>There is no classical-Marxian labor theory of value. Marx was a critic of
>the classical labor theory of value at several levels, including as
>discussed here Adam Smith's *fetishistic* criterion of commodity durable
>vendibility for productive labor (see Sydney Coontz, Productive Labor and
>Effective Demand; Ernest Mandel actually accepted this criterion), which
>threby excludes all services as surplus value generating. It's not the
>nature of thing produced but the social relation under which
>things/services are produced that determines whether labor has been
>productive of surplus value. See Patrick Murray's criticism of Mandel in
>CJ Arthur and Geert Reuten,ed. The Circulation of Capital
>
>On the other hand, Marx does seem to have accepted Ricardo's focus on only
>reproducible commodities. The price of rare works of art wholly depends on
>the whims of the buyer due to their intrinsic scarcity and would thus not
>be regulated by their socially necessary labor time--Marx freely admits
>this exception. So while Roger may be correct that an artist-genius
>working for a capitalist perhaps in a quasi putting out system produces
>surplus value, it is helpful to remember that Marx is concerned to limit
>his analysis initially to what is new and distinct about capitalism--the
>mass production of reproducible commodities by means of wage labor,
>collectively organized. Marx thus excludes from his initial analysis rare
>works of art and land--both of which predate capitalism. Some
>(neoclassical economists) would say that it is illogical to make such
>exclusions in the analysis of the price form, but Marx would say that it is
>more important to get first at the historical specificity of capitalism in
>order to disclose its law of motion, as Karl Korsch put it. There is a
>fundamental conflict here as to the appropriate method in the social
>sciences.
I agree with your qualification of what I said Rakesh, and the whole of your post too. But perhaps on a slightly different basis. I don't think the primary reason Marx is less concerned about artist's labor is because he focuses on what's new and distinct about capitalism (though of course he studied capitalism as an historical system, different than others), but rather because he initially focuses on the essential elements of the wage-labor factory system as the dominant mode of production. Using successive approximations (the abstract-deductive method), he begins with a set of facts pared to their essenyials. Volume I of Capital is thus devoted to a study of capital-labor relations in their "isolated" and "purified" forms. Capital-labor relations in the factory setting remains the focus as some simplifying assumptions are relaxed in later volumes.
As you point out, artists' work was outside the factory system. So that's one explanation for their exclusion.
For artists that do work for capital, I said Marx would consider them productive labor like anyone else. I have said that, in constructing his definition of productive labor, Marx did not explicitly recognize the question of whether the goods produced reentered the production process (as consumption of c or v). I attributed this mainly to later writers like Blake, Becker, Gough, and Harrison. This may be literally true if you pay attention only to where Marx explicitly dicusses and defines productive labor. But Capital is *about* the *reproduction* of capital and capitalist relations. As you say, "Marx does seem to have accepted Ricardo's focus on only reproducible commodities" (I would quibble with the word "only", but it was a focus). So perhaps, my distinction between the different definitions is too rigid, and I gave the later writers too much credit.
If you adopt this qualification, you get the following definition of productive labor: Labor producing luxury goods (not part of the consumption of capital or productive labor) is not productive (of the reproduction of capital) even though it produces surplus value for individual capitalists. But, noncapitalist labor that produces a product for Dept. I or II (e.g., teachers in public schools) is productive labor.
For myself, I like this breakdown. I think it gives a better sense of the relative functional importance of different kinds of labor to the continued expansion of capital. And, correspondingly, these functional relationships probably yield better ways to understand the working class as a disparate group. It changes the calculation of surplus value too.