This is the second part of a four-part (I, IIa, IIb, III, IV) series examining the historical assumptions of the popular historical grand strategy game Crusader Kings III, by Paradox Interactive. Last time we opened by discussing how CKIII attempts to simulate and represent the distinctly personal character of rule and decision-making in the Middle Ages and how this contrasted with both Paradox’s other grand strategy titles, which are far more state-oriented.
This week we’re going to turn from the person of the ruler to the structure of polity that CKIII envisages. Fundamentally, Crusader Kings III is a game about political fragmentation and decentralization, in a way that few other strategy games are. As we noted last week, most of Paradox’s titles adopt a broadly neo-realist framework when thinking about international relations, which tends (in its most common forms) to treat states as largely unitary – a single entity with a single set of interests. But there are no states as such in CKIII. The polities that do exist are highly fragmented, with power shared between a liege and many vassals who essentially function as states-within-a-state, who may contribute revenue and levies to their liege but may also oppose them and actively work against the liege’s goals and strategies.
This post ended up being very long so I have split it into two parts. This week’s part will look at how the game models ‘feudal’ vassalage in Christian Europe, while next week’s second half will look at how the game modifies those systems to try and model different forms of political fragmentation in Islamic polities, particularly in the Eastern Mediterranean.
But first, if you want to contribute levies to this project, please share what you are reading. If you have a really high opinion modifier of these posts, you can also contribute revenues via Patreon. And if you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) where I hold court.
One more thing to note before we dive in. The same as last week there is going to be a lot of ‘when we get to it’ notes as we go through here. CKIII‘s core mechanics (characters, vassal-liege relations, opinion-bonuses, religion and culture) are all fairly tightly intertwined (because they all impact character’s opinions of other characters), so teasing them apart for the purpose of analysis is at best a bit imperfect. The next post in the series will be dealing with norms (both cultural and religious) along with the broader concept of legitimacy, so all of those issues will be returned to in more detail then.
Ironically, partible inheritance is actually going to be useful here.
Fragmenting Charlemagne
We should start by clarifying some terms. In its current for, CKIII distinguishes between three government types: feudal governments (mostly the Christian polities of Europe as well as the polities of south Asia), clan governments (mostly but not entirely Islamic polities) and tribal governments (mostly Steppe and Baltic peoples with decidedly non-state polities, and also some, but not all, of the West African polities).1 Veteran readers of the blog will already know that I generally avoid the terms ‘feudal’ and ‘feudalism,’ as do many medievalists these days. The problem is that ‘feudalism’ comes to conflate two very different systems: manorialism, the economic system which structured the relations between peasant farmers (mostly serfs) and their elite landlords, and vassalage, the political system which structured relations between elites. But these systems and the relationships they structured were very different and their conflation together as a single system is a modern, not medieval, construction. Consequently I will talk about vassalage and manorialism as two distinct systems which often (but not always) co-existed, rather than a single system called ‘feudalism.’ This post is almost entirely on the vassalage half of this formulation; we’ll discuss the game’s treatment of manorialism in the final part of this series.
And indeed, the in-game definition of feudalism engages in that conflation, declaring that “the feudal system is based on the relationship between the landholder and those who provide services and labor in order to live there,” which treats together the provision of farm service and military service which we ought probably to think about separately. Nevertheless it is hard to fault a game made for the public for using a term that is almost exclusively the one used by the general public and still in use by some scholars. And the definition being used here doesn’t come out of nowhere, rather it is the definition of feudalism (which includes both vasslage and manoralism) advanced by Marc Bloch in La Société Féodale (1939; known in English as Feudal Society), one of those magisterial foundational classics of scholarship. So while these aren’t the terms I’d use in a class, it makes sense to use them in the context of a game and they are well-rooted in scholarship (albeit also well contested in the more recent scholarship).
The great challenge of any game set in the European Middle Ages (or a fantasy variant thereof) is to represent a political system whereby ‘state capacity’ was very limited and as a result political power was extremely fragmented.
While this system (vassalage, in particular) is often associated in public thinking broadly with the whole of the Middle Ages in Europe, it only really emerged in the 9th century (the Middle Ages are generally taken to have begun in the 5th century) and arrived in different places at different times. Fundamentally, the system of vassalage we see being represented in CKIII was a product of the Carolingian Empire trying to make do with extremely limited administration.
Let’s back up for a moment. As we’ve discussed elsewhere, the Late Roman Empire (particularly after Diocletian) developed a large, complex centralized bureaucracy quite different from the generally fairly ‘light’ system of Roman governance in the Early Empire. So when the Roman Empire collapsed in the West, this system – with its complex bureaucracies and officials appointed from among the landholding magnate class to offices they held at the pleasure of the emperor or king – this was the system that the ‘barbarian’ successor-states inherited. But the vibrant urban culture which produced the literate bureaucrats required to make that system work had largely faded with the empire, leaving the successor-states with a bureaucratic top-down model of governance with few bureaucrats to run it. In the immediate aftermath, they turned to the large landholders – many of them the old Roman-era aristocrats – to try to ‘farm out’ the job of administration. Each kingdom’s system was subtly different but broadly these kings hoovered up wealth and used it through a ‘gift economy’ to get their elites to do the job of administering, in some minimal way, the territories the king notionally controlled. Nevertheless these posts are still appointed posts, so a big landholder who wants these sorts of jobs needs the king’s favor to get them (and the royal ‘gifts’ that come with faithful service).
That is already a system where power is heavily fragmented, but there’s a final step necessary which comes with the Carolingians. Usurping the Merovingian kings of the Franks, the Carolingians over three generations (Charles Martel, his son Pepin and then Pepin’s sons Carloman and especially Charles (Charlemagne)) assembled a vast empire covering what is today France, Germany, Northern Italy, Austria, the Czech Republic, Switzerland, Slovenia, Croatia and parts of northern Spain. At no point in those vast conquests did they conquer a state which had developed a functional political bureaucracy, while the separation of secular authority (Charlemagne) and religious authority (the Pope) meant that the well developed religious bureaucracy was not generally available for large-scale administration.
