bitsofbusiness-blog - Tumblr blog

bitsofbusiness-blog · 9 years ago

Text

Research Proposal

The object of my dissertation is machine vision, a technology that automates perception. A promotional photograph circa 1982 of the first model of a machine vision system produced by Cognex, a firm that is still in business, shows a keyboard, and a desktop computer case, on top of which rests a monitor and a camera. This device’s function was to identify letters, numbers, and symbols inscribed on objects – a function that is known as optical character recognition. It was put to use in the manufacturing of typewriters as a form of quality control. It did so by inspecting the typewriter’s keyboard to assess whether or not the keys were in the correct positions. In the photograph, the desktop computer case is labelled with the device’s name, DataMan. On its own, this appellation evokes a superhero whose power it is to process information. Further to this point – and critically, from the perspective of the present – there is an incongruity between the compound noun DataMan, and the bundle of hardware it denotes. This incongruity points to my dissertation’s aim and methodology by bringing out the historical and particular (rather than ideal and universal) basis of the theories and concepts that subtend the appellation of this early commercial machine vision system.

My dissertation’s aim is to describe the forms of reason or the histories of thought that make possible the perception of the technology of machine vision as adequate to particular needs or uses, namely those in the milieu of manufacturing and distribution.

The preliminary basis of my dissertation’s methodology is the plural and open set of methods known as media archeology. The reason for this is that at its core this approach is the study of media from an historical perspective. But my study will be guided less so by the methods of key scholars who have undertaken media archeology, such as Friedrich Kittler. Rather, the foundation of my methodology is the work of a theorist who has influenced much of the scholarship which deploys this approach, namely Michel Foucault. His methods of archeological and genealogical analysis form the point of departure for how I will study machine vision. Thus I see the emergence of machine vision as an object of the field of science and technology, and its uptake by the economic sphere of civil society as, in Foucault’s words, a function of “the interweaving of effects of power and knowledge.”

Two germinal texts that have addressed machine vision from the perspective of the humanities are Manuel De Landa’s book, War in the Age of Intelligent Machines, and Lev Manovich’s dissertation, “The Engineering of Vision from Constructivism to Virtual Reality.” De Landa deals with machine vision in the context of his poststructuralist examination of how computers shape the practice of waging war. His anticipatory claim was that the advisory function computers had been serving in military operations was on the verge of being superseded by the deployment of computers that themselves made decisions concerning military logistics, strategy and tactics. For him, the epistemic ground on which the military adopted machine vision was the emergence of a particular intelligence practice in the early twentieth century. This practice comprised the tactic of the indiscriminate capture and collection of photographs of an enemy state’s territory, and the procedure of recognizing strategically relevant patterns within the unwieldy agglomeration of information these photographs amounted to. Initially, the military deployed the technology of image processing to facilitate the human task of pattern recognition. But this application fostered the expectation that this task could eventually be automated – machine vision was seen as the technology to do so. De Landa foresaw a time in which autonomous machines equipped with machine vision traverse the terrain of battle, distinguishing friend from foe, and thus making the decision to kill or not, accordingly. Shifting his gaze away from the war machine, and towards civil society, he noted that a contemporary application of machine vision was as a form of quality control in the manufacturing process. Here I want to mention how De Landa describes this application, because it begins to articulate how my dissertation’s aim and methodology differs from his analysis of machine vision. According to him, a machine vision system such as DataMan is “primitive.” The situations in which a machine vision device can comprehend the objects that fall within its line of sight are “limited” by the fact that these objects must share a similar form. This assessment is underpinned by De Landa’s ascription of a telos to machine vision’s development, namely its anticipated deployment as a part of an autonomous killing machine. From this perspective, he perceives machine vision’s limitation as a limitation. By contrast, I see this limitation as a contingent function that is understood as adequate to the needs of the manufacturing process, as they are known by this practice’s coeval form of reason.

