Simulated Audiences and Rhetorical Agency: How Machine Stand-Ins Help Us Feel Alone Together

In her piece “What Can Automation Tell Us About Agency,” Miller opens up by drawing attention to the computer systems of automated assessment as a new product of interest to rhetoricians. She relates the dilemma of agency to technology studies by discussing the challenge of the distinction between action and motion in agency through human and non-human, or machines and natural forces (143). She explains “by positing a machine as audience, automated assessment systems for both writing and speaking denaturalize rhetorical action, challenging and uncovering our intuitions about its necessary conditions” (140). By this, Miller suggests that rhetorical agency is at stake and attempts to illustrate how our resistance to automated assessment can inform the current debate on the nature of rhetorical agency (140).

A lot of this discussion, and particularly the examples Miller illustrates, ties in directly with the book Alone Together by Sherry Turkle. In the first part of her book, Turkle discusses the “robotic moment” and sociable robots and how they are creating new unsettling relationships and instabilities in how we understand privacy, community, intimacy, and solitude. She poses a troubling observation that robots are filling gaps in the society and we hope to use them as a solution to our own imperfections, as an easy substitute for the difficulty of dealing with others. In her examination of the relationship between humans and robots, Turkle suggests we don’t mind giving human qualities to inanimate objects, much like Miller’s comments on our eagerness to attribute agency to technologies.

Miller makes the connection to Turkle in explaining that “automated writing assessment is an applied branch of natural language processing, which in turn is a branch of artificial intelligence” (138), like robots. She continues by discussing our resistance to automation as rooted in a commitment to agency, since they remove agency from the audience and deal with the confrontation between action and motion (141). She addresses three dimensions of rhetoric as missing in this: performance, audience, and interaction. In relation to performance, Miller comments that speaking requires a living presence (145) and thus prompts agency as a kinetic energy (147).  Instead of serving that purpose, the computer instead serves as a mediator and a surrogate. This is because it does not provide a relationship between two entities that will attribute agency to each other, what she says produces the kinetic energy (149). Therefore, interaction is fundamental for agency because it creates the kinetic energy of performance and puts it to rhetorical use (150). Interaction is also an issue that Turkle addresses. She points out that computers and robots no longer ask us to think with them, but ask us to feel for and with them as they become more sociable, affective, and relational (39). She argues that their interactivity prompts our minds to start projecting consciousness and we compensate by filling the gaps. A machine only needs to act clever and people will play along. Additionally, she points out that this simulation is enough to provoke empathetic urges. This ties in directly to Miller’s discussion of ethopoeia.

Miller mentions ethopoeia and society’s tendency to have a low threshold for it to illustrate how willing we are to attribute character to an interlocutor. She comments “we go out of our way to construct a human relationship” (151). She even makes a direct connection to the Eliza effect, or the attribution of intelligence and sympathy to a computer program, which is a reference Turkle uses throughout Alone Together to explain her ideas (151). Turkle argues that the new relationships we have with robots create a loop, drawing us into the complicities that make it possible; but we are playing along, willing to defer to what the robots are able to do (100).

Miller’s article illustrates indirectly how we are motivated by technologies to live a more rushed lifestyle and expect almost instantaneous results for everything. This even includes feedback in grading assessments. This ties in directly to Turkle’s point that as technologies advance, we expect more from them and less from each other. Turkle comments that as society gradually seems to be growing more “machine ready,” it seems less shocking to put robots in places where people used to be (146). So we must ask, what are we sacrificing when we look to robots as substitutions or surrogates? Turkle says “As we learn to get the ‘most’ out of robots, we may lower our expectations of all relationships, including those with people. In the process, we betray ourselves” (125). Both authors discussion of machines opens up daunting philosophical, psychological, and moral conversations.

Overall, Miller points out that “automated assessment technologies could take advantage of our eagerness to attribute agency” (152), prompting the dilemma of where to draw the line between the human and non-human, symbolic and material. Even further, the attribution of agency concerns matters of moral judgment, human decency, respect, and acknowledgement. Therefore she asks whether or not we owe such acknowledgements and agency-granting attribution to automated assessment systems (153). Similarly, in her book, Turkle not only describes how the boundaries between people and things are shifting, but makes the reader question which of these boundaries are worth maintaining.

Turkle, S. (2011). Alone together: Why we expect more from technology and less from each other. New York: Basic Books.