In order to administer those conquests – flung widely enough that even the itinerant court could not visit them all – the Carolingians developed (and I am greatly simplifying this system) a system of royal officials drawn from the landed aristocracy and occasionally the church. The empire was divided into military-administrative districts, each of which got a leader called a comes (pronounced ‘co-MAce,’ plural comites) a Latin word meaning ‘companion’ in the sense of a military retainer of a king (like the Late Roman army which moved with the emperor, the comitatenses), which in French was comte which gives us our word ‘count’ and ‘county.’2 These comes were appointed officials who were responsible for overseeing tax collection, recruitment, and justice in their districts as well as being local military commanders in a pinch. Notionally they were then overseen by roving officials called the Missi Dominici (‘Ruler’s envoys’) to ensure they were doing their job well and effectively. Which, at least in principle serves as an extension of the appointment system the Merovingians had been using prior.
After Charlemagne, however, the strength of central Carolingian rule begins to wane. Charlemagne’s son, Louis the Pious was frequently at war with his own ambitious sons,3 who after his death warred with each other. At a deeper level, Charlemagne’s administrative system was probably unworkable from the beginning; the division of authority between the comes and the missi was never clear and the steady shift in military emphasis from an infantry-focused army of militia-farmers to an army focused on aristocratic cavalry shifted power towards the aristocrats and away from central government.
In either case, what we see in the 9th century is that the comites are increasingly able to secure their positions to their sons, leading to the positions to become hereditary, with the temporary administration of lands becoming a hereditary right or even ownership of those lands. Of course once those nobles (and we can start calling them that at this point) no longer need to rely on royal appointment to legitimate their control over their chunk of the kingdom, the king’s position relative to his nobles – now properly his vassals – grew weaker. At the same time, especially in the context of military strains both internal (civil war) and external (VIKINGS!) on the Carolingian Empire, this system – while it had negative long-term consequences for royal power – was very effective at raising the kind of military power that was needed in the moment, which was increasingly elites on horses (though a clear socially defined ‘knighthood’ is still many years away from forming).
That system in turn spread out unevenly from wreckage of the Carolingian Empire (which split into three after the death of Louis the Pious). In Germany, the Eastern chunk of the Carolingian Empire (East Francia) eventually became the Holy Roman Empire and so spread those institutions even as the Carolingian line there failed in the early 10th century, replaced by a series of dynasties with a mix of hereditary and elective traditions. Across the channel, meanwhile, the English kingdom at this time was actually quite differently organized and more effectively centralized; ‘feudalism’ only really arrives in England with William the Conqueror (of Normandy), who after seizing the English throne in 1066 imposed a more ‘French’-style system of governance.4 And of course that same sort of system would arrive in the Levant through the creation of the Crusader States in the 1090s.
Expansion here ends up being a mix of conquest to remove independent kingdoms and diplomacy, offering protected status to independent lesser nobles of different faiths in order to pull them under my rule, which of course fits with the goal of creating a multi-religious state.
And it has to be noted that CKIII is somewhat ‘fudging’ the extent of this system at both of its start dates (867 and 1066, but especially the former). Interesting at this year’s PDXCON the CKIII team was asked at their Q&A if they would consider earlier start-dates and they responded no, on the grounds that they were already ‘stretching it’ to push feudalism (their word) so early, noting that many of the recommended starting lords in the 867 start were in fact the first nobles to make their positions hereditary rather than appointed. We’ll come back to this later but part of the struggle here is that CKIII doesn’t have good systems for centralized governments or for having centralized governments de-consolidate, which makes the ‘fudge’ of having feudalism in England a bit earlier than it should and the refusal to push an earlier start date than 867 make a lot of sense, even if the developers are fairly open that these are necessary ‘fudges’ for gameplay (much the way European states are a lot more unitary in the first decades of EUIV than they ought to be).
Vasslage as a Game System
The consequence of this progressive fragmentation as the rule of comites – by now we can really call them ‘counts’ (and other similar nobles) – became hereditary was that the power of kings was diminished. Kings increasingly only exerted direct rule over their personal lands; attempting to interfere in a vassal’s administration of their lands – which was, you will recall, the point of Charlemagne’s missi – could be seen as an infringement on the traditional rights of those vassals. And the ‘royal lands’ could be very small indeed, because kings were expected to dole out lands as gifts for military service and also because partible inheritance repeatedly fragmented the royal domains, especially in France.
In theory the king still held all of the land and was merely granting it out to his vassals (thus becoming the vassal’s liege) in exchange for their military and political support but in practice the various counts (and by now dukes (from the Latin, dux, ‘leader’) and other nobles) controlled their territories directly, with the king as first among equals expected to resolve disputes between vassals and to direct the kingdom’s military affairs beyond its borders. The land the vassal holds is a fief, the process of it being granted (or its grant being recognized when it was inherited) is called enfeoffment, all of which are generally thought to derive from the Latin foedum, meaning a treaty or solemn agreement. The vassal, receiving a title either in fact (because it is the initial grant of land) or symbolically (because he has already inherited the land and the king is merely recognizing this) did so with a ritual oath of either homage or fealty (with homage being the notionally stronger form; one could swear fealty to multiple lieges, but homage in theory only to one).
Now as you can see on the map above, many of the vassals of the King of France themselves controlled vast territories, the vastness of which replicated the exact same problem that the Carolingians had confronted: how do administer a large territory when your society lacks the bureaucratic apparatus to do so directly. And so they replicated the system downward, with higher nobles (like dukes and counts) offering fiefs to lower nobles (barons, earls, and castellans; there’s a wide array of titles for these lower nobles) under the same terms: take this land in exchange for owing me military support which you will raise and supply form the revenues the land (and its peasants) generate for you. It should be noted that the peasantry is not part of this system; a peasant is not a vassal. That’s why we generally separate vassalage as a system from manoralism, the rules are different. That said, there could be non-noble vassals, most often towns that were either under the governance of a bishop (a survival or replication of Roman-era church administration) or were self-governing.
