Glitch as Infrastructural Monster

Nathan R. Johnson, University of South Florida
Meredith A. Johnson, University of South Florida

(Published May 11, 2016)

. . . . .

After a week of vacation in January of 2014, we found ourselves arguing at Chicago’s O’Hare International, a bustling nexus of national and international travelers being trafficked around the globe. Despite customer satisfaction floundering, the number of airline passengers has increased each year since 1980 (ACSI, ACSI Travel, “Benchmarks”; World Bank). Standing in O’Hare, we were two unsatisfied travelers, hardly noticeable against the backdrop of passengers, kiosks, and airport activity, perhaps except to those in line immediately behind us.

. . . . .

The Flows of Contemporary Air Travel

The airline industry, a multibillion-dollar business that specializes in travel to almost every discrete corner of the earth, ironically makes all destinations seem alike. Remove the souvenir t-shirts for the local sports teams from the gift shops and blind the telling signage and it becomes difficult to identify a city just by its airport. After all, airports tend to look a lot alike, mostly by design (Wood 28-32). The McDonaldization of the airline industry has promoted efficiency, calculability, predictability, and control (Ritzer 187). Planes travel with relative ease because they can depend on standardized operating procedures across airports. Passengers come to rely on the sameness of air travel too. Travelers have grown to expect the assembly line experience of check-in, security check, wait, board, then fly. Even minor disturbances can result in anxious passengers. Disruptions to that sameness can do more than separate traveler from their planes; they can illuminate the rhetoric within the airline industry’s infrastructural techniques and technologies.

In this article, we examine a story of air travel gone awry during the first week of January 2014, a time when the Polar Vortex upended national air travel. Our impetus for taking on this analysis is twofold:

First, we are answering previous requests to more carefully attend to infrastructure (Johnson, “Information Infrastructure” 2) during one of its most visible moments—mid- breakdown (Star 382; Star and Ruhleder 113)—as a means of connecting it more firmly to rhetoric. Seeing infrastructure is no easy task because infrastructure is typically invisible by design (Bowker et al. 98), fading into the background as it provides the necessary scaffolding for predictable interactions and operations. Susan Leigh Star and Karen Ruhleder maintain that “infrastructure occurs when the tension between local and global is resolved. An infrastructure occurs when local practices are afforded by a larger-scale technology, which can then be used in a natural, ready-to-hand fashion” (114). When infrastructure is working, the resources it provides seem natural. For instance, access to water, electricity, or air travel is taken for granted until access to the water supply, the power grid, or airplanes is compromised. Functioning infrastructure is only transparent as long as it is functional. This invisibility makes it especially difficult and important to investigate. During moments of instability, normal assumptions about how the world works are upended as the infrastructure that is the world falls away (Dourish and Bell 27–35; Edwards, “Infrastructure” 188). Rhetoric figures prominently at these relationalities.

Deliberative rhetoric is critical for coordinating sociotechnical junctures in infrastructure. In these junctures, people negotiate their relationship to each other and systems where the local and the standardized mesh. Consider, for example, how the airport’s Transportation Security Administration (TSA) procedures shape TSA agents’ and travelers’ decision-making under the most mundane circumstances. Pragmatic arguments about what constitutes good and advantageous behavior are built into the signage adjacent to the security queue, to name one example. Families with small children are afforded extra time and space at the checkpoint. The elderly need not take off their shoes. Now consider how visible this deliberation becomes when the conveyor belt jams and new, quickly improvised procedures must enable the screening of carry-on luggage.

In a broader sense, infrastructure is rhetorical in that it attunes audiences participating in the activities it supports. Rhetoric can “reattune or otherwise transform how others inhabit the world to an extent that calls for some action” (Rickert 162). Airline infrastructure hails publics into airline roles: passenger, pilot, flight attendant, luggage, agent, etc. It constrains sensibilities. Because infrastructure supports foregrounded activity, the background, although flexible, provides momentum that curbs sociotechnical activity. This article contributes to the field by highlighting the rhetoricity of infrastructure under normal and not-so-normal circumstances, both in terms of a conservative, deliberative perspective and a materialist, world-building rhetoric of attunement.

Second, this article examines the mechanisms that coalesce into airline infrastructure—or fail to coalesce, mid-breakdown—to illuminate a new aspect of rhetoric’s materiality in a posthuman age. It builds on those critics who have explored the materiality of everyday public spaces such as Starbucks (Dickinson) and extends that work by visiting new terrain: the airport. The airport is the setting where we approach agency—having an effect—as relational and dependent upon a network of actants; in short, from a new materialist perspective. New materialism enables us to better account for the symmetry of human and nonhuman agents in rhetorical practice (Rivers), “eschew[ing] the distinction between organic and inorganic, or animate and inanimate, at the ontological level” (Coole and Frost 9). As Diana H. Coole and Samantha Frost maintain:

