Chapter 10: Artifacts, Agency, and (A)symmetry
Contents
Artifacts, Agency, and (A)symmetry
(A)symmetry
Agency
Artifacts
Conclusion
Artifacts, Agency, and (A)symmetry
Actor–network theory’s attention to the agency of things comes at a good time in this age of smart machines. Modern technology behaves independently and flexibly in ways that traditional tools do not. An e–mail program, for example, can save the user the trouble of looking up and typing an e–mail address if just one message has been sent to someone and the user can remember as little as the first initial of the person’s first or last name. Even the humble fax machine has remarkable capabilities. Rae (1994) analyzed the way fax machines correct each other’s garbled transmissions. He suggested that the machines’ actions in making corrections resemble repair in human conversation (Schegloff, Jefferson, and Sacks 1977). Rae asked, “Does the capacity for repair imply that fax machines have agency?” We believe the answer is yes. The fax machines’ sophisticated ability to decode incomplete or corrupted text with no human intervention is evidence of a capacity to respond intelligently to changing conditions.
This chapter draws on the work of Andrew Pickering (Pickering 1993, 1995) as well as other actor–network theorists (Callon and Latour 1981; Callon 1986; Rose, Jones, and Truex 2005; Shaffer and Clinton in press). We argue that actor–network theory is right to direct attention to the agency of things, and that traditional accounts (e.g., Emirbayer and Mische 1998; Wertsch 1998) in which it is “only people who are doing the acting” (Shaffer and Clinton in press) are not sufficient. At the same time, we explain why we reject actor–network theory’s perfect symmetry. Using the principles of activity theory, we propose a new way to conceptualize different forms of agency.
(A)symmetry
Pickering (1993, 1995) developed a concept of symmetrical agency based on a dialectical notion of back and forth between the human and material worlds. He observed that traditional sociology sets forth “an asymmetric distribution of agency — all to human beings, none to the material world” (Pickering 1993). Pickering found such a notion absurd; humans cannot just effect whatever they want to, with no response from the material world. If ones wishes to send an e–mail but the server is down, the e–mail cannot be sent. Pickering concluded that a completely lopsided asymmetry is untenable.
What about perfect symmetry, as in actor–network theory? As Pickering (1993) noted, actor–network theory “insists there is no difference between human and non–human agents: human and nonhuman agency can be continuously transformed into one another.” As we did in chapter 9, Pickering found this problematic:
We humans differ from nonhumans precisely in that our actions have intentions behind them, whereas the performances (behaviors) of quarks, microbes, and machine tools do not ... . I find that I cannot understand scientific practice without reference to the intentions of scientists, though I do not find it necessary to have insight into the intentions of things ... . We construct goals that refer to presently nonexistent future states and then seek to bring them about. I can see no reason to suppose that DNA double helices or televisions organize their existence thus ... . (Pickering 1993)Activity theory and Pickering reject perfect symmetry for the same reason: the existence and importance of human intentions.
Pickering used Galison’s (1985) description of research in elementary–particle physics as empirical grounding for his discussion of agency. As Pickering recounted, in the early 1950s, physicists were interested in rare but interesting particles called strange–particles. The particles proved resistant to efforts to photograph them, so research was stalled. Donald Glaser, a scientist at the University of Michigan, became interested in the problem and invented a device called a bubble chamber that revealed the particles. He later won the Nobel prize for this research. Glaser began work on the bubble chamber by attempting to modify existing cloud chambers used to study other kinds of particles. The cloud chambers utilized vapor that failed to reveal the strange–particles. Glaser’s early designs with the cloud chambers failed, but he learned from each failure, eventually discovering that if he filled a chamber with a superheated liquid under pressure it would record strange–particle tracks.
