Skip to navigation – Site map

HomeNuméros1Buildings are not Processes: A Di...

Buildings are not Processes: A Disagreement with Latour and Yaneva

Graham Harman
p. 113-122


The classic and traditional source on actor-network theory and architecture is a co-authored and well-know 2008 piece by Bruno Latour  and Albena Yaneva («Give Me a Gun and I Will Make All Buildings Move»: An Ant’s View of Architecture). They argue, based on dynamic-ontology, against viewing buildings as static objects or entity and in favor of treating them as dynamic  flux or trajectory that can be broken into fully-formed individuals only through the abstractive sickness of the human mind, however necessary abstraction may be for the practical necessities of life. In this paper, arguing from the standpoint of my OOO (object-oriented ontology), I note some problems and paradoxes of Latour and Yaneva’s approach and I propose a possible solution focused on “relation”: every form of knowledge amounts to telling us either what a thing is made of or what it does, the world is not just the correlate of knowledge.

Top of page

Editor's notes

DOI: 10.17454/ARDETH01.09

Full text

1One of the typical critiques of object-oriented ontology (OOO) is that its focus on objects entails a “static” view of the world. We are told instead that reality is not made up of well-articulated objects, but of a dynamic flux or trajectory that can be broken into fully-formed individuals only through the abstractive sickness of the human mind, however necessary abstraction may be for the practical necessities of life. This “dynamicist” school of thought even has several star philosophers in its corner, including Henri Bergson (Bergson, 2001 [1889]) and more recently Gilbert Simondon (Simondon, 2005). Unfortunately, the focus on genesis that is found in these thinkers is too often confused with a rather different trend, both of them lumped together under the misleading catch-all term “process philosophy.” I refer to the philosophies of Alfred North Whitehead and Bruno Latour, who are indeed very interested in how things change, but who are by no means committed to the sort of dynamicist vision of change as a continuum found in Bergson, Simondon, and even the Aristotle of the Physics (Harman, 2014). For Whitehead and Latour, time does consist of finely articulated individual instants, whereas for Bergson and his comrades it is completely impossible to think of time in this way (Harman, 2009). We must not let the ambiguous word “process” confuse this basic difference between two entirely separate philosophical schools. This remark is in order because in what follows, I wish to consider an ostensibly “dynamic” view of architecture proposed by Bruno Latour and Albena Yaneva (Manchester), who to my awareness is the most architecturally literate person in Latour’s immediate circle. It is easy to see that both Latour and Yaneva are opposed to a static conception of architecture as the production of rock-solid final products. What remains to be seen is the nature of the alternative they offer.

2In their joint article of 2008, Latour and Yaneva try to lead us away from a static conception of architecture. As they put it on their opening page:

Everybody knows  – and especially architects, of course –  that a building is not a static object but a moving project, and that even once it has been built, it ages, it is transformed by its users, modified by all of what happens inside and outside, and that it will pass or be renovated, adulterated and transformed beyond recognition (Latour, Yaneva, 2008: 80).

3Thus, the supposed stasis of a finished building will be undercut by Latour and Yaneva by invoking both its past and its future. A building can be made more dynamic by opening up the black box and examining all the countless heterogeneous actors that went into producing it, by means of the classic ethnographic methods of actor-network theory (ANT). But the same building can also be dynamized by referring to its inevitable future changes that prevent it from remaining the same object forever. To give just one example, Chicago’s Sears Tower went from being a showcase piece of commercial SOM modernism and the tallest building in the world in the Seventies, to being known as the throwback Willis Tower and entering a state of eclipse thanks to a number of taller rivals. Tenants move in and out over the decades, renovation projects occur, and the surrounding downtown neighborhood becomes more appealing amidst a general movement of successful young Americans back into city centers. The building changes constantly. There is no individual moment that gives us the building itself, since the building  – like everything else –  is a trajectory changing constantly across time. This is the basic argument of Latour and Yaneva’s anthology chapter, though we will be able to examine it in somewhat more detail. We should also note that these arguments are not specific to architecture, but reflect ANT’s attitude towards everything that exists. Latour makes similar points when analyzing the career of Louis Pasteur and the failure of the Aramis mass transit project, and he would do much the same if he were to analyze the space shuttle or Donald J. Trump’s reprehensible 2016 march to the Presidency of the United States (Latour, 1988, 1996).

