiring information from outside our organizations? Storytelling within an organization has no role here. But listening to, and acquiring, stories from outside one’s organization can be among one’s most relevant sources of information for competitive intelligence.

Storytelling and Knowledge Claim Evaluation

Is story-telling useful for Knowledge Claim Evaluation? Here, I think the answer is mixed. Alternative stories express competing knowledge claims, and are indispensable for comparing competing knowledge claims about particular events or occurrences. By-and-large, however, except for their role as the targets of comparison, stories are not very useful for performing logical analysis, or for analytical criticism, or for comprehensive and close comparisons of competing knowledge claims. For these activities we need critical frameworks, models, or perspectives, rather than stories.

Knowledge claim evaluation is the area of knowledge processing activity in which stories are least helpful. Unfortunately this is the area of knowledge processing which most distinguishes it from information processing (See Chapter 3 of Key Issues in The New Knowledge Management).

Storytelling and Individual and Group Learning

Is story-telling useful for individual and group learning? Recalling what I’ve said earlier about problem solving, and keeping in mind that individual and group learning refers to individual and group-level knowledge life cycles nested within organizational systems, I think that story-telling is very useful for much of it. But, as with problem solving at the organizational level, it is less useful for Knowledge Claim Evaluation and Information Acquisition in individual and group learning.

Being Sensible About Storytelling and Problem Solving

So what is the bottom line on storytelling and problem solving? From the Knowledge Life Cycle perspective, story-telling is very helpful in Problem Recognition and Formulation and in Knowledge Claim Formulation. But it has shortcomings in Knowledge Claim Evaluation, and Individual and Group Learning. So, I think that we should use stories and learn to tell them skillfully. But I also think that we should all take out a membership in story-tellers anonymous, and pledge that we will not get drunk on the appeal and success of our stories in persuading others to our beliefs. Instead, recognizing that our stories are conjectural in nature, we should pledge to cultivate a critical attitude toward them, keeping in mind that if our stories survive our best criticisms they are more likely to provide a better basis for decisions.

In a future blog I’ll explore this attitude toward storytelling and problem solving further, by posting an AOK Group exchange I had with Steve Denning on “Narrative and Knowledge”.

Keelman Heaving Coals by Moonlight, (J. W. M. Turner, 1835)

Commentary on Dave Pollard’s Blog on Social Networking (continued from Part 2)

The Exchange: KM Failure or Conceptual Confusion?

The thread running through my commentary in Parts 1 and 2 is the idea that Dave’s blog and his position that KM should be re-invented as Social Network Enablement, emerge from considerations that do not dwell heavily on the nature of either Knowledge or Knowledge Management. As a result, his analysis and proposal about KM’s future is not grounded in a clear idea of what KM is about, how it is distinguished from other forms of management and how anyone can evaluate whether a new initiative is or is not a KM initiative. This issue emerged again in the following exchange on the AOK list serve:

In response to my statement:

“Mark and I think that the value propositions of KM are:

• Enhances ability to satisfy demands for new knowledge
• Enhances rate and quality of organizational learning and innovation (sustainably so), and
• Enhances organizational capacity to adapt
• Enhances corporate governance by elevating Knowledge Processing to the level of a fiduciary issue

The difference between our value propositions and yours are very striking.”

Dave said:

“Yes, they are. My experience at E&Y (and others in leading knowledge organizations have told me their experience has been similar) is that your KM value propositions, which were also E&Y’s for many years, simply have not been realized, despite the efforts of a lot of bright minds and the expenditure of millions of dollars. And it’s not just a matter of not being able to measure this ‘value’ — users tell us bluntly that KM has not delivered on any of these promises, and they’re not convinced it can or will.”

And I responded:

Dave, I’m afraid this is going to get us back into the importance of conceptualization and measurement again. My view is that the millions that have been spent on KM have not been spent on it, but rather on Information Management. This has occurred because our “best and brightest” believed that they could do KM and have a positive impact on Knowledge Sharing without being very clear about the nature of either “knowledge” or “Knowledge Management” and how they are different from information, information management, quality management, content management, collaborative management, etc.

“Let’s not worry about it, they said. No one in the last 2500 years has been able to solve the problems of epistemology, so let’s not discuss them. Let’s get away from theory. Let’s be ‘practical’ and just take all of these new tools and techniques and call them KM and see if they work”. But the consequence of this attitude is that:

(1) we’re not very clear on when something we’re doing is KM in contrast with knowledge processing, or information management, or information processing, or collaborative management, or content management, or many other kinds of management, or just good ‘ole’ business activity, so how can we (or our users for that matter) tell whether “KM’ has had any impact?

(2) we’re not very clear about whether we’re sharing “information” or “knowledge”, so:

(a) How can we tell whether our knowledge sharing programs are successful?

(b) How can we tell whether or not our lack of success is due to our sharing falsehoods (poor practices) and representing them as knowledge (best practices)? And:

(3) we’re not very clear about how we measure the impact of KM, so how can we tell whether it, in the presence of competing activities has a positive or negative impact? Let’s say, for example, that KM has had a positive impact during a particular time period, but an ERP intervention had a negative impact; if we’re not even sure what KM is, how can we disentangle those two? And, if the same enterprise is “fooling with” KM, ERP, Data Warehousing, CRM, Quality Management, and God knows what else, all in the same time period, and we’re not even sure what KM is, how can we possibly say anything about its impact?

In another current post, you named a number of tools that you said “epitomized” KM in organizations today. And, of course if those were really KM tools I’d agree that projects using them would certainly not deliver the value propositions I listed, and I’d say let’s reinvent KM as something else because it’s just not working. But, that’s just the point, I think that’s the old KM, not the New KM.

I think the New KM is different because it does offer clear notions of “knowledge” and “KM”, can distinguish between KM and other forms of management, can distinguish knowledge from information, has a clear conceptual idea of what KM impact is, is developing metrics, is developing project and program methodology, does know the difference between IT applications that support key knowledge and KM processes, and those that don’t, and has a normative vision to fulfil KM’s value propositions. So, my intention is not to persuade people to replace KM with something else, but rather to persuade them that experience with KM has taught us a few things about the foundations of our field and now we’re ready to come back with different ideas, different tools, different techniques, and much better results. People may or may not accept this view, but win or lose, we will doing KM, and not “Social Network Enablement”.

Next, you responded to a question of mine as follows:

My question: “If we don’t know what knowledge work is, how can we know if we are improving knowledge worker effectiveness?”

