a. The rich discussion around dis/ability

Being aware that readers of this journal are familiar with discussions and theories of disability, this section will, first, provide only a brief description of the origins of the medical versus the social model of disability. It will then focus on the scholarship of concern to this particular research's framework, i.e. ableism studies.

In the United Kingdom, the social model of disability was largely borne out of the strong relationships between disability rights activism and disability studies as a discipline. In the early 1970s, the Union for the Physically Impaired Against Segregation, in discussion with the Disability Alliance, proposed the fundamental principles to disability along with the definition that "disability is a situation, caused by social conditions" and, importantly, that it is "something imposed on top of our impairments" (UPIAS/DA, 1975, pp. 1 & 3). The social model of disability emerged from this foundational principle that, while impairment referred to a bodily condition (divergent from human-specific, expected abilities), disability was a social construction. In UPIAS's words, disability is:

the disadvantage or restriction of activity caused by a contemporary social organisation which takes no or little account of people who have physical impairments and thus excludes them from participation in the mainstream of social activities. (UPIAS, 1976)

Disability is therefore all of the societal aspects that restrict disabled people's abilities to fully integrate into society (Oliver & Barnes, 2006; Thomas, 2004), including inaccessibility to public transit. The social model of disability has, and continues to play, an important role in countering the medical model, which individualizes disability to a personal characteristic that ought to be remediated by medical professions.

The social model has had some critiques levelled at it, primarily due to a scholarly concern that it did not allow for a deeper understanding of the lived experience of disabilities and impairments. In other words, this model arguably lacks concern for issues of embodiment (including how pain and impairment may also restrict activity), thereby disregarding the power that disabled people have over themselves and their bodies (Crow, 1996; Paterson & Hughes, 1999; Shakespeare, 2006; Shakespeare & Watson, 2001; Thomas, 2004). These critiques have enriched the field of disability studies in unmeasurable ways, deepening our understanding and framing of the making of disability. From these critiques and deepened discussions of the place of body, I borrow some of the work in more phenomenological approaches to disability, namely through the work by Patterson and Hughes (1999) who have integrated some of Drew Leder's philosophy of the body into analyses of carnal politics. These scholars have proposed integrating the concept of dysappearance (Leder, 1990), which works in direct contrast to the idea of "disappearance". Whereas the latter would characterize something ordinarily functioning, dysappearance, on the other hands, brings focus to moments of disruption, pain, etc. This concept will be of importance to the work below.

It should nevertheless be pointed out that the social model has also always had a concern for the differences between disability and impairment and, therefore, has paid attention to the impaired body (hence the distinction between the two 2). It should be noted, however, that some of the success and subsequent adoption of the social model as a theoretical framework stems at least in part from its links to political activism (Goodley, 2014). Indeed, it is a useful tool to theorize and demonstrate the institutionalized character of disabled people's exclusion from society (Barnes & Mercer, 1997; Beckett & Campbell, 2015; Blume & Hiddinga, 2010; Oliver, 1992, 1993). It is because of this political power that the social model of disability is adopted in the framework of this research––it allows for a discussion about accessible transport that is concerned not only with the shape of infrastructures but also the ways that that they directly impact and, as I will show, influence the exclusion and inclusion of disabled people.

A more recent wave in disability studies, which draws substantially from the social model of disability, provides theoretical heft to discuss not only the barriers faced by disabled people. Rather, ableism studies inquire into the origins of these barriers, questioning the current social order and shifting concern to the relational structure of dis/ability and how or why the 'line' between them is drawn. Scholars in this wave are concerned with ableism, defined as a preference for "species-typical normative abilities" over other abilities that may result in policies that "[lead] to the focus on 'fixing' the person or preventing more of such people being born" (Wolbring, 2008, p. 253). It is argued that ableism is particularly prevalent in Western societies with neoliberal capitalist structures, wherein ideals of productivity and work depend upon an underwritten concept of compulsory able-bodiedness (McRuer, 2006). Hence, this paper will refer to neoliberal ableism––that is, a society in which an idealized able body is perceived as integrated to the workforce, and therefore socially catered to (Goodley, 2014; Goodley, Hughes, & Davis, 2012; Goodley, Lawthom, & Runswick-Cole, 2014). Consequently, where the neoliberal-able body is idealized, deviations from it are vilified, medicalized, and excluded (Campbell, 2001, 2008).

b. Work on infrastructures

While disability studies provide theoretical frameworks for an understanding of disability as a social process rather than an individualized medical phenomenon, research in the field of science and technology studies (STS) provide the analytical tools for researching infrastructures such as the transport system. Below, I briefly summarize the scholarly work that frames the concept of infrastructures/networks (the terms will be used interchangeably in this article).

