What is a service? (A note on Endnotes)

This commentary has been extracted from: “Labour” is itself a recent & ephemeral historical artefact.

Defining the notion of service is notoriously difficult. This is my brief attempt at systematising it in an historically‐specific and thus useful way, incidentally critiquing Endnotes (especially [End10]) along the way.

Typically, a “service” is defined as human social activity that’s both waged, and consumed for the sake of the activity’s immediate (unmediated) effect, rather than to produce something(s) that can be extracted from the activity and sold separately.

At stake are three analytical binaries: labour/nonlabour (= subsumed/unsubsumed = productive/unproductive, in the surplus‐value sense), waged/unwaged, and real/formal subsumption.

I’m inclined to argue that any service can be labourised — that is, subsumed — if three basic conditions are met:

1. The worker must be separated from the relevant means of production, specifically in the sense that she’s incapable of selling her service as an independent worker–seller for a price similar to or greater than the price she’d get by selling to a capitalist, agency, corporation, or what have you. The latter fulfils the role of advancer of capital.

   Note that restricted access to the means of production may take on any combination of manifold forms: legal restrictions (not limited to property), servitude, political authority, means of communication, means of transport, business connexions, etc..

2. The service (or rather, its effect) can be sold consistently enough at a price corresponding to SNLT — the latter being established through a basic level of market competition (or other MPP). This is to say that each instance of the service is a commodity: it can be directly compared to other instances of the same service or to any other commodities, within exchange, in a way that implies exchange‐value.

3. The wage is paid out of the variable capital (advanced by the aforementioned “capitalist”) roughly according to the value of the worker’s labour‐power. The worker then labours beyond her necessary labour time, thus producing surplus‐value for the capital.

In many cases, (a.) is rather unlikely. Nonetheless, it’s always possible in principle.

On the other hand, (b.) tends to be blocked for the same reasons that cause value‐dissociation to occur. No identity logic, no commodification. This barrier is more well‐described as a barrier in principle, with the practical result being that certain activities — waged or otherwise — remain nonlabour until capital can largely eliminate them as activities, if that time ever comes… which it might not.

Even when capital does subsume a service, that capital lacks the ability to restructure itself — to use accumulated capital to further rationalise the labour process — beyond a certain very early point. Doubling the capital mostly entails a mere doubling of the variable capital.^[1] In other words, formal subsumption never becomes real subsumption.

This leads us to a proper definition of service: social activity within the capitalist totality that’s waged, but that by its nature cannot be really subsumed — irrespective of whether it can be, or is, formally subsumed.

[End10] makes a similar‐sounding point, but wrongly (in my view) conflates formal subsumption with the mere inability to be really subsumed. There is, of course, another option for activities that resist real subsumption: not being subsumed at all. The mere fact that such an activity might be waged is necessary but insufficient for (formal) subsumption; to believe otherwise is just another manifestation of the wage fetish.