This lively matter associates with human actors to create agentive networks. These relationships are not fixed, however; rather the configuration of the network changes based on the motives for and kinds of action needed in a particular situation, so the result is “objects forming and emerging within relational fields, bodies composing their natural environment in ways that are corporeally meaningful for them, and subjectivities being constituted as open series of capacities or potencies that emerge hazardously and ambiguously within a multitude of organic and social processes” (Coole and Frost qtd. in McIntyre 13)

New materialist theories of agency challenge notions of the autonomous human rhetor at the helm of a rhetorical act (i.e., Quintilian’s good man speaking well) (McIntyre 10-11). Instead, “agency does not belong to any single actor, whether human or nonhuman; rather agency is the product of the relationship between actors and exists only at their point of contact. These points of contact … are rhetorical acts” (McIntyre 12). Indeed, others have successfully demonstrated that agency is a relational act between humans and vibrant matter (see Bennett 24-28 for her work on electrical blackouts). In addition to affirming that claim, this article makes three unique contributions: (1) we analyze new materialist theories of agency through the lens of infrastructure theory, an underutilized theoretical apparatus; (2) we consider the rhetorical agency of a glitching, quantifying object; and (3) we introduce terminology for describing these developments, including infrastructural monsters and critical mass agency.

To do this work, we seize upon a moment of infrastructural disorder: the italicized story we opened with and continue to share in two more parts in this article. We conduct a close analysis of disrupted airline travel during the Polar Vortex of winter 2014, our personal narrative of the story, and the workings of a glitchy luggage scale during the return leg of the trip in order to demonstrate how even carefully calibrated quantitative technologies are dynamic entities, imbricated in unpredictable agentive networks that make new kinds of rhetorical engagement possible. The luggage scale is a powerful actant in its own right that redefined travel goals along with us, the humans attempting to navigate the Polar Vortex. We chose the scale as one prominent vertex of our analysis of infrastructure because it was such a seemingly stable technology particularly adept at concealing its rhetoricity. Like the other infrastructural mechanisms that surround it, the airport luggage scale’s workings appear stable and absolute if they appear at all. It is “black boxed,” only available for scrutiny when it is rendered visible by a glitch (Latour, Pandora’s Hope 183-185).

Like Casey Boyle, who recasts glitches as “the conditions of possibility for rhetorical action” (20), we approach the glitchy scale as more than the means by which the force of classifications and standards is leveraged. We see it as a “monster” exerting its own agentive force (Haraway; Latour, “Love Your Monsters”). That is, we build from scholarship on “glitch” to forward our concept of an infrastructural monster (Boyle; Reid). The scale glitches and, like the infrastructure in which it is embedded, becomes visible “as something that manipulates and as something that can be manipulated” (Boyle 12). Monsters are misfits. They stand out. Accounts of infrastructural monsters are revelatory. By positing the malfunctioning scale as an infrastructural monster, this article takes glitching seriously “as the source of agency and thought rather than its limit” (Reid). The scale-as-infrastructural-monster is given due weight as an actant that exerts critical mass agency when translating a phenomenon into a number to match against a threshold. The scale-as-monster ultimately, though unpredictably, resolves a dispute when human actants cannot come to terms over stases or the central issues in an argument.

The profusion of connections between users and technologies that our story maps makes for an unwieldy assemblage from which to approach agency. Infrastructure theory, of which we give a fuller account in the next section, gives us one more foothold from which to examine that abundance. After an overview of infrastructure’s resources, we connect its workings explicitly to the airline industry’s treatment of luggage. We then describe how the Polar Vortex strained status quo infrastructure in 2014, anchoring our explanation in our own disrupted travel narrative. In addition to taking methodological cues from John Law’s Aircraft Stories and, to a lesser extent, Bruno Latour’s Aramis, this narrative also provides us with the base from which to explore infrastructure’s relationship to agency. The mechanisms of infrastructure, we argue, necessarily circumscribe possibility; our analysis articulates how its non-/human agents simultaneously act as sources of possibility. We consider the ancient rhetorical concept of phronesis, or applied wisdom, to explain agency during infrastructural breakdown, but ultimately find it too myopic a lens to account for agency in our story. We offer scale-as-monster as a way of understanding agency in infrastructure. We recount how a glitching scale decided the main issue at hand when humans looked to competing infrastructures during a dispute resulting from the Polar Vortex. We close by introducing the notion of critical mass agency to describe the monster scale’s specific effects and, more generally, to provide the field with more vocabulary for discussing new materialist agency using finer-grained distinctions.

What is Infrastructure?

Although “infrastructure” has often been construed as an infinitely expanding laundry list of support mechanisms, “something upon which something else ‘runs’ or ‘operates’” (Star and Ruhleder 112), a more useful way of understanding infrastructure for our purposes is to look to how documented standards, protocols, classifications, and algorithms function as an arterial system in a larger ecology (Bowker et al. 99; Johnson, “Information Infrastructure” 1-2). Infrastructure as we are approaching it here includes any “pervasive enabling resources in network form” (Bowker et al. 98). A study of infrastructure is therefore a study of dependable, circulatory mechanisms that enable action. Infrastructure frequently consists of the following:

•     Classifications, standards, protocols, and algorithms that coordinate action (Johnson, “Information Infrastructure” 1-2). For example, the timetables that govern all air travel are a form of documented classification that suggests dependability across a distributed system.

