by Jesse Powell
The WB202 podcast at Critical Inquiry has an interesting interview with one of this group’s distinguished scholarly speakers from our final webinar this year, Alexander Galloway. Titled “Defining the Digital”, Patrick Jagoda’s podcast involves discussion with Galloway about his work in general, with two general questions emerging from the conversation that I do think it is safe to say Galloway has spent a great deal of time thinking rather hard about: 1. Just what is “the digital?” 2. Just what is “the analogue?”
Reading the article linked to the podcast (“The Poverty of Philosophy: Realism and Post-Fordism”) gives a glimpse into why Galloway thinks his considerable expenditure of effort attempting to answer these questions is worthwhile: With the emergence of what he calls a “renaissance” concerning Realism in Continental Philosophy — what he calls “Speculative Realism” — has come a parallel philosophical emergence in the present state of economic organization that is prevalent in countries of the Global North. For example, in the work of Quentin Meillassoux concerned with flushing out his “speculative materialism”, Galloway finds what could be a coincidence in his ontology and the mode of production of post-Fordist capitalism that he sees as exemplified in the programing language Java, and object-oriented programming languages in general. This is perceived as a grave danger to philosophy as a whole, for if this sort of Realism truly is to be the next big thing, or at least play a major role, we stand to lose the ability to criticize our current economic situation and culture climate effectively. If our theories and critiques just reproduces the status quo, they will never be effective, and will not allow us to to imagine alternatives.
Thus, it is imperative for us to be able to judge whether or not we are simply reproducing our digital economic conditions in our philosophy that we have a clear and useful understanding of what digitality is, what it means to be analog, and here comes the real rub: we have to define these terms without ourselves making use of the resources of our economic and societal practices lest we commit the very sin we are attempting to accuse those troublesome realists of. (I think this is part of what motivated his desire for a “purely rational” definition of the digital that he mentions in the discussion, as well as efforts that he describes later on to define the digital without reference at all to computers. He is trying his best to avoid exactly what he thinks the Speculative Realists are likely guilty of).
I must say that I agree mightly, that we should be wary of our philosophical notions and what we might be taking for granted, and with the care we must take to try and avoid reproducing problematic ideas, or promoting those that our own concotions. It is precisely as HIlary Putnam says: Though we be kibitzers at the end of the day, if we don’t try to do good philosophy, all we will have left is at best philosophy and nonsense, that is always and constantly produced, and at worst will be dangerous philosophy that genuinely harms us. And to do good philosophy, we must take the sorts of precautions Galloway is in so far that as part of our job is to imagine alternatives, we should try and avoid developing what Daniel Dennett has called “Philosopher’s Syndrome”. As concerns his concerns though, is it that we have not understood the digital yet? Is it true as he says, that we have yet to really dig into its philosophical aspects, and can we do so without mention of the technologies and programming languages that does not “simply coincide with the spirit of capitalism?”
Well yes. As a matter of fact, it has already been done. But another point first; if there is one thing the Analytic tradition can offer to project like this, it must surely be its insight into the nature of definitions. From the time Russell first read Frege and from then on, without end or antebellum, they have obsessed with it, fruitfully. One of the most relevant developments was a theory of explication, as followed the work of Rudolf Carnap (1956), and Quine. What is an explication? Consider these two types of definition that we can use as definientia: 1. Stipulative definitions are those that impart meaning to the defined term, and needn’t be concerned with how the term is actually used at present, or was ever used. These sorts of definitions are necessary when we want to coin a new term or item of technical vocabulary, or to fix the reference of certain terms already in use (Kripke, 1980), or to amend the meaning of a term already possessed. 2. Descriptive definitions are similar that they try to make explicit what a term is, but they have as a design constraint that they must capture the significant features of existing usage; they cannot stray to far away from what the folk mean by the words they say.
Having a commitment to maintain the rightness of fit between the predefined meaning and the post-post defined meaning can be motivated by a number of philosophical desiderata, chief among them is a desire to analyze and thus clarify the everyday concepts central to human life, or to some specific human activity, or that is troubling the human condition in some way; were we to offer up a stipulative definition of the word “justice”, but it not refer to that sort of thing that everyday people mean it to, what good is it to them? What good is it to us? These two sorts give us the ingredients necessary for explication, the sort of defining I feel is most important; explications are a hybrid stipulative-descriptive definition, in that they aim to capture central prototypical features of the concept they aim to define as they are presently used, precisely so they can stipulate that it have this or that different feature/ different meaning. They are a sort of conceptual engineering, allowing us to change our concepts piecemeal, to shed problematic aspects of them, or to designate how we think a term should be for it to serve our ends. But when is an explication valid? When does it fit the explicatum adequately enough? What features are necessary to preserve, and how far can we alter the meaning of the term?
