An alternative to the Poly & Bournazel approach of smothering you with moralising anecdote, in order to steep you in the culture of the feudal world, is provided in my current reading by Josep María Salrach i Marès, who seems to be taking up most of my time with the printed word at the moment. In particular, in the Història Política, Societat i Cultura dels Paisos Catalans volume that he edited, which I currently have on Inter-Library Loan (bless the U. L.), he has a really useful systematic breakdown of the ideas involved in the historiography of the feudal transformation. I thought it was worth setting out here.
He distinguishes the two social systems, ancient and feudal. The former has slave-based production, a state-driven economy with redistribution by the state to its officers, and public control of justice and military power. There are other factors but that’s what he picks on especially.
The feudal system, by contrast, has production dominated by notionally-free peasants, albeit largely dependant and tied to their lands; an economy that they essentially control, because the only higher control on is the demand of their lords for surplus—increase in the demand forces them to grow more so that they eat, otherwise they’d just, as Wickham puts it somewhere in Land and Power, eat more and work less—local potentates whose status derives entirely from their own wealth and influence, not the state’s issue of a title; and locally-administered, often arbitrary, control of justice and warfare.
Having posited these two systems, he asks what would enable a society to pass from the one state to the other, and of course the answer is, the state losing its grip. This is a lot more of an answer for Catalonia than elsewhere of course, where it is possible to argue that the local rulers were left by themselves by a receding Carolingian state.
It works less well in areas where the local lords took up arms against the kings, of course, thinking Robert of Neustria or similar, and it relies very heavily on how you view the ancient state—in a lot of ways it possibly doesn’t work at all, so schematised is it—but it’s so beautifully coherent compared to Poly and Bournazel’s ramble that I love it anyway.