Pickering saw these events as a symmetrical “dialectic of resistance and accommodation” between two agencies: the material agency of the chambers and Glaser’s intention to produce a chamber that tracked strange–particles. Pickering proposed that the two forms of agency were “constitutively enmeshed in practice by means of a dialectic of resistance and accommodation.” That is, the material world offers up resistance and humans find ways to accommodate it, to work around it. Pickering felt that by organizing material and human agencies as constitutively enmeshed with one another, he could rescue symmetry, allowing him to “return to the fold” of actor–network theory. With the dialectic, he said, the “picture [could] be symmetrized” (Pickering 1993).
This argument, however, presents a contradiction. It asserts the importance of human intention, on the one hand, and then attempts to symmetrize things on the other. Pickering said that human and material agencies are enmeshed in practice, so he defined material agency in terms of human practice (which in turn is shaped by human intentions). In Pickering’s formulation, material agency is intelligible only with respect to human practice. As he said: “I argue that material agency is ... emergent in relation to practice”; “material agency is sucked into the human realm”; and “the resistances that are central ... are always situated within a space of human purposes, goals and plans” (Pickering 1993). These notions (resonant with activity theory) cannot logically be squared with a principle of generalized symmetry.
With respect to the events surrounding the invention of the bubble chamber, activity theory would suggest a history of human agency overcoming material obstacles. To say that Glaser “accommodated” the material realities is to miss the point that he kept going until he found a way to produce the strange–particle tracks. It was his passion to build a particular kind of chamber that enabled him to “circumvent the obstacles he had ... encountered” (Pickering 1993).
In activity theory terms, Glaser’s object persisted, but his actions changed to meet the resistance of the materials. Pickering’s account revealed as much, but the rhetoric of symmetry made him lose sight of the power of Glaser’s intentions and their ability to successfully meet obdurate material resistances. Miettinen’s discussion of the resistance of the wood chips in the production of ethanol is not unlike Pickering’s analysis of the invention of the bubble chamber, but Miettinen was able to use activity theory to clarify the relationship between the scientists’ objects and their actions. Miettinen explained how the scientists’ changing understandings of the wood chips permitted them to eventually find an economic use for the chips, as they continued to explore the particularities of the resistances of the wood chips within their scientific practice.
To see the relationship between material and human agency as simply one of resistance and accommodation is to miss the role of human creativity in arenas such as science.
To see the relationship between material and human agency as simply one of resistance and accommodation is to miss the role of human creativity in arenas such as science. Glaser was able to design and build a chamber that worked. Pickering seemed nonplussed by Glaser’s creativity, observing, “One can speak of scientific creativity ... that along the way [Glaser] hit on the idea that led him to the bubble chamber, and so on.” But Glaser did not “hit upon” the idea; he searched steadfastly for it, to fulfill the object of tracking the strange–particles.
Pickering argued that the dialectic of resistance and accommodation “amounts to ... a revision of plans and goals, a revision of the intentional structure of human agency.” In activity theory terms, the actions and their associated goals change to meet the resistance of the material world. But the object has not changed. Glaser is still doggedly laboring to build what finally becomes the successful bubble chamber.
As we saw in chapter 7, science studies accounts concern themselves with actions, missing the longer trajectories of objects that shape the selection of particular actions. So, while the material world did indeed provide significant push–back, forcing Glaser to keep trying till he got the bubble chamber right, his intentions and creativity are evidence of asymmetrical agencies at play. We do not think it can be said that he “accommodated” the material world; rather he manipulated the material world until he got what he wanted.
Material agency as formulated in actor–network theory makes logical sense only with respect to a human activity. Vapor is not in itself “resistant.” It is so only when it fails to behave as some human desires. Material things are not inherently, essentially resistant (or empowering, or any other quality). We describe them as such during the enactment of a particular human activity. Both a subject and a material object can potentially manifest an infinite number of properties under varying conditions. The particular properties of interest come to light in the whole context of an activity — which is oriented by a human–defined object.
We do not of course mean to suggest that humans can overcome any obstacle if they just persevere long enough. That is preposterous. We mean that humans have certain powerful resources at their disposal that must be accounted for theoretically.