In the field of STS, conceptions of networks and systems have been around for roughly forty years. More often than not, scholars' first reactions are to link back to the actor-network theory (ANT) and, indeed, some work on transportation systems have used this approach––the most notable example being Bruno Latour's book, Aramis (Latour, 1996). ANT plays an important role in the theoretical framework of this research, partly because some of its key elements are used and partly because it is critiqued. To summarize, ANT is a conceptual tool that uses a symbiotic approach to researching society, moving beyond distinctions of human and non-human actors to discuss how all of these actors need to work in conjunction in order to succeed in particular goals. ANT theorists focus on stories of alliances (or failures thereof): how actors are brought together (successfully or not) and what their functions are within the network. It is important to emphasize that, in this literature, networks are therefore understood as more than just 'things': a transport system is more than the tracks, the buses, the trains, the computerized signage system. It is also the employees, the users, the managers, and also the legislation and regulations in force.

This approach has its limitations, however. Researchers in the ANT tradition, particularly in the 1980s, focused primarily on moments of inception of networks: how actors are brought together by innovators (Callon, 1986; Latour, 1988). Less attention was given to those who were not included in the process of bringing the system together, who were excluded throughout the process, or how, when brought together, these systems were maintained/kept intact.

Other scholars have further developed ANT in attempts to address these criticisms. Among them, Susan Leigh Star and various co-authors provided the foundations for what is broadly called infrastructure studies today. Infrastructure studies began as research into processes of standardization and stabilization of systems, investigating the role that standards and categories have in shaping and maintaining them. Authors argued that the development of standards is an important step in reality-building as they unify systems through a cohesive, often technical, language (Brunsson & Jacobsson, 2000; Busch, 2011; Ritzer, 1983; Scott, 1998; Timmermans & Berg, 1997; Timmermans & Epstein, 2010). As such, for infrastructures to function and meet their goals, standards have to be developed. Yet this language is dependent on the worldviews of the actors involved in the development of these networks and their definitions of the network's goals, its users, and what constitutes efficiency. Inevitably, the oversimplification of the complexity that is reality creates "others" who are outside the boundaries of predetermined categories.

Star's work is important in this context as a critical ANT theorist. Discussing the process of standardization, she and Geoff Bowker argue that, while the process of standardization is expected and, indeed, "inescapable", a series of choices in categories are made––choices that may result in dangerous consequences (Bowker & Star, 2000). The danger these authors refer to is exclusion; when actors are left at the boundaries of a network, they can suffer in many ways. Networks, due to standardization practices, are inept at dealing with complexity beyond their categories. At these points, they become ineffective at reaching their goal, and break down (Lampland & Star, 2009; Star, 1991).

Many theorizations of infrastructure describe these networks as invisible, or at least as "designed to become invisible as [they are] stabilized" (Lampland and Star, 2009, p. 207). In other words, these infrastructures become so commonplace to its users that they have a tendency of melting into the background––only becoming visible again upon breakdown (Edwards, Bowker, Jackson, & Williams, 2009; Graham, 2010; Jackson, Edwards, Bowker, & Knobel, 2007; Lampland & Star, 2009). This definition of infrastructures has become so common as to become almost invisible itself, to the point that authors saw it as a primary characteristic that infrastructures must fulfill, i.e. "any genuine infrastructure is mostly invisible" (Edwards, Bowker, Jackson, & Williams, 2009, p. 370). Other scholars have questioned the concept of invisibility, pointing to the opposite: due to the need for constant repair and maintenance work, infrastructures are, rather, mundanely visible (Harvey & Knox, 2012; Larkin, 2008, 2013; Ureta, 2014). However, I have found this binary of invisibility and visibility profoundly unsatisfying, as it is often on the basis of breakdown or disrepair that these distinctions are made. For example, some lines are often drawn around infrastructures of the global South as particularly unstable and, as a consequence, particularly visible (Edwards, 2003). Meanwhile, infrastructures of the global North where resources are, arguably, less scarce and politics more transparent, would be more invisible. However, through this type of categorization, London's public transport system would fall under the invisible category: it is a network with significant financial subsidies, that enables billions of journeys every year. It would, from a macro-analysis, be seen as functioning and, thereby, invisible.