Your response: “If the user, i.e. the knowledge worker’s ‘internal customer’ says that the knowledge worker is doing great work, then to me the argument about whether it qualifies as ‘knowledge work’ is not terribly important.”
And I responded:

Dave, your statement of the KM value proposition implies that the distinction is important. If so, then my question holds as stated. If not, then you won’t mind me restating your value proposition as: “improving worker effectiveness”. Stated this way, it’s very clear that this value proposition is not specifically related to Knowledge Management, but to Management, more generally, and you and I are not talking about KM anymore at all, so we have no disagreement.

That completes the account of my commentary and exchange with Dave Pollard about Social Networking Software and its relation to KM. This exchange along with a number of others I’ve had over the past three years, raises the question of whether formal interventions identified as KM “interventions,” “projects,” “programs,” and “endeavors,” are really examples of “KM practice.

Has KM Been Done?

In other words: Has KM Been Done? This is a trick question. Of course, KM has been done. KM is a natural function in human organizations, and it is being done all of the time in an informal distributed way by everyone undertaking activity in order to enhance knowledge production and integration activity. But whether formal interventions claiming the label “KM” are instances of KM practice is another question entirely. To answer that question, we need to have clear, non-contradictory ideas about the nature of knowledge, knowledge processing, and Knowledge Management. And to have those, we need to get beyond the notion that we can do KM by just doing anything that may have a positive impact on worker effectiveness while calling that thing “KM.”

Instead we need to recognize that the immediate purpose of KM is not to improve either worker effectiveness (though it may well do that) or an organization’s bottom line. Its purpose is to enhance knowledge processing that solves problems and to enhance the diffusion of these solutions, in the expectation that such enhancements will produce better quality solutions, which, in turn, may, ceteris paribus, improve worker effectiveness and the bottom line. And when we undertake KM projects, we must evaluate the contributions of our interventions to the quality of knowledge processing and knowledge outcomes. That means tough, precise thinking about knowledge processing, knowledge, and the impact on these that our interventions are likely to have.

The question I am asking here is whether KM practitioners are, in fact, providing this tough, precise thinking as a basis for KM practice, or whether, instead, they are “practicing KM” by helping fields or techniques such as Information Technology, Content Management, CRM, Social Network Analysis, Story-telling, Communities of Practice, and “Knowledge” Cafés to “colonize” it? I don’t propose to answer that question now, but just to raise it and to suggest that my friendly exchange with the perspicacious Dave Pollard points to at least one case where there is not only a tendency, but a proposal, to substitute the means of social network enablement, for the end of high quality KM practice. Are there other cases you can point to? If so, I’d be interested hearing about them, and together we can perhaps address the larger question of whether such conceptual displacement is not so widespread that we can conclude that, generally speaking, at least, KM as a formal endeavor has, indeed, not yet been done.

Owentsia Hunt Ball — Chicago (1904)

Commentary on Dave Pollard’s Blog on Social Networking (continued from Part 1)

KM, Social Network Management (SNM) and Conceptual Drift

You then went on:

“Four important unanswered questions:

1. What role can Social Network Enablement and social software play in enhancing individual and organizational learning?”

Social network enablement and social software can provide a stronger foundation for more intensive and connected social interactions. However, this may not have a uniformly beneficial effect on knowledge processing and its outcomes. Much depends on the underlying social psychological preconditions in the organization receiving such software.

If mistrust and division, are present to begin with, SNE software may produce greater integration within conflicting groups and factions and may lead to an increase in inter-group conflict within an organization. Moreover, it is not clear what impact either decreased or increased conflict would have on knowledge processing. Much would depend on the initial state of conflict in an organization and the context of the SNE intervention. I think one thing is fairly certain, however, the impact of SNE and social software on knowledge processing will be beneficial in some respects and harmful in others, and it will be one of the concerns of Knowledge Managers to track the impact.

“2. How do you measure and reward contributions to a network (a) by full-time knowledge workers (people in the organization, like researchers and help desk staff whose sole value is contributing to the network) and (b) by network ‘players’ outside the organization?”

This is an interesting question, but it is not clear that it is a KM question. Isn’t it a Social Network Management (SNM) question? Also, the question implies the desirability of introducing a formal incentive system to “reward contributions” to a social network. But this assumes that it is desirable to reinforce participation in the network beyond the reinforcement provided by participation itself. This may be unwise, because it involves a managerial imposition of a perceived desirable outcome on the network. Enabling the network with software is one thing; manipulating the incentives to participate in it is another. If we want to take advantage of natural tendencies to self-organize, we should avoid the second.

“3. How do organizations equip and foster networks without unduly controlling their actions and membership and therefore crushing them?”

Very carefully. And the real question is, how do Managers, Knowledge and otherwise, equip and foster networks without impairing the organization’s ability to adapt and remake itself by incenting or imposing behavior that the organization taken as a pattern would not incent? I think the answer to this question, is that managers should equip and foster networks, and then do whatever else is necessary to enable people to use them. After that, managers must trust to people themselves to use their social networking tools to do their jobs including solving the problems that occur in the course of doing them.

“4. How do we capture summaries and abstracts of organizational conversations that occur in other than written form (voice-mail, teleconferences and meetings), so that the blog record of networks is complete?”

We wait for the technology that makes it convenient to do this. Until then, we do what we can with what is in the written record.

Of these four unanswered questions, only the first of them relates to either knowledge processing, or KM directly. I think this illustrates where too great a focus on Social Network Enablement and Social Software will take us, namely away from KM and into Social Network Management.

One of the continuing problems in KM is that of “conceptual drift”. Since the foundations of KM as a discipline are relatively undefined and we are in disagreement over what we mean by Knowledge, KM, and the distinctions between Data Management, Information Management and KM, as well as distinctions among a number of other basic concepts, we find ourselves subject, from time-to-time, to claims that KM “really is”, or should be “reinvented as” (take your choice): Quality Management, CRM, Data Warehousing, Organizational Learning, Collaboration Management, Library Management, Information Management, Human Resource Management, Communities of Practice, Content Management, and now Social Network Enablement. I think such advice is incorrect, and, thus far, at least, always based on a superficial account of the nature of KM. In fact, it is because those who offer such proposals do so without a careful analysis of “knowledge” and “Knowledge Management” that their ideas often initially seem plausible.

Even though I don’t agree with your suggestion that we should re-invent KM, I do think that Social Network Analysis is important for KM, and Social Network Enablement and Social Software are important trends that we should incorporate into KM interventions as appropriate and necessary. But as with every other KM intervention alternative, tool, or technique, both before and after we undertake projects that use SNE, we should try to assess what the impact of our intervention will be or has been, as the case may be. And to do this, what we really need, on an urgent basis, is a better network of concepts, indicators, and metrics that will allow us to talk more precisely about the direct impact of our interventions on knowledge processing and Knowledge Management, as well as their indirect impact on other organizational outcomes (for which better metrics may already exist).

