The Shelter that Wasn’t There: On the Politics of Co-ordinating Multiple Urban Assemblages in Santiago,Chile

1. Urban assemblages, power and a missing bus stop shelter

This paper is about a missing bus stop shelter. This shelter should be located in downtown Santiago, Chile. More precisely, it should be located on Alameda Bernardo O’Higgins, Santiago’s main avenue, a dozen metres from the corner with MacIver Street. As a replacement there is a single metallic bus stop sign that indicates that bus lines 210, 412a and 418 stop there. Also, a rectangular area closer to the street is located several centimetres above the sidewalk level and covered by grey and white striped tiles; yet the rest of the structure—roof, maps, seats, etc.—is missing.

While telling the story of why this particular bus stop shelter is missing, the paper will also tell another story; a story about co-ordination practices between assemblages enacted by two Chilean public organisations. The usage of this term is not casual. Deriving from Giles Deleuze and Felix Guattari’s (1988) concept of agencement, in the past few years assemblage has gained an important degree of momentum in the social sciences (Bennett, 2005; De Landa, 2006; Farias and Bender, 2009; Area, 2011 special issue 43.2; City, 2011 special section in issues 2 and 3–4; Marcus and Saka, 2006; Ong and Collier, 2008). In trying to elucidate the concept, an important starting-point should be the recognition that

Therefore, instead of a single and consistent social theory, assemblages should be seen as “part of a more general reconstitution of the social that seeks to blur divisions of social–material, near–far and structure–agency” (Anderson and McFarlane, 2011, p. 124). When using the term assemblages, social scientists are usually referring to multiple things, from merely a particular research focus to a completely new ontology of the social (for a more detailed comparison, see the table in Brenner et al., 2011, p. 231).

Following the strongest, or ontological, sense of the term as proposed by Manuel De Landa (2006) assemblages can be defined as “wholes whose properties emerge from the interactions between parts” (p. 4). Therefore what defines assemblages as entities is not their internal wholeness or coherence but, on the contrary, what De Landa calls their “relations of exteriority” (p. 10) or the way in which the components of assemblages are not exclusive to them but “may be detached from [them] and plugged into a different assemblage in which [their] interactions are different” (p. 11); therefore the components are autonomous; they have agency and commonly belong to several assemblages at once. Such exteriority also implies, centrally, that “the properties of the component parts can never explain the relations which constitute a whole” (p. 11). So assemblages are never fully stable and well-bounded entities; they do not have an essence, but exist in a state of continual transformation and emergence. They exist “as a process of putting together, of arranging and organising the compound of analytical encounters and relations” (Dewsbury, 2011, p. 150). In this sense, the concept “offers the possibility of grasping how something … heterogeneous … holds together without actually ceasing to be heterogeneous” (Allen, 2011, p. 154).

Several characteristics derive from this conceptualisation. First, the notion of assemblage

Secondly, agency is not located only in key actors, but largely distributed in several components of the assemblage whether human or non-human. Even in the case of

Thirdly, and because of these multiple agencies, assemblages are always in between what De Landa (2006, p. 12) calls processes of territorialisation (a relatively defined and stable identity) and deterritorialisation (a mutable state, undefined identity). Or, we can say, they exist but do not exist too much, too solidly; existence is a constant accomplishment not a fact and must be constantly reaffirmed. For this reason

Finally, this relational ontology implies, on the opposite side, that “the notion of assemblage involves no outside” (Farias, 2011, p. 369). An assemblage has no environment; it is not submerged inside a general framework such as a structure or ideology. Everything that matters to the assemblage is related to it in some way or other; for this reason they call “for a positive description of their becoming, not external explanations” (Farias, 2011, p. 369).

From this perspective, cities such as Santiago can be seen as the sum of multiple “assemblages of people, networks, organizations, as well as of a variety of infrastructural components, from buildings and streets to conduits for matter and energy flows” (De Landa, 2006, p. 4). Such assemblages “do not form wholes or totalities, in which every part is defined by the whole, but rather emergent events or becomings” (Farias, 2009, p. 15). Therefore, assemblage urbanism changes the focus of enquiry from stable structures that relate to each other in unequal ways to the question of

Such a perspective “allows us to move away from a notion of the city as a whole to a notion of the city as multiplicity, from the study of ‘the’ urban environment to the study of multiple urban assemblages” (Farias, 2011, p. 369).