4In doing so Latour and Yaneva take aim at no opponent in particular, no exemplary devotee of the “static” architecture they condemn. Nor is it likely that they had object-oriented ontology (OOO) in mind, despite their use of “object” as a pejorative term by comparison with “thing.” This heideggerian use of the two terms has a decade-long history in Latour’s work, dating back to his co-curated 2005 Karlsruhe exhibition on the parliament of things, and even back to his work in the 1990s on the politics of nature (Latour, 2004; Latour, Weibel, 2005). More importantly, OOO had no presence to speak of in architectural discourse until at least 2011, three years after the publication of Latour and Yaneva’s piece. Yet I would like to volunteer OOO to serve as a model of what they call static architecture, since I find myself wanting to stand up for much of what they attack.

5Let’s begin with the troubles I see afoot in their article, of which there are at least four. First, Latour and Yaneva assert that a building is a project composed of numerous agents rather than simply a unified product. Since everyone knows this, they are obviously not trying to report a newly discovered fact. Instead, they are trying to claim ontological priority for the swarm of actors who pre-date the completion of a building; Yaneva herself did a fine bit of fieldwork in this respect, as an embedded ethnographer working in the office of Rem Koolhaas (Yaneva, 2009). In more familiar Latourian terms, the argument is that a building like anything else is a “black box”, and that a black box can always be opened. Second, Latour and Yaneva make the same claim about the future of a building. Since the best laid plans of architects often go far, far astray, and since no one can predict the surrounding environment or socio-political context of a building’s future, we have to extend the trajectory of its existence forward as well as backward. But since everyone knows this too, it is again not the discovery of a novel fact about the world. The novelty of their claim comes from the ontology they derive from it: one in which no actor can be confined to a single instant, since each makes sense only as a line of flight across time. Third, there is the paradox that despite posing as remorseless champions of flux against the stasis of motionless buildings, Latour and Yaneva adopt a metaphor that is not normally associated with Heraclitean flow. I speak of cinematic frames, which our authors describe in glowing methodological terms:

[Jules] Marey had the visual input of his eyes to be able to establish the physiology of [a gull’s] flight only after he invented an artificial device (the photographic gun); we too need an artificial device (a theory in this case) in order to be able to transform the static view of a building into one among many successive freeze-frames that could at last document the continuous flow that a building always is (Latour, Yaneva, 2008: 81).

6Here we encounter an important point that is always effaced whenever there is loose talk of a group of “process” philosophers who favor change over stasis. The Latour-Yaneva model wants to replace the static view of buildings with one that accounts for their dynamism. Yet they want to do this by means of a series of “successive freeze-frames”, just like Marey reproducing the flight of a gull as a series of hundreds of instantaneous poses. Yet this is precisely what Aristotle and Bergson both forbid: the former when he holds in the Physics that movement is continuous and cannot actually be split up into a given number of static poses, the latter when he argues against the “cinematic” conception of movement that Latour and Yaneva endorse before our very eyes (Aristotle, 1996; Bergson, 2007 [1907]). Those who want to agree that they truly offer a dynamic model of architecture need to be aware that, by their own acknowledgment, it is a dynamism of hundreds of static freeze-frames laid out in a row, not a continuous non-decomposable flux as in the Bergsonian style of philosophy. My fourth objection to the article is that it is not clear that what Latour and Yaneva are offering is a new theory of architecture, since what they seem to be constructing instead are principles for a new ethnography of architecture: how to study it, not how to do it. Let’s consider each of these points in turn, since all four touch on topics of philosophical significance.

7The first concerns Latour and Yaneva’s view that buildings should be treated «as movement, as flight, as a series of transformations» (Latour, Yaneva,  2008: 80). This is analogous to Manuel DeLanda’s claim that atoms must be individuated by the history of their formation in the cores of various stars (DeLanda, 2006). In both cases we are asked to forsake the notion of a unified individual entity in favor of the historical process that gave rise to them. I have made my objections to DeLanda’s point in print, and would make the same objection to Latour and Yaneva here (Harman, 2008; DeLanda, Harman, 2017). An object becomes an opaque black box not just through some arbitrary human act of oversimplification. Reality itself creates black boxes: once a molecule of water is formed, the history of its formation and the cosmic backstory of its hydrogen and oxygen atoms becomes mostly irrelevant. Yaneva, during her years observing Koolhaas and his employees at OMA, sees countless things happen, only some of which are important enough to record in her book. The truth of an object is not a highly complicated process that becomes censored or muffled by its simplification into a black box. Rather, the black box itself represents a new reality, much of whose backstory is irrelevant to the black box itself. The technical name for this in philosophy is “emergence”, and by decomposing architectural objects into historical detail, they miss emergence as much as the hardcore physical reductionist who thinks that nothing exists but the ultimate layer of the universe. In the terminology of OOO, thinking that an object’s history tells us everything is a form of undermining, hardly different in kind from the reductionism of the pre-Socratics (Harman, 2013).

