A New Mode of Production
Back in very early summer when the Lukacs reading group was dominating real_philosophy, I didn't have the time to read all the material. However, the incident made an impression on me for a number of reasons, although probably not the impression that was intended. Rather than rebutting contemporary analytic philosophy of mind, it illuminated for me ways in which it might be truly progressive in ways that would have been impossible for Lukacs to anticipate. The reading group appeared to end on a depressing note because the vision of the future presented in the material failed to materialize, with seemingly only humanity to blame. But what if the failure of the vision is due not to people failing to take up their dialectical responsibility, but rather to the obsolescence of the theory that begot the vision? There may be more than one way out of the historical bind. We may already be out of it (and into another).
Those were the hunches at the time, anyway. So now I'm going back and (re)reading the Lukacs to be thorough about them. Since any imagined pressure is off, this is a distinctly mid-burner project, but it's a nice one in that it's forces me to stick close to a text, which is something I'm rarely forced to do these days. And the project is bearing fruit already--more so than I expected. By happy coincidence, I'm working now in an industry that, as it has matured, has invented for itself a new mode of production that is radically different from those of either capitalist investment or proletarian labor: software engineering. Buying the assumption for the moment that the mode of production really has the sorts of effects on society that Lukacs believes it does, these differences should be socially and philosophically significant.
There's a lot I need to start backing up now. To the text! Take this passage, from the Reification of the Proletariat section:
"If we follow the path taken by labour in its development from the handicrafts via cooperation and manufacture to machine industry we can see a continuous trend towards greater rationalisation, the progressive elimination of the qualitative, human and individual attributes of the worker. On the one hand, the process of labour is progressively broken down into abstract, rational, specialised operations so that the worker loses contact with the finished product and his work is reduced to the mechanical repetition of a specialised set of actions. On the other hand, the period of time necessary for work to be accomplished (which forms the basis of rational calculation) is converted, as mechanisation and rationalisation are intensified, from a merely empirical average figure to an objectively calculable work-stint that confronts the worker as a fixed and established reality."
This describes the change in the mode of production leading up to the state of the world in the 1920's. It strikingly fails to capture the mode of contemporary software production.
As I've said, I've got some experience in this industry, and I have friends and family in it as well. And nowhere have I found evidence of their work being "broken down into abstract, rational, specialized operations." There is an excellent reason for this: the whole point of software engineering is to automate abstract, rational, specialized operations. There is nothing more frustrating and antithetical to the thinking and productivity of a programmer than to perform the same abstract task more than three or four times in a row. So when faced with this sort of situation, programmers more often than not either (a) write a program which does this operation for them, or, increasingly, (b) find the tool that has already been written to do the operation, and learn to use it.
So, except for uncharacteristic exceptions, the programmer's work is not abstract or able to be rationalized. This contributes to the second departure from the kind of labor Lukacs was looking at: software development time is notoriously unpredictable. There is a book which I've never read but is considered a seminal one for the industry called The Mythical Man Month, which is precisely about the hopelessness of getting a new software system out in any kind of predictable time. Again, this makes perfect sense: if it were possible to rationalize the process of software production, then somebody would write a program that would do it faster and better than the programmers. So software production always happens on the frontier of the irrational--as soon as the land is cultivated, the engineers move on.
According to Lukacs, because of the rationalization of factory production, "the finished article ceases to be the object of the work-process. The latter turns into the objective synthesis of rationalised special systems whose unity is determined by pure calculation and which must therefore seem to be arbitrarily connected with each other." Not so with software production, because the work-process cannot be rationalized into arbitrarily connected components. Rather, under any reasonable conditions the software is an organic result of the programmer's act of rationalization.
The rest of Lukacs's dire logic is similarly dismantled for software engineering, since it is "this fragmentation of the object of production" that "necessarily entails the fragmentation of its subject." But what if the object is not fragmented? We find a reversal of the factory situation. In the factory, "the human qualities and idiosyncrasies of the worker appear increasingly as mere sources of error when contrasted with these abstract special laws functioning according to rational predictions." In the modern software company, the idiosyncrasies and human qualities of the worker are nurtured and cultivated, as these are the irrational source of the worker's inspiration. We see this most expressed in environments where the company has the resources to spend on its workers. A few weeks ago I got to go to Google's new office in New York City because a friend of a friend works there. They have a gym there. They have ping pong tables and foosball and Xbox's connected to enormous flat screen monitors mounted on the walls. They have dedicated Lego areas; workers are encouraged to go and spend time building whatever they want. There is the famous 80-20 rule, where it is recommended that 20 percent of everyone's time should be spent on personal projects. The same sort of irrationalities are part of other tech companies I know of--you pick your own hours, it is acceptable to spend company time organizing softball games, there are times set aside for lectures on technical topics of general interest but which are probably unrelated to ones work.
The result is that the programmer, contrary to the factory laborer, is "the authentic master of the process," and not at all "a mechanical part incorporated into a mechanical system." And, rather than "the subjects of labour" becoming "rationally fragmented"--with their labour-power objectified "into something opposed to their total personality," the goal of the management is to seduce the entire personality of the programmer into becoming oriented toward the object of his or her labor. The contemplative stance here is counterproductive; the personality must be actively engaged and, moreover, socially engaged, as the interactions between different programmer's code/actions are just as impossible to rationalize.
-------------------------------------------
What are the consequences of this? I have to do some more reading. However, there are a few leads I want to note here for now:
(1) Not everything is idyllic. In most tech companies, workers have to sign contracts where they sign away the rights to the code they produce--it becomes property of the company. In many cases, the product is then licensed and "sold" as if it were a normal commodity--think copies of Windows, etc. This is a remarkably uncreative way of making profit out software which is only possible because of some very artificial monopoly rights provided by the government. Of course, open source development doesn't have this problem. And then there is the fascinating case of web services like Google (or Amazon, or Ebay, or Yahoo, or Facebook) is different--here, the product is held hidden from consumers except as a service which is funded by advertising dollars. Note here that consumption here becomes an integral part of the mode of production, which we should expect in an era of flexible accumulation (cf. David Harvey) and which, I think, leads to some other (possible independent) unanticipated by Lukacs--specifically, the power of consumer action described by Ulrich Beck.
(2) Looking ahead, if programmers do not have their mediate status inflicted on their consciousness via their mode of production, the prospects for class consciousness here are grim. However, the unique relationship with rationality and its limits does make it very clear to programmers the impossibility of having a systematic account of the totality. I need to look into this more, but my hunch is that those philosophies most closely tied to the programmer class--namely, contemporary functionalism in philosophy of mind--may entail a conception of subjectivity that is much more of a departure from bourgeois thought than some people give it credit for.
(3) I just need to say this: The Industrial Age was the era of capital accumulation. Now, we are in the Information Age--the era of rationality accumulation.
(4) There are consequences of (3). It seems central to Lukacs' work that the commodity structure permeates all of society. For example, he writes:
"But this implies that the principle of rational mechanisation and calculability must embrace every aspect of life. Consumer articles no longer appear as the products of an organic process within a community (as for example in a village community). They now appear, on the one hand, as abstract members of a species identical by definition with its other members and, on the other hand, as isolated objects the possession or non-possession of which depends on rational calculations."
What happens when "consumer articles" aren't commodities--can't be commodities--because they are "intellectual" goods, like software (or music, or text). With digital reproduction and transmission the try nature of all "intellectual property" as being only in the most artificial and constructed sense "property" of anybody becomes increasingly clear. Each good is no longer a "member of a species," but a unique form that, though manifest in many 'copies'--on your computer and mine--really is the same object, at least in terms of its relation to production.