•     Time- and space-spanning techniques that exert force in the world. For example, airlines classify bags as “carry-ons” because their dimensions and weight are treated uniformly no matter where they are measured and weighed. Though bags actually weigh slightly less at 35,000 feet (i.e., cruising altitude) than they do at sea level, scales and measurement procedures provide certainty, establishing a vacuum that occludes time and space. Properly functioning baggage scales provide consistent information about size across different environments, which in turn provides a logic for how to stow measured objects (i.e., luggage). Fine-grained distinctions are subordinated for the sake of stabilization.

•     People, particularly human labor, who help coordinate action by enacting time- and space-spanning techniques or adjusting them to help smooth over inconsistencies that challenge the veneer of regularity that infrastructure creates. For example, pilots use their flying expertise to keep schedules by navigating changing weather and compensating for delays. They make in-air choices that are implicated in keeping the airline’s logic of tempo. Meanwhile, the ground crew moves baggage from one location to another. When it snows, the ground crew’s comportment shifts to compensate for hazardous runway conditions. Finally, passengers abet air infrastructure by listening to and obeying the instructions for orderly boarding given to them by airline personnel. Flying today means becoming part of the airline infrastructure. Passengers board their planes, and the airport also boards its passengers. Infrastructural people toggle between being agents of infrastructure supporting the network and organizing new infrastructural action.

A study of infrastructure is a study of relations between these people and mechanisms as they impact organized practices. Information scientists, computer scientists, historians, anthropologists, and sociologists interested in science and technology studies (STS) have worked extensively with the concept of infrastructure to explore communication standards for software and hardware (Hanseth, Monteiro, and Hatling), the mythology surrounding infrastructural technologies (Dourish and Bell), the social impact of science infrastructures (Edwards, A Vast Machine), interactions of culture and infrastructural technologies (Ong and Collier), and systems development (Star and Ruhleder). However, very few rhetoricians have looked to STS for its approach to infrastructure (Brown et al. 525; Dyehouse, Pennell, and Shamoon 336-38; Selber 12). And when they have, applications have been limited.

For example, in Danielle Nicole DeVoss, Ellen Cushman, and Jeffrey T. Grabill’s discussion of how software in composition classrooms creates constraints for production, the authors ultimately seem to engage infrastructure in terms of particular technologies rather than as an entire circulatory system that enables action, which is an approach that we think too narrow in its scope (21-22). DeVoss et al. do cite the importance of standards, classifications, and protocols, and note that standards are involved in software and university networks. But they do not follow that analysis further than observing that standards limit composition and that changing those limitations changes composition (e.g., bandwidth speed limits the types of files that can be saved or downloaded). DeVoss et al. acknowledge that networks that connect different localities change practice in each locality; however, this observation is largely left unexplored. From our point of view, we are left questioning how the specifics of network standards are implicated in the pedagogical problems that DeVoss et al. take on. We get a sense of how changes to software and server space can help create videos, but less of a sense of how standards were important actors in those changes. In addition, we remain curious about how these issues might take shape in rhetorical situations outside of a composition classroom (which is admittedly beyond the scope of their article). We recognize the important work DeVoss et al. do as we seek to continue to thicken infrastructure as a useful methodological concept (Jasinski 256-257). Ultimately, DeVoss et al. provide a reading of infrastructural issues that we think is notable but could be rethought with a different analytic sensibility that accounts more fully for its mechanisms and actants.

We turn, then, to infrastructure theory as an apparatus for “tracing the actors in a vital, rhetorical ecology” that can deftly accommodate human and nonhuman actants alike (Rivers). Consider the number of actants involved and the complexity of the standards at play in moving baggage onto an airplane. Southwest Airlines, for example, tells customers,

Checked baggage will be screened and is subject to physical inspection as mandated by the TSA. Customers may be required to present identification. Regulations require name identification on the outside. We recommend placing identification on the inside of the baggage, too. Name labels are available at Southwest Airlines ticket counters. Once you've checked your luggage, make sure you receive a separate claim check for each piece of baggage you've checked. Check to make sure that the city shown on the claim check(s) matches your final destination. (“Checked Baggage”)

Baggage is one of many spaces where the airline industry’s standards converge. Standards, or accepted and enforced rules that span more than one site and are designed to make things work together over time and distance (Bowker and Star 13-14), comprise a pivotal part of airline infrastructure. Airlines enact standards to help guide bags from location to location, and these procedures are enforced across hubs. In a perfect world, a checked bag would fit into a flow chart that seamlessly classifies the parcel, charges the customer a fee, and sends bags and passengers to their destination without a problem. Yet rhetoric and STS scholars have both pointed out that plans rarely work out as they have been modeled (Keränen 232-34; Suchman 13). Ours certainly didn’t when, in January 2014, Meredith’s luggage took an unexpected week-long vacation in Flint, Michigan, while we were visiting Chicago, Illinois, though we had intended to vacation in Indianapolis, Indiana.

