Jpeg: more than accidents, relations and qualities

Having established that jpeg can be treated as an object, it becomes necessary to explore what characterises an object from Harman’s point of view. What characteristics does jpeg as object have that enable us to understand its nature and its workings? For Harman, objects must be seen as beyond accidents, beyond relations and beyond qualities{Harman 2010@148-9}. Here an object has an ontological existence as well as analytical power and usefulness independent of its transient, shifting facade.

A Kodachrome slide is an object regardless of whether the cardboard mount is in pristine condition or weathered and dirty. It does not stop being a Kodachrome object because it is a dirty Kodachrome or a clean Kodachrome. The accidents of its condition or transient changes in its form do not change its object status. Similarly its status as object is not dependent on its relations. Whether it is in my shoebox, in an envelope en route from Switzerland; my photograph or the curator of MOMA’s,  it is a Kodachrome object – merely in a different position to the world. Finally the object is beyond its qualities. As Harman puts it when discussing the table object, these are not the “accidental streaks of dirt or the random amount of sunlight in which it bathes at any moment. Instead […] the qualities of the table [are what…] it needs in order to be precisely that table”{%Harman 2010b@149}. The qualities of the Kodachrome slide might include its transparency, its 35mm size. These are the things that we could not change without the Kodachrome object ceasing to be the same object. We will address the way this relates to the idea of an essence in a moment. But the slide is more than those qualities. The Kodachrome’s power maybe even presence (even leaving aside whatever image it contains) exceeds those qualities. This is what Jane Bennett refers to as an thing’s vibrancy{Bennett 2010}.

For Harman, the object as an ontological category, a philosophical concept and a tool goes beyond these incidentals. Unlike Whitehead whose occasions become and perish, objects are not tied to their accidents. Unlike Latour where objects are constituted in their relations or David Hume and the empiricists where  objects are nothing more than a bundle of qualities, for Harman, the object is more.

It is fairly easy to see how this can be applied to physical material objects such as tables and Kodachrome slides, but what about software or even more elusively, jpeg?

The first question is whether jpeg is more than its accidents, something beyond any transient changes. Just as a table is more than the dust on its surface, so jpeg is more than any particular instantiation of it within a piece of software, a device or a business plan. That is what makes it a standard. It remains regardless of whether it is active in any particular photographic moment. My use of jpeg in my camera as I take a photograph or in my archiving as I file my images does not exhaust what jpeg is. The incidental moment of Jpeg compression as I take a photograph or that when Flickr encodes my PNG screen-grab as a jpeg, are ‘accidents’, they are the dust on the table. They do not tell the whole story of jpeg. As we shall see, Harman uses this to provide a new framework for understanding time.

Jpeg can also be seen as beyond its relations. Here is where Harman moves beyond Latour. To say that jpeg has an ontological status, an analytical and political power beyond any particular relationship is to acknowledge that the standard has a form of ontological independence from the software with which it works, the business strategies that it in part enables, the images it encodes and the practices it sets in motion. This is not to say that there is no connection. As we will see, Harman acknowledges the importance of that and develops his account of internal relations to discuss that necessary connection. It is simply to say that even if there were no connections, even if image processing and archiving software did not use jpeg; even if Flickr and Facebook switched to archiving in another format, that would not stop jpeg being. Whether it would ‘matter’ as much is another question. The issue is whether one needs to address the object or the relations. For Harman the object needs to be accounted for independently of any particular instantiation or connection. It is here where Harman uses the tension between the object and its relations as a way into discussing space.

Jpeg can also be addressed independently of its qualities. Jpeg has qualities – the algorithms it uses to encode image data. It uses a series of codecs, markers and transforms to do its (standard) work. But they do not exhaust jpeg. There is more to Jpeg than the some of its parts. As a protocol utilising particular compression algorithms, that standard remains regardless of how those algorithms change. If, for instance, Huffman coding should change, that does not effect jpeg’s position as object insofar as it is addressed within software, practice, business plans my photography or now my PhD. Those accidents or changes are incidental to its status as object. Just as the layer of dust may change the look and even the function of the table (unsuitable for a posh dinner party for instance) so a ‘new Huffman algorithm may change the function of jpeg, but in the sense that jpeg remains the standard I use, it is still jpeg. This tension between the object and its qualities allows Harman to develop his understanding of essence.