In studying infrastructures and placing it at the forefront of analysis, I will argue that an analysis that deems infrastructures as "invisible" is one done from a privileged and partial perspective. Indeed, networks flitter back and forth between function and dysfunction, and, importantly, their function and dysfunction are also dependent on the passenger and their journey. As dynamic monsters––a collection of technologies, materialities, interests, actors––infrastructures produce many paradoxes (Howe et al., 2016; Larkin, 2013). An analysis that takes these paradoxes into consideration can afford to grapple with the complexity of infrastructures, beyond the in/visibility binary.

c. Bringing STS & Disability Studies together (again)

Readers may be wondering here how these literatures, from two rich and diverse disciplines, are being brought together to provide the analytical framework for this article. From that perspective, this article is not the first of its kind: other authors have worked towards building the bridge between these two disciplines before. Mol, for example, has explored how bodies and illnesses are made multiple through the variety of contexts that they enter (Mol, 2002), and Blume has studied assistive technologies such as cochlear implants and their adoption or rejection by Deaf communities (Blume, 1997, 2009). Similarly, ableism scholars have discussed assistive technologies through a critique of human enhancement and the transhumanist movement that conceivably create the possibility of a new social class, the "techno-poor disabled" (Wolbring, 2008, p. 254).

There have also been ventures of STS scholars into disability studies to attempt to use STS concepts as a framework to explain the creation and ordering of disability (Galis, 2011; Moser, 2000). Often based on critiques of the social model regarding embodied and enacted experiences of disability, these STS-inflected "models" of disability are, generously read, an attempt at hybridizing the two fields. Galis (2011), for example, sees in ANT a way of thinking beyond "modernist divides" (p. 828), allowing for an analysis that simultaneously accounts for diverse barriers, environments, policies, and bodies. A decade prior to Galis' piece, Ingunn Moser had also explored the theoretical contributions that ANT might hold for disability studies. She and Galis agree on some of the promises of ANT as applied to disability, namely that it allows for an understanding of disability beyond being "a condition", but rather a "result of specific relations and configurations" (Moser, 2000, p. 224). Moser also identifies some limitations in this application which are worth highlighting: in order to follow an ANT model of disability, one must dissolve bodies––agency becomes distributed and that encumbers analyses that require a more traditional focus on individuals as subjects. This includes a category important to this article, namely that of users and passengers of public transit. As Moser eloquently puts it, "Given the demands disabled people are confronted with daily to live up to the standards of normative subjectivity, I also want to acknowledge and attribute to [them] these subject positions." (Moser, 2000, p. 225). 3

Both of the examples above are interesting applications of STS theories in their own right, but are named here as examples of an analytical approach that this article does not wish to follow. Indeed, Galis argues that his article intended to propose a "theoretical alternative" to disability studies scholars and activists alike (Galis, 2001, p. 835), a response to critiques to the social model through the use of STS literature. I wish to be wary, however, of entering disability studies to suggest renewed theorizations of the field's main concept from the perspective of another discipline that has different goals of understanding and analysis. As discussed above, the social model of disability has a rich history and it continues to evolve and be refined within the discipline through the scholarship produced in journals such as this. Rather than use the concepts of one field to provide alternative models for another, my aim here is to create a fruitful partnership between them as they each have their own strengths. STS therefore provides robust theorizations and concepts to tackle the process through which infrastructures and networks are made, shaped, molded, and maintained. However, as has been pointed out by many critics of ANT specifically, STS has not always provided apt analyses of power relationships. As John Law has put it, "STS finds heroes to be more interesting than ordinary folk." (Law, 1991) And––whereas it is true that the stories explored by STS in the past quarter century have changed thanks to critiques––disability and, more particularly, the impact of disabled people in the shaping of artifacts and networks, is still a less oft discussed topic. The rich literature from disability studies can offer robust concepts, being a discipline historically concerned with power and exclusion.

Hence, while this article is largely focused on infrastructures and networks––through the case of public transit in London––and uses STS categories and theories to analyze these structures, the work is done on a disability studies canvas that embraces the recent theoretical developments of the field to conceptualize exclusion of disabled bodies in contemporary society. Rather than defaulting to punctual application of one field's concepts to another field's subject of study, the aim is to fortify bridges between these fields while embracing each one's analytic strengths. This is done in the hopes of creating a truly symbiotic, interdisciplinary approach.

a. Data collection

The primary source of data was in-depth, semi-structured interviews, recruited by a combination of purposive and snowball sampling. Calls for participants were emailed to stakeholder groups, posted on my personal social media account (and retweeted over 200 times by a variety of accounts), and distributed by charities and organizations to their members and other interested parties.

A total of thirty-four participants, of which twenty-seven were wheelchair users, were interviewed. Non-wheelchair users included an engineer in public transport, policy advisers, and partners of wheelchair users who accompanied them in the interview. Wheelchair interviewees used a variety of types of wheelchairs––manual, electric, and hybrid––and some owned more than one wheelchair. Distribution in terms of both gender and age of participants was relatively spread though soon in the interview process it surfaced that the differences among age and gender were relatively minimal in experiences and impact discussed with the interviewer. Interviews were undertaken until it was deemed that a saturation point had been reached in the narratives told, i.e. until ''no new information or themes [were] observed" in the process of doing new interviews (Guest, Bunce, & Johnson, 2006). All interviews were audio recorded and transcribed by the author, and interviewees chose their own pseudonyms used here (some chose to waive anonymity altogether, a choice respected according to recommendations by Kaiser, 2009, and Giordano et al., 2007).

