Recently, economic sociologists have tended to view markets as embedded in social relations and social networks, the structures they see as defining markets and framing economic action
(e.g. White 1981, 1993, 2002; Baker 1981, 1984; Baker et al. 1998; Granovetter 1985; Swedberg and Granovetter 1992; Swedberg 1994, 1997; Burt 1983, 1992; DiMaggio and Louch 1998; Uzzi 1997, 1999; Podolny 2001).
This chapter draws a distinction between two types of markets: those based on a network architecture where social relationships carry much of the burden of specifying market behaviour and of explaining some market outcomes, and markets that have become disembedded and decoupled from networks and exhibit what I shall call a flow architecture.1 As illustrated elsewhere (Knorr-Cetina and Bruegger 2002a, 2002b), flow architectures are “microstructured” rather than simply network/relationally-structured. They are more richly structured than the relational vocabulary allows for, and display patterns of coordination and behaviour that are global in scope and microlevel in character. As Fligstein notes (1996: 657), networks are sparse social structures, and it is difficult to see how they can incorporate the patterns of intense conversational interaction, the knowledge flows, and the temporal features observed in some areas of practice. Though flow architectures may include networks, these networks are not the salient structuring principle of a global microstructure. Flow architectures, I shall argue in this chapter, also involve global “scopic” systems that project market reality while at the same time carrying it forward and allowing it “to flow”. The suffix “scope”, derived from the Greek “scopein”, to see, when combined with a qualifying notion means an instrument, etc. for seeing or observing, as in “periscope”. Social scientists tend to think in terms of mechanisms of coordination, which is what the network notion stands for; a network is an arrangement of nodes tied together by relationships which serve as conduits of communication, resources, and other coordinating instances that hold the arrangement together by passing between the nodes. Cooperations, strategic alliances, exchange, emotional bonds, kinship ties, “personal relations”, and forms of grouping and entrenchment can all be seen to work through ties and to instantiate sociality in networks of relationships. But we should also think in terms of reflexive mechanisms of observation and projection, which the relational vocabulary does not capture. Like an array of crystals acting as lensesFrom pipes to scopes 123 that collect light, focusing it on one point, such mechanisms collect and focus activities, interests, and events on one surface, from whence the result may then be projected again in different directions. When such a mechanism is in place, coordination and activities respond to the projected reality to which participants become oriented. The system acts as a centring and mediating device through which things pass and from which they flow forward. An ordinary observer who monitors events is an instrument for seeing. When such an ordinary observer constructs a textual or visual rendering of the observed and televises it to an audience, the audience may start to react to the features of the reflected, represented reality rather than to the embodied, pre-reflexive occurrences. In the financial markets studied the reflexive mechanism and “projection plane” is the computer screen; with the screen come software and hardware systems that provide a vast range of observation, presentation, and interaction capabilities sustained by information and service provider firms. Given these affordances, the pre-reflexive reality is cut off and replaced; some of the mechanisms that we take for granted in a lifeworld, for example its performative possibilities, have been integrated into the systems, while others have been replaced by specialized processes that feed the screen. The technical systems gather up a lifeworld while simultaneously projecting it. They also “apresent” (bring near, see Schutz and Luckmann 1973) and project layers of context and horizons that are out of reach in ordinary lifeworlds - they deliver not only transnational situations but a global world spanning all major time zones.
As I shall argue below, they do this from trading floors located in global cities (Sassen 2001) which serve as the bridgehead centres of the flow architecture of financial markets. Raised to a level of analytic abstraction, the configuration of screens, capabilities, and contents that traders in financial markets confront corresponds to a global reflex system, or GRS, where R stands for the reflexively transmitted and reflex-like (instantaneously) projected action and other capabilities of the system and G stands for the global, scopic view and reach of the reflex system. For the present purpose, which is that of distinguishing between forms of coordination relevant to understanding markets, the term is intended to denote a reflexive form of coordination that is flat (non-hierarchical) in character while at the same time being based on a comprehensive summary view of things - the reflected and projected global context and transaction system. This form of coordination contrasts with network forms of coordination which, according to the present terminology, are pre-reflexive in character - networks are embedded in territorial space, and they do not suggest the existence of reflexive mechanisms of projection that aggregate, recontextualize, and augment the relational activities within new frameworks that are analytically relevant to understanding the continuation of activities. With the notion of a GRS system, I am offering a simplifying term for the constellation of technical, visual, and behavioral components packaged together on financial screens that deliver to participants a global world in which they can participate on a common platform, that of their shared computer screens. On a technological level, the GRS mechanism postulated requires that we must understand as analytically relevant for a conception of financial markets not only electronic connections, but computer terminals and screens - the sorts of teletechnologies (Clough 2000: 3) that are conspicuously present on trading floors and the focus of participants’ attention - as well as the trading floors themselves, where these screens cluster and through which markets pass.In the following, I begin with an analysis of financial markets as focused upon computer screens as the centrepieces of such a GRS form of coordination. I will also briefly sketch the historical innovation and emergence of the relevant systems in the 1970s and 1980s and point out how they led to a replacement of network markets. I address some temporal features of the foreign exchange market which I take as my exemplifying case. A flow architecture, I shall argue, results from a combination of these temporal features with the GRS form of coordination, clustered in time-zone-specific bridgehead centres.