FormalPara Definition

Knowledge brokering is the process of ‘moving ideas from where they are known to where they are not’ (Hargadon 2002: 44).

Knowledge brokering is the process of ‘moving ideas from where they are known to where they are not’ (Hargadon 2002: 44).

A growing stream of literature argues that all innovations – in science, art, philosophy or technology – ultimately stem from a knowledge brokering process: ideas developed in one domain diffuse to domains where they are not yet known (Collins 1998; Hargadon 2002). The knowledge brokering perspective stands in contrast to theories of innovation that emphasize the role of ex nihilo creativity (Weisberg 2006: 57), resting on the opposite view that new ideas always result from the novel combination of old ideas (Schumpeter 1947; Hargadon 2002; for a discussion of ex nihilo versus combinatorial views of creativity see Perkins 1988). Similarly, it distances itself from ‘heroic’ views of innovation as born of exceptional genius, focusing instead on the processes by which knowledge diffuses across domains (Burt 2004: 387). The view of innovation as knowledge brokering spurred novel insights into the mechanisms that lead to the generation of innovations at the individual (Burt 2004), team (Hargadon and Beckhy 2006), organizational (Hargadon and Sutton 1997) and technological community level (Carnabuci and Bruggeman 2009).

Knowledge brokering provides opportunities for innovation because the social world is fragmented into pools of specialized and situated knowledge, or ‘domains’, such that it is ‘difficult to disentangle and recombine the resources from one domain into another’ (Hargadon 2002: 44; Felin and Hesterly 2007). While domains define knowledge boundaries, they are situated in social structure. Accordingly, domains are typically supported by thick pockets of densely interconnected actors, while structural holes separate such dense pockets across domains (Carnabuci and Bruggeman 2009). In line with this view, social network scholars have demonstrated that social actors bridging holes in the social structure have a ‘vision advantage’ that allows them to identify knowledge brokering opportunities invisible to others (Burt 2004; Fleming et al. 2007). While social networks are a key knowledge brokering channel, however, they are not the only one. In the knowledge-based economy a great deal of knowledge is codified and made public, for example through patents and technical literature, which often act as vectors through which knowledge can be brokered across domains. Thus, Operti and Carnabuci (forthcoming) showed that firms in the semiconductor industry broker technical knowledge by learning from the patented knowledge of their competitors. Similarly, Fleming and Sorenson (2004) showed that published scientific knowledge may help engineers broker technological knowledge from one domain to another.

The claim that the root cause of innovation is knowledge brokering received support from a wide range of studies at various levels of analyses. At the cognitive level, knowledge brokering has been argued to occur through ‘a process of analogical reasoning, in which ideas from one domain are used to solve the problems of another’ (Hargadon 2002: 45). In this literature, some researchers have taken the position that analogical reasoning represents the most important cognitive mechanism behind the generation of innovative ideas, while others have adopted the stronger position that it is the only means. At the organizational level, a rich stream of studies brought evidence that the distinguishing trait of innovative teams and firms lies in their ability to systematically broker knowledge across disparate domains, for example by using technical solutions developed in one industry to address technical problems arising in other industries (Hargadon and Sutton 1997; Fleming 2002). Systematizing this insight, Hargadon (2002) developed a general process model linking knowledge brokering to organizational innovation. The author posited that, at the organizational level, ‘knowledge brokering involves exploiting the preconditions for innovation that reside within the larger social structure by bridging multiple domains, learning about the resources within those domains, linking that knowledge to new situations, and finally building new networks around the innovations that emerge from the process’ (Hargadon 2002: 41). At the level of technological communities, furthermore, Carnabuci and Bruggeman (2009) showed that knowledge brokering is a key mechanism for understanding why certain technology domains grow faster than others. Using a large-scale patent dataset, the authors showed that the growth rate of a technology domain depends on the extent to which knowledge is brokered into that domain from other technology domains. The larger the volume of knowledge brokered into a technology domain at any given time point, the higher the domain’s growth rate in subsequent years.

While these varied works provide support to the argument that knowledge brokering is the engine of innovation, brokering knowledge across domains is rarely a smooth process (Carlile 2004). Burt identifies four levels of difficulty in knowledge brokering (Burt 2004: 355, n. 3). The simplest type is the mere communication of information about brokerage opportunities. A second, more difficult type involves transferring best practices across domains. A third type of knowledge brokering consists of making analogies between distant and apparently unrelated groups. The fourth and most difficult type is what Burt calls ‘synthesis’, where ideas are integrated across domains. A window into the difficulties inherent in knowledge brokering is presented by Sverrisson (2001), who conducted a series of in-depth case studies to examine how environmental knowledge was brought into the domain of industrial firms in Sweden. The author concluded that knowledge brokering always requires a ‘translation’ process, whereby ‘abstract categories and quantification techniques’ are turned into ‘practical knowledge’, that is, knowledge framed to solve practical problems in the new domain (Sverrisson 2001: 318). A quote from one of his interviewees captures this point well:

If you talk to a forest person and say bioenergy, he means [that is, interprets this as] a heap of chipped wood. If you talk to a bacteriologist, he means bacteria that produce hydrogen. If you talk to a mechanical engineer, he means a steam turbine in which you burn wood to produce electricity, etc. … This is the hard part, when you are talking to your contacts, and all the time you must continuously interpret what is being said. (Sverrisson 2001: 318)

Generalizing this point, Carnabuci and Bruggeman (2009: 616–617) notice that domains develop idiosyncratic ‘embedding circumstances (for example, with regard to the technical jargon, instruments, and testing criteria used)’ and that ‘brokering knowledge means de-embedding knowledge from one domain and re-embedding it in another, which entails passing more arduous cognitive and cultural barriers (Brown and Duguid 2001), and it may trigger political intricacies and irrational factors whose effects are hard to predict (Latour 1987)’. Testifying to the many difficulties inherent in knowledge brokering, Hargadon (2002: 57) noted that ‘when Edison invented the light bulb, he was accused of “the most airy ignorance of the fundamental principles of electricity and dynamics”’.

Summing up, the knowledge brokering perspective spurred many novel insights into how innovations come about, as well as into the obstacles that might prevent individuals, teams, firms and even communities from developing them. Further, it allowed researchers to integrate – or perhaps one should say broker – knowledge across literatures as diverse as sociology, organization theory and cognitive psychology. As such, knowledge brokering represents a bridging concept that promises to offer useful and innovative theoretical developments in the years to come.

See Also