A second source of data were observation sessions that took place on three occasions: a training session where wheelchair users are taught new ways of maneuvering their wheelchair, a "Disability Roadshow" at a bus depot where bus drivers and disability charities were invited to discuss accessibility concerns, and an afternoon spent travelling with a wheelchair user, Alan, who will figure prominently in discussions below.

Finally, a documentary corpus was collected, consisting of official documentary records (such as those created by the English government, including legislations and Select Committee inquiries) and commercial media accounts (including newspaper articles and public-facing websites). In total, the corpus of data for this research consisted of over 274 000 words of transcribed interview material, 10 000 words of observation notes, and over 150 documents.

b. Analysis

An interpretative data analysis approach was used to sort through the vast corpus of data produced during fieldwork, inspired by a grounded theory approach (Lewis & Ritchie, 2003, p. 201). All documents and transcriptions were uploaded to the CAQDAS software, NVIVO, to facilitate the work of analysis in two phases. Firstly, an initial first and second reads of the materials (focusing on interview and observation data) allowed for the identification of categories and tropes in the data. Secondly, these categories were reassembled into larger thematic and theoretical groups through the use of coding groups and memos (Saldaña, 2015 [2009]; Yin, 2015) The work there focused on identifying broader narratives of how wheelchair users use public transport in London, and the impact it has on their lives. The following sections first provide the historical context of London's transport infrastructure to then discuss the empirical findings of the data collected while weaving it with theoretical contributions.

a. On Dysappearing Bodies and Visible Infrastructure

Describing the experience of using public transit in London, it was apparent in interviewees' narratives that they often felt unwelcome or not thought of in that space. I began this article purposefully using Michael J.'s quote in which he refers to feeling outright invisible. In this section, through empirical data, I will illustrate some of the manifestations of unwelcoming disabled bodies in transportation, then relate these examples back to a theoretical discussion about the invisibilizing of disability through infrastructures that act as "society made durable" (Latour, 1991). Namely, I will argue that this is a case of disabled bodies being made to dysappear as a symptom of being invisible to the infrastructure, whereas, paradoxically, the infrastructure is highly visible to them.

Obviously it is a Victorian system, and obviously back then, people with disabilities weren't really considered important. –– Carl

Carl was among numerous other interviewees who brought up the legacy of London's public infrastructure as a source of numerous barriers to the inclusion of disabled people, broadly speaking. And, indeed, as we can gather from the abridged history of transport above, transportation in London was developed and largely consolidated during an era where disabled people did not figure as potential passengers of the transport system. As discussed, the disconnection between the history of transport and the history of disability in the United Kingdom has meant that it wasn't until the rise of the disability rights movement that changes in the system began to occur to render the network more accessible.

Scholars in the field of infrastructure studies remind us it is incredibly difficult to rebuild systems from scratch, and they "[do] not grow de novo" (Star, 1999, p. 382) –– decisions from the past embed and literally become material in infrastructure, making radical deviations difficult. In the case of public transport, its development in 1800s Victorian London, and the continued existence of neoliberal ableist ideals, there is little surprise that the system continues to demand largely compulsive able-bodiedness from its users (McRuer, 2006). This isn't to say that changes are impossible within infrastructures: various improvements in accessibility have been won in the past decades in London, from the gradual, 11-year, introduction of low-floor buses (though imperfect, much more accessible than the Routemaster alternative), to the addition of elevators and manual boarding ramps to Underground and Overground stations, to the development of new training programs for employees. These gradual changes, attempts to bring the system up-to-date with new stakeholder demands, might be dubbed retrofits (Howe et al., 2016). Retrofitting becomes a manner through which to embed new demands and new technologies in spaces that can accommodate them.

Despite these retrofits towards a more accessible transportation system, wheelchair users still experience a number of frustrations surrounding these fixes, including broken technologies, space negotiation with other passengers, and ill-tempered drivers. 5

The wonder of technology is the wonder that it ever works. –– Michael J.

Since 2005, then, all routes in London are serviced by low-floor buses with mechanical ramps and a wheelchair priority area onboard. Yet, despite this, buses are still prominent in conversations with wheelchair users about their struggles with transportation. Michael J., for example, worries about the ramp breaking:

The ramp might get stuck so I'm stuck on the bus for 10-15 minutes until they can get me off… So there's lots of things that I have to think about. – Michael J.