. . . . .

Weeks in advance of her trip, Meredith booked a two-hour nonstop flight from Tampa International Airport (TIA) to Indianapolis International Airport (IND) for Monday, January 6, 2014. Nate had made hotel reservations for a vacation in downtown Indianapolis. Nate would drive about an hour from his home in Indiana to pick Meredith up at IND and would drive them into the city. The Polar Vortex of 2014 made for impressive winter storms, and many flights, including Meredith’s original flight from Tampa, had been delayed or canceled. So, at about 2:00 a.m. on the morning of the original flight to IND, Meredith checked the schedule and found her flight out of Tampa had been completely canceled. By 2:15 a.m., she was on the phone with the airline looking for alternative routes to Indianapolis.

The customer service representative checked Indiana and all four contiguous states—Michigan, Illinois, Ohio, and Kentucky—for flights and found exactly one available option that would put her within driving distance of Nate that day: a flight around noon from Tampa into Flint, Michigan. Meredith agreed, left the house immediately, and was going through security at TIA before 3:30 a.m. The airport was already crowded with its share of underslept, displaced travelers, and soon more would be added to the fray. Meredith checked her not-quite-50-pound bag to Flint, Michigan, and waited for her new flight. However, when she heard the last call over the PA system for a missing passenger on a different, completely booked flight boarding for Chicago’s Midway International Airport at 6:00 a.m., Meredith ran to the counter, asked for the seat, and got it. Meredith was at Midway in Chicago with only the backpack on her back before Nate woke up in Indiana that morning, but it would be another 30 hours before we saw each other and another week before we saw Meredith's bag.

Nate awoke to a series of frantic, confusing messages from Meredith and had his own questions: “Are you in Flint, Michigan? You want to rent a car and drive from Flint? Wait, Michigan? Why are you in Michigan?” But the question of how to get Meredith from Chicago to Indianapolis still lingered and became increasingly fraught with complication as the day progressed. The weather had steadily declined that morning, and by 8:00 a.m. Chicago time the temperature had hit a record-breaking sixteen degrees below zero (“Record of Climatological Observations”) with a wind chill as low as forty-two degrees below zero (“Chicago Record Low”). The eleven-plus inches of snow and ice already on the ground were impervious to salt mixtures and normal snow-removal procedures (“Record of Climatological Observations”). Black ice and blowing snowdrifts combined with historically low temperatures to bring giant swaths of the Midwest, including Chicago, to a standstill. At both Midway and O’Hare, almost 3,200 flights were proactively canceled that day and the day prior (“Record-Breaking Cold”). Public transportation of all stripes in and around Chicago was compromised to varying degrees (“Cold Temperatures”). This was all unknown to Meredith, who bundled up and caught a cab across town from Midway to O’Hare with surprising ease before the road closures caught up with her.

The ever-changing trip now required Meredith to cab over to Chicago’s other airport (O’Hare) to catch the Lafayette Limo Bus to Indiana. We note that the infrastructural breakdowns that marked every phase of our trip are only clear in retrospect. News of rail delays, intermittent road closures, and the volume of canceled flights was trickling out in the early morning of January 6, 2014. The Lafayette Limo was still scheduling bus service out of O’Hare and would bring Meredith to Nate in Indiana, saving him a four-hour round-trip drive. So Nate purchased a limo ticket for Meredith, and she made her way across the city optimistically (at least at first).

Chicago’s dicey roads gave the cabbie plenty of time to clue Meredith in to just how bad things had gotten since she left her home in Tampa so early that morning. Meredith checked in with Lafayette Limo personnel as soon as she got to O’Hare and found they were no longer guaranteeing any bus service to Indiana that day because the Indiana Department of Transportation (INDOT) had closed Interstate 65, Interstate 80, and Interstate 94 for hundreds of miles in both directions (Wang). Northern Indiana was inaccessible to all traffic except emergency vehicles (“Chicago Record Low”). With Meredith now at O’Hare and Nate still at home in Indiana, we obsessively refreshed INDOT’s Twitter stream for updates, hoping the rumors that the travel ban would be lifted by 3:00 p.m. were true. They weren’t. By 2:30 p.m., Illinois Governor Pat Quinn declared a state of disaster (“Governor Quinn Provides”). So Meredith found a very cold but highly prized piece of outlet-adjacent floor, plugged in her laptop, and Skyped with Nate while she waited. By 7:00 p.m., we gave up for the night. By 10:30 p.m., she had made her way to the closest vacant hotel room. It was right up the street from Midway (not O’Hare).