8The opposite problem becomes visible when Latour and Yaneva try to reduce the building to its future rather than its past, as when they say that «even once [a building] has been built, it ages, it is transformed by its users, modified by all of what happens inside and outside, and that it will pass or be renovated, adulterated and transformed beyond recognition» (Latour, Yaneva, 2008: 80). The point seems to be that a building will constantly change its features, and that it therefore cannot possibly remain the same building. Yet this is a simple non sequitur that might well fuel the opposite conclusion, as it does for me. The doctrine of British Empiricism that an object is merely a bundle of qualities with nothing hiding behind it can be found again in Whitehead, one of Latour’s favorite philosopers, who speaks of entities instead as bundles of relations. An entity is the sum total of its prehensions (that is, its relations), and it follows as a matter of course that every shifting relation will result in a new entity: closely related to its predecessor, but really only bearing a certain family resemblance to it. This is why the religious side of Whitehead’s thought can have no traffic with notions of an immortal soul, since entities for him cannot last more than an instant, let alone for eternity. We are left with something called “objective immortality” as our only hope, in which the me of 1973 continues to live on in the me of 2016, even though nothing has endured from one to the next. They supposedly belong together as part of one and the same trajectory, but how do we establish that two distinct entities belong to a single trajectory? Since there is nothing in each of the freeze-frames that links it with the others aside from a supposed resemblance, some external entity (a human being, no doubt) must be responsible for ensuring that they belong together. But none of these problems arise if we simply reject the initial presumption that a thing is exhausted by the sum total of its relations (Whitehead) or its actions (Latour). The Whitehead/Latour problem is one that OOO calls overmining, since it reduces entities to their external effects, appearances, or the events in which they participate. The fact that a building will have numerous future uses hardly proves that the building is nothing more than the sum of its uses. Instead, it indicates that all of these shifting functions and contexts do not exhaust the reality of the building, which remains more than any of its current, past, or future uses.

9The third issue concerned Latour and Yaneva’s endorsement of a model of time made up of innumerable freeze-frames. We wanted to note, first of all, that this converts time from a continuum into a series of static poses, and would thus be viewed by the Bergsonians as a philosophy of stasis in its own right. Now, there is no particular reasons why anyone should feel compelled to follow Bergson’s views (or those of Aristotle’s Physics), we need to recognize that Whitehead and Latour belong to an entirely different tradition from that of the philosophers of radical becoming. It is, in fact, the occasionalist tradition of continuous creation, with each instant of time vanishing before being instantly recreated by God in a similar but not identical state. Rejecting the supposed stasis of realism and replacing it with a freeze-frame model of time simply multiplies the problem of stasis, then tries to solve it by the fiat of claiming that all of these standalone instantaneous moments are linked by something called a “trajectory”. But no convincing account is given of how this would be possible. There is still another problem with this trajectorized model of a lengthy series of closely related freeze-frames. Namely, it has a difficult time distinguishing between important moments of a trajectory and unimportant ones. Since any new act or relation changes what an actor is, then the least movement along the trajectory (a hair falling from someone’s head, the murder of Julius Caesar) is equal to any other least movement along the trajectory. All actions are ontologically equal for ANT, but this is not what we learn from history or biography, or indeed from the history of art or architecture: in all of these domains, certain events are clearly transformative while others are trivial. Yet there is no good way to account for this in a theory where all actions are excessively equal.

10The fourth and final point concerned whether Latour and Yaneva’s chapter can actually be useful for architects, or only for ethnographers of architecture. They clearly believe the former, and indeed they write rather eloquently about the shift in architectural thinking that they foresee as resulting from their new outlook:

Instead of analyzing the impact of Surrealism on the thinking and design philosophy of Rem Koolhaas, we should rather attempt to grasp the erratic behavior of the foam matter in the model-making venture in his office; instead of referring to the symbolism implicit in the architecture of the Richards Medical Research Laboratories in Pennsylvania as a scientific building, we should follow the painstaking ways its users reacted to and misused the building after the fact of its construction, and thus engaged in thorny negotiations with its architect Louis Kahn, with glass and daylight; instead of explaining the assembly building in Chandigarh with economic constraints or with the trivial conceptual repertoire of Le Corbusier’s modernist style and his unique non-European experience in master planning, we should better witness the multifarious manifestations of recalcitrance of this building, resisting breezes, intense sunlight and the microclimate of the Himalayas... (Latour, Yaneva, 2008: 88).