Of the twenty-seven wheelchair-using interviewees, fourteen explicitly mentioned that they had been unable to board or alight a bus due to broken ramps. Broken elevators at train stations (both of the Underground and Overground varieties) also featured in conversations:

[Y]ou go on the TfL website, you find out that it's wheelchair accessible from platform to street, and then you get there and find out that the lift's not working. – Marie

And, while the static Transport for London website might claim these stations as accessible, it invisibilizes the brokenness of many of these access artifacts. Other websites and Twitter accounts, such as Up Down London, attempt to collect up-to-date information and make these infrastructures visible again to users who need them, publicizing the status of elevators at the stations. Yet, barriers like ramps and elevators are easy topics of conversation: they are easily recognized as useful assistive tools for wheelchair users and indeed are often the first "solutions" proposed. Though they are sources of unreliability, there are less discussed aspects of transportation that render disabled bodies particularly visible in transportation, too:

So I don't like the fact that there's the siren that starts wailing at you, or at everybody, when you're about to get on or about to get off the bus. It's all a bit of a big faff, but you get used to it. I mean, public humiliation seems to be… you've got to be able to deal with it if you're disabled. – Sophie

The siren Sophie is referring to in this extract is a loud alert to other passengers and passers-by that the ramp is being deployed––perhaps to ensure that space is made for them to board or alight, or so that others are not caught off-guard on the street by the ramp's deployment. Whatever the designer's intention, this loud siren demands attention––attention that, for wheelchair users, can be unwanted and turns attention not only to the ramp that is put in motion, but also the person who requested it. It is, in many ways, a marker of difference, clearly pointing to a passenger that requires non-standard attention in a system that hasn't seamlessly integrated their needs.

It would be too simple a story to focus on various bits of infrastructure hardware. If that were the case, our story would become one of "poor design" where one might be able to technologically "fix" the issue and "solve" the problem. Yet, accessibility issues are not that simple or straightforward and using the frame of infrastructure studies on this topic demonstrates the full complexity of questions of access. As discussed in the literature review, infrastructures are made up of more than technology: they are made of maintainers, users, employees, policies, among others. Bus drivers, for example, were an often-cited character:

The good thing is that all London buses are wheelchair accessible, the problem is the drivers. [Laughs] You can have the best bus, but if the driver is not willing or whatever to push a button to open the ramp, then the best high-tech bus is not worth the money, and that's exactly what happens. – Kerstin

Whether it is for boarding or alighting the bus, stories of drivers not acknowledging wheelchair users abound in interviews. This spanned anywhere from perhaps not hearing that the ramp had been requested to what, from the interviewees' perspective, seemed like purposely ignoring their presence at a bus stop––actively rendering wheelchair users invisible. In general, staff involved in the functioning of transportation in London were often brought up in interviews, including in cases in the London Underground or Overground where staff are needed to deploy the manual boarding ramps at specific stations. Unlike the buses' automatic ramps, these are specifically created for trains where the platform is lower or higher than the train itself. As the name suggests, they require manual deployment by a station staff member with a key to unlock them from holders. These ramps therefore require careful navigation on the part of passengers who need them as they must make their presence known (making themselves visible) to ask for the ramp to be deployed at their first station, and often have to remind the staff member to call ahead to the station where they intend to alight for the ramp to be deployed there, too, if required––again, making them visible in advance:

Ah, yeah, I had an experience where the guy with the ramp wasn't there and I attempted to disembark the train and the front caster wheel got stuck in the gap and the main wheel were fine but the front one got wedged in and it was kind of a case of, you know, I hope the guy doesn't just drive off. – Carl

In Carl's case, we can again see the fear of being invisible in moments where visibility to the infrastructure is key. If the train driver doesn't see him (something entirely possible given the length of London trains) or another passenger doesn't see him in need of assistance, Carl is in an extremely dangerous position. In other cases, even when assistance did show up, the ramps may not have been secured properly, or staff put passengers on the wrong carriage which meant they were unable to disembark at their intended station. A plethora of problems can arise from this seemingly simple "solution" of the manual boarding ramp, and neither solely the technology nor solely human error can be cited as the primary guilty party.

Another manner in which wheelchair passengers in London's transport system deal with the inconsistent categories of visibility and invisibility to the system are what I have lovingly nicknamed the "Battle of Wheels'' in my coding process. Two other interviewees used similar terminology during our interview, such as Kerstin:

So that's an issue that is a massive issue with the buggy war [laughs] as I call it in London, so that people take up the wheelchair space and are not willing to move when a wheelchair user wants to board. – Kerstin

The story repeated itself many times over, and it is a simple one: a wheelchair user will board a bus and, in doing so, will find a person with a baby carriage in the wheelchair area. The debate, ultimately, boils down to what policy applies to that space: is it "first come, first served" or do wheelchair users get priority over that space? Is the person with a buggy required to move? 6 Yet, while passengers with baby carriages were often mentioned, at times they were defended by interviewees who saw bus drivers as particularly unhelpful in the space negotiating process:

But occasionally it's the bus driver who doesn't even give me the chance to negotiate with the parent in the space, they just say, "No, there's somebody in the space. You can't get on, you'll have to take the next one." My favourite phrase. – Sophie

In essence, what surfaced in interviews that mentioned struggles with negotiating the use of the wheelchair space was a profound sense of injustice of this happening in the first place:

[They'll say] "Well, they wanted equality, first come, first served." And the whole point is it's not equality if there's only one space on the bus that you can safely travel in. – Alanni

It's not about who's more important, it's about who has a choice; so I do not have a choice about my use of the wheelchair whereas a baby can be got out of its buggy. – Diana

People with children who use public transit might balk at this quote, knowing how difficult it is to navigate transportation themselves. It bears considering how, then, these passengers are rendering one another invisible to the infrastructure––if a buggy user is in the space, the disabled person who needs it may more easily become invisible and ignored. If the disabled person is there, the same can be said for the buggy user. Others framed their point by a discussion of historical political struggles:

Then wheelchair users started campaigning for a wheelchair space and the wheelchair spaces were implemented, and that dragged in its wake the possibility of bigger and bigger buggies to use that space. – Anton

According to Anton, the existence of a wheelchair space at all was clearly due to the political engagement of disability activists for accessible public transit––a significant movement in the 1990s in London. Prior to that space, passengers with small children might have struggled, but also had recourse to smaller foldable carriages and alternative child-carrying devices that still enabled them to use public transit. In some ways, Transport for London agrees with this perspective, as the space on the bus is titled the wheelchair priority area on its driver's handbook (also known as the "Big Red Book"), and its public awareness campaign with posters that read, "Buggy users, please make space for wheelchair users". A campaign that can potentially be interpreted as a way of making wheelchair users visible to others on the system.

The category of invisibility of an infrastructure, then, quickly becomes a fraught one. In some cases, wheelchair-using passengers are scared that they will be invisible to others, or that they will become hyper-visible to the point of verging on "public humiliation", as pointed to by Sophie. Wheelchair-using passengers, and all other passengers whose access needs were not materialized in the system from its outset, thereby find themselves on the networks' edges (Star, 1991). Their presence crystallizes the paradoxical nature of infrastructural invisibility, I argue, in a way that creates states of dysappearance (Leder, 1990) wherein the body, particularly the disabled body, appears as the "thematic focus of attention, but precisely in a dys-state" (ibid, p. 84). Said otherwise, being constantly confronted with the variety of barriers discussed earlier in this section, the wheelchair users' experience of this system is fraught with constant reminders of difference that are often painful or humiliating. The homophony of dysappearance and disappearance is particularly apt in our case, as I propose a closer analysis of the idea of in/visibility of infrastructure: it is not the network that is invisible to the user, but rather the user that is invisible to the network. Or, in some cases, so beyond the categories that the infrastructure is used to catering to that the infrastructure needs to draw such attention to it that it nearly sounds broken, as the case of the loud sirens in the deployment of the ramps on the low-floor buses.

In their turn, wheelchair users are easily pointing to things that they dislike in the system itself: sirens, elevators, other passengers, drivers, broken ramps. These constitutive parts of the system are anything but invisible to them. To argue that infrastructures are invisible, then, from their perspective, is to willfully ignore the ways in which that they are constantly confronted by barriers and issues in their daily travels. And, where the infrastructure does not provide flexibility for their passengers, it is the "non-standard" users that need to demonstrate malleability. In the next section, we will explore the tactical responses to infrastructural failure on the part of wheelchair users as recourse to active visibility.

b. Rendering oneself visible

Let us recall the question that originally framed this research: "How do wheelchair users use public transport in London?" It is clear from the discussion above that the use of this infrastructure is not a straight-forward one by wheelchair-using passengers. The various barriers discussed above demonstrate how these large systems, through materializing social conceptions of who its users are, pushes wheelchair users (and disabled passengers more generally) to its margins. Yet wheelchair users still do use this transit system. In this section, we unpack the ways in which they do use it through considering the various tactics developed to counter some of the barriers encountered, seeing a common theme of active visibility through them.

We're always coming up with ideas. When somebody says, "you can't do that", 1) Can't doesn't exist in the English dictionary, and 2) Watch me. – Adam

Among the more daring interviewees, Adam summarized rather well the spirit that was present in all interviewees of finding ways of making the infrastructure work in their favor, whatever anyone else might say about accessibility issues. One might easily spend the rest of this paper listing the variety of ways in which this might take place, but in interest of brevity, we will focus on a small selection of individual tactics, then discuss some collective strategic work.