At 10:00 a.m. the next day, the interstates were reopened. Having decided Meredith had done enough traveling for one trip, we opted to scratch our Indianapolis plans altogether and spend the week together in Chicago. Our nonrefundable reservations at the Omni Severin in Indianapolis were mercifully refunded due to the weather crisis. By 10:10 a.m., Nate was on I-65. By 10:30 a.m., INDOT reclosed it. The slick roads from Indiana to Illinois were still lined with slide-offs and blowing snowdrifts. But Nate continued his white-knuckled drive towards Chicago. It was normally about a two-hour trip. Six hours later, he arrived.

. . . . .

Agency and Infrastructure

By the time we saw each other in Chicago, our trip was almost 36 hours and 200 miles off course from our original destination: Indianapolis. Flight times and destinations, crucial parts of the air industry’s infrastructure in the form of classifications, didn’t stand a chance against crippling weather. Geoffrey Bowker and Susan Leigh Star define a classification as those mutually exclusive boxes into which we put the world, be they temporal, spatial, or both (10-11). Classifications, along with standards, protocols, algorithms, time-/space-spanning techniques, and the people who enact them, provide the necessary continuity of practice required to enable action. As Nathan R. Johnson notes, “Infrastructure comes into being as a side effect of social dependence on a stable background. Infrastructure allows communities to assume stability wherever and whenever they are, which produces infrastructure’s flattening of time and space” (“Protocological Rhetoric” 385).

That is, infrastructure provides stability to facilitate a relatively predictable experience. Returning to Star and Ruhleder’s assertions about infrastructure’s essential dimensions (e.g., embodiment of standards), we cite their claims about infrastructure’s embeddedness, its reach or scope, how it is learned as part of membership in a community of practice, its linkage with conventions of practice, and how it is built on an installed base (113). All of these dimensions speak to the ways in which infrastructure digs in.

To illustrate this point, consider the Boeing 737 family of jets, the best-selling commercial planes of all time. At any time, around 1,250 737s are in the air, and a new one is taking off approximately every two seconds (Brady, The Boeing 737 1). The original 737 models were designed in the mid-sixties with the engineering techniques of the era. Newer 737s are still built on the same core techniques. And while 737s are by no means obsolete, they weren’t built to accommodate some more extreme contemporary weather patterns that have evolved since the 1960s, like the one that shifted our course in 2014. Instead, they continue to reflect tentative midcentury climatological models and theories of engineering that limit the altitudes 737s can achieve, the maximum crosswinds they can fly through, and the maximum precipitation they can take off in (Brady, “Limitations”). So why not just adjust aspects of the 737 design? Updating 737s to account for the increasingly violent weather is no simple enterprise. It would require an enormous number of changes to entrenched engineering and manufacturing techniques. Any given part of a 737 sits at the juncture of multiple economies. It is “‘sunk’ into, inside of, other structures, social arrangements and technologies,” its production reaches “beyond a single event or one-site practice,” it represents the work of those laboring in a community of practice, and it is built on an installed base of manufacturing techniques (Star and Ruhleder 113). Changes to infrastructure have a cascading effect because infrastructures are inherently relational and highly entrenched (Bowker 111). The entire relational structure can be compromised when any constituent part is thrown into flux (Star and Ruhleder 113). And so infrastructures are self-perpetuating networks (Bowker et al. 105) with far-reaching consequences because they can materialize “the situated knowledge work of the past and [use] it to organize the present and future” in ways that may or may not make sense (Johnson, “Information Infrastructure” 1).

The Airline Passenger Bill of Rights Act passed by the 111th Congress in 2009 and expanded in 2011 is instructive here (“Legislation”). This legislation prohibits planes from sitting on the tarmac for more than three hours and requires airlines to provide access to water and bathrooms in the event of delays (“U.S. Department of Transportation”). The bill arose in response to a series of horror stories of passengers trapped on immobilized, hot planes sitting on runways for hours, able neither to deplane nor take off (see Zeller, for example). But as Amy Cohn and Peter Belobaba argued before the bill was even passed, the solution isn’t exactly a panacea. They went so far as to claim, “It won’t work.” And subsequent lengthy tarmac waits suggest they have a point (see Boroff; Pawlowski). Cohn and Belobaba were quick to note that the common-sense provisions that permit longer delays for safety, security, and air traffic control render the legislation ineffective. Flight delays for ongoing thunderstorms, for example, are excluded because it’s not safe for ground crews to guide planes back to gates. Even if it wasn’t extremely hazardous, chances are there would be no available gate to return to since they would all be occupied by other stranded planes. Non-international airports aren’t equipped to handle customs and passport control, so international flights diverted to non-international airports are also exempt. We think Cohn and Belobaba’s observations testify to more than legislative impotence. They also speak to the cascading effects of challenges to established infrastructure and its endemic constraints, even under the most extreme circumstances.

We offer these examples as proof of infrastructure’s inertia and weight. It’s held in place by powerful economic incentives and historical inheritances. Which is not to say infrastructures are monoliths that completely delimit all decision-making possibilities. In fact, they are at their most flexible during moments of strain when rules and protocols are more open for interpretation. And so we also explore the agentive possibilities of infrastructure instead of limiting our analysis to the ways infrastructure circumscribes.