We do not of course mean to suggest that humans can overcome any obstacle if they just persevere long enough. That is preposterous. We mean that humans have certain powerful resources at their disposal that must be accounted for theoretically. So potent are these resources that the Earth — a system billions of years in the making — has been drastically altered in only a few centuries of human activity. Vitousek and his team of biologists at Stanford University reported in the journal Science:
Human alteration of Earth is substantial and growing. Between one–third and one–half of the land surface has been transformed by human action; the carbon dioxide concentration in the atmosphere has increased by nearly 30 percent since the beginning of the Industrial Revolution; more atmospheric nitrogen is fixed by humanity than by all natural terrestrial sources combined; more than half of all accessible surface fresh water is put to use by humanity; and about one–quarter of the bird species on Earth have been driven to extinction. By these and other standards, it is clear that we live on a human–dominated planet. (Vitousek, et al. 1997)Part of our concern with establishing clear concepts of human and material agency is that we see a grave danger (to all life on the planet) in underestimating the force of human agency. Actor–network theory has drawn attention to the power of things, but its suspension of the concepts of human intentionality and creativity to attain symmetry is too limiting. Our position is not a “humanist” one in which humans are seen as superior by virtue of tipping the agentic seesaw over to their side; we simply seek a characterization that draws attention to the particular potency of human agency.
Agency
We will apply the principles of activity theory to derive typologies that we believe capture essential differences between different kinds of agents and varying forms of agency.
For Leontiev, the primary type of agency was the agency of individual human subjects. In chapter 3, we defined human agency as the ability and the need to act. Now we want to revisit and refine this definition. In light of the principles and ideas deliberated so far, we will explore exactly what “the ability and the need to act” means. If we want to know whether we can extend agency to nonhuman entities, we need to formulate the question in a more general way: What are the criteria of agency? Once we know that, we can examine the fit between the criteria and different kinds of entities. We begin by examining criteria of human agency. We then compare these characteristics to those of other kinds of entities.
Following standard dictionary definitions, the most basic meaning of the “ability to act” is the ability to produce an effect. But this meaning is too broad for our purposes. If “acting” is understood as just producing an effect, then the ability to “act” is a property of anything that exists, either physically or ideally; any object, process, or idea. A narrower definition of acting is producing an effect according to an intention. Acting in this sense can be applied to, for example, a Mars probe, since the probe, as opposed to, for instance, stones on Mars, produces an effect according to the intention of the people who created it. Humans can act in both these ways: producing effects, and producing effects according to the intentions of others. But humans can also do something that neither the stones nor the probe is capable of: they can develop their own intentions on the basis of their needs, and meet their needs by acting on other entities, both human and nonhuman.
The “need to act” encompasses both biological needs and cultural needs. It is important to note that we use this distinction exclusively to refer to the origins of needs. Of course, all human needs are social in the sense that the way they are manifested and experienced is determined by individual’s development in a social context. The criteria of what an individual considers healthy, attractive, prestigious, and so forth, are determined by the immediate and general cultural environment. The meaning of objects as things (or beings) that can potentially meet the needs of an individual is established socially. For instance, religious norms can prescribe that potentially edible objects are not perceived as food.
However, humans are animals too, and, as any animals, they must meet their biological needs. Our biological needs derive from our heritage as evolved biological organisms. Throughout biological evolution, a driving force of development has been meeting the basic needs of organisms. By “biological needs” we mean the needs that ensure survival and reproduction. The invisible evolutionary background of human agency means that biological needs are deeply ingrained in the very nature of human agency. The human adaptation of culture created a unique new set of needs. Cultural needs have the potential to change rapidly and to proliferate in number far beyond basic needs.
Intentions are driven by biological needs to ensure survival and reproduction, and by cultural needs that are established socially. Humans themselves are, recursively, the realization of cultural needs expressed in the intentions of others. Our abilities to act are shaped by such cultural needs. We embody cultural needs as a result of our activity and the activity of others who act on us. The surgeon’s hands, the attorney’s mind, the athlete’s body — such agencies are the result of culturally specific object–oriented activity. Those who act on us to make us who we are include family, friends, peers, teachers, coaches, and coworkers, as well as the wider culture. We might also act according to another’s intention that challenges culture or is outside culture, even including antisocial activity outside the culture.