Lastly, I hope you don’t consider this or other posts I’ve offered in this AOK session as hostile to you or your work. Actually, I’m very favorable to Social Network Analysis as a perspective, and I think you’ve done a great job with your blogs and the contributions you’ve made here. I’ve offered my posts, because in many respects, my views are different from yours and both of us may benefit by exchanging communications on these differences. I’ve tried not to use any ad hominems or unsupported assertions in what I’ve written, in hopes that you would see this in the way I intend it, as an attempt at discussion.

In Part 3 of “Has KM Been Done?” I’ll finish this series of blogs with an exchange between Dave and I and a preliminary consideration of the question providing my title for the series.

Rain, Steam, and Speed (J. W. M. Turner, 1844)

In March of 2004, David Pollard served as the Star Moderator in the Association of Knowledge Work’s (AOK) Star Series. Many subjects were covered during the very fruitful exchanges of Dave’s tenure. One of my posts (message 1484) was a commentary on one of Dave’s blogs called “Social Networking, Social Software and The Future of Knowledge Management.” In this blog, I’ll first reproduce (with a few minor revisions) a good part of my commentary, then supplement it with another post (message 1498) recording an exchange between Dave and myself on KM Value propostions; and then I’ll discuss the significance of these posts in suggesting the question: “Has KM Been Done?” This first of three installments will cover the portion of my commentary dealing with the issues of “KM’s failure” and Dave’s proposal to reinvent it as Social Network Enablement.

Commentary on Dave Pollard’s Blog on Social Networking

The Burning of Parliament (J. W. M. Turner, 1834)

I believe that my friend and collaborator Mark McElroy (2001) was the first to point out the inherent conflict in the very popular, but we think mistaken, idea that Knowledge Management should be aligned to organizational strategy and that its purpose should be to mobilize knowledge in the service of it. But what’s wrong with this widely held idea? Isn’t knowledge just a “tool” or instrument of strategy and KM just a method of upgrading the quality and effectiveness of the tool? And doesn’t this mean that KM is itself just a hand maiden of Strategy? Well no, not really. Things are a bit more complex than that.

The Three-Tier Model: Where Is Organizational Strategy?

It may be helpful to keep in mind the distinctions made in The New Knowledge Management’s three-tier model (See Figure 1). The top KM-tier includes process activity and process outcomes aimed at enhancing the way we perform problem-solving (where we define problems as “knowledge gaps”) processes in the second, knowledge processing tier. This tier, includes process activity and outcomes aimed directly at solving problems originating in process activity in the bottom tier. This tier includes all other process activity and outcomes including all non-KM management activities.

Figure 1 — The Three-tier Model

So, here is a three-way distinction that is very important to thinking through the scope of KM. “Doing”, including everything outside of KM itself that uses knowledge to attain strategic goals and objectives, mostly occurs in the bottom-tier. “Learning” (in the sense of problem-solving) and integrating the outcomes of learning mostly occurs in the middle-tier. And Knowledge Managing, which includes both “Doing” KM, and “Learning” about how to do “learning” and KM, occurs in the top-tier.The middle-tier is where the responsibilities of management, in general, and KM, in particular, meet. KM has the responsibility for enabling knowledge processing in the middle-tier, but the knowledge workers who perform middle-tier processing are responsible to managers, other than knowledge managers, for their performance of operational business processing including their use of knowledge in that performance. So, where is strategy?

Organizational Strategy (excluding KM Strategy), considered as a process, is a bottom-tier phenomenon in the three-tier model. It can occur only after strategy, considered as a network of knowledge claims, is created through knowledge processing in the middle tier. The KM component of organizational strategy, on the other hand, is performed in the top-tier, as is knowledge processing creating KM strategy.

Organizational Strategy is about setting goals and objectives for the organization and specifying high-level plans for getting there. But KM strategy is not and cannot be about achieving any of these goals or objectives, because the production of specific goal and objective-oriented knowledge that would be effective for fulfilling strategy is not the kind of result that can be caused and therefore planned by management, knowledge or otherwise. Instead KM strategy focuses only on organizational goals and objectives related to performing KM and enhancing knowledge processing (including knowledge processing at the KM Level). Its goals and objectives should be limited to enabling knowledge processing,and the self-organization of knowledge workers around problem solving and knowledge integration, that alone can bring sustainable adaptive success.

The Conflict

The idea that Knowledge Management should be aligned with organizational strategy, and that therefore Knowledge Managers should be subordinate to managers concerned with fulfilling the goals, objectives and plans of strategy carries with it an irreducible conflict that harms not only Knowledge Management, but also, knowledge processing, operational management and operational processing, as well. The conflict arises from the three-tier model.
Here’s an outline of its dimensions.

• It is the purpose of KM to enhance knowledge processing and its outcomes. Any organizational authority structure that prevents knowledge managers from performing their role of enhancing knowledge processing reduces their effectiveness in maintaining and adding to the adaptive capability of the organization. So, any organizational arrangement mandating (for knowledge managers) a higher priority for strategic goals at the expense of the KM goal of enhancing knowledge processing will reduce the effectiveness of KM in accomplishing its primary goal.
• It is the purpose of knowledge processing to produce and integrate knowledge, including the organizational strategy that is implemented in operational business processing. Any organizational authority structure that prevents knowledge workers, including managers, from performing their problem solving role reduces their effectiveness as knowledge producers, and indirectly the effectiveness of operational processes as well. So, any organizational arrangement mandating (for knowledge managers) a higher priority for strategic goals at the expense of the KM goal of enhancing knowledge processing, will also generally result in less effective knowledge production.
• It is the purpose of operational processing to fulfill strategy, and to manage and perform the day-to-day work of implementing operational processes and attaining the goals and objectives of the organization. Any organizational authority structure that prevents managers charged with implementing strategy and operational processing from performing their roles detracts from the effectiveness of the organization. So, any organizational arrangement mandating (for knowledge managers) a higher priority for strategic goals, at the expense of the KM goal of enhancing knowledge processing, will generally detract from operational managers’ performance in their roles. It will do so because it injects Knowledge Managers into operational processes by involving them in issues of knowledge use, an area of responsibility of operational management.