Agency and the Possibilities of Phronesis

For example, our multiple course corrections on January 6-7, 2014, suggest a great deal of agentic momentum. Meredith changed flights from Indianapolis to Flint to Chicago before 8:00 a.m. She moved between Midway and O’Hare and back again. Nate drove on intermittently closed and dangerous roads. Many of these choices required the assistance of human agents of authority (e.g., the airline’s customer service representative who searched four states for available flights). Normal infrastructural constraints became more flexible mid-breakdown; different practical possibilities emerged and these outcomes were influenced and changed through rhetoric (e.g., Meredith talked her way onto a flight to Chicago and no fee was imposed). Infrastructural junctures create rhetorical possibilities that benefit from wisdom. Because of the impromptu quality of these occasions, the rhetorical concept of phronesis plays a pivotal role in these deliberative moments.

Phronesis has most generally referred to deliberate action that results in virtue or a well-ordered state (Warnick 305). It is understood as a form of knowledge that complements nous/dianoia (general intelligence), episteme (scientific knowledge), sophia (theoretical knowledge), and techne (craft). Phronesis has a flexible relationship with rhetoric, which is defined in the Aristotelian tradition as a techne: an art for making change through argument. Rhetoric, at least virtuous rhetoric, is tied to phronesis in that phronesis provides its grounding for ethical decision-making. From her reading of the Nicomachean Ethics, Barbara Warnick suggests the “functions of phronesis are to use the products of techne wisely, to deliberate well about what is good and advantageous, and to command right action” (305-306). Moreover, phronesis is grounded in experience (306). Phronesis, as applied and advantageous wisdom, is predicated on rhetoric while also providing the foundation for future rhetorical action. Aristotelian argument is founded in phronesis, which uses other types of knowledge to decide on practical and sensible action.

In our story, infrastructural disruptions created rhetorics which invited renewed phronesis. The catastrophic weather disrupted flight schedules. The airplanes were forced to make their flights at times different than originally planned. Many of the flights were canceled or rerouted in air to safer locations. As a result, the semi-predictable behavior of the human labor involved—passengers, crews, technicians, communicators—changed to accommodate the unexpected disruptions. Passengers were kept in safe locations, crews were kept on call, technicians and communicators worked longer hours. People were scared, angry, and frustrated (“Arctic Blast”; Lawrence). As a result, the normal rules and procedures for conduct were altered because they no longer made sense for employees or passengers. Boarding a plane in a timely manner only works when the plane can be boarded, for instance. As part of the new accommodations, in some cases passengers were rerouted to safer locations, delayed at airports for days, or sent to hotels. The airline’s support for safely transporting passengers to different locations required new standards, classifications, protocols, and algorithms that could handle the catastrophic events and resulting cascade of infrastructural changes. These are moments when agents were making difficult decisions that demonstrated wisdom. Phronesis helps explain human agency in our story, but it is theoretically insufficient for understanding the vast agentic network of actors and actants that determined the course of Meredith’s bag’s return trip to Tampa.

. . . . .

Meredith's lost checked bag was one of thousands of dislocated bags that week, fallout from the historic weather disruptions that had wreaked havoc on the entire travel system across the country. And when it became apparent that the airline couldn’t confirm her bag’s location or promise its return, we bought replacement clothing and toiletries and vacationed in Chicago without the bag. It wasn’t until six days later, the day before her return flight to Tampa, that the airline managed to locate the bag in Flint and send it to O’Hare. The airline was too overwhelmed with luggage complaints to physically return the bag to Meredith at the hotel, so they left it lined up at baggage claim along with hundreds upon hundreds of other black rolling bags that had been on their own adventures that week.

Meredith returned to Midway for her flight home to Tampa. Now, in addition to the original not-quite-50 pounds of clothing and toiletries she had originally packed, Meredith stuffed all of the replacement items purchased in Chicago into her suitcase for the return flight. That bag, which originally weighed just a little below 50 pounds, now weighed in at 58.1 pounds on the airport luggage scale. The airline’s customer service representative explained the bag would be assessed an additional $75 fee for being overweight. Meredith attempted to convince the airline representative to take the checked luggage without a fee even though it was over the weight limit.

Meredith tried a number of tacks—all unsuccessful—to convince the clerk to waive the fee: the extra weight was the result of the airline’s original error, the bag was only about 8 pounds over the limit, the airline “owed” us some goodwill for the weeklong hassle of going without the bag. He didn’t budge. By the time Meredith asked to see the supervisor, a considerable backup of travelers was queued up behind us. The supervisor held firm. Meredith asked to contact corporate on the spot, but headquarters was in another time zone and wouldn’t open for several more hours. The on-site supervisor was the end of the line, at least for now.