The final tricky task here was getting into the right ‘struggle phase’ because everyone needs to be in a conciliatory mood for this to work too. This whole process is a lot more complicated than had I simply aimed to end the struggle by raw conquest, but that’s part of the fun.
So how does CKIII simulate this complex system?
Regardless of government type (tribal, clan, feudal), every polity is composed of an independent lord (often, but not always, a king or emperor), and their collection of vassals and the vassals of their vassals and so on. Each noble in the chain raises two kinds of resources from their holding, taxes and levies. Each vassal then sends a percentage of those taxes and levies upward one level in the chain to their de facto liege (not their de jure liege if the two are different); for feudal vassals, the percentages are fixed by the formal obligations in their contract, inherited from one generation to the next, rather than subject to vassal opinion, though contributions are reduced if the vassal’s de facto liege is not also their de jure liege (it’s hard to insist on someone fully carrying out their ancestral obligations to you if they don’t, technically, have any ancestral obligations).
I’ve mentioned this before but to reiterate, the tiers of titles are baron (of a single holding), count, duke, king and emperor in order of lowest to highest. A ruler is considered to have the status of the tier of their highest title. Vassals must be a lower tier than lieges, no vassal may have more than one liege (we’ll come back to that) making the system a fairly strict pyramid. Moreover, vassal contracts are unique to them (though inherited by their heirs) and may specify non-standard levy or tax obligations, lock in certain inheritance rules, provide special privileges and so on, making each liege-vassal relationship effectively unique.
There is a necessary bit of ahistorical standardization in this model of clearly delineated titles with clear ranks; in actual historical practice titles were not nearly so standardized – indeed, much of the standardization of ‘standard’ systems of titles happened later as the titles were increasingly detached from actual direct governance of territory. By way of example, the game titles Matilda of Tuscany as a Duchess, which fairly reflects her territory and position in the empire, but not her actual historical title as la Gran Contessa (‘the great countess’). But Tuscany was not a county either, nor yet a duchy. Tuscany’s status as a ‘Grand Duchy’ is a later phenomenon of the 16th century. Instead to the best I can tell it was regarded as a Margraviate (from the German Markgraf or ‘March-Lord,’ with ‘march’ here meaning ‘a boarder military district’) in this period. The game works in some recognition of these complexities with culture-specific variations on titles, so lords with ducal-level titles might go by a wide range of names: Petty-king (various feudal), Grand Prince (Russian), High Chieftain (various tribal), Mahasamanta (Indian), Atabeg (Oghuz Turkic), Emir (various Muslim), Doux (Greek) and so on. I think that, combined with the gameplay impacts of religion and culture (next time!) serve to alert the player to at least some degree that there’s a lot of variety here being compressed for gameplay’s sake.
A core part of this system in turn are factions, which reflect informal coalitions of vassals within a single polity trying to achieve political objectives. At present, vassals have four factions to choose from: independence (aiming to break away from the kingdom), dissolution (aiming to dissolve the kingdom entirely), liberty (aiming to lower ‘crown authority’ – essentially pushing the rights of vassals over the prerogatives of kings) and claimant (pushing for a different ruler). A faction with enough military power between all of its vassals will make demands and if refused, rise in revolt to enforce those demands. Keeping vassals out of factions thus becomes crucial to maintaining the coherence of large realms (we’ll get into next week some of those strategies), but the upshot here is that even though vassal opinion doesn’t modify levies or taxes for feudal lords, its role in the faction system means it cannot be ignored. A liege must keep most of their vassals happy.
Because vassals only provide a percentage of their military strength to their liege but will field all of their strength in the event they rebel (typically through forming a faction whose demands are refused), large realms can be dangerously unstable unless the ruler can maintain a large personal territory (the demesne) from which they can draw the backbone of their own army independent of their vassals; doing so under partible inheritance is particularly hard (as you can see concerns about keeping the royal possessions together dominating a lot of my House al-Yiliq strategy). Consequently to maintain stability, rulers have to keep their vassals happy (through mechanisms we’ll be exploring mostly next week. Meanwhile vassal holdings are, as noted last week, personal, meaning they may bear relatively little relation to the de jure internal divisions of a kingdom. ‘Double-dukes’ (ducal-level vassals controlled two duchies) are particularly common and such large and powerful vassals become major movers in internal politics because of the tremendous amount of power they can deliver to a faction if they join it.
In addition, rulers need to stock a series of administrative and military positions. For king and emperor-level rulers, those are the five members of their council (one each for the game’s major attributes), plus a selection of ‘knights’ (here effectively unit and army commanders, whereas household knights are reflected in men-at-arms regiments) and typically at least a few court positions. Some of those positions can be filled with landless courtiers, but generally the major vassals expect these jobs, will be angry if they don’t get some of them and in any event reflect the largest pool of talent for them. Consequently in addition to handling the extraction of revenues locally, vassals are also the largest source of the king’s subordinates in the very limited central government.
Finally, there is ‘crown authority,’ which reflects in an abstract way the level of centralization a kingdom has. Lower levels of crown authority make vassals happier and higher levels make them actively unhappy (which will funnel them into the ‘liberty’ faction to lower crown authority). At the same time, higher crown authority raises substantially the levies and money vassals owe to their liege (improving the king’s power at the expense of vassals) and provides more powerful options for shaping who-owns-what, like title revocation and the ability at maximum crown authority to designate an heir in defiance of inheritance norms. Just about every polity starts with very low crown authority; over time rulers are going to try to increase that while vassals will attempt to keep it low.