To see the conflict more clearly, consider the following questions. What if strategy doesn’t work, what if it turns out to be mistaken? And what if the managers who are attempting to implement an erroneous strategy won’t accept that it is erroneous. What if they require knowledge managers to “adjust the rules of knowledge making”, so that knowledge workers will neither find nor seek a better strategic solution; but will, instead, devote all of their energies to learning better ways of executing a counter-productive strategy — one destined to produce bad results, regardless of how well it is implemented? These questions are close to being rhetorical, but they still illustrate the main points.

There is a KM function in organizations and it should not do knowledge processing in response to operational problems ultimately related to strategy, or attempt to command such knowledge processing. There is a knowledge processing function in organizations, and if KM adjusts its rules in order to save a favored set of knowledge claims, its adaptive function, along with the adaptive function of the knowledge workers who perform it, will be compromised. Lastly, there are many operational business processing functions in organizations, and if KM adjusts the rules of knowledge making so that the strategic solutions that are closest to the truth are not produced by knowledge workers, and therefore are not available to managers and workers who need to use knowledge in business processes, the general result will be a decline in the effectiveness of operational business processing, not a happy result for investments in KM.

Why?

If it is true that the principle that KM should be aligned to organizational strategy has such a plain inherent conflict, then why is it that KM practitioners don’t acknowledge the conflict? Why is it that they, instead, continue asserting that the first step in establishing a KM program is to create an alignment with organizational strategy? I think the answer is that they don’t distinguish among Knowledge Management, Knowledge Processing, and Knowledge Use, and don’t visualize KM-related concerns in terms of the three-tier model. They therefore confuse operational business management interventions, focusing on how previously produced and integrated knowledge should be used, with KM interventions focusing on how knowledge production and integration may be enhanced. It is Knowledge Use, occurring in the bottom-tier of the model, that should be aligned to strategy, and Knowledge Processing that may be performed in support of strategy provided that the problem or problems initiating it do not call strategy itself into question. But, again, KM should never be aligned with any aspect of organizational strategy, except KM’s central function of creating and enhancing sustainable knowledge production and innovation.

The Implication

The idea that KM is independent of strategy and should not be subordinated to it has an important implication, also first pointed out by Mark McElroy.That implication is that KM is a fiduciary activity in organizations that must not be under the line authority of those, including the CEO, who make and enforce strategy. In private sector organizations, both profit and non-profit, KM should, like the CFO, be responsible to the Board of Directors. In the public sector, KM, like other fiduciaries, should be responsible to the legislative authority.

I know that, given the historical development of KM, this implication of the relationship between KM and strategy will make many of you groan. It’s tough enough to sell KM, you say, without having to tell the “powers-that-be” in your organization that they need to establish the KM function as independent of their own authority, rather than as their faithful servant. In response, all I can say is: there exists a management process, or set of management processes, in organizations that can create and enhance sustainable innovation. Call it (or them) KM, or call it something else, it still exists. If you want to practice it, you have to recognize that its “best practice” can only be accomplished if it is independent of, and autonomous with respect to strategy, and that implies that it should have fiduciary status. So there it is. If you want to practice KM, as opposed to practicing knowledge use management, you need to make the case for its independence and autonomy.

Light and Color (J. W. M. Turner, 1843)

In the opening blog of “All Life is Problem Solving”, I gave a general account of how problem solving occurs in living things including humans. In my second post I focused on problem solving at the organizational level. Here I want to develop some ideas about Personal Knowledge Processing and Knowledge Management.

Personal Knowledge Processing

If we want to get beyond the trivial view that all individual decision making and action involves knowledge processing, in the sense that every decision we take is based on knowledge that we use, we need to make a distinction between knowledge use and knowledge processing. Much of personal knowledge processing is routine in the sense that it involves the use of memory, expectations, and predispositions to actions in situational contexts, to supplement and change our beliefs, and to make decisions. But sometimes situations don’t meet our expectations, both because we find ourselves not meeting our objectives, and also because we view these situations as inconsistent with our expectations and all of our other knowledge related to them.

As individual persons then, we sometimes reach a point where it seems to us that we need new knowledge to accomplish our objectives and goals. We also believe that we can’t proceed toward them without a period, however small or large, of search and inquiry, during which we “figure out” what to do; that is, we go about solving our problem. So I want us to begin by recognizing that every decision and action we take involves knowledge use, but, in addition, some subset of decisions and actions, those involving inquiry, search behavior, or problem solving, are instances of knowledge processing.

In my first blog on “All Life is Problem Solving”, I wrote about Popper’s theory of problem solving and presented a graphic showing his model and its relationship to expectations and behavior. In Figure 1 below, I make a slight modification of that model to visualize the distinction between knowledge use and knowledge processing. At this point, some of you may ask: Where do problems come from? How do we come to recognize them? What motivates us to doubt our expectations, even in the presence of “objections from reality” to even consider that we have a problem?

Figure 1 — Knowledge Making, Knowledge
Processing, and Knowledge Use

My answer is that we are, with variations caused by genetic endowment and previous individual experience, built that way. That is, our predispositions are complex. We are predisposed to pursue our goals and objectives, but to varying degrees we are also predisposed to look for problems and to recognize when we lack knowledge to decide. We have previous knowledge that lets us suspect that, given a situation, we may have a problem. And when we suspect that we may have a problem, we may then either use previous knowledge to recognize what that problem is, or, alternatively, we may use such knowledge to decide that we must solve another problem of determining what the first problem is. In short, there is no magic here. The recognition of problems emerges from knowledge use and problem solving. There is no third category of activity in personal knowledge processing.

The “bottom line” here, applying Popper’s Theory, is that personal knowledge processing is analyzable into recognizing problems, formulating attempted solutions, and eliminating errors. What about Personal Knowledge Management (PKM)? What is it and where does it fit?

Personal Knowledge Management

I think, in accord with the theory of problem solving, that Personal Knowledge Management is activity we perform in order to improve our problem recognition, formulating attempted solutions, and error elimination activities. And I think everyone does it in some measure, in the sense that everyone performs some activities to help themselves perform activities in each of these areas.

Personal knowledge management can overlap with interpersonal knowledge management, group-level knowledge management, and organizational knowledge management. For one thing, we identify with groups and organizations we are participants in, and sometimes take group and organizational level problems as our own. For another, our efforts at personal knowledge management, may be part of a more comprehensive pattern of group or organizational knowledge management without our knowing that they are, and without our intending to make such a contribution. So, personal KM, group-level KM, and organizational KM, are not disjoint sets of activities. Moreover, groups and organizations may decide to reinforce personal KM in order to enhance KM at the group or organizational levels; and some analysts, like myself, are of the opinion that enhancing personal KM enables individuals to improve self-organization around knowledge processing, in each of the three areas of problem solving.

