Expressed simply, an agent is one who acts. The power granted or effected through that action is the quality of agency. The OED gives for “agency:” “The faculty of an agent or of acting; active working or operation; action, activity;” and for “agent:” “Acting, exerting power, as opposed to patient” (OED online, 2005). A simple enough concept when put in these terms, but the conceptual, theoretical, metaphysical and historical armature which this concept has generated and upon which it now rests is vast and anything but unified. The simple definition is not wrong; it is, rather, the matrix out of which a host of questions and complications emerge.

The specific definition of “agent” as “one who acts for another, a deputy, steward, factor, substitute, representative, or emissary” finds echoes in the definition of “medium” as “a person or thing which acts as an intermediary” (OED online, 2005). If such a convergence between agency and media is meaningful beyond the merely denotative, the significance is not that agency and media cannot be distinguished, it is that they cannot effectively be thought apart. The task of this essay is to substantiate that point of convergence, to situate theories of agency within some of their relations to media and media theory.i

Any definition of agency is complicated from the start by disciplinary differences in its conceptualization and use. On the one hand, there is a vast literature, ranging from the discipline of History to the disciplines of Artificial Intelligence and Sociology, in which “social agents” are empirical or quasi-empirical entities (see for example: Arthur et al, 1997; Castelfranchi, 1998). On the other hand, there is a body of work which understands itself to be acting within an explicitly political framework and which views agency as a tool for conceptualizing the modern subject. As such, this literature allies itself with the broader theorization of subjectivity and marginalization (Fraser, 1992, p. 16-19) and takes as its point of critical departure the liberal notion of the person as naturally (“inalienably”) endowed with the power to act, whether that power is conceptualized as agency or “natural rights” or democracy (Warner, 2002; Macpherson, 1962; Skinner, 1978). In the middle, or to one side, is another vast literature: the philosophical work on autonomy, agency and free will, which tends to understand agency as a self-reflexive relationship–a relation of self to self–and to take as its central, motivating task the clear differentiation of free will from coercion (Buss, 2002).ii

We can begin to think media and agency together by specifying what is meant by action. The mise-en-scène of action has two main elements or aspects: 1. the action itself, and 2. the context for action. Additionally, the resultant action is experienced not only by the actor, but by witnesses, any of whom might differ in their evaluation of the efficacy of that action (i.e. its agency).iii Which is to say, embedded within the concept of agency is always an assessment of efficacy, and within any scene for action, there are people who are more and less empowered to make that assessment. Some acts are seen to be effective, and those we grant agency; some acts are seen to be ineffective (e.g. coerced) and those appear to lack agency. If an action is seen not to be efficacious, then there is a source or site of blockage, about which we can say that agency, in that situation, rests there, at the point of resistance. Thus, the conditions for acting effectively, the possibilities for being an agent, change over time (historically) as the conditions for action change. To put it another way, the notion of agency is radically relational (i.e. contentious), and difficult to think apart from concepts like individuality, society, culture, subjectivity, experience, narrative, embodiment, power.iv With reference to Michel Foucault’s work on the last term in this series, we might also say that agency is a theory of power; in any case, it is a concept whose meaning is not intelligible outside of a field of power relations (Foucault, 1982). The relevance of this concept for media lies broadly in the fact that media and theories of media (whether understood as a material entity–e.g. the radio–or the Media, or a theoretical concept referring to mediation) register changes to all of the above conditions of possibility for agency — the conditions of possibility for being or becoming someone whose actions matter.

Heidegger’s “Age of the World Picture” offers an epochal theory of agency: it defines the essence of the modern age by a new form of agency resident in that age, an agency which Heidegger defines in relation to new media and new technologies of representation (Heidegger, 1977). Heidegger’s claim is that the world as such–the world in its essence–has become a picture because it has become that and only that which human subjects perceive and measure. In other words, the world now exists only as an effect of the twinned acts of representation and spectatorship. This places human subjects centrally in relation to the world; in effect, human agency engenders the world as such. But here, agency lies less with the individual than with a kind of collective agent (the set of all scientists, all observers, all subjects) and the effect is not necessarily to make individuals more powerful.

