> The transformation of "value" into "production
> prices" is literally meaningless.
The term "transformation" is very unfortunate. A better term would be "relation." How are relative value magnitudes related to prices (because they are)? And the answer concerned Marx so seriously that he devoted much attention to the issue. (But that is *not* what the "transformation" issue is about. See below.)
Both value and prices are "real," in the sense that the content of both concepts is *objective* -- *socially* objective. (For my take on social objectivity, please refer to my old post on productive labor: http://mailman.lbo-talk.org/2006/2006-January/001158.html) Value refers to the deeper reality of the phenomenon. Prices refer to the superficial appearance of same phenomenon.
Real social life viewed as unfolding over time is a single stream where everything is chaotically intertwined with everything else, yet -- at once -- things appear disjoint and differentiated. If you look at capitalist societies as societies where stuff is traded, then prices and their changes can be more immediately perceived. Value is an aspect of the same phenomenon that requires a significant rational effort to elucidate and grasp, since it's not immediately available to our senses.
Prices are noise, value is signal. Both signal and noise are real. But it takes a human mind (not just our sensory and perceptual mind, but our rational side as well, manipulating abstractions) to decompose the complex reality of sound into its noise form and its signal content. The rational effort is aimed at connecting the dots, making sense of things that appear disconnected, of finding the continuity and commonality of things that appear discrete and distinct, and grasping the necessity of the distinctions and variations.
> To try to reduce it to a short sentence: value is
> not a quantity underlying price; value is a social
> relationship expressed through price.
Marx distinguished between the "substance" and the "magnitude" of value. The substance refers to *labor* in a setting where production is conducted by private, independent producers. It's alienated labor.
Labor that can only find an objective expression in the value of commodities. Thus, the exchange of values is, in fact, when looked at as a whole, the social metabolism that enables the reproduction of a society in which production is a private and independent activity. So value is form, the social form of labor necessarily associated with private, independent production. (In turn, the form of value is the content of other forms, e.g. prices, and so on and so on.)
Value magnitude expresses itself as *relative* value, i.e. reflected in its relation to an equivalent. Anwar Shaikh called the center of gravity of prices under simple commodity production the "direct price," a price that reflects the values of the two commodities compared or exchanged, a term to be distinguished from the "price of production," which refers to the specifically capitalist case.
The "direct price" (or, if the reader doesn't like Shaikh's term, the relative expression of value of a commodity under simple commodity production) has to the "production price" a relation analogous to the relation between simple commodity production and capitalist production proper. To the extent capitalist production resulted from the historical generalization of commodity production, then this transformation was an actual historical process. How we grasp it logically is another question.
Formally, the relation between "direct price" and "production price" (the so-called "transformation") is not the same as the relation between "value" and "price." Both direct prices and production prices are *prices*, relative expression of values. To push my signal/noise analogy, the direct price is the sound of violins in a Mozart concert, while the production price is the sound of the whole orchestra. Both sounds are the unity of signal and noise. Violin sound as a phenomenon can be rationally decomposed into its signal and noise. Orchestra sound as well, but the latter is more complex, because it includes the violin sound mixed up with the sounds of other instruments in very complex ways.
Personally, I believe there are many possible ways in which one can relate production prices to direct prices. In the abstract, one necessarily has to make assumptions to make the explanation easier to communicate. But success or failure in the tasks of relating violin sound to complex orchestra sound is no reason to deny the fact that violins play a key role in the sound of a well-orchestrated Mozart concert. It'd be silly for me to say that, because a given acoustic physicist was unable to show me how the sound of violins logically "transforms" itself into the sound of the orchestra in a Mozart concert, then there's a "formal contradiction" between violins and orchestra, and -- therefore -- the whole notion that sound can be decomposed into signal and noise, and the reality of signal must be questioned. A neo-Ricardian would conclude that we should not care about signal, only about the *real* aspect of sound -- namely noise.
>From my (distorted) viewpoint, things look thusly: If showing the
logical relation between violins and orchestra were practically
necessary for my enjoyment of Mozart's music, then I would be more
interested in that "transformation" problem, and I'd want to spend
more time to get it all straightened out. Since it's not, then I am
not. Having said that, I don't exclude that hard thinking on the
"transformation" of violins into orchestra by those who continue that
debate could lead to some insight with actual applications helpful to
me and others. That's why I tend to respect the work of people who
keep grappling with that topic.
Finally, I agree that I.I. Rubin's 1920's work (http://www.marxists.org/archive/rubin/value/index.htm) is stupendous.