Steve Barth, has led the charge for personal KM. His view of it was first stated in KM Magazine (2000) in “The Power of One“, and more recently was amplified in KM World in “Three Thousand Communities of Practice“, and a post he offered to the AOK Group’s discussion on PKM and Interpersonal KM (IPKM). Steve is also the author of a KMWorld column on PKM. Denham Grey has made influential statements on the subject emphasizing the non-social aspects of PKM in his blog. More recently, David Gurteen and Lilia Efimova have added well-written statements in AOK’s discussion of PKM/IPKM.

The view emerging from these and associated discussions is that PKM may be focused on the individual, but that it is not a lone individual that is its focus. Rather it is an individual at the nexus of various social networks and information streams. Moreover, even though Steve Barth’s emphasis in various articles has been on tools for supporting PKM, he, as well as everyone else who has commented on the subject, emphasizes that the core of PKM is not tools and techniques; but rather individual processes of acquiring, creating, and communicating knowledge or information. In the wake of this vigorous discussion of PKM, I want to comment on a number of its aspects.

PKM and the Interactions Among Cultural, Explicit Mental, and Non-Conscious Knowledge

In individual systems, the theory of knowledge making suggests that problem production, formulating alternative solutions, and error elimination, are important sub-processes in problem solving or knowledge production. At the organizational level, a second knowledge process, knowledge integration, is very important for distributing knowledge. At the individual system level, however, knowledge must be organized and maintained through a process of knowledge storage and organization. The names of the sub-processes immediately suggest only explicit knowledge processing (of cultural products) is involved here. There’s more to this than meets the eye, though. Individuals learn explicitly by forming, testing and evaluating beliefs as well as claims. And each of the above processes has a psychological (and not just a cultural/linguistic) side in which new beliefs are formulated, tested and evaluated.

In fact, in personal knowledge processing, our efforts to formulate new knowledge claims are in constant interaction with our efforts to formulate new beliefs. In addition to the explicit learning of both the cultural and mental varieties that goes on in these sub-processes, theory suggests that individuals also learn non-explicitly from all activities they engage in, because the capacity to do so is inherent in our brains and associated biological systems. In fact, evidence from neuro-science suggests that all our conscious experience is accompanied by non-conscious learning that produces knowledge that is inaccessible to consciousness. And evidence from psychology suggests that this same experience is also accompanied by non-conscious learning producing changed mental predispositions (changes to our attitudes and values). So, even though personal knowledge processing is about problem solving and explicit learning, we must keep in mind that its products include non-conscious biological and mental knowledge as well.

PKM and Complexity

In performing these sub-processes and all other activities as well, individuals are embedded in a social, cultural, economic, and ecological, context containing multiple complex adaptive systems in which they participate throughout their lives, which affect their decisions and actions, and which they affect, with those same decisions and actions. Their families, enterprise environments, voluntary associations, nation-states, the international system, the world eco-system, are all CASs in which they participate intermittently from time-to-time. Individuals are at the nexus of these systems and must integrate all of their transactions with them in such a way as to maintain the individuals’ coherence and identity. They use their knowledge to do this, and that is a large part of what adaptation is about.

The important direct outcomes of personal knowledge processing are knowledge claims and beliefs and reasons for thinking that one’s knowledge claims and beliefs may be relied on in decision making. These claims and beliefs are stored in one’s media (documents and information systems) and one’s brain (memory). The combination of the two, I call The Personal Knowledge Base (PKB). It is the mental aspect of the PKB that individuals use in decision making.

A classification of PKM activities includes:

1. It suggests that PKM is “social” in the sense that all decisions and actions, including those involved in knowledge production, occur in, and are influenced by, a social context, and therefore “personal knowledge is socially constructed”. But
2. it also suggests that PKM does not imply that personal knowledge must be socially validated. For one thing, the social construction of personal knowledge is not dependent on one social context, but on many. For another, the individual must integrate a variety of socially constructed perspectives in evaluating knowledge and in deciding which knowledge claims and beliefs “match” experience better than others. In performing this sort of activity, the individual acts as an autonomous system. Its knowledge claim and belief evaluation activity is “emergent” and is influenced by differing social contexts, biological factors (e.g. synaptic structures and brain functioning), psychological predispositions (attitudes and values), and tacit, implicit, and explicit situational orientations.
3. The framework also suggests some ideas about the current focus of Personal Knowledge Management Practices and Tools. Much of the literature of PKM focuses on: (a) tools for organizing information resident on one?s computer and retrieving that information as an aid in using it in decision making or developing new ideas, (b) tools for enhanced searching and retrieving on the web, and (c) tools for visualizing conceptual relationships that are useful in clarifying one’s ideas and developing new ones. The first and third of these overlap somewhat, especially in such areas as portal interface navigation tools such as The BrainEKP.

PKM has thus far experienced almost no focus on tools, practices and procedures for evaluating knowledge claims and beliefs. This area is vital to PKM because it is about the quality of one’s knowledge, about eliminating errors in it, about distinguishing one?s knowledge from one’s information, and ultimately about making good decisions.

Another area that I don’t think is much focused on in PKM is resource allocation. Resource allocation is one of the most important activities in KM. At bottom, it involves prioritization and risk assessment. But I haven’t seen much discussion of tools such as ExpertChoice, and techniques such as the Analytic Hierarchy Process in PKM.

PKM, IPKM, and Enterprise KM (EKM)

IPKM is about systems containing two or more individuals relating in a peer-to-peer fashion on an on-going basis. The formal hierarchy that exists in the enterprise is not there, and the scope of interaction in systems practicing IPKM is much narrower than the scope of interaction in enterprises. Knowledge processing and KM activities in the IPKM context are more similar to EKM processes than PKM processes. In the IPKM context, knowledge production includes: problem production, information acquisition, individual learning, knowledge claim formulation, and knowledge claim evaluation. Knowledge integration is relevant to the IPKM context. iIs sub-processes are: knowledge and information broadcasting, searching and retrieving, teaching, and sharing. The Distributed Organizational Knowledge Base (DOKB) replaces the PKB used in the PKM framework.

The KM activities in the IPKM context include: Building External Relationships, Symbolic Representation of Authority, Leadership, KM-Level Knowledge Production, KM-Level Knowledge Integration, Crisis Handling, Resource Allocation, Negotiating Resources with others, and Changing Knowledge processing Rules. Mark McElroy and I have discussed this framework in our book Key Issues in the New Knowledge Management, Burlington, MA: KMCI Press, 2003, Chs. 1-6. You can also see some graphics at KMCI.

