The basic concern is the distinction between Formal and Real Subsumption. This can be broken down in the following way:
Labor-extensive : Labor-intensive ::
Formal subsumption : Real Subsumption ::
pre-capitalist production : capitalist production
The basic narrative arc is: capital encounters pre-existing modes of production and subsumes those workers to the demands of capital but without changing the nature of work itself. Any extra surplus-value is generated by working harder, or longer (what some economists call an "industrious revolution," see Kenneth Pomeranz). Once capital becomes the central, organizing motor of economic production, however, it begins to repeatedly revolutionize production by introducing techniques such as enlarged scale, machinery, overseer systems, etc. This is capitalist production proper.
And so a few things to consider:
1) This demonstrates that within Marx's conception, a set of producers can belong to capitalist competition and circulation without technically operating under terms of capitalist production. We saw this many times in Volume II: an individual, non-capitalist producer that belonged to a total social capital marked by a mode of production that was capitalist in the proper form. E.g. slavery in the American south competing with cotton produced by wage laborers elsewhere around the world.
2) What does it mean to be subordinated to capital? One could always say that merchants have always existed, middlemen have always existed. What marks them as specifically capital at any one moment, if the rest of the features of capitalist production are missing? This is what I'm trying to figure out. Marx will occassionally reference how even family production that owns the means of production still can produce for merchant capitalists. This is in fact the premise of Kautsky's The Agrarian Question. Still, the majority of "formal subsumption" passages assume a separation of wage laborer from means of production.
3) The most thorough explanation is in the Appendix to the Penguin edition of Volume I, entitled "Results of the Direct Production Process." It's also part of the 1861-1863 Manuscripts, which is where most Marxist scholars attribute the concept to. So I didn't read this section the first time around reading Volume I, but I re-read it recently.
4) There are hints of history here. Besides the modular nature of Marx's descriptions, there are references to encounters between capital and pre-capital:
Despite all this, the change indicated does not mean that an essential change takes place from the outset in the real way in which the labour process is carried on, in the real production process. On the contrary, it is in the nature of the matter that where a subsumption of the labour process under capital takes place it occurs on the basis of an existing labour process, which was there before its subsumption under capital, and was formed on the basis of various earlier processes of production and other conditions of production. Capital thus subsumes under itself a given, existing labour process, such as handicraft labour, the mode of agriculture corresponding to small-scale independent peasant farming. If changes take place in these traditional labour processes which have been brought under the command of capital, these modifications can only be the gradual consequences of the subsumption of given, traditional labour processes under capital, which has already occurred. The fact that the labour becomes more intensive, or the duration of the labour process is prolonged, that the labour becomes more continuous and more systematic under the eyes of the interested capitalist, etc., none of these things changes the character of the real labour process itself, the real mode of labour. This therefore forms a great contrast to the specifically capitalist mode of production (labour on a large scale, etc.) which, as has been shown, takes shape as capitalist production progresses, and which revolutionises the kind of labour done and the real mode of the entire labour process, simultaneously with the relations between the various agents of production. It is in order to mark the contrast with the latter mode of the labour process that we call the subsumption of the labour process under capital examined so far — which is the subsumption under capital of a mode of labour already developed before the emergence of the capital-relation — the formal subsumption of labour under capital. The capital-relation is a relation of compulsion, the aim of which is to extract surplus labour by prolonging labour time — it is a relation of compulsion which does not rest on any personal relations of domination and dependence, but simply arises out of the difference in economic functions. This capital-relation as a relation of compulsion is common to both modes of production, but the specifically capitalist mode of production also possesses other ways of extracting surplus value. If, in contrast to this, the basis is an existing mode of labour, hence a given level of development of the productive power of labour and a mode of labour which corresponds to this productive power, surplus value can only be created by prolonging labour time, hence in the manner of absolute surplus value. Therefore, where this is the sole form of production of surplus value, we have the formal subsumption of labour under capital.5) Despite that, there are reasons to believe that even if real subsumption occurs, as Antonio Negri emphasizes, one should hold out the possibility that it is incomplete, that formal subsumption and precapitalist production ("survivals" as anthropologists term it) survive. Here I think it is important to reference the only section that really discusses formal and real subsumption in the actual text of Volume I. The Chapter is number sixteen, entitled "Absolute and Relative Surplus-Value." In it, Marx describes the distinction then goes on to say that hybrid forms persist:
The prolongation of the working-day beyond the point at which the labourer would have produced just an equivalent for the value of his labour-power, and the appropriation of that surplus-labour by capital, this is production of absolute surplus-value. It forms the general groundwork of the capitalist system, and the starting-point for the production of relative surplus-value. The latter pre-supposes that the working-day is already divided into two parts, necessary labour, and surplus-labour. In order to prolong the surplus-labour, the necessary labour is shortened by methods whereby the equivalent for the wages is produced in less time. The production of absolute surplus-value turns exclusively upon the length of the working-day; the production of relative surplus-value, revolutionises out and out the technical processes of labour, and the composition of society. It therefore pre-supposes a specific mode, the capitalist mode of production, a mode which, along with its methods, means, and conditions, arises and develops itself spontaneously on the foundation afforded by the formal subjection of labour to capital. In the course of this development, the formal subjection [subsumption] is replaced by the real subjection [subsumption] of labour to capital.
It will suffice merely to refer to certain intermediate [hybrid] forms, in which surplus-labour is not extorted by direct compulsion from the producer, nor the producer himself yet formally subjected to capital. In such forms capital has not yet acquired the direct control of the labour-process. By the side of independent producers who carry on their handicrafts and agriculture in the traditional old-fashioned way, there stands the usurer or the merchant, with his usurer’s capital or merchant’s capital, feeding on them like a parasite. The predominance, in a society, of this form of exploitation excludes the capitalist mode of production; to which mode, however, this form may serve as a transition, as it did towards the close of the Middle Ages. Finally, as is shown by modern “domestic industry,” some intermediate forms are here and there reproduced in the background of Modern Industry, though their physiognomy is totally changed.
6) Finally, as noted elsewhere by Daniel Buck, the discussion of formal and real subsumption earlier appears under the heading of original accumulation in the Grundrisse, written before Marx had coined these new phrases. In that sense formal/real subsumption together with original accumulation occupy the heart of the problem of history and capital, pre-capital and transition. Here is where one should start in order to think through the problem of the gap between capital's history and capital's logic:
Marx apparently understood primitive accumulation and subsumption to be intimately related: he discusses the incorporation of weaver labor into the circuits of capital in the context of primitive accumulation in the Grundrisse (1973:510), but in the context of formal and real subsumption in Volume I of Capital (1976: see “Results of the immediate process of production”, especially pp 1019–1038).