Another way of putting my dissertation’s aim is that it comprises a description of the conditions of existence of statements that validate the substitution and supplementation of human by machine vision. Manovich’s dissertation represents an important advance towards this objective. In the twentieth century, he explains, a number of environments arise in which vision serves a productive function. His dissertation is a history of thought that responds to this state of affairs by remaking vision as a productive tool. Machine vision is the object of the field of artificial intelligence (AI). He asserts that a shift in the perception of the location of human productivity from the body to the mind, which coalesces during and after World War Two, engenders this field’s emergence. The arrival of tasks such as radar operation that primarily call upon mental faculties brings about this epistemic shift. The discipline of cognitive psychology emerges out of the need to generate knowledge that makes the mind and its functions the basis of labour. The development of AI is, in part, an effort to meet the demand of automating this form of labour. Cognex’s name is emblematic in this respect, as it is an abbreviation of cognition experts. However, Manovich fails to note that, as Margaret Boden shows, the histories of these two research areas are closely intertwined. Perhaps this is due to the fact that he examines machine vision, and vision as an object of cognitive psychology from different points of view. Whereas he shows that cognitive psychology conceptualizes vision as a communication channel in relation to the discipline’s theorization of the mind as an information processor, he situates machine vision within the history of visual technologies that map space – a history that begins with the genesis of linear perspective. Still, his account of how machine vision fashions vision, namely as a technique for identifying objects in terms of their position within a measured space, is compelling. From the standpoint of the needs machine vision has been made to satisfy, this seems to be one aspect of how vision is conceptualized. For example, an article circa 1990 from the journal Computers and Industrial Engineering reports that a firm adopts a machine vision system as a form of quality control in the manufacturing process of surface mount device resistors. How the system assesses whether or not the device is operative is by visually measuring the distances between its tips.

My dissertation’s focus on quality control as an application of machine vision brings to light one limitation of Manovich’s history of vision. Namely, his account of the needs cognitive psychology addresses is constrained to kinds of work that arise only with the advent of the post-industrial economy. These kinds of work comprise the object of the field of human-computer interaction. The human labour in the manufacturing process for which machine vision was being seen as a substitution or a supplementation is not equivalent to the jobs Manovich discusses. In another article reporting on the implementation of machine vision in a manufacturing process, the discursive positivity that makes sayable the judgement of why human inspectors failed to notice faulty parts seems to be a kind of hybrid of theorizations of productivity as based in the mind, and the body. The fallibility of the inspector is the result of a decline in their “attention level,” evoking the concept of vigilance as it is understood by cognitive psychology. But the cause of this decline is ascribed to “the mechanical, repetitive nature” of the job, bringing to mind the concept of the fatigue of the body as machine.

Perhaps this hybrid form of reason is connected to the nature of industry in the post-industrial economy. For, as Michael Hardt and Antonio Negri assert, the “informatization” of the economy involves not simply the emergence of jobs that comprise the manipulation of information, but also the remaking of the pre-existing industrial sector. Although this point is raised at a highly abstract level, it does open onto another aspect of how I conceive my dissertation’s aim. I’m not only concerned with the history of thought that takes the human in the manufacturing process as its object (a history that includes the field of AI). Another part of my dissertation involves a study of the history of thought whose object is the manufacturing process itself. The article in Computers and Industrial Engineering notes that the machine vision system happens to produce information valuable to statistical process control, a method of managing the manufacturing process. This theory is as equally historical and particular as cognitive psychology’s and AI’s conceptions of vision. Therefore, this affinity was by no means certain.

My dissertation also corresponds in terms of methodology and aim to a more recent study of applications of machine vision, namely Kelly A. Gates’ book about what she calls “automated face perception technologies.” Critically, Gates understands vision not as an essential fact, but rather a discursive object. However, what principally I want to adopt of her approach is her objective of describing the conditions in which the realization of such technologies is seen as inevitable. For example, she asserts that the emergence of security as an ideal of human conduct management in the 1990s gave rise to the applicability of facial recognition technology, the kind of automated face perception technology her book primarily focuses on. Moreover, she shows that the sense that facial recognition technology serves this need is not a given, but rather a result of concrete discourses and practices, such as the established technique of correlating a person’s face with their identity.

To end I’d like to put forward three more speculative propositions about how I will undertake my dissertation, and what subject matter it will cover, as well as suggest what academic study it will contribute to, and in what spirit it will be written.

Both De Landa’s and Manovich’s analysis of machine vision is based upon the archive of the field of artificial intelligence. Nonetheless, I think that my dissertation’s aim absolutely entails further exploration of this archive towards discerning the descent of the field of artificial intelligence’s conceptualization of vision, in order to converge on this concept’s historical specificity.

I also think that my dissertation’s exploration of the historically-delimited group of theories, concepts and objects that make coherent the deployment of machine vision in the manufacturing process should not be outright constrained to the histories of thought which are immediately related to this application. So there is, in the undertaking of my dissertation, the potential for examining histories of thought beyond those of cognitive psychology, AI and the manufacturing process.