However, such a conception has also been criticised, especially in relation to how assemblage urbanism deals with the issue of power and inequality in urban dynamics. Following Sayer (1992), Brenner and Wachsmuth et al. (2011, 2012) criticise the assemblage approach for being based on a ‘naive objectivism’ or the belief that

For this reason, “this approach offers no clear basis on which to understand how, when and why … some possibilities for reassemblage are actualised over and against others that are suppressed or excluded” (Brenner et al., 2011, p. 235).

Answering such criticism, the proponents of this ontology claim to recognise from the start that urban assemblages are “structured, hierarchised, and narrativised through profoundly unequal relations of power, resource, and knowledge” (McFarlane, 2011a, p. 655). The very possibility of emergence of a particular assemblage always

Given that inequality is not explained by the recourse to structural argumentations, the focus on urban assemblages is better prepared than traditional political economy of cities to unveil “the actual practices, processes, sociomaterial orderings, reproducing asymmetries in the distribution of resources, of power and of agency capacities, opening up Blackboxed arrangements” (Farias, 2011, p. 370). Along with this emphasis on opening black boxes, the assemblage approach is useful for critical analysis on the basis of its parallel focus on potentiality, or how it shows

In this paper, I would like to contribute to this on-going discussion by exploring in detail the way in which power emerges in urban assemblages through the in-depth study of why there is a missing bus stop shelter in front of Santiago’s Biblioteca Nacional. In doing this, I will change the usual analytical focus on the emergence of assemblages to the controversy arising when two large and complex urban assemblages share component/s that each of them claims as exclusive. Following De Landa (2006 p. 9), we could say that such controversy is between two contrasting totalisations of each assemblage in which the component/s under dispute is/are seen as being “constituted by the very relations they have to other parts in the whole. A part detached from such a whole ceases to be what it is, since being this particular part is one of its constitutive properties”. It is in the shifting management of such controversies—sometimes involving a closure, sometimes remaining open and unstable—that a key materialisation of power and inequality on urban assemblages emerges.

In analysing such an encounter I will bring into the analysis one particular conceptual device that could be helpful to enhance the way assemblage theory deals with the issue of power in urban settings: co-ordination. As Annemarie Mol (2002, especially ch. 3) has explored in her study of the treatment of atherosclerosis in a Dutch hospital, the recognition that multiplicity is irreducible from ontology does not mean that anything goes. In order for entities to hang together across sites, a constant work of co-ordination between their multiple enactments is required to prevent “distribution from becoming the pluralizing of a disease into separate and unrelated objects” (p. 117). Leaving aside the usual realist recourse to “refer to a preexisting object”, co-ordination becomes the practices through which “the various realities of atherosclerosis are balanced, added up, subtracted. That, in one way or another, they are fused into a composite whole” (p. 70).

As in any other displacement, moving Mol’s conception of co-ordination from her Dutch hospital to my particular case in Santiago necessarily implies some adaptations. After all, Mol’s conception refers to one multiple/single object (atherosclerosis) being co-ordinated when moving along the wards of a single hospital. In my case, there is no one multiple/single object but, at least, two and their mobilisation involves several different locations, such as governmental offices, documents, sidewalks of the city, pictures, etc. As a result of this situation, co-ordination practices in the case under study involve the performance of multiple assemblages of the involved entities in which the components in dispute are presented as being divided and/or shared in new ways. A key point is that such assemblages are always multiple, usually proposing quite different arrangements and attributions of the entities involved. Co-ordination in this case relates to the different practices and techniques involved in the stabilisation of one particular assemblage over the others, a process in which different tactics of power and inequalities are very much present. Such stabilisation, however, is never definitive and solid. Like any other assemblage, the selected assemblage is in a process of constant emergence, always open to be transformed, challenged or discarded.

