I want to begin with an observation, obvious but frequently ignored: states are complex things. The apparatus by which a state gathers revenue, raises armies (with that revenue), administers justice and tries to organize society – that apparatus requires people. Not just any people: they need to be people of the educated, literate sort to be able to record taxes, read the laws and transmit (written) royal orders and decrees.

(Note: for a more detailed primer on what this kind of apparatus can look like, check out Wayne Lee’s (@MilHist_Lee) talk “Reaping the Rewards: How the Governor, the Priest, the Taxman, and the Garrison Secure Victory in World History” here. He’s got some specific points he’s driving at, but the first half of the talk is a broad overview of the problems you face as a suddenly successful king. Also, the whole thing is fascinating.)

In a pre-modern society, this task – assembling and organizing the literate bureaucrats you need to run a state – is very difficult. Literacy is often very low, so the number of individuals with the necessary skills is minuscule. Training new literate bureaucrats is expensive, as is paying the ones you have, creating a catch-22 where the king has no money because he has no tax collectors and he has no tax collectors because he has no money. Looking at how states form is thus often a question of looking at how this low-administration equilibrium is broken. The administrators you need might be found in civic elites who are persuaded to do the job in exchange for power, or in a co-opted religious hierarchy of educated priests, for instance.

Vassalage represents another response to the problem, which is the attempt to – as much as possible – do without. Let’s specify terms: I am using “vassalage” here because it is specific in a way that the more commonly used “feudalism” is not. I am not (yet) referring to how peasants (in Westeros the “smallfolk”) interact with lords (which is better termed “manorialism” than as part of feudalism anyway), but rather how military aristocrats (knights, lords, etc) interact with each other.

So let us say you are a king who has suddenly come into a lot of land, probably by bloody conquest. You need to extract revenue from that land in order to pay for the armies you used to conquer it, but you don’t have a pile of literate bureaucrats to collect those taxes and no easy way to get some. By handing out that land to your military retainers as fiefs (they become your vassals), you can solve a bunch of problems at once. First, you pay off your military retainers for their service with something you have that is valuable (land). Second, by extracting certain promises (called “homage”) from them, you ensure that they will continue to fight for you. And third, you are partitioning your land into smaller and smaller chunks until you get them in chunks small enough to be administered directly, with only a very, very minimal bureaucratic apparatus. Your new vassals, of course, may do the same with their new land, further fragmenting the political system.

This is the system in Westeros, albeit after generations of inheritance (such that families, rather than individuals, serve as the chief political unit). The Westerosi term for a vassal is a “bannerman”. Greater military aristocrats with larger holding are lords, while lesser ones are landed knights. Landed knights often hold significant lands and a keep (fortified manner house), which would make them something more akin to European castellans or barons than, say, a 14th century English Knight Banneret (who is unlikely to have been given permission to fortify his home, known as a license to crenellate). What is missing from this system are the vast majority of knights, who would not have had any kind of fortified dwelling or castle, but would have instead been maintained as part of the household of some more senior member of the aristocracy. A handful of landless knights show up in Game of Thrones, but they should be by far the majority and make up most of the armies.

There’s one final missing ingredient here, which is castles, something Westeros has in abundance. Castles – in the absence of castle-breaking cannon – shift power downward in this system, because they allow vassals to effectively resist their lieges. That may not manifest in open rebellion so much as a refusal to go on campaign or supply troops. This is important, because it makes lieges as dependent on their vassals as vassals are on their lieges.

Bret Devereaux, “New Acquisitions: How It Wasn’t: Game of Thrones and the Middle Ages, Part III”, A Collection of Unmitigated Pedantry, 2019-06-12.