11Everywhere we find “thorny negotiations”. This is the trademark of ANT, the greatest method ever developed for dealing with the heterogeneous human and non-human elements that must be enlisted for any project to succeed.

12Then again, ANT is an ultra-relational philosophy that strongly insists that a thing is composed of its relations with other things, and nothing more. Let’s imagine for a moment that we tried to apply this method to the arts. Picasso’s landmark proto-Cubist painting Les demoiselles d’Avignon would no longer be treated as a “static” artwork enduring across the ages. Instead, we would reduce it to a set of thorny negotiations between Picasso and past artists, critics and admirers of the work, and even thorny negotiations between pigment and canvas, then between the canvas and the aesthetic taste of Gertrude and Leo Stein and later of Georges Braque. Looking to the future, Les demoiselles would not be an identical substance enduring across the decades, but a series of thorny negotiations between the finished canvas and Miró, Pollock, and others indebted to Picasso’s apparently heroic risk with this painting. There would be no lingering Picasso-in-himself, but only a series of ever-ramifying Picassos depending on what uses were made of him in later times.

13As I argued in my 2012 pamphlet The Third Table, to do this would be to reduce the table simultaneously to its generative components and its outward effects, thereby leaving us with no table at all (Harman, 2012). While it is true that every form of knowledge amounts to telling us either what a thing is made of or what it does, the world is not just the correlate of knowledge. Les demoiselles d’Avignon is more than its physical and genetic components, and more than its various outward impacts. So too is a table, a nation, and a building. Latour and Yaneva are in fact asking us to consider the building as being nothing more than the story of its birth and its effects, with nothing left over for the black box to have some reality that belongs neither to its pieces nor to its relations with others.

Top of page


Aristotle (1996), Physics, trans. R. Waterfield, Oxford, Oxford University Press.

Bergson, H. (2007 [1907]), Creative Evolution, trans. A. Mitchell, London, Palgrave Macmillan.

Bergson, H. (2011 [1889]), Time and Free Will: An Essay on the Immediate Data of Consciousness, trans. F.L. Pogson, New York, Dover.

DeLanda, M. (2006), A New Philosophy of Society: Assemblage Theory and Social Complexity, London, Continuum.

DeLanda, M., Harman, G. (2017), The Rise of Realism, Cambridge, Polity.

Harman, G. (2008), DeLanda’s Ontology: Assemblage and Realism, “Continental Philosophy Review”, 41 (3), pp. 367-383.

Harman, G. (2009), Prince of Networks: Bruno Latour and Metaphysics, Melbourne,

Harman, G. (2013), Undermining, Overmining, and Duomining: A Critique, in J. Sutela (ed.), ADD Metaphysics, Aalto, Finland, Aalto University Design Research Laboratory, pp. 40-51.

Harman, G. (2014), Whitehead and Schools X, Y, and Z, in N. Gaskill, A. Nocek (eds.), The Lure of Whitehead, Minneapolis, University of Minnesota Press, pp. 231-248.

Latour, B. (1988), The Pasteurization of France, trans. A. Sheridan, J. Law, Cambridge (MA), Harvard University Press.

Latour, B. (1996), Aramis, or the Love of Technology, trans. C. Porter, Cambridge (MA), Harvard University Press.

Latour, B. (2004), Politics of Nature: How to Bring the Sciences into Democracy, trans. C. Porter, Cambridge (MA), Harvard University Press.

Latour, B., Weibel, P. (eds.) (2005), Making Things Public: Atmospheres of Democracy, Cambridge (MA), The MIT Press.

Latour, B., Yaneva, A. (2008), «Give Me a Gun and I Will Make All Buildings Move»: An Ant’s View of Architecture, in R. Gesier (ed.), Explorations in Architecture: Teaching, Design, Research, Basel, Birkhäuser Verlag, pp. 80-89.

Simondon, G. (2005), L’individuation à la lumière des notions de la forme et d’information, Grenoble, Millon.

Yaneva, A. (2009), Made by the Office for Metropolitan Architecture: An Ethnography of Design, Rotterdam, NAI010.

Top of page


Bibliographical reference

Graham Harman, “Buildings are not Processes: A Disagreement with Latour and Yaneva”Ardeth, 1 | 2017, 113-122.

Electronic reference

Graham Harman, “Buildings are not Processes: A Disagreement with Latour and Yaneva”Ardeth [Online], 1 | 2017, Online since 01 October 2017, connection on 20 July 2024. URL:

Top of page

About the author

Graham Harman

Southern California Institute of Architecture –

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search