The supervisor reviewed our options with us. We could pay the fee. We could remove the extra weight and relocate items to a carry-on. We could purchase a second bag, sold by the airline at the counter, and check a second bag for no additional fee (beyond the cost of the bag itself). What we could not do, though, was continue to hold up the line. So we were asked to move to the side, and management relocated the bag to an unoccupied station at the counter also equipped with a scale. That second scale read 47.1 lbs. When the second scale gave a different reading than the original, the formerly unyielding supervisor became apologetic and willing to negotiate. We were instructed to move a few items out of the bag and into the backpack Meredith had worn upon arriving in Chicago. The bag now weighed in on a third scale at 52.2lbs. Though still technically overweight, it was checked without an additional fee.

. . . . .

Agency, Infrastructure, Time, and Nonhumans

During Meredith’s return flight, two different working infrastructures collided: (1) the emergency infrastructure that had governed Meredith’s outbound flight from Tampa and that she was still operating under, and (2) the status quo infrastructure of standard airline travel that the airline officials had reverted to as quickly as possible. During the emergency infrastructure, passenger safety became the priority. Passengers were moved (or not moved) to keep them safe. Stuff got lost. Classifications, standards, protocols, algorithms, and time- and space-spanning techniques were reconsidered and some procedures were subordinated to focus on life-threatening weather conditions. As quickly as possible, however, the status quo infrastructure kicked back in. Its resilience is unsurprising given that status quo infrastructures are governed by logics (like their emergency counterparts), but, as we’ve argued, they’re also powerfully entrenched.

These two infrastructures were brought into competition during Meredith’s disagreement at the baggage check on her return flight. The temporary emergency infrastructure allowed for more phronetic reflection. Meredith argued that the same deliberate phronesis of emergency infrastructure should govern her baggage on the return flight because the bag’s weight was a consequence of emergency infrastructure. The employees reasoned that the weather had stopped interfering with flights and normal operating procedures had resumed. The parties disagreed about stases, the issues in dispute, and invoked their favored infrastructure to substantiate their claims.

The airline representatives relied on arguments of fact, focusing on conjectural stases and the pertinent standards, protocols, and classifications that best supported their position. For example, they pointed to rules (i.e., protocols) governing the weight allowance (i.e., standard) for any individual bag over 50 pounds (i.e., classification) and noted the bag in question exceeded that allowance. Meanwhile, Meredith insisted on qualitative stases and argued about what she perceived to be fair. Her argument relied on the protocols of the emergency infrastructure, positioning her baggage as an effect of that infrastructure so that it would be treated as part of the emergency (i.e., classification). When the glitching scale, an infrastructural monster, emerged as an actant in the discussion, the airline representatives’ classifications were called into question, as was the status quo airline infrastructure they had returned to. The scale’s glitch produced logics more appropriate for emergency infrastructure. The airline employees then took up positions as emergency infrastructure labor once again and exercised phronesis to renegotiate operating procedures.

Figure 1

When the status quo infrastructure is fully operational and the scale is in compliance with that infrastructure, it classifies inconspicuously, translating shipping service into pounds and dollars.

The scale-as-monster’s deviance rendered it and these moves more plainly visible, throwing the dominant infrastructure back into view and, consequently, into question. This was “a shift from seeing technical objects as static entities to conceptualizing them as temporary coalescences in fields of conflicting and cooperating forces” (Hayles 86). The scale exerted force in the argument. This by itself isn’t necessarily interesting as new materialism theorists have been making this argument for decades. What is more interesting is what can be deduced about a type of agency being exerted by this particular scale.

The scale was used to determine oversized bags. Bags that were over 50 pounds would be charged at $75. Bags over 100 pounds were prohibited. We suggest thinking about this type of material agency as critical mass agency. It is exerted when material objects translate a phenomenon into a number to match against a threshold. In this case, part of customer service was defined as baggage shipment. Baggage was translated into weight in pounds. That weight was then checked against two thresholds. If the weight met threshold #1, customer service would be valued at $75 more. If the weight met threshold #2, customer service would be denied. Valuation was necessarily built into the normalized airline infrastructure and enforced through the scales deposited at each station. When the scale calculated unpredictably, its ability to abet status quo infrastructural values receded as well. The scale had been serving as a source of objective ethos, a position it could no longer occupy as a monster.

Conclusions and Implications

What can we learn from this research? We are suggesting that infrastructural breakdowns created multiple instantiations of infrastructure, each with their own unique rules. Invoking and bridging these instantiations created rhetorical exigencies for humans and nonhumans alike. The baggage being shipped acted as a boundary object moving between infrastructures (Star and Griesemer). The scale, one of the primary tools for interpreting this boundary object across infrastructures, haphazardly exerted critical mass agency. This glitch brought infrastructural flux back into perspective and reauthorized airline representatives’ phronesis (see table 1).

Table 1: Summary of Events and Infrastructural Effects

Event	Infrastructural Effects
Meredith books air travel	Status quo infrastructure.
Outgoing flight interrupted by Polar Vortex	Phronesis performed by agents to initiate emergency infrastructure. For example,     plane change fees waived.     baggage shipping subordinated to threatened passenger safety.
Meredith checks in for return flight	Status quo infrastructure enacted by airline agents. Meredith is still operating under emergency infrastructure. Parties cannot come to terms over stases and invoke mechanisms of favored infrastructure to support opposing positions.
Airport scale glitches	Scale exerts critical mass agency unpredictably and resolves human dispute.

