xhxhxhx:

You don’t need taxes to run a polity, @argumate​. Chris Wickham:

In an ideal-type tax-based state, where wealth is taken from (nearly) everyone, the fiscal system provides an independent basis for political power, separate from the goodwill of the aristocracy, for the army is paid directly from public coffers, and complex bureaucracies, themselves usually salaried, handle the process of tax collection (as well as other aspects of administration and law, which can also, as a result, in principle operate separately from aristocratic interest). 

This separation of the state from the aristocracy is seldom complete, for the aristocracy tend to dominate the fiscal and military administration too, if for no other reason than that anything to do with taxation is such a reliable source of enrichment, legal or illegal; public office may also be, in fact usually is, an important element of aristocratic status itself. But the complexity of the state is such that there are many levels of mediation between the interest of the ruler and the interests of aristocrats; and the wealth of the state is so great that it will keep aristocratic loyalty and commitment for a long time. 

Tax-raising rulers have one crucial advantage over their dependants, too: in the case of unreliability, whether through disloyalty, corruption, or simple ineptness, they can simply dismiss them, and stop paying their salaries. This process works, however badly the other checks and balances work, in any tax-based system. Subjects have only one practical recourse in return: the replacement of the ruler, by rebellion or coup. 

Regional autonomies, in particular, are hard to create, unless the state structures themselves can be regionalized, because any ambitious regional leader would regard separating himself from tax-raising powers as a pointless exercise. In practice, in fact, rulers were overthrown, and provinces did break off, in the history of any tax-raising system, from the Roman empire to the ‘Abbasid caliphate and beyond. But the state machinery continued to be central, even as its rulers were removed or defied.

Contrast an ideal-type land-based (or rent-based) state: here, the bulk of the wealth of a ruler is derived, not from a whole population, but only from the rent-paying inhabitants on the land he (very rarely she) directly controls, and that wealth is also the major support for all political aggregation. The administration is simpler, for the tax system is absent or rudimentary; the ruler’s principal officials are his local representatives and his army leaders, and they too are based on the land, as indeed is the whole of the army. All political reward is dominated by the ‘politics of land’—cessions of land and its rents, to officials or to other powerful aristocrats, in return for loyalty. 

Rulers have two basic problems here. The first is that land is finite, except in  periods of political expansion: essentially, the more I have, the less you do. Rulers may achieve loyalty from one round of land-gifts, but they have less to give as a result, and may become less attractive over time; furthermore, land, once given, on whatever legal terms, is hard to get back, except by force. Rulers were often very good at force—societies of this type tend to be highly militarized, with a good deal of respect paid to ruthlessness; but still, in the long run, in the absence of wars of conquest, or the sort of civil war that is won so totally by one side that the victor can confiscate on a large scale and start again, polities risk becoming weaker. 

The second problem is regional fragmentation: in the absence of capillary administrative controls,and unless the centre is particularly powerful and effective, regional officials have little preventing them from increasing their autonomy, and maybe in the end breaking free altogether. This can again be countered by force, and often has been, across the centuries. Even more often, it has been countered by the generation of a political culture, a set of assumptions about legitimate political action and how to characterize and symbolize it, which favour central rather than regional power—whether this is the social position conveyed by being a king’s guest, at his banquets or his Easter celebrations, or the ideological importance of unconditional political loyalty, or the greater status attached to being a dealer in central-government politics rather than being a regional leader—political systems have differed about which of these, and others, mattered more, but all polities have had to deal in one or more of them if they were to avoid total failure. Still, the standard risks in a land-based system are, other things being equal, structural weakness and centrifugal tendencies.

That’s from Wickham’s Framing the Early Middle Ages (Oxford, 2005).

Wickham contrasts the tax-based regimes of the Mediterranean core with the land-based regimes of the European periphery. Byzantines and Arabs raised taxes. Franks and Lombards did not. 

Land-based states are not always weaker than tax-based states. But they tend to be. When they were not, they tend to fragment. The Franks were stronger than Byzantium in 800, but the Franks had fragmented by 900. The Byzantines hadn’t.

I’m not actually sure that this wouldn’t be an attractive solution for @sadoeconomist, who wants a weak and fragmentary state.