As a simulation of vassalage-based political organization, the system is broadly effective. Most games – and even much fantasy long-form fiction – continues to imagine medieval polities as unified states or at most as divided into only a handful of constituent parts which are not themselves then subdivided further. CKIII is one of the few games that leans into the complexity of actual political systems; having dozens of direct vassals and hundreds of indirect vassals is not uncommon for large realms. To take one example, Emperor Henry IV (in game as Kaiser Heinrich IV; I love the culture-specific titles and names in the game, by the by), begins the game in the 1066 start with sixty-one direct vassals, including more than a dozen dukes with their own large collections of vassals (as many as fifteen direct vassals for some). When it comes to scheming and politics, all of these figures matter, though some matter more than others (historically minded players beginning at this start will especially note the powerful Duchess of Tuscany, Matilda and realize the trouble she could cause, as indeed historically she did).
Adding to this complexity is that while the king may be a feudal lord, that does not mean all of his vassals are. Indeed, ‘republic’-type vassals, reflecting the civic governments of towns, will be a factor in almost every polity (albeit these aren’t as extensively developed in-game yet) and theocratic vassals – reflecting territories controlled by bishops or monasteries – will be a factor in most. And it is entirely possible for a feudal polity to end up with some vassals whose government is clan or tribal in nature. That complexity is actually a fairly good representation of the often very complex polities this system created, where each liege-vassal relationship was a unique and bespoke bilateral relationship.
Moreover, the system of ‘crown authority’ provides a framework – if abstract – for two related tensions that defined a lot of the politics in the period. For vassals that are the de jure subjects of their liege, the normal channel of discontent is towards factions to lower crown authority (which the liege is going to keep trying to raise), creating the cycle of tensions between central authority and vassals which defined much of the politics in the period. Rather than attempting to simulate every small dispute (which might get properly difficult from both a gameplay and a game development standpoint), the abstraction of crown authority and struggles to decrease it can cover a wide range of such conflicts like the Baron’s Wars in England, or the wars related to the Investiture Controversy (which had as much to do with the powers the King could wield as it did the religious position of bishops).
At the same time as the game advances, for most of these polities the opportunities for power-centralization will grow, leading to a slow, halting march towards progressively more centralized polities. Crown authority levels are gated by innovations which are gated by years (more on this later) and in general the progression through innovations, realm development (building new holdings or expanding old ones) and so on will tend over time give rulers more of a margin on the opinion and power of their vassals, which they will naturally ‘consume’ by raising crown authority. That creates a noticeable ebb-and-flow, since poor rulers or unlucky succession (especially where the heir is a minor) will create windows for vassals to push crown authority back down (by force, if necessary), but over time more effectively centralized realms will emerge. And because of the levy benefits of those more centralized regimes, they will tend to be stronger than their neighbors and thus expand although expansion increases the number of vassals as compared to the largely static royal demesne, increasing the risk of factions and potentially re-fracturing the kingdom.
We’ll come back to this point in the last part of this series but I think this system works better for some polities than for others. It does a good job of reflecting, for instance, the steady centralization of power within the Kingdom of France under the Capetians; France went from a badly fragmented, highly decentralized kingdom where the king’s powers were extremely limited in the 10th century to a substantially more centralized and effective kingdom by the 14th century and that process of very early stage state-formation is fairly well reflected in the systems here. But the steady pressure towards state centralization doesn’t seem quite as effective in modeling the Holy Roman Empire which spent much of this period (especially following 1200 or so) making the same journey in the opposite direction, steadily becoming more fragmented and less centralized. The culprit here is the innovation system, which pushed many of the game’s systems towards a fairly teleological chronology where innovations must come to all cultures in a fairly set sequence. Put a pin in that, we’ll be back to it in a couple of weeks.
Finally there are some elements where complexity was clearly sacrificed for playability. The most obvious of these is that each vassal can have only one liege. By contrast actual medieval vassals often had multiple lieges, with different titles owing allegiance to different notional lords. The distinction was thus sometimes made historically between homage owed to one’s primary liege and fealty which might be offered to several lords, though the terms and nature of these relationships were not at all standardized. While CKIII opts to stress the personal nature of rule and thus divides polities into rulers rather than titles, the medievals themselves often thought of these questions in terms of titles rather than rulers. Thus as the Duke of Normandy, William of Normandy was the vassal of the King of France, but as the King of England, William the Conqueror (same fellow) was an independent ruler. Perhaps the most dramatic example of this is what is sometimes termed the ‘Angevin Empire‘ – the vast territories in France of the Plantagenet kings of England. The contrast between de jure and de facto does provide some simulation of the tensions this creates, since the king of England holding de facto territories which de jure should be under the king of France both creates severe opinion penalties between the two monarchs and also an excuse for war between them. That’s not a terrible representation of the Capetian-Plantagenet conflicts, but also not a perfect one.
Nevertheless as a system for modeling decentralized rule through a system of vassalage as practiced in Western Europe during the high and late Middle Ages, I think Crusader Kings III does better than effectively any other game I have played. The commitment to developing the system in something approaching its full complexity (albeit with some of the sharp edges sanded down a bit) brings home to the player quite effectively just how decentralized these states were. The first time a player realizes they have a troublesome vassal – perhaps an ambitious double-duke – and that they cannot actually get rid of them despite being king (or even emperor!) is a fantastic learning opportunity for understanding the nature of personal rule in a vassalage-based system. Likewise, players working as vassals will find they resort to many of the same inheritance and marriage strategies (backed up by military force) which assembled things like the Angevin Empire in the first place. Quite accurately for the period, the balance of mortality and male-preference succession in much of Christian Europe means that there are bound to be several heiresses or brother-to-brother successions in each generation of aristocrats in any large kingdom, each providing the opportunity for vassals to expand within the kingdom in a way that, again, their king may be powerless to stop.
On to Part IIb!
So while there is certainly some flattening of complexity in CKIII‘s system for representing fragmented European polities – particularly in how the complexity of different titles and vassals with multiple lieges is condensed into a fairly simple pyramid of obligations – overall this system is quite good. In its terminology and implementation it reflects fairly well the older Marc Bloch vision of feudalism as a social system. One thing we haven’t covered here yet is how the game translates this into practical actions the player takes in order to maintain relations with their vassals (that’s in Part III), but this system provides a good foundation for it.