In order to understand Harman’s threefold (in 2008 and his later fourfold{%Harman 2009}) account of how we can address time, space and essence in terms of objects, it is necessary to take a step back to Husserl’s “intentional objects”. In Harman’s reading (and of course as with much of Harman’s reading of the classics, there are other readings), what is interesting about Husserl is not his bracketing of the real world and focus on the phenomenological realm, but what happens when he does this, i.e. when one brackets the real world and focuses on experiences, one is dealing with objects. Harman goes as far as to say “phenomenology is above all an object-oriented school”{Harman 2010@150}. The real flower we smell or bee we hear are bracketed, they withdraw. What we encounter are the intentional objects that are “something different from the profile through which they become manifest”{Harman 2010@152}. These intentional objects (Harman also variously refers to them as images or even simulacra)[ref]There are obviously issues over Husserl’s relation to Platonic accounts of the ideal that I will not address here[/ref], are what we deal with, what we encounter and what we connect with. The experience of colour or sound or smell when we encounter something is the basis for realities.

As an example, in my encounter with the 2012 Fence, I and the Fence can be addressed as objects. I do not encounter the ‘real’ fence, I encounter an intentional object, the image of the Fence. I cannot encounter the real Fence. It is always more, not only in the sides or sub-atomic level I cannot see but that which is beyond the qualities and accidents that are accessible. In this framework, the Fence also encounters me[ref]This is not an attempt to anthropomorphise material objects or develop a new form of pansychism (although Harman does discuss this). Nor is it to attribute agency, let alone perception to everything. It is simply to work from the basis that objects by necessity share the same space and time and therefore can be seen as having some form of relation. It is not that the Fence is aware of me but it is affected by my presence in terms of my breath contributing to its rust, my gravitational pull and other myriads of trivial ways[/ref]. Except the Fence does not encounter me, it encounters an image of me. As an object I too withdraw. There is more to me than the breath that moves the Fence etc. The real Fence and the real me withdraw from each other, but there is a relation, a connection between us. For Harman, that connection is itself an object because that encounter can be observed and described by another object (a watcher or the security camera for instance). For Harman the relation is not external to objects, it happens on the inside (in the molten core, as he puts it) of another object. We don’t need to talk of relations or context or flux or plasma to discuss relations, we can do it all with objects. And the reason to do this is, from Harman’s point of view, so that we can develop an object-oriented account of time, spaces and essence and for me so that we can open the Exploit which is at its most powerful when dealing with objects.

Before we discuss that, let us redraw our example using the jpeg object. The jpeg protocol-object encounters the Facebook database-object (remember, in this account anything that has or appears to have a form of unity can be addressed as an object). Jpeg encounters an image of the Facebook database-object. Firstly it can never encounter the whole database-object, it connects with only a part of that code and secondly its connection in the moment of uploading does not exhaust what that database is – it has a position as object before and after, it is a site of other connections and workings and it is being constantly reconstituted as new data is entered, new searches conducted, new datamine connections made or the database moved from one server farm to another. The jpeg-object encounters the intentional Facebook database-object. At the same time the Facebook database-object encounters the jpeg protocol (within the Facebook software)[ref]I am obviously drawing a slightly arbitrary distinction here between different components of Facebook’s software assemblage in order to illustrate the objects in play[/ref]. In fact it encounters an intentional image of Jpeg, a particular instantiation. It encounters a moment of encoding, a particular position of the standard. It does not encounter the complete complex reality of that standard in terms of its history, nature and other connections and, as we shall see, essence. It does not need to. It needs to encounter which is necessary to establish the database image and data point. This encounter happens not in some extra space or context, in Latour’s plasma or Whitehead’s space of becoming. It happens within another object – what we might call the Facebook photo-object. Within an object-oriented framework the encounter between jpeg and the Facebook database can be addressed as an object because it has a unity. It can be described by another object. Within Facebook’s business and architecture the photo-object describes and is constituted by the encounter between an instantiation of jpeg and the database. When I upload a PNG image file, jpeg encodes it as a jpeg/JFIF file and encounter the database where it files the data and the metadata. This encounter is described or observed by the Facebook photo-object in the sense that that photo-object holds that relation together it defines its limits. In concrete terms of other software or even human engineers it says “here is a data point, something to mine”. In its turn, this photo-object encounters other objects – datamining software, human engineers, Facebook friends etc. The thing is that all those connections happen within objects.

In one sense this is a form of nested objects but it is important to emphasise that these are not nested in any hierarchical let alone value-laden sense. There is no sense in which objects connecting with other objects should be seen as leading to a foundational macro or micro object. Just as DeLanda rejects macro and micro-reductionism{%DeLanda 2006}, so this model not only refuses to leave the object but also refuses to find the single object. There is no Facebook-object or Surveillance-object or Capitalism-object that can take the place of ‘context’ or ‘relation’ as foundation for all connections. Nor is there some machine code-object or electrical charge-object that can substitute for a founding object or fundamental particle.