W.V.O. Quine offers these helpful criteria and a rule of thumb. What is truly important is that an explication preserve function in a specific context, differences in meaning beyond this being irrelevant. (1960) If the term differs before and after only out in the “dont-cares”, difference in meaning is not an objection. This definitional invariance of function is what gives explications their real usefulness for Quine. What this allows us to do is start with some objectionable concept that we would like to see replaced, but that we cannot do without, nor can we replace it with a whole sale different concept. We can then formulate some explication, or a series of them, that we can freely interchange, on more and more occasions as we pin down what we consider the central functions of the original term to be across context, developing explications that eventually game full isomorphy with respect to function, yet lack all of the disagreeable parts of our original concept. We can then just stop using the original all together. To be sure, he understands explication always to elimination, but it needn’t be conceived of happening all at one time, meaning whatever theories we have being dependent on the bad concept will be altered to drastically or unnecessarily, allowing us to more or less go about deploying them with success (something akin to what he calls in “Two Dogmas Revisited”, The Principle of Minimal Mutilation).
What this all means for Galloway, is that he might be going about this all the wrong way. We needn’t object to using the terms and concepts from computer science, or from our current economic theories, etc. We should be aware of when we are borrowing concepts surely, but if their deployment truly is a heinous thing, something that should be out of the question for the good critic and philosopher, we can just explicate the bad parts away. Besides, it is the concepts being used in economics and computer science, in business, and so on and so on, and these concepts precisely, that Galloway is concerned with coming to understand and define.
We want to capture the central features of digital and analog as they are used in the real world, not conjure up two separate concepts, digital* and analog* wholesale. If he cannot adequately define these conceptions of digital and analog he cannot criticize this new wave of realists for deploying them, because he cannot say if they truly are out of ignorance. By starting with the accepted notions, we can deploy them in humanities work to the functions they serve elsewhere, and slowly shape them to be discipline specific and most useful for us.
First point made. Now then, it just so happens that there exists a taxonomy of terms within which the digital and the analog have been defined within the philosophical tradition, that do not depend on reference to computers, and that I think are perfectly suitable for Galloway’s purposes. They can be found in the wonderful article by Corey Maley “Analog and digital, continuous, and discrete” (2011). Drawing together the central features of these terms as they have been used in the cognitive science and philosophy of mind literature concerning mental imagery and representation, as well as electrical engineering, communications, and computer science and mathematics, he produces a quadrumvirate of illuminating definitions.
They all concern representation, and I will spare you more exegesis, though I will list my personal tweaks to his definitions that I believe will fit the role that Galloway has in mind for them. To begin with, it is important to recognize that though often we conflate analogue with continuous, that they are not necessarily synonymous, and we treat digital and discrete dually. The main difference between the two sets of terms is that the one, analog and digital, has to do with representations, and the other can apply to anything whatsoever. Maley defines a representation to be “Analog” if and only if it changes in value in response to a change in value of what it is representing, i.e. they covary, and that the difference in the representation be a linear function of the change in that represented. Thus there is a clear sense in which the changes are “analogous”, in that they correspond, match, and so on. He defines a representation as “Digital” if and only if it is a series of digits in a positional notation system. The position determining the value means that there can only ever ben countably many digits in a sequence, so digital representations are necessarily discrete, but the converse does not hold. And that’s all.
Thus the mystery has been solved. Galloway need not stress about computers, and silicon valley, and the dangers of realism. True enough a computer can be a digital computer, but it also can be analog, and many have. I’d like to close with how I think these definitions have helped clarify the “Digital” in the Digital Humanities for me personally. Out of all the things I could say it is, I am going to follow Karl Popper and say that the Humanities is distinctive in that it is primarily concerned with studying the intended and unintended products of human actions and thoughts. Now to make that Digital, we take those objects of our study, represent them digitally, as sequences of numbers, and study the properties and structures of the representations in hope that they reflect features of the original objects.
If the original objects were “born digital” all the better; saves us one step.