Conclusion

The frictionless paradigm in the Internet research literature predicts the elimination of hierarchies and the spread of new network-based forms of governance, based on the assumption that the Internet will diffuse access to information and relationships and the new transaction-cost-free networks will foster both dynamic exchanges and cooperation. We have challenged this view providing secondary and primary empirical evidence for new hierarchies created in networked social systems, and giving theoretical justifications for these empirical findings.

New hierarchies in network society are created out of the cognitive complexity of the systems, coupled with human bounded rationality and bounded sociability. Another complex effect comes from the contemporary impact of the network technologies on the economies of the infomediary-content infrastructures and the economies of relationships, such as economies of scope and networks externalities. The resulted new social organizations, however, are not simply more hierarchical, because the increased interconnection and the dynamics of social capital also decrease cognitive hierarchies.

Networks are not simply more democratic and socially rich alternatives to hierarchies, coordinated through trust instead of price or authority. They are complex systems in which different coordination mechanisms and different types of hierarchies are present. Hierarchies are not eliminated because transaction costs are not eliminated. Also in a network economy we are rationally bounded, and we can also find the same scarcity effect in sociability processes. Transaction costs, and the need for cognitive and social hierarchies, might be reinforced by the uncertainty and the complexity of the new social systems. The real novelty of the new social and organizational networks does not seem to be the different degree of hierarchy, but the different degree of interconnection (complexity) and flexibility (dynamism of connections) of the systems.

The social side of these transformations is not captured by the approaches to network society that emphasize deterministic and radically liberal views of the relationship between technology and social change. On the social side of these transformations, Powell (2001) sees a polarization of society groups and emerging a "winner take all" ethics, "in which success creates increasing returns; that is, the capabilities, skills, and experiences of those who have prospered rebound such that they are vastly better positioned and qualified than those left behind. This reinforcing cycle is virtuous for the winners, vicious for the losers." Analyzing the negative social effects of the flexibility of networked organizations, at the expense of a downsized workforce, Harrison (1994) sees a "leaner, but also meaner" social organization emerge from these changes. Calling this phenomenon concentration without centralization, Harrison argues that firms that have mastered global network production couple a core-ring structure (characterized by a privileged solid center connected to a disadvantaged, and overexploited networked periphery) with the use of computerized technology for coordinating flexible intra-firm and inter-firm networks. In his view the governance mix comprises both cooperation with the strategic core of employees and partners, and domination of the powerless periphery of the network.

But Powell (2001) does not think that this unequal and hierarchic outcome of the network revolution is intrinsic to the network society. It is an option, which is important to be aware of, but not automatically emerging from the distributed capitalism. He writes: "... it is wrong, I believe, to argue that the new system is just a kind of decentralized Fordism or a wolf in sheep's clothing. We are undergoing a period of 'creative destruction,' in which the established practices of one regime are being replaced by new ones. ... but, it is not at all clear that these inequalities are a necessary component of the new form" (p. 29). Social order and social organization are political outcomes, not technological recipes. Technology can foster flexibility, but also coordination. It can help periphery to be connected with other peripheries and with the center, without loosing flexibility. It is a matter of design of the network; it is a matter of negotiated structures of the networks.

The network society can be a more hierarchical society or a more democratic one, depending on the social and policy decisions of individuals and groups, since also in the digital world there are trade-offs and alternatives, shaped by economic and social constraints and opportunities. Therefore also in the digital society freedom and quality of life are built out of ethical and policy agendas and arenas.

We accept the call of Rullani (1998) to challenge the fundamentalist idea that technology liberates the invisible hand of markets in societies, beyond the need for collective choice and action, and Rullani and Vicari's (1999) call for a more complex understanding of the complex issues posed by the exploding connectivity in societies. With them, however, we do not propose a conservative view of the impact of new communication technologies on society and economies. We also believe that network technologies can open very powerful opportunities for change and improvement of the quality of individual life and societies.