Moreover, I am interested in industry’s initial adoption of machine vision, which coincides roughly with the timespan (the mid-1980s to the early 1990s) in which De Landa’s book and Manovich’s dissertation were published. However, another aspect of my dissertation’s aim is to tell the history of vision as a productive tool in cognitive psychology, AI, and other fields, in relation to machine vision’s application in the manufacturing process, from this time up until the present.

I see my dissertation as a contribution to the academic study of what Jonathan Crary called the “historical construction” of vision, a line of inquiry of which his book Techniques of the Observer is an exemplary text. The purpose of how I study the episode of this construction crystallized in a particular application of machine vision is, in Crary’s words, to bring to light the “historically fabricated and densely sedimented makeup” of “the shape of the present.”

Bibliography

Boden, Margaret. Mind as Machine: A History of Cognitive Science (Oxford: Oxford University Press, 2006).

Chapman, Kenneth W., Johnson, W. Carroll, McLean, Thomas J. “A High Speed Statistical Process Control Application of Machine Vision to Electronics Manufacturing.” Computers and Industrial Engineering 19, no. 1-4 (1990): 234-238.

Crary, Jonathan. Techniques of the Observer: On Vision and Modernity in the Nineteenth Century (Cambridge: MIT Press, 1990).

Foucault, Michel. The Archaeology of Knowledge (London: Routledge Classics, 2002).

— Power/Knowledge: Selected Interviews and Other Writings 1972-1977 (New York: Pantheon Books, 1980).

Gates, Kelly A. Our Biometric Future: Facial Recognition Technology and the Culture of Surveillance (New York: New York University Press, 2011).

Hardt, Michael, Negri, Antonio. Empire (Cambridge: Harvard University Press, 2000).

Huhtamo, Erkki, Parikka, Jussi, eds. Media Archaeology: Approaches, Applications, and Implications (Berkeley: University of California Press, 2011).

De Landa, Manuel. War in the Age of Intelligent Machines (New York: Swerve Editions, 1991).

Manovich, Lev. “The Engineering of Vision from Constructivism to Virtual Reality.” PhD diss., The University of Rochester, 1994.

“Company History.” http://www.cognex.com/company-history.aspx (accessed November 16, 2015).

Anonymous. “Real-Time Machine Vision: It Takes Two.” Production 100, no. 8 (August 1988): 69-70.

8 notes · View notes

bitsofbusiness-blog · 9 years ago

Text

Here’s a thing I wrote about object-oriented ontology and Leviathan:

Graham Harman asserts that the French philosopher Quentin Meillassoux’s principle contribution to the field of philosophy is to trigger a destabilization of the status quo of its continental tradition. Correlationism is the term Meillassoux ascribes to what he claims is a shared assumption underlying twentieth century continental philosophy. Correlationism, Harman writes, holds “that we can think neither of human without world nor of world without human, but only of a primordial correlation or rapport between the two.” This fundamental limitation occasions continental philosophy to explore the grounds on which we obtain access to the world, foreclosing the possibility of thinking the outside of thought. Thus our relation to the world attains a unique and privileged position from an ontological standpoint. Meillassoux’s diagnosis initiated the consolidation of a philosophical position called speculative realism, which attempts to move beyond the constraints of the correlationist circle. I will argue that Leviathan, a nonfiction film about a commercial fishing vessel harvesting hauls of fish, directed by Lucien Castaing-Taylor and Verena Paravel, and released in 2012, is animated by the objective of speculative realism. To do so I will draw on the work of Timothy Morton and Levi Bryant, two scholars whose thought cultivates the speculative realist point of view.

Global warming is an effect of human activity, namely the disposal of carbon dioxide throughout the biosphere principally due to the combustion of fossil fuels. Ironically, Timothy Morton argues, to best understand global warming we must adopt a philosophical approach that discards the idea that humans sit atop the order of being. Global warming is a unique example of what he calls hyperobjects in that it enjoins humans to acknowledge its existence as such. Hyperobjects, he writes, “refer to things that are massively distributed in time and space relative to humans.” He objects to the idea that reality is featureless stuff, hiding beneath a layer of qualities intended by humans. Rather, reality is composed of discrete, unique things. For him then hyperobjects are real. They exist apart from, their being cannot be exhausted by, both our ideas of them and their appearance for us. The core of Morton’s thesis is that coming to terms with the existence of hyperobjects entails a reconsideration of how we understand the world. Leviathan performs specific ways in which hyperobjects do this.