The missing bus stop shelter will be analysed as a humble casualty of the co-ordination practices between two urban assemblages that emerged during the development of Transantiago, the new public transport system of Santiago. ¹ In the next section, I describe these two urban assemblages: the Biblioteca Nacional building and the bus stop shelters, or Estaciones de Transbordo. Then the paper will analyse the practices of co-ordination carried out when both assemblages collided on the corner of Alameda Avenue and MacIver Street. Finally the paper will conclude by exploring the utility of the concept of co-ordination to the understanding of the issue of power in urban assemblages.

2. Two Urban Assemblages of Santiago

3. Co-ordinating Assemblages

After the designs were accepted internally by the CdT, the construction of the transfer stations started in several locations throughout the city. Most of them were built without any more hassle than that expected when dealing with the obduracy (Hommels, 2005) of the existing urban infrastructure. Some of them, however, encountered a different kind of obduracy, much harder to deal with. This was the obduracy of national monuments protected from any transformation in their ‘authentic’ disposition by the CMN. As a consequence any transfer station located in the vicinity of a monument would have to be approved by them before being built.

Such a situation required a work of co-ordination between both organisations. ³ Its starting-point was a meeting in March 2004 at the CMN offices where actors from CdT provided antecedents about Transantiago and the transfer stations that they were planning to build. After studying this information, in June of that year the CMN sent a letter back demanding, among other issues, detailed plans of any transfer stations located near monuments.

Along with sending the plans in November 2004, the CdT sent a lengthy letter providing the arguments behind the particular designs of transfer stations. It concluded with the following two paragraphs

In this extract we again can see the performance of a version of the transfer stations as highly visible entities, an element of “contemporary design” that materialises and unifies the modernity promised by Transantiago, besides its functional use as a guide to passengers. The use of italics and bold characters in the last lines of the paragraph show the importance that the CdT gave to this particular version of the assemblage. Even in a context where national monuments are located, the transfer stations should remain relatively immutable. They should always keep a certain degree of visibility to stand out in each one of their concrete urban locations.

After checking the provided plans, the actors at the CMN asked for several stations to be relocated, something that the CdT accepted and most of the controversial issues were settled relatively quickly. There was only one exception: shelter 7 from Santa Lucía transfer station that, in accordance with the plans, had to be built in the middle of the block occupied by the Biblioteca Nacional.

At the beginning of 2005, the CMN sent a letter to the CdT asking for the removal of this particular shelter “in order to avoid an obstacle in viewing the patrimonial building”. Here we see a performance of Biblioteca Nacional as an assemblage that has to be protected. However, this protection did not refer to the material structure of the building, but to its view. This is a much more ambiguous claim. Strictly speaking, law 17.288 talks about the CMN having regulatory attributions regarding the “conservation, reparation or restoration” of monuments, not about their view. From the words of Duran, quoted earlier, we can conclude that view was certainly included by the members of the CMN in the ‘conservation’ attributes. Yet this did not solve the problem; it does not clarify what the view of the Biblioteca Nacional exactly is and/or how it might be affected by building transfer station 7 in its surroundings. Given the lack of any previously set standards, the characteristics of this view and how both buildings were going to relate to it needed to be determined in the process of argumentation itself.

Visibility, as any other component of an assemblage, is always relational. It does not belong to the object but is a relation between itself and other entities in its vicinity. Especially in a cityscape populated by multiple objects, to increase the visibility of one (or to add a new visible one) normally speaking means to decrease the visibility of other/s. This is fine when the visibility of such objects is not considered relevant, but it becomes an issue when it is protected as in the case of Biblioteca Nacional, or forms a central component of its design and attached function as happened with the transfer station. In this case, we can see how the territorialisation of an assemblage, understood as the “processes that define or sharpen the spatial boundaries of actual territories” (De Landa, 2006, p. 13), is directly related to a certain deterritorialisation of other/s. The territorialisation of the transfer station as a highly visible entity in this particular location implied, from the point of view of the CMN, the deterritorialisation of the Biblioteca Nacional building as a monument with certain heritage views that must be protected.