Since Knowledge Processing and KM sub-processes and activities in IPKM are highly similar to those in EKM, it?s not surprising that the techniques, procedures, tools and practices of IPKM are also highly similar. This is especially true in the area of informal structures and processes where CoPs, Story-telling, Knowledge Cafes, Facilitation methods, Collaboration Tools, and Social Network analysis would all be relevant. In addition, practices, tools, etc. that are important in PKM would also be relevant to IPKM since anything that enhances Personal Knowledge processing is likely to have a positive impact on IPKM.

The present state of IPKM is similar to the state of PKM in that knowledge claim evaluation and resource allocation and prioritization practices, tools, etc. are not well-developed. I think progress in these areas would be important for the further progress of IPKM.

Finally, though PKM and IPKM are often contrasted with EKM, I think the progress of EKM is related to the progress of both of these areas. The “bottom-up” approach to EKM is essentially based on a similar insight, which recognizes that formal organizations are dependent on the enhanced functioning of individuals and groups, for their own enhanced functioning.

For More Information

In addition to the book referred to earlier, you’ll find much more information on the theories and models offered in this paper, and on training in the New Knowledge Management at three web sites: dkms.com, macroinnovation.com, and kmci.org. Many papers on the New Knowledge Management are available for downloading there. Our Excerpt from The Open Enterprise . . . may also be purchased there. Our print books are available at Amazon, Barnes and Noble, or Butterworth-Heinemann/Elsevier. Finally, there will be many more blogs coming and these will apply the point of view expressed here to many of the major issues in Knowledge Management today.

In the opening blog of “All Life is Problem Solving,” I gave a general account of how problem solving occurs in living things including humans. But are organizations living systems, or, at least, are they like living systems in their problem solving patterns? How does organizational problem solving happen?

In seeking an answer to that question, I’d like you to recall Popper’s theory of knowledge making as I explained it in my first blog. In living systems, ordinary activity sometimes leads to situations that don’t match predispositional knowledge and expectations. This leads to formulating tentative solutions, and then to error elimination producing new knowledge and expectations.

With that theory in mind, let’s begin with decisions and actions, the ordinary activity of organizations. Decisions are part of a sequence that has been described in slightly varying terms, using many names (e.g., the organizational learning cycle, the experiential learning cycle, the adaptive loop, and others). I call it the Decision Execution Cycle (DEC). It includes planning, acting, monitoring, and evaluating behaviors. Decisions are produced by planning and are embodied in acting. Decisions produce actions. And actions – activities – are the stuff that social processes, social networks, and (complex adaptive) social systems are made of. These are built up (integrated) from activities in ways my collaborator and friend Mark McElroy and I have described in our books (See McElroy, The New Knowledge Management, KMCI Press/Butterworth-Heinemann, 2003; Firestone, Enterprise Information Portals and Knowledge Management, KMCI Press/Butterworth-Heinemann 2003; and Firestone and McElroy, Key Issues in the New Knowledge Management, KMCI Press/Butterworth-Heinemann, 2003, and Excerpt #1 from The Open Enterprise: Building Business Architectures for Openness and Sustainable Innovation, KMCI Online Press, 2003).

Now, let’s distinguish three tiers of organizational processes: (1) operational processes, (2) knowledge processes, and (3) processes for managing knowledge processes. Operational processes are those that use knowledge but, apart from knowledge about specific events and conditions, whose acquisition doesn?t require problem solving, don’t produce or integrate it. For that, we turn to knowledge processes. There are two knowledge processes: knowledge production, the process an organization executes that produces new solutions to its problems; and knowledge integration, the process that presents this new knowledge to individuals and groups comprising the organization. Knowledge production and problem solving, on this account, are one and the same. There are at least 9 processes related to managing knowledge processes. I won’t cover them in this blog, but I’ll have occasion to return to these Knowledge Management processes in other blogs.

Incidentally, the three tiers of organizational processing I just distinguished, are part of a high-level model Mark and I developed at the Knowledge Management Consortium International (KMCI). We use the model as one of the organizing constructs in our Certificate course in Knowledge and Innovation Management (CKIM). You can see a graphic on p. 8 of the KMCI Brochure at: http://www.kmci.org/KMCI_Brochure_as_of_3.04.pdf.

Knowledge Processes

What corresponds to Popper’s theory of knowledge making at the level of organizations? To begin, problems occur for organizations just as they do for other systems. Organizations go through problem production processes to produce their knowledge of problems. Here is how things work.

Operational business processes are performed by individuals and groups using previous knowledge, both their own mental knowledge, knowledge claims in organizational repositories; and also situational knowledge, the result of on-going non-problematic learning, to make decisions. Sometimes previous knowledge and situational knowledge do not provide the answers they need to perform their roles in organizational processes. If so, a problem has arisen — an epistemic gap between what an agent knows and what it needs to know to perform its role in the organizational process. Such a problem initiates knowledge processing: specifically, a new knowledge production process.

Popper’s second step of attempted solutions corresponds to three sub-processes of knowledge production at the organizational level:

• Information Acquisition
• Individual and Group Learning
• Knowledge Claim Formulation

That is, once an organizational problem is produced, there is a need to formulate tentative solutions in response. These can come from new individual and group learning efforts to address the problem; or from external sources through information acquisition; or from entirely creative knowledge claim formulation efforts; or all three. Where the tentative solutions come from, and in what sequence, is of no importance to the self-organizing knowledge processing pattern of knowledge production.

Individual and group learning is itself knowledge processing. It produces knowledge claims for consideration at higher levels of analysis of knowledge processing. But at the individual and group levels, learning is knowledge production and also problem solving. Let’s call this the recursive “nesting” of problem solving in organizations. Knowledge claim formulation uses the results of both information acquisition, and individual and group learning, and its own sub-process interactions to produce alternative solutions for:

• Knowledge Claim Evaluation (KCE); the organizational sub-process corresponding to error elimination in Popper’s theory.

Organizational knowledge is not produced until the tentative solutions, the previously formulated knowledge claims, have been tested and evaluated. And KCE is the way in which organizational agents select among tentative solutions (competitive alternatives), by comparing them against each other in the context of perspectives, criteria, or newly created ideas for selecting among them.

KCE is at the very center of knowledge processing and knowledge production. Think about it. Without it, what is the difference between information and knowledge? How do we know that we are integrating knowledge rather than just information? Or that the “knowledge” we’re using in operational business processes is of high quality? Absent a social process in organizations, be it formal or informal, through which competing claims can be held to tests of veracity or verisimilitude, how can we possibly make judgments about truth versus falsity? Knowledge claim evaluation, then, is what gives us the ability to know knowledge when we see it.