Levi Bryant is concerned with how a speculative realist ontology can ameliorate our tools for not only analyzing how social orders endure, but also for identifying points or relations in their compositions that can be acted upon so as to precipitate their becoming otherwise. For him existence is made up of machines. Being is machinic because what a thing does constitutes what it is. Echoing Morton, Bryant notes that advancing a machine-oriented ontology amounts to disposing of the notion that a thing just is a bundle of qualities and properties. Conversely, he writes, “a machine is a system of operations that performs transformations on inputs thereby producing outputs.” A machine is divided along two dimensions: its virtual proper being, and its local manifestations. Virtual proper being denotes a machine’s power to operate, which exists whether or not it is actualized. Local manifestations refer to the outputs produced by a machine. He uses this term because it points to the fact that a machine’s outputs are a function of the machines it is structurally coupled to, or mediated by. Finally, machines become. That is to say, they are capable of gaining or losing powers. What counts as given to us is a machine’s local manifestations. However, he emphasizes that not only do we fail to grasp the sum of a machine’s outputs at any one time, but also we cannot access the reservoir that is its virtual proper being. Machine-oriented ontology is the foundation of what he calls onto-cartography, a methodology in the same vein as Foucault’s analytics of power, and Latour’s sociology of associations. Briefly, “what onto-cartography attempts to analyze” Bryant writes, “is the way in which relations between machines… organize social or ecological relations.” I argue that Leviathan constitutes a kind of onto-cartography.

Hyperobjects

Hyperobjects, Morton explains, forge a particular form of relation with humans. This form occasions a rethinking of aesthetic experience. Insofar as Leviathan furnishes this kind of aesthetic experience, it exemplifies hyperobjective art. Hyperobjects are viscous. This capacity is contrary to our normative disposition towards nature. He argues that we hold steady a distance between ourselves and nature as a psychological salve that conceals the fact that nature is actually menacingly close to us. Viscosity collapses this distance, referring to our unnervingly intimate relation with hyperobjects. He writes, “I do not feel at home in the biosphere. Yet it surrounds me, and penetrates me.” Hyperobjective art then traverses the space between itself and its experiencer, piercing their skin and affecting their physiology. He argues that this capacity brings to light a hidden conditioning moment of all aesthetic experience, anterior to the subject’s act of intention. As he puts it, “there must already be a sticky mesh of viscosity in which I find myself tuned by the object.” At one point in the film the camera accompanies fish carcasses, guts and blood as they are dumped into the ocean. An irregular rhythm is established as the camera dips in and out of the water. Suddenly, it is violently wrenched to the bow of the ship, where it bobs up and down, in tune with the undulating sea. This section of the film activates the viscous mesh occasioning aesthetic experience, viscerally affecting the viewer by imparting them with an unpleasant, queasy feeling. Morton models how we are disposed towards the world as our relation to objects in a mirror: we are in effect ontologically separated from them. In the era of hyperobjects, however, the mirror is no longer mere intermediary, but rather assumes the form of an oozy, sticky entity. Hence Leviathan’s seeming ability to reach out and touch the viewer.

According to Morton, our acknowledgment of hyperobjects brings about the end of the world. He argues that the idea that the weather is a neutral stage on which human affairs take place is destabilized by global warming. Global warming prompts us to recognize that the weather is in fact not a thing in itself, but rather the appearance of climate for us – climate’s virtual proper being is necessarily not given to us. Therefore, global warming renders the weather present at hand - it ceases to function as a background. Thus the significance the weather acquires in light of global warming renders illusory the horizon before which things appear as meaningful to us – the horizon occasioning our sense that we dwell in a world. Morton argues that the world is an aesthetic effect. It requires fabrication in the same sense as the seamless transparency of Classical Hollywood Cinema necessitates an elaborate technical apparatus. In light of this, it seems to me that the end of the world can be realized via aesthetic means. This opens the way for Leviathan, which does so in two main ways. First, the film refuses to put things in perspective for us. The unfolding space and time of the film is chaotic. It fails to afford the viewer a stable orientation towards the variety of phenomena it documents. The film’s frame registers less as an a priori window, and more so as a local manifestation of the camera’s mediation of the marine world. Throughout the film, the camera plunges beneath the surface of the ocean. In one of these sections, we see a random alternation of almost pure black with a snow-like swirl of white air bubbles. This passage altogether lacks a sense of spatial depth. Second, the film resists intelligibility. Its images attest to a dynamic in which iconicity regularly veers into abstraction. For example, the film’s first section shows a haul of fish being deposited on the ship’s topside. Or rather, this describes its representational content. Alternatively, if we describe the images in themselves, what we get is something different. Initially, the shot is almost entirely black. We catch a brief glimpse of orange and blue netting. Then a black “band” is dragged diagonally across the frame, circumscribed by churning green, and a patch of blue. That this passage evokes celluloid bearing the traces of catastrophic scratching, despite the fact that Leviathan is a digital film, is a measure of the degree to which it arrests legibility. Therefore, the film embodies, to paraphrase Morton, the elimination of the world-picture, the refusal of the reduction of nature to a use value, to an object of aesthetic experience. The film, however, is not only a negation of the world. What it affirms connects Morton’s idea of the situation after the end of the world to what onto-cartography maps. I’ll illustrate this after laying out several ways in which the film functions as a sort of onto-cartography.