In their reply to this letter, in July, the CdT argued that it was impossible to move the shelter from this block given that it “will imply to significantly increase the distances that the public transport user has to walk, because there is no other position close by to locate the bus stop”. So, instead of agreeing to deal with the issue in terms of a monument whose view has to be protected, they maintained an assemblage in which the bus stop in this location represented not only a highly visible entity, but mainly a materialisation of the users’ right to accessible public transport.

This move was acknowledged by the CMN who, in a letter sent in August, argued that this particular bus stop must be moved for three reasons

In this extract, the CMN brought new elements into the Biblioteca Nacional’s assemblage. While in point III we can again see it as a site of a heritage building whose view needs to be protected, the other two points are much more connected with the argument proposed by the CdT. Instead of bringing entities to argue about the validity of a view in itself as a reason to move the bus stop, they portrayed the area surrounding the building as also containing an important number of pedestrians who were going to have problems in their mobility due to the presence of the bus stop.

This strategy proved to be the wrong one for the purposes of the CMN. It is in a city of moving people where CdT moved more at ease, having several kinds of technical device from transport planning to bring in support of their position. This was clear in the letter in answer to the request of the CMN sent in October. Using several different quantifications they argued that the flow of people in front of the Biblioteca Nacional would not be affected at all by locating the shelter in the block and, besides, that it was impossible to move it anywhere else. In order to mitigate such a situation they offered to enlarge the sidewalk by 1.5 metres.

Regarding the performance of the library as an heritage building whose view should be protected, their strategy was to minimise the visual impact of the shelter. First, they offered to divide bus stop 7 into two smaller sub-shelters named 7a and 7b and move them to the two extremes of the block in order not to obstruct the central body of the building. Secondly, they included an image in order to visualise the impact of the shelter on the view of the library, as shown in Figure 1.

OPEN IN VIEWER

Figure 1.

The CdT’s visualisation of the proposed bus shelters, 7a and 7b, in front of the Biblioteca Nacional.

This was not merely illustrative. As a long line of studies of scientific and technical practices have shown, visualisations have a certain agency of their own. First of all, they “are irreplaceable as documents which enable objects of study to be initially perceived and analyzed” (Lynch, 1985, p. 37). However, this is not their only attribute. Scientific and/or technical visualisations are also endowed with what Söderström (1996) calls external efficacy or “the persuasive power of representations” (p. 62). Given the complex technical processes needed to produce them, these visualisations are usually seen as more valid than other forms of representation, especially by actors outside the particular expertise that produced them. For both reasons contemporary urban planning has become highly dependent on such devices to the degree that very few working practices are done without the active presence of different kinds of visualisations, from plans to models (for a detailed exploration of this issue, see Dühr, 2007). Especially in controversial issues, visualisations are quite helpful because they “render what are highly positioned notions of space … as universally applicable and desirable” (Lepawsky, 2005, p. 707). ⁴

In this sense, Figure 1 can be seen as offering a particular kind of visual assemblage of the space surrounding Biblioteca Nacional. A first thing to note is the character of this visualisation. Instead of presenting a proper plan or model, they choose to use as the basis for the visualisation an actual photograph of the building taken from the other side of Alameda, superimposing over it a computational representation of the proposed shelters 7a and 7b. By doing this, they create a hybrid in which both the existing building and the future infrastructure co-exist, endowing a higher degree of plausibility to their proposal. In parallel, the inclusion of several other entities in the picture looked to weaken the position of the CMN. Future shelters 7a and 7b, especially the one on the right, are also represented as partially hidden behind several objects that currently block the view of the Biblioteca Nacional, such as passing cars and buses and trees growing on the divide in the middle of the avenue. At the top right-hand side of the library we can see two new buildings and a piece of infrastructure also interfering with the view. In all, the image represents the “staging of a scenography in which attention is focused on one set of dramatized inscriptions” (Latour, 1990), in particular the happy co-existence of the Biblioteca Nacional and the shelters, especially given that the existing purity of the view of the Biblioteca Nacional was quite questionable.