Once knowledge is produced by KCE, and an organizational problem is solved, the process of knowledge integration begins. Knowledge integration is an organizational process that may have a loose analog at the individual level in the processes of reinforcement learning, which fix knowledge in memory and synaptic patterns. Knowledge integration distributes organizational knowledge claims across knowledge repositories and also distributes beliefs about these knowledge claims across the organization.

Knowledge integration is made up of four more sub-processes, all of which may use interpersonal, electronic, or both types of methods in execution:

• Knowledge and Information Broadcasting
• Searching/Retrieving
• Knowledge Sharing (peer-to-peer presentation of previously produced knowledge), and
• Teaching (hierarchical presentation of previously produced knowledge)

There is no particular sequence to these integration sub-processes. One or all of them may be used to present what has been produced to the organization’s agents or to store it.

Those agents receiving knowledge or information through knowledge integration don’t receive it passively. For them, it represents a communication that may create a knowledge gap and initiate a new round of their own problem solving. The integration of knowledge, therefore, doesn’t signal its “acceptance.” It only signals that the instance of knowledge processing initiated by the problem is over and that new problems, for some, have been initiated by the solution. For others, the knowledge integrated is knowledge to be used: either to continue with executing the business process that initiated the original problem, or at a later time when the situation calls for it.

Either way, the original problem that motivated knowledge processing is gone. It was born in the operational business process, solved in the knowledge production process, integrated throughout the organization afterwards, and, in this way, it ceased to be a problem — i.e., it died. This pattern is a life cycle, a birth-and-death cycle for problems arising from business processes and through which new knowledge is also produced. Since the life cycle gives rise to knowledge, both mental and cultural (linguistic), Mark McElroy and I call it the Knowledge Life Cycle (KLC) (See p. 12 of the KMCI Brochure for a graphic at: http://www.kmci.org/KMCI_Brochure_as_of_3.04.pdf).

Every organization produces its knowledge through myriad KLCs that arise from its problems. KLCs occur at the organizational level and also at every level of social interaction and individual functioning in the organization. It is through these cycles that knowledge is produced, and the organization acquires the solutions it needs to adapt to its environment.

The KLC is another critical organizing concept used at KMCI in our CKIM classes and developed extensively in our books. KCE is discussed in some detail in Chapter 5 of our Key Issues in the New Knowledge Management, KMCI Press/Butterworth-Heinemann, 2003, and I suspect this is the most complete treatment of KCE in the recent KM literature, where KCE is, most often glossed over, or ignored entirely.

Outcomes

Organizational knowledge and problem solving processes, of course, produce outcomes. From my point of view, knowledge is an encoded, tested, evaluated, and still surviving structure of information that helps the system (agent) that developed it to adapt. There are two types of knowledge important in organizations: (1) tested, evaluated, and surviving beliefs or belief predispositions (in minds) about the world; and (2) tested, evaluated, and surviving, sharable (objective), linguistic formulations (knowledge claims) about the world. There are also other outcomes of knowledge processes, the most important of which are knowledge claims about the performance of other knowledge claims during knowledge claim evaluation. Mark McElroy and I call these “meta-claims,” or claims about claims. Organizational linguistic knowledge consists of claims that have survived our tests, criticisms, and evaluations, along with their corresponding meta-claims, whose content records the performance of such claims.

The various outcomes of knowledge processes may be viewed as part of an abstraction we call the Distributed Organizational Knowledge Base (DOKB). The DOKB in organizations has electronic storage components. But it is more than that, because it contains all of the outcomes of knowledge processing in documents, and non-electronic media. And since it includes beliefs and belief predispositions, as well, it also includes all of the mental knowledge in the organization as well.

For More Information

You’ll find much more information on the theories and models offered in this paper at three web sites: dkms.com, macroinnovation.com, and kmci.org. The last web site has the most complete information on the CKIM class team-taught by Mark and myself, and on our new KM strategy and methodology class called K-STREAM(tm). Many papers on The New Knowledge Management are available for downloading at all three sites. Our Excerpt from The Open Enterprise . . . may also be purchased at any of them. Our print books are available at Amazon, Barnes and Noble, or Elsevier/Butterworth-Heinemann. Finally, there will be many more blogs coming and these will apply the point of view expressed here to many of the major issues in Knowledge Management today.

I’ve named this blog after a statement from a lecture of Karl Popper’s, delivered in 1991 near the end of his long life. “All Life is Problem Solving” also became the title of a book of his essays published posthumously in 1999 by Routledge. I love the phrase because it sums up his wonderful work in epistemology, ontology, philosophy of science, and political theory performed over a period of nearly 70 years — his general theory about how new knowledge is made, or, if you like, how learning occurs.

The Theory of Knowledge Making

Knowledge is made, he thought, through a simple three-step process found in evolution, in individual psychodynamics, and in social interaction. That process is:

• the problem
• the attempted solutions
• the elimination.

Living things have expectations. Problems have their origin in events that run counter to those expectations. The response of life, however primitive, to a failure of expectation is to search for another way around. Living things engage in search behavior and testing to “replace the wrong expectation with a new one.” We “make” solutions whose successful application would create new expectations. And we “match” those solutions against aspects of our subjective reality, that, in turn, if we are to be successful in action, must have some correspondence with actual reality.

We see this three-step pattern in Darwinian evolution, where a failure of expectation caused by environmental change creates problems of survival for species which are solved through genetic recombination and mutation (attempted solutions) producing individuals that are better adapted to the changed environment than individuals of the old species were. The old species, along with most mutations and recombinations are eliminated by the environment. They are errors. The species that survive embody genetic knowledge — encoded information with adaptive value relative to the changed environment.

We see the three-step pattern again in the area of learning resulting in cognitive knowledge. Changes in the environment of living creatures result in failed expectations (problems). Search behavior leads to the discovery of new solutions, which if they match the changed environment are then encoded into the memories of living creatures. In humans this pattern is seen in the development of changes in synaptic structures and changes in beliefs as we discover new solutions, test them, and then encode the successful ones in our brains, and, we think, in our minds, as well. So, for living individuals, the three step pattern, the learning process, produces biological, and in some species, mental knowledge (beliefs).

Humans are unique, or at least, close to unique, on earth in enjoying evolution’s gift of language. Through language we can and do create sharable encodings that help both ourselves and our societies and cultures to adapt. The process of creating such sharable, adaptive encodings, or cultural knowledge again fits the three- step pattern. We recognize problems, we formulate tentative solutions (but now with the aid of language we can formulate and consider many and much more complex solutions than can animals who are not able to “objectify” their thinking), and we attempt to eliminate errors in those solutions so we can arrive at the solution that is the strongest in the sense that it has best survived our tests, i.e. our matching of it against those aspects of reality we think are important for its evaluation. Of course, our attempts at error elimination are also much stronger because of the gift of language. We can take the stories we tell about tentative solutions, write down those “stories”, or knowledge claims, or “theories”, or “models” and do a much better job of comparing them and evaluating them because language is the handmaiden of our comparison.

