Knowledge, Diffusion of

views updated

Knowledge, Diffusion of


Diffusion suggests a drop of color in water, spreading until uniformly distributed. Knowledge, such as the cause of a disease, can diffuse until everybody knows. Diffusion is less specific than communication, and seems driven by the inherent tendency for the differentials between knowledge and ignorance to disappear, just as heat differences decay and entropy rises. But managing diffusion may be very different. Instead of knowledge moving under its own impulse, seeking equilibrium, knowledge needs to be pushed or pulled. It may be sticky, as if in a frictional medium; pressure is required to make knowledge flow. At other times, knowledge seems slippery, and one struggles to prevent leakage. Knowledge is inherently diffusible because of its nonrivalrousness, there being no loss of knowledge by those who already have it as it is diffused to others who do not.

Stickiness and slipperiness seem inherent properties of different types of knowledge. But diffusion may be more shaped by the relationship between the knowledge being moved and the context or medium through which it is moving. Stickiness implies antipathy, whereas slipperiness implies a mutuality of context, a readiness to take up and move the knowledge along, as when rumors fly. Contexts can be local, as when one department resists sharing knowledge with another, or broad, as when new political ideas spread like wildfire.

The heterogeneity of diffusion contexts is explored in network theory. Early research reflected communications technology and distinguished star or radial networks from wheel and more complex interconnections. Recent research focuses on social systems and social capital, when the context of knowledge flow is shaped by political, social, or economic power, or by interconnectedness (centrality). Horizontal and vertical knowledge diffusion imply transfer between equals versus transfer between those with different social power or degrees of centrality. Small world or scale-free structures are particularly interesting, suggesting people are closer and knowledge flows more rapid than were social networks randomly connected (Watts 2003).

But people ignore questions like what is knowledge? and what flows? (Ryle 1949). Knowledge cannot be understood simply in contrast to ignorance, for that would require a person to know ignorance too. Knowledge is revealed by contrasting types of knowledge. The explicit/tacit distinction is often invoked to explain stickiness (e.g., Szulanski 1996; Boisot 1998), presuming a normal context is sympathetic to explicit messages but less so to tacit knowledge. Yet in other contexts, explicit knowledge may be sticky and resisted, and tacit knowledge slippery. The explicit message smoking kills is well known, but apparently not in ways that greatly affect the behavior of smokers. Child abuse is often based on cycles of slippery tacit knowledge-diffusion that seem difficult to break. In short, analyses of the diffusion of the knowledges shaping human behavior must consider both the heterogeneity of the modes of knowing and of the contexts of that knowing. Thus the analysis of knowledge diffusion calls up epistemological assumptions about human beings and the world to be known, and their interaction.

For many, knowledge is about gathering data about the world out there. Data can be diffused or moved so long as the underlying notion of that world is shared and stable. Knowledge-as-data contrasts with knowledge-as-meaning, the lens that is put over data to make sense of it, to make it accessible to analysis. Meaning is about people rather than the world of facts, and diffusing it is different from diffusing data that fits within a shared meaning system. Language can only inform those who already speak it, and Ludwig Wittgenstein (18891951) argued that practice is the key to meaning. Data and meaning are mutually supporting and are tied together in information. Both are mental and imply stepping back and reflecting on ones experience of the world. But practice, in contrast, is in the world and the instant. Tacit knowledge is important because it draws attention to this third type of knowledge, knowledge-as-practice, which lies beyond data plus meaning.

Diffusing data means helping the receiver distinguish, within an agreed field of possibilities, the noted from the unnoted. But to diffuse meaning, you rely on receivers to add something of their own construction; you can only send data to another. Even more creative, to diffuse your practice, you rely on receivers to cope with the unique moment and context of their activity and construct practices that emulate yours. Thus practice is always embedded in its context and cannot be transferred to another. Diffusion means creating a new practice, guided rather than determined by prior practice.


Boisot, Max H. 1998. Knowledge Assets: Securing Competitive Advantage in the Information Economy. Oxford: Oxford University Press.

Ryle, Gilbert. 1949. The Concept of Mind. London: Hutchinson.

Szulanski, Gabriel. 1996. Exploring Internal Stickiness: Impediments to the Transfer of Best Practice within the Firm. Strategic Management Journal 17 (Winter): 2743. Special issue: Knowledge and the Firm.

Watts, Duncan J. 2003. Six Degrees: The Science of a Connected Age. New York: Norton.

J. C. Spender

About this article

Knowledge, Diffusion of

Updated About content Print Article