However, this picture was the closest that shelter 7 ever came to existing. As Latour (1990) has stated, visualisations do not work on their own. To be really effective, they need to mobilise along with them several other entities that contribute to re-enact their ‘facts’ in new locations. In this case, the image in Figure 1 was presented only with the arguments that the CdT has been stating (and the CMN rejecting) all along. In the end, the visualisation failed to convince the CMN and in November 2005 they sent a letter making the same argument as the one in August, but using stronger terms and concluding that the bus stops must be moved “definitely”.

The authoritative tone of this letter was based on their inclusion in the controversy of a metrology in which the positions of both organisations could be finally balanced: legal bodies. By doing this, they not only established a common ground for discussion but also, and centrally, a hierarchy. Such hierarchy was based on the different weight of the legal bodies behind each organisation. While the CdT was supported by a quite diffuse presidential decree of 2002, that did not even detail its legal attributions, law 17.288 supported the CMN in full. Even more, in its eighth article, this law establishes that “civilian, military and police authorities will have the obligation to co-operate with the fulfillment of the functions and resolutions that the council adopts in relation with the conservation, care and vigilance of National Monuments”. So, in any controversy involving the conservation of a national monument, even in the case of its ambiguous view, the CMN was always going to have the last word.

Notwithstanding, the position of the CdT was not completely lost, as can be seen in the following extract from their final letter on this subject sent to the CMN in June 2006.

Bringing the position of the CMN again into the performance of the library as surrounded by flows of moving people (“the central worry [of the CMN] is the important influx of people …”), they were able to use their greater fluency in transport issues to maintain the location of the bus stops, concluding (possibly not without irony) that such influx was “sadly … going to exist, with independence of the bus stops being installed [there]”.

However, in locating the transfer stations there, they could not ignore the performance of the Biblioteca Nacional as having a view that must be protected by the CMN and including law 17.288. In this respect, the CdT ended up proposing the removal of only the platform, but not the rest of the structure of the transfer station in this particular location. In order to illustrate their proposal, they added a new visualisation, shown in Figure 2. Here we have a different composition of the view of the Biblioteca Nacional from that shown in Figure 1. This time, the picture on which the visualisation is based was taken standing on the divide in the middle of Alameda Avenue, 50 metres or so down the street. It was also taken at a moment with less traffic—there is only one small truck coming down the street. The shelter from transfer station 7 is also missing and the only remnant of Transantiago’s proposed infrastructure is the pale red line marking the area where the sidewalk would be extended. As a consequence, the view of the building appears to be much clearer, with only a few palm trees and passing pedestrians blocking it.

OPEN IN VIEWER

Figure 2.

The CdT’s visualisation of the Biblioteca Nacional without the shelters.

By doing this, and contrary to the CDT’s earlier claims about the immutability of its design, transfer stations proved to be a quite fluid entity (de Laet and Mol, 2000), being able to adapt, to be transformed in order to create a new assemblage including both the transfer stations and the Biblioteca Nacional. However, such fluidity was not without costs. In order to form this new assemblage, the station had to dispense with most of its structure above ground level. By doing this, it was able to keep being the point at which pedestrians would turn into public transport users, but sacrificing its related embodiment of the modern city promised by Transantiago. Or, in other words, Transantiago’s modernity was deterritorialised in order for the library to maintain its territorialisation as a heritage building whose view is protected. Therefore, in this particular context, the view could not be shared—as an element in common between the Biblioteca Nacional and the shelter of the transfer station—but the totalising claim over it of the former prevailed.

Conclusions

When Transantiago started its operation in February 2007, most of the 35 original transfer stations were already built. Beyond a few pieces in the media with architects criticising their shape in aesthetic terms, they attracted very little public attention. Also, they were effective in becoming highly visible pieces of urban infrastructure, contributing to the disappearance of the former practice of drivers letting passengers on and off the bus anywhere they wanted. Taking into consideration that almost no other aspect of the plan seemed to be functioning as expected, transforming Transantiago into one of the biggest controversies in the country since the return of democracy in 1990 (Muñoz and Gschwender, 2008), transfer stations could be seen as one of the very few uncontroversial entities of the early implementation.