The advantage of this approach is not only that escape from a context or macro/micro-reductionism. It is, appropriately enough, four-fold. It provides a way of escaping the problem of the subject; it lets us find a way of talking of space and time without having to choose between time chunks and time as arrow; it allows us to talk of essences and technological determinism without a sneer; and finally it enables us to open up Exploits for intervention.

Firstly this perspective escapes correlationism, Quentin Meillassoux’s term for the tendency to focus on the subject-object relation, to see everything in terms of the human-world connection{Meillassoux 2009}. Here there is no world without the human nor human without the world. It is this separation (yet partnering) of subject and object that drags us away from focusing on objects, their connections and their working.  In terms of jpeg, it demands we address jpeg and the Facebook database in terms of the humans using it. At the very least this means it becomes difficult to explore machine vision systems where computers ‘see’, ‘file’ and ‘analyse’ with no human intervention, a situation an object-oriented approach could happily discuss in terms of photo object connecting within face-recognition object within a surveillance-image-evidence object.

Secondly, for Harman, an object-oriented approach allows us to approach time and space not in terms of whether time should be seen as an arrow, a continuous flow or as a series of discrete chunks but rather as an emanation of the tension between objects and accidents. Again rejecting any idea that we need to move beyond objects to relations to account for time, Harman sees the passage of time as the difference between the table and its (accidental) dust in one ‘moment’ and the table and no dust in the next. Here time is the tension between jpeg and the CCD signal in one moment and the lack of it in another. Time is the experience of accidents. Similarly space can be understood in terms of the tension between objects and relations. It is this difference between the Kodachrome slide in my shoebox and in a lab in Lausanne; the difference between jpeg in my camera alongside the Fence and on a server in Texas. There is no need in this perspective to go beyond the object.

Thirdly, an object-oriented approach allows us an account of ‘essence’ that does not close off debate, connections, change and power. For Harman what we experience as essence – and we do in the sense that we experience something about tables, Kodachrome slides and even jpeg that does not change – is actually the tension between objects and their qualities. Objects have an independent reality, more importantly a specific reality insofar as a Kodachrome slide is not the same object as a table or a jpeg protocol. “Hence there is no avoiding a concept of essence”{Harman 2010@163}. That does not mean that essences are eternal or natural,“to defend essence is not to conspire in a sinister plot by the Party of Reaction. It is nothing more than to insist that objects are not exhausted by the relations to other objects”{Harman 2010@164}. What we experience as essence is the outcome (or emanation as Harman calls it) of the tension between the object and its  qualities. We experience certain qualities as essential. There are things about a table, a photographic slide or a protocol that are ‘necessary’ for it to be that table, slide or protocol[ref]This is perhaps even more so with software insofar as if those qualities were not present, the software would not work[/ref]. But as we have discussed these qualities are not identical with the object. They do not exhaust it. What we experience as ‘essence’ is what is left when we strip inessential qualities away. This is significant because it means we can talk of jpeg, of things. We can say: “yes there is jpeg” and then trace its connections within objects. We can use that essence as a space for Exploit. An object-oriented essence is a starting point not an end.

Even more controversially perhaps, this rescuing of ‘essence’ allows a similar embracing of ‘technological determinism’. As Geoffrey Winthrop-Young puts it: “to label someone a technodeterminist is a bit like saying that he enjoys strangling cute puppies”{WinthropYoung 2010@121}. This non-reductionist, object-oriented reading of essence however allows us to say: “yes technology determines”. The issue become how that determination is drawn in terms of causality or what DeLanda calls “catalysis” for instance{DeLanda 2006}. But leaving that debate to one side, the issue is that again an object-centred approach can explore determinations as connections within objects rather than as reflections of something more basic, foundational or powerful. It allows us to proudly and openly say that the connection between jpeg and the Facebook database (within the Facebook photo-object) does things.

Finally, the escape from the subject, from the context, the relation, the continuum and the occasion – the focus on the object – allows a space for Exploit, a critical practice that draws on Galloway and Thacker’s rethinking of networks.

  • Bennett, J., 2010, Vibrant Matter: A Political Ecology Of Things, Duke University Press, Durham.
  • De Landa, M., 2006, A New Philosophy Of Society: Assemblage Theory And Social Complexity, Continuum, London ; New York.
  • Harman, G., 2009, Prince of Networks: Bruno Latour and Metaphysics, Anamnesis, Melbourne.
  • Harman, G. 2010, Space, Time, and Essence: An Object-Oreinted Approach (2008), in Towards Speculative Realism : Essays And Lectures, Zero Books, Winchester and Washington, pp. 140-69.
  • Meillassoux, Q., 2009, After Finitude: An Essay On The Necessity Of Contingency, Continuum, London.
  • Winthrop-Young, G., 2010, Kittler And The Media, Polity Press, Cambridge, UK.