[A]ll of the ramps on the network are secured in holders and to get access to those, you need a thing called a T- Key which is just a square spanner, really. So we've got one of those because the number of times that you go places and people say, "I've lost it, I can't find it." […] So we carry [it] around, and it costs 3 quid from eBay, you can just say, "There you go, we've got ours." – Alan

I've been out and bought a two-foot ramp so I'm going to have a ramp on my wheelchair. If I do get somewhere and I need help, I can just have somebody to hold the ends while I get out. – Marie

Developing a personal toolkit to counter inaccessibility was one common tactic discussed by interviewees. This toolkit may contain anything from the T-Key and a screwdriver all the way to a lightweight portable ramp carried on the back of the wheelchair. It became clear that these tools are a comfort to their users, as well as time-savers, after spending an afternoon travelling with Alan (who, like Marie, carries his own portable ramp):

Alan explained to me that if they asked for the ramp at Richmond on the District line, [TfL staff] would not allow them to get off at Hammersmith because it wasn't officially step-free there and they would have to call ahead to inform staff where they were getting off. [Personal observation notes]

Wheelchair users are actively creating accessible stations for themselves through the use of these technologies––even stations that would not otherwise be labelled "step-free" become step-free for Alan when he uses his portable ramp. Effectively, they are making their needs legible to the system: they are making themselves visible to a system that, without their intervention, would not have seen them and catered to their needs. They rewrite the rules in their own ways or, as Alan told me:

They're in charge, but you're in control. - Alan

It also transpired that wheelchair users use their own bodies and wheelchairs as one of the tools handy in difficult situations:

How to scatter a crowd: accelerate towards it. Especially effective in a powerchair. – Anton

And, while a drastic step that might potentially cause bodily harm, it was found to be an effective way of making space for themselves, or ensuring that doors do not close without them having boarded or alighted at their desired stations. Here, too, wheelchair passengers make themselves hyper-visible in a system that might otherwise keep moving without heeding them were it not for these interventions. Individual tactics are also not limited to "active" physical maneuvers such as these:

I've got strategies, I manage my disability. And this is it, people manage it […] because people want to do things, they can get out and do it. – Adam

There is an onus on us to perform disability, so we need to make sure that we're not getting out of our wheelchairs, that we're not moving our legs in ways that would suggest that we're not disabled. – Chiara

Many interviewees discussed the need for close scrutiny of oneself and one's environment to balance their internal state (managing disability) with external social interactions (performing disability). Managing one's disability can be anything – from recharging power chair batteries, to choosing journeys with more transport mode changes but fewer distance rolling, to choosing to use the wheelchair over crutches one day and the other way around the next. It requires a deep knowledge and engagement with one's body, including one's wheelchair and assistive devices, to gauge what next steps might be taken during a trip. The performance of disability, on the other hand, is a demonstration of how observant wheelchair users must be of their surroundings to avoid potentially negative situations––it is a form of emotional labor, or emotional management that requires work on the part of one individual to soothe and ensure the happiness of others in effective ways (Federici, 1975; Hochschild, 1979, 1983; Zapf, 2002). Examples of this work include, like Chiara points out, avoiding movements that might flag oneself as non-disabled to avoid confrontational situations (like being told they are faking their disability to receive government benefits). It may also be a performance to acquire sympathy so that users make space:

I put on my best little-lady-in-the-wheelchair face and ask really nicely. – Jo

Jo uses her disability to enroll allies throughout her journey, in order to get people to perceive her as someone who requires (or might even "be deserving") assistance. Others made use of public friendliness as a constant shield:

Be friendly to everyone even when you don't want to be. – Anton

This friendliness is not just an act of kindness. It is an approach that strives to weed out negative social situations or to frame requests for help in a more positive light. Many interviewees expressed feeling like they were being watched and judged as more than individuals, but rather as representatives of a broader group––wheelchair users, generally, or perhaps even more broadly as disabled people. They felt the need to be "ambassadors'' and "not hav[e] a chip on [their] shoulder." We see, yet again, wheelchair users actively playing with the categories of visibility and invisibility here: making themselves more apparent in the system when it might enable them to get access more easily (using one's little-old-lady voice) or playing up a disability to avoid drawing attention to oneself to avoid instances of dysappearance, should a non-disabled person question them.

