Every once in awhile someone in Knowledge Management will state the view that organizational knowledge is the set of beliefs on which there is an organizational consensus. This viewpoint fits with constructivist approaches to knowledge because, according to the constructivists, knowledge comprises the belief filters we humans create to interact with the world, and then label with the honorific, “knowledge.” And organizational knowledge is just the subset of these beliefs that humans in organizations agree should have that label. There are lots of difficulties with this view. I’ll review some of them in the rest of this post
First, many in KM accept the classical definition of knowledge as Justified True Belief (JTB), including Nonaka and Takeuchi in their noted book presenting the very popular SECI Model. This definition represents a problem for the constructivist view because in the JTB view, knowledge is a belief that must be justified as certainly true by an appeal to another belief that is certainly true. Mere consensus, however, is not such a belief. It’s just a fact about what people have created and believe. It certainly doesn’t justify a belief in the sense that it proves that it is true. So, from the point of view that knowledge is JTB, the idea that knowledge is mere consensus is clearly unacceptable.
Second, another point of view on knowledge is that, since all our beliefs are fallible, there is no belief that can be justified in the classical sense of JTB; but our various beliefs can often be “justified” relative to certain foundational beliefs that are accepted for purposes of discourse and inquiry as beyond question. So, in this view, we can work with a relativistic notion of knowledge as JTB. This conception of knowledge may be consistent with the consensus knowledge outlook provided that the foundational belief in one’s system is that a belief is knowledge as long as a consensus that it is knowledge exists. However, for those who believe that the foundational beliefs ought to be about observation, or expert opinion, or what one’s manager agrees with, or what a religious authority thinks, or what one’s intuition is, the view that “knowledge” is belief justified relative to a consensus among individuals in a collective is not acceptable.
Third, there’s the view of knowledge that says that no belief can ever be justified, but that knowledge is belief that has survived our criticisms, tests, and evaluations (our experience) over time and in the context of alternatives. According to this view, mere consensus on certain beliefs doesn’t define knowledge because people and communities can insulate their beliefs from experience in various ways that preclude criticism, testing, or evaluation. So, the resulting beliefs are not knowledge, but only “propped up” beliefs. This cuts very seriously in the many human situations where an individual is told to act in a particular way because that way is prescribed by organizational knowledge, and the individual knows very well that the so-called “knowledge” is just “propped up” belief that the organization has decided to insulate from experience and uphold come what may.
This third view on knowledge doesn’t preclude the fact of organizational knowledge being “consensus knowledge” in many specific contexts. According to it however, this “consensus knowledge” is not “knowledge” because of the existence of consensus. Rather consensus has formed out of individuals deciding that organizational knowledge processes that have subjected the beliefs in question to experience have produced a record showing that the “knowledge” has survived better than other conceivable alternatives. In other word, in this view, consensus is sometimes a by-product of the emergence of organizational knowledge; but in no way justifies the beliefs at issue as “knowledge.”
It’s this third view of knowledge that I accept. To the fact of consensus on knowledge, I have no intrinsic objection. What I do object to is the use of an existing consensus to claim authority and justification for that knowledge, as if consensus is or ought to be one of our criteria for critically evaluating knowledge claims or for deciding what we, individually, should accept as knowledge.
Using consensus as a validation criterion for knowledge claims continuously leads us into great difficulties. For example, appeals to consensus led Congress to support the Wars in both Iraq and Vietnam. Appeals to consensus, led us to accept rather uncritically the religion of de-regulation and market fundamentalism that has led us to the current economic slaughter. Consensus at NASA on the idea that there was no need to track down the causes of deviation in Columbia’s performance from expectations, was one of the factors that led to the Columbia disaster. I could, of course, keep on pointing to cases where appeals to “knowledge” justified by consensus served to short-circuit much needed criticism and evaluation that may have resulted in avoiding disasters. There’s a very good reason why philosophers through the ages have avoided definitions of knowledge as belief consensus. And that reason is that the term “knowledge” is meant to signify a special kind of belief that is of higher quality than competing beliefs, and one of the most clear things emerging from historical experience, obvious even as early as the Presocratic Greeks, is that belief consensus neither defines nor is a measure of belief quality.
So, it’s not that consensus knowledge doesn’t exist. Rather, it is that it can be wrong and very often is. Like other knowledge it is fallible, and because it is always fallible we need always to treat it just like other knowledge claims” always to criticize it, always to evaluate it, always to test it, and always to challenge it. We need to do that to learn its weaknesses and to learn about where it is likely to fail us. Once we know about that, we can know when to avoid using it in practice and when we have to develop newer and better knowledge. By the way, in case you think the above approach is impractical, it’s the approach George Soros followed to amass his substantial fortune and to predict the collapse of the great real estate bubble of the past few years.
Finally, there is a major difficulty with the very idea of “consensus knowledge” as “organizational knowledge” that shouldn’t be ignored. Generally, consensus knowledge is viewed as agreement of the individuals in a collective that a belief or set of beliefs is “knowledge.” However, the mere fact of such agreement doesn’t mean that the belief on which there is agreement is organizational knowledge. The reason is that organizational knowledge is a global, collective property of organizations emerging from the interaction of the individuals comprising. The idea that collective knowledge of the organization is the mere aggregation of agreement among its individuals is an unwarranted inference (sometimes called the “aggregative fallacy”) that fails to take into account the effects of social organization, power structure, and social interactions on the emergence of collective knowledge. Generally, organizational knowledge is not what people agree upon. It is what emerges from social interaction, and it usually varies quite a bit from belief consensus.