The Unified Theory of Knowledge

The solutions that survive error elimination constitute, once again, our cultural knowledge. As Popper pointed out this knowledge is objective because (a) it is sharable among those who have language, and (b) once made by us, it is autonomous, in that its continued existence can effect our future mental states, and through them our behavior. In contrast, mental knowledge is subjective because we cannot directly share it. However, this in no way dminishes its importance, since it is our mental knowledge which we use in order to behave, make decisions, and act, and since we create our cultural knowledge through action, it is also true that we use our subjective knowledge to create objective knowledge. So mental knowledge, while subjective, and also influenced by cultural knowledge, is also partly autonomous and responsible for the occurrence of cultural knowledge.

By now it should be plain that Popper used his three-step learning process (see Figure 1) to explain how three different kinds of knowledge are made: adaptive encodings in the material world (e.g. genetic encodings and synaptic patterns), adaptive encodings in the mind (attitudes, values, beliefs, etc), and adaptive encodings in cultural products (stories, arguments, theories, models, knowledge claims, propositions, etc.). Though Popper never used this term, this is a unified theory of knowledge (thanks are due to Art Murray of Tel-Art Technologies for the name), because each type of knowledge identifies encodings that are adaptive for the systems that use them relative to their environments. At the same time, the unified theory of knowledge acknowledges that “knowledge” is diverse in chraracter, and suggests that the ambiguities and variations we experience in using this term are due to this diversity.

Figure 1 — Popper’s Theory of Knowledge Making

Evolutionary Epistemology and Complexity Theory

Popper’s Theory of learning and problem solving along with his associated unified theory of knowledge emerge from an evolutionary persective. In his later years, he was associated with a movement called evolutionary epistemology; and it is important to recognize the connection between Popper’s epistemological work and modern Darwinian Theory. Popper’s perspective is also close to complexity thinking. He believed strongly, as I do, in the emergence of complex systems from simpler ones as a fact of life in the universe. And he believed, as I do, in the importance of downward causation as a factor in the emergence and maintenance of complex systems. Interesting work is being done today in the area of merging evolutionary epistemology and the sort of complexity theory that we find in the work of Maturana and Varela and Fritjof Capra. That work (see especially Mark Bickhard’s papers) will reinforce Popper’s view that “all life is problem solving”.

The Critical Perspective and Fallibilism

The last of the three steps in making knowledge is error elimination or “matching”. This step is the gateway to knowledge. But it is, as Popper pointed out, fundamentally negative in character. It is about eliminating mistakes and not about supporting any of one’s tentative solutions. In animals lacking consciousness, mistakes are eliminated, when the animal receives negative reinforcement from the environment for selecting the wrong solution. That is, the animal in question can only learn by experiencing the negative consequences of its mistaken expectation and ensuing decision. Often the wrong choice means that the animal making the choice is eliminated along with its mistake. Animals with consciousness and especially sharing language have a great advantage over other animals. We can eliminate errors and learn by testing our solutions through the surrogate processes of criticism, controlled testing, and comparative analysis, before we take a decision. We, unlike other animals, can manage our knowledge making so that “our worst ideas die in our stead”, and our best ones inform our decisions and actions. But to do so, we must use our gift of language and be diligent in criticism, testing, and evaluation of our tentative solutions.
So, in problem solving and in life, the critical perspective is the key. It is responsible for the elimination of errors, the growth of knowledge, and for adaptation in individuals and society. But why is this so, why has nature and biology relied on error elimination to get us closer to the truth rather than a process of justification or proof of our ideas? The answer is that all of our knowledge, including our biological, mental, and cultural expectations, is uncertain, and no amount of positive support can prove’ beyond doubt, that any proposition or idea is surely correct, or that any piece of genetic encoding, will allow us to adapt to changes in environmental conditions that are yet to occur. This idea, called fallibilism, also espoused by the founder of Pragmatism, Charles Sanders Peirce, before Popper, is skepticism, but it is not relativism. It doesn’t deny that we can find the truth, or that we ought to seek it, but only that we can never know with certainty that we have found it. Xenophanes expressed fallibilism in a wonderful way that Popper liked to quote:

The Gods did not reveal, from the beginning,

All things to us; but in the course of time,

Through seeking, men find that which is the better.

But as for certain truth, no man has known it,

Nor will he know it; neither of the gods,

Nor yet of all the things of which I speak.

And even if by chance he were to utter

The final truth, he would himself not know it;

For all is but a woven web of guesses.

The connection between fallibilism and error elimination is this. Since justification and certain proof is not attainable, the obligation to find a method that will produce certainty does not exist, and the obligation to pursue certainty ourselves without such a method is also gone. What remains is the problem of selecting among our tentative solutions, “our guesses” according to a method that is open to us. This method is error elimination through criticism of competing ideas and beliefs in light of various critical perspectives (fallible ideas themselves) we develop and use.

This whole perspective may be summarized by the concluding line of a brand new article by Deborah Blackman, James Connelly, and Steven Henderson called “Does Double Loop Learning Create Reliable Knowledge?” The Learning Organization, 11 (2004), 11-27. The line, which may take off from Xenophanes, by way of Popper’s Conjectures and Refutations, is:

” . . . what a wonderful web we weave, when first we practice to critically believe.”

This Blog and Me

In future installments of “All Life Is Problem Solving” I will use the perspective you’ve just read to treat a variety of subjects. Many of them will be in the fields of Knowledge Management and Organization Theory, where I do much of my work using an approach developed by myself and my friend and close collaborator Mark McElroy called the “The New Knowledge Management” and a normative model called The Open Enterprise. Sometimes though, I will write about Politics and Open Societies, and Physics, and Philosophy, and, as is appropriate for a blog, anything that comes into my head. Whatever I write about is likely to reflect the perspective of “All Life is Problem Solving” and that’s why I’ve given that name to my blog.

If you’d like to learn more about me, what I do, Knowledge Management in general, my collaborator, and our organizations, Executive Information Systems, Inc., KMCI), and (Macroinnovation Associates), please visit our web sites. There you’ll find lots of information about us, and lots of free papers and presentations about Knowledge Management, Enterprise Information and Knowledge Portals, and Data Warehousing. You’ll also find information about our books, both printed and electronic.