That said when it came to modeling the Christian polities of the period, CKIII had a fairly good starting position; CKII‘s mechanics already did this fairly well. There have certainly been some modifications here in how crown authority works, the way factions function and the role of the council (some of which is implementing as core features things which in CKII were in DLCs), but the CKII bones set the mechanical foundation here pretty clearly.
Next week we’re going to come to the second half of this topic, however, where we look at how well the game handles Muslim polities. CKII‘s system for this was, a bit of a mess, particularly the decadence system, but also innate modifiers like hostility between brothers and bespoke mechanics packed into the ‘Iqta’ government form (tax changes as well as greater latitude to revoke ducal titles) which made Islamic polities more effectively centralized by nature rather than as a consequence of their structure or development (which in turn tended to mean they didn’t fragment to the degree that their historical parallels did). I think it’s fair to say that this was one area of significant player dissatisfaction in the design of CKII (in part because of the feeling that these mechanics were ‘gamey’ and thus ahistorical), so does CKIII do any better on this score?
- West Africa is not my area of expertise, but what I do know about the polities there would make me hesitant to designate many of them as ‘tribal,’ though the game largely reserves the feudal government type there for the most centralized West African states like the Ghana Empire. One hopes a more detailed system with bespoke West African mechanics is on the CKIII DLC to-do list.
- Remember when I said the base political unit of Crusader Kings III was counts?
- The issue being that the birth if his fourth son, Charles (to be Charles the Bald) and Louis’ efforts to provide an inheritance for Charles upset the other three sons (Lothair, Pepin and Louis) leading to conflict
- Once again, we are sliding over a lot of complexity here.
A correction – Atabeg is actually a title exclusive to the Oghuz Turkic culture, not a Indo-Aryan title.
Thanks, will fix!
Hmmm, are there any inheritance strategies that Kings had to keep their Domains together? I know that in game one early-game strategy to consistently keep counties over generations is to set feudal electives for duchies (while owning most or all counties in such a duchy) to force all titles in such a duchy to go to a preferred heir. It is very “gamey,” but looking at later Capetian Dynasty monarchs even after they were having multiple sons, they were remarkable in their ability to keep extending the royal desmene. Does the above strategy have any basis in history?
A fairly common strategy in France and Germany was to appoint your favoured heir as co-king during your own lifetime. This was a way of controlling elective monarchies (ensuring that the title didn’t pass outside your line on your death) but also helped solidify patrominies under your heir. It helped, though, if you had one clear heir, and not multiple ones who (under Frankish customary law) would demand a slice of the pie. A strong king could still carve up his assets with some certitude, but it risked war. William the Conqueror, for instance, set things up so that Normandy would go to his eldest son Robert, England to his second son William, and his youngest son Henry would get a big pile of cash. But on the Conqueror’s death, Robert and William inevitably went to war over each others’ territories, eventually William was assassinated, Henry grabbed William’s territories and managed to conquer Robert’s completely.
Henry II of England did the French thing of making his eldest son Henry king of England within his own lifetime, apparently with the intention that his second son Richard would get Aquitaine, third son Geoffrey would have Brittany (made duke within Henry’s lifetime) and John would get Ireland. But none of them was entirely happy with this and this prompted a massive rebellion aided and abetted by the king of France. In the end, thanks to Henry and Geoffrey selfishly dying, and Richard having no heir of his own, John copped the lot in the end (and then promptly lost most of it).
It also helped if you had heirs at all. The Capetians got the French throne in 987 and it descended in a direct line father to son to 1316 (and didn’t leave the dynasty until the monarchy was abolished in 1792). Germany was never able to conjur a dynasty capable of replicating the feat, so despite arguably holding better together than France in the initial aftermath of the Carolingian disintegration, it never pulled itself back together as the French monarchy was able to do starting in the 12th century: the Ottonians and Salians died out after three or four generations, and the Hohenstaufens were deliberately targeted for extermination by the Papacy.
You talk about the council system – I haven’t played 3 but one of my favorite things in 2 was how you wanted to pick councilmembers for three characteristics: being prominent enough to deserve a spot on the council, being skilled enough to do the actual job, and being loyal to you personally. But, of course, you never got all three and indeed are really lucky to get two of them for each slot. It really makes it an interesting mechanic and also makes medieval monarchs leaning on “favorites” suddenly parseable: the monarchs are prioritizing the latter two characteristics over the first one, with predictable results.
The council works pretty much exactly the same way in CK3 as it does in CK2, right down to powerful vassals having an opinion malus with you if they’re not on the council.
Thank god for the disinheritance option! I keep my kingdoms centralized by disinheriting all but the best and the brightest. Cost a *lot* of renown, but worth it. Not very historical, of course – I doubt that the disinherited first son of a noble would just say “it’s a fair cop” and let their claims fall.
If it’s really historical, that’s a good way to end up with no heir. Of Æthelwulf’s five legitimate sons, four succeeded him to the throne before the youngest finally had an heir. Henry II had four legitimate sons and the youngest son still ended up inheriting (though only after fighting with his nephew). Etc.
As Dr Deveraux points out though, the Saxons had a different system of inheritance than the Carolingian aftermath, and so didn’t partition their kingdom wildly in a way that would require effort in reconsolidating the realm.
The deaths are still a danger
Yeah. I’m pretty sure in practice that usually leads to the disinherited heir rallying a faction of nobles around them as a rival claimant, by the expedient of promising the nobles assorted concessions.
Well, that or the disinherited heir coming back as the figurehead of a foreign army planning to install them as the new puppet ruler.
Also, as Mary points out, in practice it’s a really bad idea to rely on “needing” only one heir; in medieval times people often died suddenly or unpredictably.
