John Law: Introduction: monsters, machines and socoptechnological relations

s. 18:

Once we understand that entities and their relations are continuous; once we understand (as sociology does not) that they are heterogeneous; once we understand that the differences and distributions that are drawn between them could be otherwise; once we understand (as STS finds difficult) that their histories and their fates vary widely; then we will come to appreciate that we are all monsters, outrageous and heterogeneous collages.

Michel Callon: Techno-economic networks and irreversibility

s. 132:

Technology rarely grows in a predictable and unilinear manner within a relatively stable social and industrial context (Foray 1989). Models which assume this cannot explain its radical, and sometimes revolutionary, character. Instead, the new sociology and economics of technology suggest that science and technology are a product of interaction between a large number of diverse actors

s. 134:

Unlike economics, sociology does not start with a stylized image of the actor. Instead it assumes that actors are only intelligible when they are inserted into a common space which they have built themselves. For instance, Crozier and Friedberg (1977) speak of actors and systems, Bourdieu (1980) of agents and fields, and Parsons (1977) of roles and functional prerequisites. In their different ways sociologists thus assume that every actor contains a hidden but already social being: that agency cannot be dissociated from the relationships between actors.

Economists teach us that interaction involves the circulation of intermediaries. Sociologists ‘teach us that actors can only be defined in terms of their relationships. But these are two parts of the same puzzle, and if we fit them together we find the solution.

/…/

Technical objects as networks

Skills as networks

s. 139:

In the way I initially want to define the term, an ‘actor’ is any entity able to associate texts, humans, non-humans and money. Accordingly, it is any entity that more or less successfully defines and builds a world filled by other entities with histories, identities, and interrelationships of their own. This initial definition suggests that intermediaries are synonymous with actors. For instance, a scientific text seeks to create a reader with the skills needed to mobilise, consolidate, or transform the network described in the paper. Thus it acts: it is an actor.18 And the same is also true for other intermediaries. Like intermediaries, actors may be hybrid. They may but need not be collectivities. They may take the form of companies, associations between humans, and associations between non-humans.19 In this ontology actors have both variable content and variable geometry.

s. 142:

All groups, actors and intermediaries describe a network: they identify. and define other groups, actors, and intermediaries, together with the relationships that bring these together. When such descriptions include an imputation of authorship, then actors emerge in the stopping places, symmetries, or folds (Deleuze 1989). But the network of intermediaries accepted by an actor after negotiation and transformation is in turn transformed by that actor. It is converted into a scenario, carrying the signature of its author, looking for actors ready to play its roles. For this reason I speak of actor-networks: for an actor is also a network.