This success, however, was not complete. Shelter 7 of Santa Lucia station proved to be too visible, too remarkable, in a space where visibility was already monopolised by another entity. Under these conditions, the path taken was to transform the station into a fluid object, removing the shelter and most of its elements above ground level, and hence its visibility, thus allowing the co-existence of both entities in the same block. At first, we can say that the CMN finally won over the CdT, with the Biblioteca Nacional remaining immutable/total while the transfer station was forced to become fluid. Yet this is only part of the story. As explored by the practitioners of actor–network theory, the fluid character of entities, their capacity to change continually, might also be an advantage. As affirmed by de Laet and Mol

Therefore, if we take contemporary cities as constituted by shifting geographies, by ever-evolving assemblages that relate to each other in multiple ways, the immutability/totality of a building like the Biblioteca Nacional does not constitute an asset; on the contrary it “is likely to lead to rupture, difference, and distance” (Law and Mol, 2001, p. 614). This is exactly what happened here, when the versions about the city of the past and the one of the future could not be put together into an assemblage that included both, ending up sacrificing one of them.

Along with exploring this particular case, the analysis made in this paper looked to contribute to the development of assemblage urbanism in several ways. First of all is the issue of methods. Moving away from grand analyses or urban ‘trends’, a focus on urban assemblages leads us “to investigate previously neglected dimensions of capitalist urbanization” (Brenner et al., 2011, p. 231). Given its emphasis on the dynamic and ever-shifting emergence of urban entities, assemblage urbanism tends to favour the study of the very concrete practices through which the urban is continually produced, no matter how small or context-specific. In particular, this article dealt with the practices of co-ordination between two organisations enacting contrasting assemblages of a very particular urban location. In doing this, it focused on following the territorialisations/deterritorialisations occurring in the process, or the ways in which each utterance made by the involved actors implied the territorialisation of certain assemblages and the deterritorialisation of others. The documents and visualisations produced by the actors involved in the controversy were studied as complex spatial narratives on their own, ordering a variable number of entities in the shape of a particular territory that included and excluded certain assemblages. Such a focus reorientates the research not only as the analysis of existing urban entities, but also as the study of “the multiple practices through which urbanism is achieved as a play of the actual and the possible” (McFarlane, 2011a, p. 652).

In the second place, the paper hoped to contribute to current debates about the (supposed) problems of assemblage urbanism in dealing with issues of power and inequality. When we move from just describing the performance of single assemblages to the co-ordination practices between two (or more) of them sharing elements over which they claim exclusivity, power and inequality appear in all their force. Through the ensuing co-ordination practices, the competing assemblages become entangled in multiple hierarchical relations. Such hierarchies are given by their different position in metrologies introduced as a way to balance their relative weights, as seen here in the case of legal bodies. It is exactly in these multiple hierarchies that a great deal of power in assemblages resides, ⁵ enhancing or restricting, even forbidding, the capacity of assemblages to territorialise in different times and/or spaces, to re-enact in some ways and not others. This kind of power itself is always relational, it exists not as something intrinsic to urban entities but always as an ‘exteriority of relations’, using de Landa’s term, or as the sum of unequal relationships in which a certain urban assemblage becomes entangled.

Finally, the paper has hoped to contribute to the development of a more fluid understanding of urban spaces. As the case under study showed, urban assemblages are always emerging, continually entangled in relations of territorialisation/deterritorialisation. Even hierarchies themselves change continually, readapting to the emergence of new metrologies and the multiple assemblages whose relations they regulate. So does power. Then the space constituted by such assemblages, their power and hierarchies, should be seen as ever-shifting, even if its pace is so slow that it appears as completely static. Even monuments such as the Biblioteca Nacional, the epitome of urban stability, are continually changing as they come to be related to new entities and react to them, even such humble ones as a single bus stop shelter. Then assemblage urbanism invite us to see spaces not as composed of stable and well-bounded objects which naturally embody power, but as fluid entanglements of entities whose power is a temporary achievement that must be continually reasserted through complex co-ordination practices.