I think I should point out that all three of those scenarios are in fact potential consequences of the disinheritance strategy. Disinherited heirs are removed from the line of succession, but they keep their claims, and therefore they can become the nominal leader of a claimant faction or go to another kingdom and ask them to go to war to support their claim. And of course there are plenty of ways for an heir to die unexpectedly.
i’ve been unfashionably obsessed with ck3 for like a year now. over time i’ve started to question certain pieces of orthodoxy about the game. Among them I’ve come to doubt that keeping your domain together is a big deal.
Yeah, I know. I used to do the disinheritance thing very aggressively, but then realized that long-term this is not great for my dynasty, since I don’t get to invest the renown in anything. So I thought: “well. what if it fractured?” After the first moment mourning the loss of domain income to my heir once he got knocked down to one county, I decided to think in terms of the sum total of all the incomes raised by me and all my vassals and all their vassals throughout my entire realm instead of in terms of the income earned by my heir. like — I don’t care how much of the money gets all the way up to me, all I care is that my realm is loaded.
So I busted out some spreadsheets — the fact that I did this may indicate that if I were a ck3 ruler my main stat would be stewardship — and came to realize that unless your realm is already very well-developed), most of the time a big chunk of the total income of your realm comes from the +1.00 gold bonus to realm capitals.
I know, right?
So let’s say it’s 900 and your domain includes a county in a kind of crappy desert region or whatever where the main castle is bringing in 0.4/month and where there’s not that much income coming in from cities or temples. You could build that place up yourself, _or_ you could give the county away (ideally to someone high-stewardship, but we’ll bracket that off for now).
Previously, the realm as a whole was making 0.4 off that castle and you were making 0.4. If you give it to some unlanded person and make them a new vassal, that castle is now making 1.4. If you’ve got good contracts you’ll get about the same amount of money you were pulling in earlier (a bit more, even, if your new vassal is a high-stewardship spreadsheets type himself). You break even — but your realm as a whole is 1.00 gold a month up. That’s not your gold, exactly, but it’s real gold, and if you ride herd on your vassals enough to keep them from getting into stupid wars with each other, a big chunk of that gold is going to go toward improving that county.
So that’s how I’m playing now, and it is _so fun_. I only have to direct-manage one little domain, but that’s not a big deal because I’m pulling in a fair bit of change from all my little vassals, and I can pull in more money by making good personnel decisions (my high-stewardship vassal’s idiot son took over? I’ll do a little tyranny, gank his title, and find some four-eyes halfway across the world who wants to visit my prestigious court. Meanwhile, my realm is just kind of… growing. vertically, as my nerd-vassals make bank and build junk, but also horizontally, cause sometimes it’s worth it to give a high-martial high-prowess beeftank a county or two, bar him from attacking inside your realm, and just sit back and watch as he incorporates more counties and duchies into your realm every few years.
heh, god, I can’t believe I just wasted 20 minutes typing this up.
okay, okay, one more thing for ck3 players who like to think of themselves as the kind of monarch who’d invent double-entry bookkeeping: you can take the reasoning behind the “have as many vassals as possible to get your cut of the +1.00 realm capital bonuses, don’t worry about having a small domain” strategy a little bit farther. When I started doing this I aimed for a totally flat vassal hierarchy, so that I’d get my full cut of the realm capital bonus. But then I realized that it’s worthwhile for some vassals to have a high-stewardship vassal of their own. Why? Because a count who’s their lieges’s steward gets a +2.00 gold bonus a month. A clear win in terms of total-income-in-whole-realm, and so long as you give the subvassal a kind of crappy county you can end up actually breaking even in terms of your personal income, even though you’re only getting a bit less than 10% of their income.
ck3 is a stupidly deep game. I love it.
Strategic success through willingness to accept a smaller slice of a bigger pie reminds me of https://slatestarcodex.com/2017/03/16/book-review-seeing-like-a-state
Yes! And this even carries over into the early portion of EUIV; in that game you’re limited to a relatively small number of vassals (and vassals take up an alliance slot) but filling that cap constantly is the key to any strategy. There’s also a very metagamey method of dividing e.g. Brazil into nine carefully balanced separate colonial dependencies and getting 9 times the global trade bonus from it.
I found the mechanics you’ve described to work well in theory, but fail in practice: you get more than enough tools to keep your polity stable and your vassals loyal or cowed. Scheming and swaying are incredibly powerful, and metagaming allows you to avoid the historical pitfalls.
So far, this would be my main criticism with this series as well. It seems to engage with CK3 as it should work ideally rather than with how the game functions in actuality. The mechanics encourage, and kind of demands, that the player not rely on their vassals, for anything, and only their personal holdings, which can be quite extensive, especially if the player takes active steps to maintain a high domain limit. You don’t even need to rely on vassals to effectively fill your council either. And, personally, I’ve found that I can maintain the highest levels of crown authority fairly easy even with an empire covers a greater territory extent than either the Roman or Achaemenid empires at their height. Really enjoying the series and finding out the origin of many of the game’s systems though.
These articles aren’t really about how the metagame of CK3 has evolved, though. It’s about what the game’s mechanics are saying and communicating to the player.
The problem is that while whats in this post is broadly true at the beginning of the game it ceases to be true fairly quickly which kinds of undermines the point of the post. A liege has to be able to fight faction wars on their own since they involve all of the vassals in the faction vs their liege on his own which leads to the opposite conclusion – that lieges were not reliant on their vassals. The same dynamic applies to military units. Vassals only provide levies and never men-at-arms or knights so a liege isn’t reliant on them for military strength. Again, at the beginning of the game this isn’t true and what little levies you get from your vassals does help but it doesn’t take long before your military strength negates anything you get from you vassals, which sends the opposite message of the post.
A large portion of this issue is that the AI was really really bad at managing their personal holdings and so it was very easy to simply outscale them by basic intelligent holding selection and construction.