Onto-Cartography

Leviathan shows that the ship’s instruments emanate a gravitational field that mediates the operations of the fishermen. Bryant defines gravity as the ways in which a machine affords or constrains another machine’s movement, local manifestations, becoming, and relations with other machines. He uses this term because it sidesteps the anthropocentric connotations of power, which limit explanations of how the social materializes to the discursive level. “There is,” he writes, “a gravity proper to phenomena as diverse as chemical, biological, meteorological, and semiotic processes.” A fundamental concept of the correlationist paradigm is the subject-object relation. The subject, a position only ever occupied by humans, exhibits the faculty of taking action, whereas the object is only ever acted upon. In Bryant’s flat ontology, however, subject and object no longer designate different kinds of being. Rather, they become roles that any entity whatsoever can inhabit in specific circumstances. According to him, “a subject is a catalytic operator that draws together machines in particular gravitational relations.” A machine occupying the subject position functions as the centripetal force that initiates and organizes a social assemblage. On the film’s map the various instruments the fishing vessel-machine puts to use occupy a subject position because they condition the actualization of the fishermen’s powers. Three moments in the film illustrate this. Looking over the side of the fishing vessel as a load of fish is being hauled in, we see an obstinate chain impelling a fisherman to grab hold of it; he endeavours to untangle it so that it doesn’t impede the process. On the deck, the fishermen are facilitating the mechanical drum winch as it coils the trawling wire. But to us they seem to be feeding an alien entity. Later, we see preparations for the unloading of a haul of fish. The netting’s movements are awkward and unpredictable; the fishermen strive to create the appropriate conditions for the haul’s release. Bryant notes that correlationism privileges perception as the seat of knowledge production. The subject passively intends an object. But, he argues, given that a machine is what it can do, to know it we need to put it to use. Seemingly animated by this idea, the film plunges into the fray, showing us the chain, winch and netting in action, absent prejudicial mental pictures of the situation provided by, for instance, the fishermen. In so doing, it allows us to grasp these machines as subjects.

In its mapping of the relations in which humans are engaged, Leviathan topologically positions the human as necessarily beyond-human – as a mesh rather than an impervious surface. In terms of Bryant’s ontology, qualities or attributes become local manifestations produced by a machine. This means that a machine’s colour is not something it has, but rather something it makes. The film encourages us to see the fishermen’s bodies in this way in a scene in which they process scallops. For a portion of this scene the camera is pressed up against a fisherman’s forearm. Its skin, emerging from an orange-red, fuzzy abstraction, is covered in dirt and abrasions, bearing protruding veins – all qualities speaking to the effects of associating with the environment of commercial fishing. His forearm then is a mark of a trans-corporeal body. Trans-corporeality, a concept Bryant borrows from Stacy Alaimo, refers to the idea that a body is co-constituted by the milieu in which it operates. Alaimo writes, “’nature’ is always as close as one’s skin – perhaps even closer.”

According to Bryant, machines are assemblages of other machines. The emergence of a distinct machine is the discovery of powers in such an assemblage not attributable to any of the machines of which it is composed. In the film’s mapping of the contributions of the fishermen to the harvest of hauls of fish, they appear to us as anonymous, faceless figures, human-shaped bodies draped in yellow, orange and green rain slickers. This depiction renders the fishermen as pure exteriority, which is reinforced by the lack of both intersubjectivity and indications of their psyches in the film. In their activities on the topside of the ship, the fishermen are actualized less so as humans, and more so as human-rain slicker assemblages, as unique machines separable from the bodies concealed by the outerwear. Bryant argues that the recognition of human-nonhuman assemblages such as these as distinct entities is an important affront to humanistic claims that our powers always stem directly from us.