A Typology of Agents
We use dimensions of human agency as a framework within which to categorize different kinds of agents, as shown in Table 10.1. These dimensions come from the basic principles of activity theory, as well as extended implications of the principles, to specify dimensions of asymmetry between different kinds of agents.
As can be seen in Table 10.1, we do not propose a dichotomous scheme such as human–machine agency (Rose, Jones, and Truex 2005) or human–material agency (Pickering 1993). We consider several dimensions that distinguish agents and suggest that under varying circumstances, different kinds of agents may exhibit similar agencies. For example, humans and social entities share with some nonhuman entities the possibility to act through delegated agency. We believe a more expansive treatment of agencies is needed to capture the complexity of phenomena related to modern technologies, especially intelligent machines.
Table 10.1: Forms of agency Agencies Agents Things (natural) Things (cultural) Nonhuman living beings (natural) Nonhuman livings beings (cultural) Human beings Social entities Examples tsunamis, Northern lights, vernal pools, Martian rocks speed bumps, sewing machines, teapots, adzes grizzly bears, California poppies, truffles, protozoa house cats, Dolly the sheep, GMO corn, Bourbon roses Spinuzzi’s traffic engineers, Miettinen’s scientists, ANT’s princes World Trade Organization, ISO, Doctors without Borders, United Nations Conditional agency Produce effects Need-based agency Act according to own biological needs Act according to own cultural needs Delegated agency Realize intentions of (other) human beings
Table 10.1 first of all establishes a distinction between agents that are living and nonliving entities. In chapter 3 we established that not any entity is a subject. A subject lives in the world and has needs. Interaction between the subject and the world cannot be symmetric because nonliving things do not have needs.
Next, the table distinguishes between human and nonhuman living beings according to the kinds of needs they have. Humans have both basic and cultural needs while other living beings have basic needs only. We distinguish between different kinds of nonhuman living beings: those that are the product of cultural needs and those that are not. The category “nonhuman living beings (cultural)” includes organisms such as domestic animals, plants, and fungi; live vaccines; clones; and genetically engineered plants and animals (like the knockout mice in chapter 7). The distinction is between organisms that have evolved outside human intention and those that have been cultivated, cultured, husbanded, bred, cloned, or genetically modified. The latter are a direct result of human activity, of some cultural need.
The dimension called “realize intentions of humans” suggests things (natural) and things (cultural). Things that are the result of some human intention produced in a cultural milieu are artifacts. A speedbump slows a driver because it is designed to do so, a fence keeps in the sheep, a vaccine deters a disease, a field of corn is harvested to feed the pigs. By contrast, the ocean currents that move the scallops in the study by Callon (1986) are a form of agent with no intention. The vapor that resists the physicists’ efforts is another. Crusoe’s footprints produce an effect, but with no intention. We of course recognize that California poppies might be cultivated in a garden, that grizzly bears might be managed in a park, and so on. However, such living things have had, and continue to have, existence beyond human intention in certain contexts. This dimension also applies to humans, who may realize the intentions of other humans, and to social entities.
Should we also call nonhuman living things (cultural) artifacts? Do knockout mice have subjectivity or are they so engineered, has so much been knocked out, that they are simply artifacts? We do not have a ready answer to this question. In research labs where we interviewed for the study described in chapter 7, we talked with lab technicians who clearly regarded lab animals as living creatures to be treated as kindly as possible (before the inevitable end). The technicians spoke of how they clipped the animals’ nails for their comfort and fed them cranberry treats. On the other hand, the animals were exposed to hazardous chemicals and lived in extremely limiting circumstances that did not permit them autonomy to meet their own basic needs. The point here is not to decide the question, but to suggest that the power of human agency may shape cultural categories in complex ways.