This, however, is something P dox identified and have claimed to have addressed in some regard in the most recent patch. Meaning that going forward, factions will be rather more dangerous as the AI better develops their own lands and so has more and better troops available
I’d say that’s the game being too easy if you play well, but the mechanics work. It’s got mechanics for wanting vassals (they’re a large part of my income base and levies) and various mechanics for them making your life difficult. It’s just set up so you can generally succeed at things kings wanted to do but were unable to pull off, though in my case I tend to have devastating civil wars at successions because I paper over structural factors with various personal opinion boons.
I’d go as far as to say it’s one of CK3’s big failing in comparison to CK2. Vassals are so much less militarily significant in 3 that I don’t really care about them, only about my own personal holdings.
Yes but recent patches made the AI significantly more ambitious and competent so it shouldn’t be as easy now
I think that’s mostly an artifact of the game rule complexity. I’m much more familiar with the EU series than I am with Crusader Kings, but in those games too, you can often work your way around the systems that the game intends to constrain your behavior, if you’re good and you know what you’re doing. Certainly far better than the AI can.
But OGH reminded me there and I’ll repeat it here; there’s a huge section of the gameplaying population who *arent* superb players who can make the game system dance to their puppeteering. And for them, these systems do work. Analyzing their efficacy through the lens of the master player is probably a mistake when most players are not in fact masters of the game.
Paradox games seem to have a problem where they notice that dynamic “X” isn’t represented in the game so they come up with a mechanic that they call “X” and consider the problem solved. The issue is that the dynamics exist in an echo system while the game mechanics all seem siloed.
Gavelkind for instance seems like it would be crippling to a kingdom since your titles keep dividing. But a player and often even an AI who is even modestly expansionist will keep having a plentiful influx of new titles to the royal crown, making it quite easy to have plenty of titles for younger sons, and indeed even for younger daughters. Rather then fragmenting realms, gavelkind means you’ll have homogeneous realms as one dynasty occupies the titles at lower and lower levels and schemes against the other families of the realm to drive them out.
To me CKIII seems to illustrate history more in it’s breach then it’s observance. We can see that gavelkind by itself is not enough to produce weak monarchs and fragmented realms, we see that you need more in addition to gavelkind for that to happen. And that seems useful in understanding why monarchs adopted the system in the first place, they didn’t think it would lead to weakness and infighting because in theory by itself it doesn’t need to produce those results.
It’s funny, but I actually tend to find larger realms easier to run, except for continual popular revolts because I impose rulers of my own culture and religion on deeply hostile territory. When you have many vassals each individual vassal decreases in importance, and you’ve got a number of tools to raise opinion realm-wide. I’ll put a minor vassal or landless character in as chancellor because they have 22 diplomacy and set them to domestic affairs. This makes a powerful vassal very mad but all my powerful vassals together can’t hit the faction military threshold.
Then I get an unlikable heir with a reduced demense who hasn’t had time to build up opinion modifiers or land new vassals and it all goes to hell in a handbasket.
“Adding to this complexity is that while the king may be a feudal lord, that does not mean all of his vassals are.”
Do you have any recommendations for how common the various forms of local government were? For example, in France in 1200, what fraction of the people lived in counties vs bishoprics vs republics?
I’m interested in this question because there is a common belief that republics were a feature of Antiquity and the Modern Era, but were uncommon during the Middle Ages. I suspect that this is false, and that republics were decently common at least in the High Middle Ages (maybe 5-10% of the population of Western Europe), before declining as state capacity increased in the Early Modern Era. This myth probably originated to avoid the observation that the French Revolution destroyed a lot more republics than it created.
In the Holy Roman Empire, it is estimated in the 1700s ~12% of the population lived in ecclesiastical states and ~3.5% in the imperial towns. It would have been slightly higher in the High Middle Ages, but not enormously. However, this will tend to underestimate it as those are just those with imperial immediacy so there would have been a lot of church lands and cities (including some members of the Hansatic League!) who owed fealty to an intermediate lord but more or less self-governed. But conversely, the ecclesiastical states often had feudal vassals.
France in 1200 would have mostly been baronies and had no real republics, although a lot of largely self-governing cities and church-owned lands. Both tended to increase over the course of the Middle Ages. England by the late Middle ages had roughly 25% of land owned by the church and 20% of population in the self-governing towns (who often were vassals of a vassal rather than the king). I haven’t played CKIII really, but in CKII ~half the holdings in feudal areas (increasing over the course of the game) would be cities or temples but the overall tenor was very feudal and that seems relatively appropriate (though the church should have mattered much more, Pope was a major player, but local clergy rarely mattered much).
As I recall, the Pope was at least moderately important for Catholic rulers, in that he could give you boons if he liked you enough, and you wouldn’t get anything from your vassal bishops if they liked the Pope better. But there wasn’t much framework for conflicts between crown and cross, except on the individual level and whether the current Pope would lend you money. (Unless you pissed him off enough to sanction your enemies invading your kingdom and taking your crown.)
There were also the Buddhist republics of India, but I suppose they count as being of antiquity as well in your schema (fun fact, the Buddha is commonly referred to as being a prince, but the more detailed story in Buddhist tradition is that he was the son of the prime minister of the Sakya republic).
Anyway, Italy was ass-deep in republics during medieval times, so that’s another thing to consider. I was looking through this list of republics in Wikipedia and I forgot that the Tlaxcala in Mexico and the Haudenosaunee in Eastern North America would have been considered as having their origins in medieval times: https://en.wikipedia.org/wiki/List_of_republics
It’s going to depend what you mean by ” a republic”. There would have been many functionally self-governing polities with a fundamentally republican local government but which still owed fealty to a monarchical ruler (often, the emperor). Even Venice, arguably the most successful of the medieval republics, was still nominally subject to the Byzantine Empire for a large portion of the Middle Ages.
Of course, “republic” as now used is a pretty vague term and not necessarily a helpful qualitative measure as it focuses almost exclusively on the identity of the head of state while ignoring the rest of the constitution. The Polish-Lithuanian Commonwealth – a kingdom – was much more “republican” in setup than, say, the English republic of the 1650s, which didn’t have a king but was essentially a quasi-monarchical military dictatorship.
