Strength Of Weak Ties

Granovetter's 1973 paper in the American Journal of Sociology pioneered the use of actor-based [Page 635]social networks for linking the micro and macro levels of sociological theory. It introduced the idea of the strength of weak ties, that is, personal relationships (“weak ties”) are often more important than institutional links (within an organizational hierarchy) in getting things done. Many of the networks he and later writers analyze are related to studies of power; these range from very large networks, concerning mass action or social movements, to policy-forming networks consisting of only tens of elite actors. Granoveter's original article offers support for many intuitive and counterintuitive notions of power in networks by exploring the importance of weak links that connect one part of a network to another part.

The importance of social networks in understanding power is largely related to the transmission characteristics of the network. Whether the transmission involved is that of influence, information, or other tangible or intangible resources, the formal analytical questions are similar. The strength of weak ties offers an insight into the reachability of actors in the network and, crucially, the length and route of the path from one actor in the network to another.

Granovetter realized that in many network studies it is the strongest ties that are considered crucial to understanding the nature of the network. He disagreed for two key reasons: first, the tendency of social networks to be clustered (one is more likely to know the friends of one's friends than an actor drawn at random), and second, the implications of clustering for the distribution and strength of the links. He concluded that weak links are in fact extremely important in connecting clusters of actors through shorter and more reliable paths than a route that consists only of strong links.

The work is based on a simplifying assumption that, given any three actors, if two are joined to the third by a strong link there will be at least a weak link between these two. In friendship terms, this would mean that if B and C were both friends with A, then B and C must be at least acquaintances. From this simplification, the conclusion is reached that no strong tie is a direct bridge and that strongly connected, clustered subnetworks will always be linked to each other by weak ties. More precisely, for large networks there may be a long convoluted path around the network, but the weak ties provide a local bridge.

The weak ties must then be considered important because with transmission along longer paths there is always a cost, whether real costs such as time, money, and so on, or costs in terms of risks of distortion in transmission. In power studies, for example, this distortion of transmission could relate to a lack of precision, which would have implications for the results of influence.

The links between cliques can be illustrated by the example of a mass action network creating conditions suitable for a movement to succeed. The utilization of strong ties, perhaps with friends or work colleagues, may never provide a bridge between clusters based on age, ethnicity, faith, or similar factors. Weak ties, such as working or having attended school together, will be much more likely to bridge these clusters. Evidence is cited by Granovetter from some of the experiments in small world theory that attempt to get a message delivered from actor to actor using the shortest chain possible.

...