The final column in table 10.1 is the category of social entities. We are least sure about the status of social entities relative to the other categories. Social entities comprise entities from all the other columns. They produce effects, and they can be said to have cultural needs (if they are to survive and reproduce themselves, certain things have to happen), and they realize human intentions. But because they are a composite of the other four entities, they have perhaps evolved to a different level of abstraction for which the dimensions of the table are insufficient. However, the notion of macro–actors in actor–network theory suggests that social entities have “interests” and can be seen as agents in their own right. In actor–network theory, large–scale entities — for example, the European Union, Silicon Valley, the space program, high–tech organized crime — can be said to have interests. While interests are most certainly associated with these macro–actors (Callon and Latour 1981), we are unsure how they are the related to the interests of living organisms, and leave this question for future investigation.
The cells in the leftmost column of the table identify different kinds of agencies. Rows 3 through 6 identify dimensions of these agencies. Row 3 indicates that all agents can produce effects. Row 4 indicates that when producing effects, some agents realize biological needs. Row 5 indicates that when producing effects, some agents realize cultural needs. Row 6 indicates that an agent may realize the intentions of (other) humans. The Mars probe realizes the intentions of the scientists who built it. A schoolchild learning to read realizes the intentions of parents and teachers (and of course her own intentions as well, in most cases).
Living beings are a special kind of agent in that they strive to meet needs in the world, engaging other entities as they do so in a patterned way.
Living beings are a special kind of agent in that they strive to meet needs in the world, engaging other entities as they do so in a patterned way. An erupting volcano has tremendous agentic power, but it simply explodes, affecting whatever lies in its path without regard for where its cinders rain down, where its lava flows. By contrast, a plant with its biological needs reaches for the sun, it produces chlorophyll, its flowers attract bees of a particular kind, and its seeds are eaten by certain birds who scatter them in the woods.
As we saw in chapter 3, Leontiev searched for a concept to describe the context in which mind evolved, considering “life,” but rejecting it as too general. He settled on activity, defining it in terms of a subject’s relation to a world in which it attempts to fulfill its needs. Nonliving things are inert in not having needs. Phenomenologists have also noticed this, observing that things do not “care,” as Heidegger (1962) proposed. Subjects engage in activity because they care about what will happen to them in the future (see also Emirbayer and Mische 1998). This caring is the condition of the tiniest one–celled animal struggling for a mite of algae, to that of humans attempting to solve the most difficult scientific or social problems. In her novel Housekeeping (1980), Marilynne Robinson put it poetically: “And there is no living creature, though the whims of eons have put its eyes on boggling stalks and clamped it in a carapace, diminished it to a pinpoint and given it a taste for mud and stuck it down a well or hid it under a stone, but that creature will live on if it can.”
Thus agency should not be considered a monolithic property that is either present or absent in any given case. Producing effects, acting, and realizing intentions, while potentialities of certain kinds of agents, vary within the enactment of a specific activity. Extending the notion of agency beyond human subjects may appear to be a deviation from the asymmetry of the subject and the object postulated by activity theory. However, what we propose is a combination of (1) a strict subject–object dichotomy (and resulting asymmetry) and (2) the notion of levels of agency, an understanding of agency as a dimension rather than a binary attribute.
With this in mind, we can differentiate between types, or levels of agency. Our analysis, as depicted in table 10.1, suggests a preliminary typology including:
Need–based agency Human beings have both biological and cultural needs. To meet their needs, they form intentions and act from these intentions. Similar types of agency are manifested by social entities (even though they do not have biological needs) and higher animals (even though they do not have cultural needs).
Delegated agency Various things and living beings can be said to clearly realize intentions, but these intentions are delegated to them by somebody or something else. These things and living beings are agents in the sense of acting on somebody else’s behalf. For example, an animal such as a thoroughbred race horse realizes the intentions of its breeders when it wins a race. But at other times the horse realizes its own intentions, while grazing in the pasture or resting in its stall. Human society is of course set up so that humans delegate their intentions to other humans, even in the simplest societies where children help their parents, marriages are arranged to advance a clan’s fortunes, and so on. The intentions of the individual subject and intentions delegated to a subject by others may be in accord, or they may create conflict and dissension.