Beyond personal approaches, whether physical or social, wheelchair users also engage in much broader movements to improve accessibility in London's public transport. Some work as insiders, either in policy-making or within the industry, from members of the Parliament (two interviewees for this research were members of the House of Lords) to creating positions for themselves as disability coordinators in transport companies. This is important to state particularly in journals that do not focus on disability studies, as it still seems to come as a surprise elsewhere that disabled people not only constitute roughly 20% of the population but also that many are gainfully employed. Yet other wheelchair users work as "outsiders", engaging in a variety of strategies from writing emails to Transport for London when an elevator is broken, or when a bus driver does not stop for them, all the way to activist demonstrations to push for improvements. These activist movements, borne of the rich history of disability activism in the UK, were particularly powerful in the 1990s, having successfully demanded the phased introduction of low-floor buses from 1993 onwards (the last Routemaster, an inaccessible model of London bus, was taken off the streets in 2014). Today, some disabled activists in the English capital work with a charity called Transport for All, a London-based group that demands accessible transport for disabled and elderly passengers. It is important to point out both the insiders and the outsiders––those that have a direct hand in changing policies, and those who demand those changes. It is this powerful combination that creates what one interviewee called "the pincer movement":

So that joint thing, of disabled people and their organizations pushing and someone inside the organization going, "We have to do this, this is what we need to do." So that, I think, is what makes change happen. – Alice

It is in combining all of these tactics and strategies, from the day-to-day tools to the big activist movements, that wheelchair users and disabled groups more broadly conquer the transport system. Despite the stigma that continuously labels disabled people as passive members of society, this work recovers the constant labor undertaken by wheelchair users in order to get around the city––anything but passive. These findings will be anything but surprising to disability scholars, but, as we have seen throughout the discussion section, they cast doubt on some key concepts about infrastructures developed by scholars in STS. We need to radically revisit the notion of invisibility and who, or what, is invisible to whom in infrastructures.

c. The Paradox of Invisibility

It was discussed in the literature section above that one of the main characteristics ascribed to infrastructures is invisibility, or at the very least a tendency to become invisible (Lampland & Star, 2009). Though there have been recent challenges to this definition, much of the work has been centered in studying infrastructures in different geo-political and sociohistorical contexts (Ureta, 2013, 2014). One example of this is the research undertaken on infrastructures in the Global South that reframes infrastructures as visible in the day-to-day due to economic constraints that render daily care and maintenance difficult. However, research on infrastructures in the Global North that counter this definition are rarer, and this might potentially imply a more economic or geographic taxonomy of infrastructure, e.g. those in the North are more stable and invisible while those in the South are more unstable and visible. This article goes beyond this, having studied an infrastructure in the Global North. Rather, we can see here how the (in)visibility of infrastructures is linked to particular understandings of users in light of social values. If, as discussed in the section above, there are so many barriers and struggles faced by disabled passengers, could one ever claim that infrastructures are invisible to them? Questions of visibility are, therefore, not reducible to characteristics of the infrastructure itself. Rather, they become a question of one's experience of said infrastructure. Analyzing infrastructure from the perspective of marginalized users, those whose needs are not embedded or materialized in the system, demonstrates how limited an analysis focused on the thing-itself can be.

Furthermore, analyzing infrastructure through the experiences of disabled users allows to highlight another often-invisible aspect: invisible labor. Through discussing the various tactics developed by wheelchair users, rather than leaving these stories hidden, we give them primacy in this research. I argue that the tactics and strategies described above are a type of invisible, and unrecognized, work––wheelchair users are constantly repairing and making the transport system work for them, despite its disruptive and exclusive nature. As such, there is a particularly interesting tension between visibility and invisibility that plays out when wheelchair users tinker in the system, carrying their toolkits and engaging with passengers and employees alike:

By being out and visible, I'm actually going out, like lots of other disabled people are, and changing people's opinions and impressions. – Alan

Ultimately, while the work undertaken through these tactics might go unrecognized and might be deemed invisible or even irrational in the face of the infrastructure itself, wheelchair users interviewed argued that it is only by interacting with these systems, engaging with them, being seen, that changes might be made. It is a way of breaking a negative feedback loop, where little accessibility provisions are given due to perceived low numbers of disabled users, which service providers then use to justify a lack of investment in accessibility refurbishments. When wheelchair-using passengers use transportation, they argue, they are making the infrastructure visible––its limitations and inadequacies––while also countering social expectations. It is thanks to these interventions, both in the daily work and in large social collectives, that wheelchair users impact infrastructures and conquer incremental changes towards more accessible transport systems. Infrastructures, thereby, are not of themselves invisible, but are experienced as invisible by those perceived and embraced as users. Those on the margins have diverse, complex, experiences of infrastructure, and work to make themselves legible.

In a world of neoliberal ableist systems, this demand for space through a deep engagement with these systems––and subsequent awareness of said engagement––is a powerful act towards reclaiming one's abilities, and opposing established social rhetoric. The development of inclusion mechanisms is dependent on significant labor, but it allows the subversion of social narratives that would otherwise label wheelchair-using passengers as passive. As the work they engage in demonstrates, these users are, if anything, much more active within the system than their non-disabled counterparts. Through this, while infrastructures might succeed at the placing of barriers and disabling individuals, wheelchair users find ways of reclaiming and demonstrating their capabilities through carefully crafted instances of subversion. As Adam puts it: "Watch me."