Marshall McLuhan also conceives an epochal theory of media and agency, but runs the lines of force in the opposite direction: here, technology acts upon human agents by “extending” our bodily capacities (e.g. perception, cognition), altering our minds and bodies. Agency for McLuhan is predicated on self-recognition; in order to act effectively, we first must clearly perceive the changed conditions of our own existence. In Understanding Media, McLuhan argues that one of the dangers of new media is to blind us to the very ways that media impact our conditions of existence, which are the conditions of possibility for effective action (McLuhan, 1964).

This theme of recognition, or self-understanding, is a trope one finds in discussions of both media and agency. Recognition here refers to the idea that people cannot control the conditions of their existence until they learn to clearly perceive those conditions, a clarity which media and technology are thought either to hinder or to enhance, although the emphasis is commonly on the former (e.g. in critical or pessimistic or techno-phobic accounts). Althusser’s theory of ideology is one of the most influential formulations of the relationship between agency and recognition. It posits ideology as a form of self-misrecognition, a blindness to one’s own material conditions (Althusser, 1971). In a similar spirit, but with a clearer interest in media, Baudrillard claims that contemporary media, which appear as new communicative technologies, offer only the illusion of true communication. We can understand “true” communication here as effective speech acts (speech which offers the possibility of reciprocation) and thereby, as a form of agency (the capacity to speak meaningfully, efficaciously). Baudrillard argues that contemporary media erase the possibility of agency by erasing the possibility for people to perceive that they do not have agency (Baudrillard, 1981).

If agency is action, what specific forms of action are most directly implicated in theories of media? For McLuhan, understanding itself is a form of agency: the capacity to clearly perceive the conditions of our own lives, and specifically, the changes wrought on our bodies by media. Baudrillard makes a similar claim when he argues that true or “reciprocal” communication, another form of agency, is rendered impossible by our inability to perceive that we lack the power of effective, political speech, and that media themselves rob us of this capacity. But in all of the above cases, agency is something which, paradoxically, appears to be granted from the outside, by media or by technology or by public intellectuals. Against the evidence of our own experience, we are often told that we lack agency, or alternatively, that we possess more agency than we think we do. This attributive power is what many people consider to be critical aspect of critical theory. Agency, however, is also an experience that one has of oneself: one feels empowered or disenfranchised in certain contexts for action. Whatever McLuhan says about technologies making users numb, or what Baudrillard says about media blocking true communication, new media and new technology are often experienced as empowering, as granting their users new capacities for action, for speech, for communication, and for throwing off old forms of domination (e.g. Turkle, 1995). And even if agency is never quite a fully self-determined relation, its self-determined or self-assessed (phenomenological) form at least sits in productive tension with its externally-determined forms. Deleuze provides a philosophical grounding for this conceptual shift with his theory of the simulacrum, which reverses the polarity of Platonism by installing difference at the center of thought and meaning (Deleuze, 1990). This removes the conceptual foundation that theorists like McLuhan and Baudrillard rely on to distinguish true from false (or illusory) experiences of agency. On Deleuze’s view, the central question shifts from how do we discern true from false agency to how can we engender and value as many manifestations of agency as possible, a shift which prioritizes agency as a self-evaluated or self-determined experience.

Nancy Fraser describes this tension between structural constraints (external limits placed on one’s capacity to act) and agency (defined as the self-determined power to act) as especially relevant to feminist politics, and differentiates feminist theorists according to where they lay the stress of their analyses: with the individual agent, assumed to be a site of resistance, or with the limiting structure within which the individual is forced to live (Fraser, 1992). It might seem that an optimistic politics would side with the individual while a more pessimistic politics would attend to hegemonic structures which oppose the individual, but it may be more helpful to notice how agency cannot exist (as a concept or as a potential for action in the world) without a paired force of opposition. Foucault puts it this way, in what amounts to one of his strongest claims for the power or agency of individuals: “…there is no relationship of power without the means of escape or flight” (Foucault, 1982, p. 225).

Despite the various ways in which the literature cited above complicates the liberal notion of the self as endowed with agency in the form of “natural” rights, in all of this literature the (effective) agent is still understood as the powerful self, the assertive self, the self who is able to impose her will, if only over the terms and conditions of her own life. Baudrillard is an exemplar: for him, to be a passive receiver is to give up all possibility of power, and even the capacity to recognize whether, in the first instance, one is or is not powerful or agential (Baudrillard, 1981). But the experience of selfhood is never simply one of absolute enfranchisement versus absolute disenfranchisement, agency versus the utter lack of agency. The experience of agency is graded and responsive to shifting conditions; in certain extreme cases, even the experience of powerlessness might be experienced as powerful or agential. Some queer theories of agency have questioned the implicit value that is granted to assertiveness, to assertive power as such, and have thereby contributed to a far more agile conception of agency (Bersani, 1993). This is especially relevant to theories of media because, if one of the effects of new media and new technology is to change the conditions for subjecthood, as most theorists of media argue that it does, then we might also expect a radical change in the conditions under which we experience and understand agency.