Conditional agency Anything and anyone can produce unintended effects. The Russian winter of 1812 did not target Napoleon’s army but undoubtedly contributed to its defeat. Truck drivers do not intend to create obstacles on highways, but they repeatedly do. Even without intentions, something or somebody may constitute a force — or condition — to be reckoned with.
Artifacts
Artifacts are special agents that are the product of cultural needs. Humans have gained some control over our needs through the design and deployment of artifacts that embody our intentions and desires. We are able, in the lifetime of a single individual, to create new solutions to meet needs as conditions change. As we saw in chapter 3, artifacts empower people through the use of technical and psychological tools. Activity theory conceptualizes the potency of human agency in part through the principle of mediation: tools empower in mediating between people and the world. People “appropriate” tools in order to empower themselves to fulfill their objects (Bakhtin 1981; Wertsch 1998). The principle of mediation clearly suggests that things have agency, because if they did not, they could not act as mediators. In a vivid example, Zinchenko (1996) observed, “Communist and fascist symbols acquired such fanatic energy that ... they nearly devoured the great cultures of Russia and Germany.”
Functional organs are a special kind of mediator. Zinchenko (1996) discussed the cellist Rostropovich and the quality of mediation provided by his cello. When asked in an interview about his relationship to his cello, Rostropovich replied:
There no longer exist relations between us. Some time ago I lost my sense of the border between us ... . In a portrait [by the painter Glikman] ... there I was — and my cello became just a red spot at my belly ... . (Quoted in Zinchenko 1996)A functional organ is a relation between human and thing that is different from that between nodes in an actor–network. A functional organ brings human and thing close together, in a relation more intimate than a system of like nodes. A red spot at the belly is an apt metaphor for how we experience our most cherished technologies. As discussed in chapter 3, functional organs are subsystems that are integral parts of a subject who still decides when to use functional organs and whether they have to be altered or even abandoned.
Actor–network theory speaks of artifacts as having “delegated competences” (e.g., Latour 1993). Interests are “translated” between elements in the network, that is, performances and competences move back and forth between nodes in a symmetrical network. For example, a user’s competence at typing a URL can be translated to a computer that can guess at the string intended by the user (as long as a few clues are provided) and complete the typing for the user.
In the scheme proposed in table 10.1, delegation flows from humans to all other kinds of agents (including other humans). But in actor–network theory, a nonhuman entity might delegate to a human. So a cell phone demands fresh batteries. It enlists the human to put them in by beeping when the battery is low. Activity theory sees this scenario differently. There is no delegation from thing to human; the human has decided that in order to use the mediating technology, she will supply the batteries. The human has other possibilities. She can turn off the beeper; she can throw away the cell phone; she can cancel her contract with the phone company. At any time she can resist; she can modify; she can alter her relationship to the technology based on her desires. The cell phone, by contrast, does what it was programmed to do; it is without desire and intention. The cell phone’s agency is manifest in its ability to beep but it is an agency designed and delegated by humans.
Artifacts can be designed to produce effects to replace human labor at the level of operations and actions. Leontiev (1978) observed that operations are destined to become the functions of machines, and indeed the proliferation of such machines created since the Industrial Revolution has continued unabated (as the Luddites also foresaw). Automatic gearboxes, electric mixers, dishwashers, and text completion are examples of artifacts taking on some particularly tedious operations.
At the level of actions, programmable artifacts have proved enormously efficient and even intelligent. As early as the beginning of the nineteenth century, Jacquard looms driven by punched cards created elaborate fabrics (and provided the inspiration for the design of Babbage’s Analytical Engine as well as Hollerith’s Census Machine used to conduct the U.S. Census in 1890). A modern washing machine can carry out an impressive sequence of actions. And of course computers are the exemplar of programmable artifacts, interpreting languages that permit flexible sequences of actions (and operations) on a new scale of complexity.