Onto-cartography, explains Bryant, posits two kinds of relation between machines. Relations take place either on the plane of content or the plane of expression. The plane of content is the plane on which corporeal machines mediate one another. By contrast, on the plane of expression semiotic machines affect not the machine itself, but rather how other machines relate to it. He writes, “the two planes of content and expression are distinct and autonomous domains… While they can intermingle and interact in all sorts of ways, one plane cannot be reduced to another.” Leviathan is primarily focused on mapping the plane of content. For example, the film depicts the movements of the fishermen as they process fish as mechanical, systematic and habitual. More specifically, as a fisherman sorts fish in terms of edibility, the camera fixates on his shoulder, articulating it as a hinge, as an instrument composed of bone, muscle and other tissues, locally manifested to efficiently perform the task at hand. Similarly, in the scallop-processing scene, the activity of shucking is characterized by a regularity of tempo and motion, in terms of an individual fisherman, as well as a sense of uniformity, in terms of the whole group. These examples show a particular “form” and “set of dispositions,” actualized through the folding of the fishermen’s bodies as corporeal machines with the powers of the media ecology of the fishing vessel. That the fishermen populate this environment because they agreed to sell their labour in exchange for wages, and that they have ostensibly been trained to perform the tasks depicted by the film - that is to say, that what I have described is a function of the plane of expression – is crucial. However, Bryant would emphasize, a map of these relations is not completed simply by considering structuration by semiotic machines.

What is the situation after the end of the world? Following Graham Harman’s object-oriented ontology, Morton refers to the discrete, unique things that he contends make up existence as objects. Beyond the world’s horizon then he finds that “objects compose an untotalizable nonwhole set that defies holism and reductionism. There is thus no top object that gives all objects value and meaning, and no bottom object to which they can be reduced.” Here he dismisses both the positions of phenomenology and scientism. For his part, Bryant replaces the concept of the world with that of multiple worlds. In this light, he writes, “A world is an ecology of loosely coupled machines linked by machines without any of these machines totalizing world.” How are humans emplaced in a world? According to him, “To be in a world is to be decentralized, to lack all mastery, and to be a participant in an assemblage, network, or composition that exceeds society, culture, and one self.” So far, I’ve touched on some of the ways Leviathan embodies this non-correlationist conception of world. To end then I’d like to highlight how the film carries out the onto-cartographical imperative to remain faithful to the anarchic reality of worlds. Beyond the engagement of humans with nonhumans, the film maps a variety of encounters between nonhumans, critically from a point of view more or less inaccessible in reality to our perception. These include the mixture of fish waste with ocean water, marine life’s disturbance by and entanglement with fishing nets, the brief habitation of fish in the containers in which they are processed, and the capture of seagulls by the ship’s gravitational field. The film’s final section evokes the idea that after we discard the notion that we ground the world, we find that, in Bryant’s words, “Worlds are indifferent to our existence.” In this section, the camera is thrust into the ocean, where it remains for the rest of the film. Thus we see a field of mediations absent the human. Holding a disconcerting point of view, the camera offers us roiling and churning white caps, which appear to emerge from an opaque vertical plane. The ceaseless productivity of this motion seems to allude to the unthinkable power of the ocean’s virtual proper being. Seagulls hover in front of the ocean surface, arrayed in an almost geometric pattern – as if together they become a machine, thereby able to persist in the deep sea environment. This section also maps the ship’s efforts to contend with the ocean’s immense well of kinetic energy. The camera dives below the surface where it is assaulted by the ocean’s propulsive flow. The glimpses we catch of fishing net and trawling wire, coupled with the whir, hum, creak and groan of motorized equipment, serve to evoke the intense struggle of the ship with the ocean’s wave power. In this world, the ocean is the dominant gravitational force, constraining the actualizations of the fishing vessel-machine. Conversely, the ship’s apparent inability to perturb the operations of the ocean constitutes the ocean’s lack of care for the ship. In other words, the ship does not make an impression on the ocean, and by proxy neither do us humans.

#cinema #Leviathan #speculative realism

4 notes · View notes