Kris R. Cohen


i Because the objective here is a critical definition of agent/agency, what follows will necessarily be biased towards theories which conceptualize agency as such, rather that simply assume its stable and self-evident existence. This, then, places the emphasis on self-consciously political theories of agency, as these have done the most work to conceptualize what agency is and how it works.

ii A separate literature has recently extended the notion of agency to apply beyond people to things. A collection of writings and writers, loosely affiliated under the name Actor Network Theory (ANT), have been concerned to account more fully for the role played by things-and especially by media and technology as things, e.g. scientific instruments, writing implements, war technologies, and various media-in the formation, experience and transformation of society (Law and Hassard, 1999).

iii Aristotle, in the context of Physics and the study of motion, particularizes the scene of action slightly differently. He specifies the “agent” (active) and “patient” (acted-upon), where both elements are, in their own ways, active, and where both contribute to the overall effect of “Motion.” The field of linguistics provides a third characterization: it identifies the scene of action as a sentence, and introduces a third term: the agent is the subject, the patient is the object, and the “instrument” is whatever is used to perform the action. The instrument here is a kind of medium: it is what mediates between the agent and the patient, subject and object, actor and acted-upon.

iv My thanks to Abigail Baim-Lance for help on this point.


Althusser, Louis. “Ideology and Ideological State Apparatuses.” Lenin and Philosophy and Other Essays. London: New Left Books, 1971.

Aristotle. Physics. Trans.: Richard Hope. Lincoln, NE.: The University of Nebraska Press, 1961.

Arthur, W. Brian, Steven Durlauf, and David Lane, eds. The Economy as an Evolving Complex System, II. Reading, MA.: Addison-Wesley, 1997.

Baudrillard, Jean. “Requiem for Media.” For a Critique of the Political Economy of the Sign. St. Louis, MO.: Telos Press, 1981.

Bersani, Leo. “Is the Rectum a Grave?” AIDS: Cultural Analysis, Cultural Activism. Ed. Douglas Crimp. Cambridge, MA: The MIT Press, 1993.

Buss, Sarah. “Personal Autonomy.” Stanford University. 2002 .

Castelfranchi, C. “Modelling Social Action for AI Agents .” Artificial Intelligence 103.1 (1998): 157-85.

Deleuze, Gilles. “The Simulacrum and Ancient Philosophy.” The Logic of Sense. Ed. Constantin V. Boundas. New York, NY: Columbia University Press, 1990.

Foucault, Michel. “Afterward: The Subject and Power.” Michel Foucault: Beyond Structuralism and Hermeneutics. Ed. H. Dreyfus and P. Rabinow. Chicago, IL.: University of Chicago Press, 1982.

Fraser, Nancy, and Sandra Lee Bartky, eds. Revaluing French Feminism: Critical Essays on Difference, Agency and Culture. Bloomington, IN.: Indiana University Press, 1992.

Heidegger, Martin. “The Age of the World Picture.” The Question Concerning Technology and Other Essays. 1977.

Law, John, and John Hassard, eds. Actor Network Theory and After. Oxford, UK: Blackwell Publishing, 1999.

Macpherson, C. B. The Political Theory of Possessive Individualism. Oxford, UK: Oxford University Press, 1962.

McLuhan, Marshall. Understanding Media: The Extensions of Man. New York, NY: Signet Books, 1964.

Simpson, John, ed. Oxford English Dictionary. Oxford, UK: Oxford University Press, 2005.

Skinner, Quentin. The Foundations of Modern Political Thought, 2 Vols. Cambridge, UK: Cambridge University Press, 1978.

Turkle, Sherry. Life on the Screen: Identity in the Age of the Internet. New York, NY: Simon and Schuster, 1995.

Warner, Michael. “Chapter 1: Public and Private.” Publics and Counterpublics. New York: Zone Books, 2002.