While mediators can be designed to autonomously assume human operations and actions, they cannot, in themselves, create meaningful activities. Artifacts cannot decide what they want; they cannot form an intention, or say what is meaningful and what is not.
While mediators can be designed to autonomously assume human operations and actions, they cannot, in themselves, create meaningful activities. Artifacts cannot decide what they want; they cannot form an intention, or say what is meaningful and what is not. Rae’s fax machines, clever though they were, were not subjects with needs; they took no pleasure in correcting one another’s mistakes and would never have done so without a human having programmed them.
Though we can’t think of any actual artifacts with intention or desire, such artifacts are alive and well in the human consciousness. The Golem, Frankenstein’s monster, the robot Maria in Lang’s Metropolis, HAL in 2001, and the Terminator, to name a few, inhabit a narrative universe in which cyborgian desires exert powerful influences in human life. These humanoid characters, while compelling, appear to be rather obvious projections of human desires and fears. A more likely source of artifacts with intention may be research in artificial life, an area to which we will look for interesting future developments (Reynolds 1987).
Mediators empower in ways specified by human designers, but the agency of artifacts may also be conditional, producing unintended effects. Winner (1977, 1986) has explored the unintended consequences of technology in several empirical investigations. Shaffer and Clinton (in press) pointed to the studies of Postman (1993) and Tenner (1997) in observing that “[things] have a way of exceeding or changing the designs of their makers.” Unintended consequences of artifacts may be of value to people (such as the discovery of penicillin) or they may be tragic (such as an explosion at a nuclear power plant).
Conclusion
Actor–network theory has fruitfully challenged traditional accounts that restrict agency to humans. Using activity theory’s basic principles and definitions, we have proposed a formulation that builds on actor–network theory’s accomplishment but offers a different categorization of agents and agency, one that retains the asymmetry of the subject–object dichotomy. The subject–object dichotomy does not imply that activities should be understood in terms of a lone subject acting in a passively resisting environment. An extended notion of agency provides for creating richer representations of real–life settings. It takes into account human beings, who can help and support or resist and oppose: as representative of organizations, as individuals, and as physical bodies. It includes individual human beings and social entities, composing, for example, complex networks of activity systems (Engeström 1999b). And how about animals? Any pet owner would probably agree that animals have agency (to spare). Finally, the notion of things having a kind of agency can be found in virtually all cultures, from the ancient tradition of giving things (such as ships) and natural phenomena (such as hurricanes) human names, to the modern anthropomorphizing of the computer and the use of metaphorical language to describe machine behavior.
We identified different kinds of agents according to whether they are driven by the demands of life and whether they embody cultural needs. We found that these qualities indicate that a simple symmetry between people and things leaves out far too much. In particular, it leaves out the power of human agency and the way humans regulate their own activity through the creation and use of new agents, both living and nonliving. Table 10.1 also suggests an interesting tension: every agent is capable of producing effects for which there is no intention. The more cultural things (living and nonliving) we have in the world, based on human design and intention, the more possibilities we introduce for conditional agency, that is, for new kinds of unintended effects.
About the authors
Victor Kaptelinin is Professor in the Department of Informatics at Umeå University, Sweden.
Bonnie A. Nardi is Professor of Informatics in the Donald Bren School of Information and Computer Sciences at the University of California, Irvine.
Return to the Acting with Technology: Activity Theory and Interaction Design Table of Contents
Copyright ©2007, First Monday.
Copyright ©2007, ictor Kaptelinin, Bonnie Nardi, and MIT Press.
Acting with Technology: Activity Theory and Interaction Design by Victor Kaptelinin and Bonnie Nardi
First Monday, volume 12, number 4 (April 2007),
URL: http://firstmonday.org/issues/issue12_4/kaptelinin/chapter10.html
