> This may be true, but doesn't it also work the opposite way? I mean,
> the most complex thing in the whole book is the commodity, which Marx
> puts at the beginning and is hardly a "simple starting point." And
> doesn't the building of relationships come in fits and starts, moving
> backward and forward, from complex to simple? For instance, the last
> chapter of the book goes back to the beginning of the story, how
> capital was created.
Yeah, you're right. I think good theory does go both ways.
> Again, I don't think this describes good empiricism. Effects don't
> reveal some hidden structure. They are the structure (or whatever you
> want to call it), and empiricism investigates them.
Maybe an example will clarify what I mean. The basic microeconomic idea of a typical demand curve is that the demand for a good falls when its price rises, all other things being equal, because people substitute other goods, can afford less, etc. We can't (very often) directly verify this with reference to actual price and sales data, because 'all other things' hardly ever remain equal. Only single points on the demand curve are actualised at any given point in time, and they only have 'effects' as part of a conjuncture with other factors. Yet I would say there is something in reality that corresponds to a demand curve at any point in time, even if it's not as smooth as our abstract representation, and even though it is impossible to collect direct evidence about it - it has its reality in the state of preferences, incomes etc.
As Shag says, my thinking on this comes from Bhaskar. I don't really have much background in philosophy and I understand Bhaskar induces eye-rolling in some quarters. But I've found him useful anyway.
Cheers, Mike