We observe that airline infrastructures are designed to account for many of the disruptions and glitches we document in our story. Infrastructure balances a tension between standardization and flexibility in the name of resilience (Hanseth, Monteiro, and Hatling 407). It would be plausible to interpret our case as that of one normally functioning (though straining) infrastructure responding to catastrophe as intended rather than the result of two infrastructures at play. We acknowledge the appeal of a simpler explanation, but we ultimately think that explanation would be insufficient.

We are not the first to note how multiple infrastructures can compete to background the same action. In her work on bioterrorism preparation, Lisa Keränen noticed a similar issue involving multiple infrastructures. Keränen investigated TOPOFF 3, a “congressionally mandated weapons of mass destruction (WMD) preparedness drill” (227). During TOPOFF 3, scripts and staging were used to prepare participants for a potential bioterrorism attack. These materials foregrounded one possible set of responses by the nation’s defense infrastructure (to the exclusion of other responses). Keränen argues that the documents used for the preparation were “fantasy documents,” organizational planning rhetorics that differ markedly from the actual lived scenarios they model (233). Sociologist Lee Clarke says that these kinds of documents can help prepare publics for disaster, but that they are incomplete (86). When actual organizational catastrophe occurs, participants and technologies frequently behave differently than fantasy documents invited them to act. In other words, when catastrophe happens, a different infrastructure emerges, one that is not the fantasy world staged in preparedness documentation. We suggest that the catastrophic events we encountered created the same diverging infrastructures that Keränen notes surround TOPOFF 3.

Further, we reject the notion that a single infrastructure was stretching to do its job because the resources involved in the deliberation met none of the established criteria for resilient infrastructure. Resilient infrastructures are characterized by their resourcefulness, redundancy, robustness, and rapidity (Scholl and Patin 34). Resourcefulness refers to infrastructure’s ability to identify problems and mobilize resources fluidly. Neither human agents nor the weigh stations could effectively narrow the issues of stasis, hence we found no evidence of resourcefulness. Redundancy indicates that resilient infrastructure has backups that can take the place of failing resources. Although there was more than one scale, the scales were duplicate technologies. When one scale malfunctioned, it was difficult for the airline employees to know which scale was correct. In addition, there were no means for assessing the cost of luggage other than the scale. Hence there were deficiencies in redundancy. Robustness points to resilient infrastructure’s ability to withstand a given level of stress without losing functionality. The scales were not under stress but lacked in functionality. The glitch’s cause wasn’t detectable to any of the people involved in the argument. Finally, rapidity refers to the speed with which an infrastructure can recover from catastrophe. The infrastructure wasn’t able to respond quickly enough to resolve the issue before Meredith needed to board the plane to make her flight home. And so the suggestion that our story features one resilient system operating as intended, while appealing in its simplicity, is unsatisfying.

Yet while resiliency cannot explain away the multiple infrastructures at work in our case, it does provide us with a better understanding of deliberation and phronesis in infrastructure. When infrastructure breaks down, rhetors may guide deliberation by considering which attributes of resilient infrastructure could synthesize competing infrastructures in terms of resourcefulness, redundancy, robustness, and rapidity in both their social and technical aspects. Guiding phronesis through these heuristics invites participants to consider problems through the lens of infrastructure, asking them how it is an important partner in deliberation. In our case, we might have asked how a third party’s scale could introduce infrastructural redundancy and thus help produce deliberative evidence to more rapidly resolve the dispute. This article’s research thus adds to rhetorical scholarship by suggesting that infrastructural phronesis and deliberation can be guided by considering the attributes of resiliency. In addition, this article contributes to rhetorical theory with the introduction of the term critical mass agency. While the concept of “rhetorical actancy” is useful for acknowledging distributions of agency (Gries 73-75), the field also needs notions such as critical mass agency as it begins to identify the nuanced roles that nonhuman objects play in a specific phenomenon. Latour helps rhetoricians understand that things bring different degrees of power to any given relation. Critical mass agency helps the field tune into these degrees of power by moving towards figuring out which precise roles those things play.

Questions to investigate in the future include: How can long-term infrastructural change be initiated by adding material critical mass agency? How can we better understand critical mass agency? Are breakdowns special moments for change, or do they just highlight possibilities better than most occasions? How can phronesis be better described to account for nonhuman agents? Finally, because our analysis was performed with post hoc data, we suggest that future studies that analyze scalar devices would help to better understand infrastructure and rhetoric. In particular, conversation analyses of arguments involving scalar devices could highlight when and how scale becomes an important infrastructural rhetoric.

. . . . .

Epilogue

A few days after returning to Tampa, Meredith called the airline’s corporate headquarters and began recounting her trip. Before she could make it very far into the story, the airline offered her a $250 travel voucher. She happily accepted.

. . . . .