I think for a sensible definition of “republic”, they were reasonably common throughout the Middle Ages, although most of them were pretty small and some of them quite short-lived.
According to Witold Kula, noblemen were 8-10% of the population, and 6.6% according to Norman Davies (at the time when the typical figure in Europe was 2%). Some estimates are as high as 15-20%.
The system everyone, not just the first son, but even female children got an equal share of possesions, had predictable results. Individual nobles had less and less power over time, and many even became landless(but still with noble rights, such as to carry a sabre). Meanwhile a small number of them became ultra rich by clever manipulation and management of their assets.
The form of government doesn’t neatly fit into any modern category. It could be called an oligarchy, but it was often called “Demokracja Szlachecka” (Noble Democracy).
You could also make comparisons to European Union(!), because many of the regions joined willingly. Lithuania joined because Lithuanian nobility wanted to enjoy the same rights as Polish nobility, and they felt helpless about the powerful magnates in their country. Prussians also joined at one point, and they kept their laws and the language. The Ruthenians also wanted to join, but their Orthodox religion was a point of contention and they were never treated fairly.
In all the cases though, it was always about the nobles. The Polish-Lithuanian Commonwealth was only looking to invite nobles from other regions, and include them the decision-making process. A few exceptions were prominent cities – they were allowed to have their own council and representatives.
There were also elements of federation and confederation, elective monarchy because the king was elected, and constitutional monarchy (pacta conventa).
What I find disappointing is that the last efforts to reform it – before the Partitions – were definitely in the direction “back” to the “enlightened absolutism”. For example the Constitution of 3 May would exclude the landless nobles from voting, on the grounds that they’re susceptible to bribery. I mean, they weren’t wrong…
This whole subject of Carolingian governance and its development into vassalage sounds utterly fascinating to me. What books could I read to get deeper into this topic, Bret?
Bloch’s Feudal Society is nice, but touchs much more on the development into vassalage/manorialism, and the ways they developed, than Carolingian society itself.
Reynolds S. Fiefs and vassals. The Medieval Evidence reinterpreted. Oxford, 1994 would be a good suggestion.
Eastern European historians have always known that the whole manoralism+vassalage system didn’t really apply to their medieval predecessor polities. For a long time that was considered a sign of “retarded development”, an aberration that had to be explained, but slowly evidence showing that manoralism+vassalage wasn’t the norm up until the height of the Middle Ages was collected in Northern Europe, Central Europe and even Western Europe, including France itself.
On the note of vassals, a random thought; What’s the practical (” “) white-room limit of feudalism?
Defining feudalism specifically as “a king’s royal-to-vassal ties of loyalty are based solely on personal relationships”, which, in the real world will have more exceptions than examples (infact, a 100% exception rate if I had to guess).
We’ll use Dunbar’s Number (150) to set a general limit on the number of people one person can know.
This sets a purely theoretical limit of king > 150 dukes (or other title), and each of those 150 could have 149 earls (barons, etc) (one slot taken by the king), and so on in an infinite pyramid.
Maintaining a stratification where none of the peers know each other seems incredibly unlikely – and due to explode on contact, unknown ties will result in conflict and a population unable to self-replenish.
So, in a still strictly stratified and brutal* world (the king has only vassals/relationships of one rank lower to avoid “losing” space to having someone with no corresponding underlings, and doesn’t have any relationship with his own servants and/or family (maybe he passes remembering them off on his wife?)), let’s draw an outline.
We’ll call the rank procession ; king > duke > count > baron > all lower ranks.
(Crusader Kings’ format. Not strictly accurate, but this is just for fun and trying to parse out just one country’s nobility rank pyramid would be an undertaking.)
Every noble (and royal) has to have a wife (/husband) and two children to be self-replacing.
Every noble has to know the king, and every noble in the chain directly above them. (Eg. Barons know their Count, and the Count’s Duke.)
The opposite is not true. (Eg. Dukes do not need to know which Baron they’re talking to. Assumably the relevant Count will intercede for them.)
The king therefore can only know a maximum of 147 dukes.
A Duke can know … 146 Counts
A Count can know … 145 Barons.
Barons know … 144 lower ranks.
A “perfect” pyramid encloses :
1 King
147 Dukes
21,462 Counts
3,111,990 Barons
448,126,560 Other
So, that’d be 488 million in our Modron-esque feudal kingdom, which is an absurdity in human terms.
So, lets try saying that 50% of their Known quota has to be taken up by same-rank same-fealty persons. So each Duke spends 74 (73.5) of their 147 people on fellow Dukes, and so on. (Admittedly requiring an unlikely perfect spread to keep everyone balanced in known/known-by numbers.)
1 K.
147 D.
10,878 C.
794,094 B.
57,174,768 Oth.
57 million is just short of today’s Italy (58m), and around x4 medieval France (14m, 1348AD – from a cursory internet search).
So, lets refine that again; removing 20% of the known from each, in order to let royals and nobles know their own household, plus a margin of error (kingdom-famous people, etc) taking up spaces.
1 K.
118 D.
5,546 C.
260,662 B.
11,990,452 Oth.
Medieval France is actually ~3m ahead of this model feudalism! Of course, that France, being a real kingdom, has many different systems running inside of it – some of which probably create certain (in)efficiencies, but also probably deals with larger groupings of people.
*Although oddly caring, as there’s no (limited, anyway) administration with everything being based on personal ties – no one is just a number, politically at least. Economically, militarily, and such, you’re out of luck.
There’s a lot of problems with trying to estimate this via Dunbar’s Number. First off, the fact that nobles need to know a lot more people to run their fiefs than just direct vassals and lieges. You left out all sorts of advisors, merchants, military officers, priests, family members of the other people you know, etc.
Second, you’re equating the relationship between a baron and his peasants to that between a count and his barons, which is not how that works. A noble doesn’t need to know all the serfs farming his demense.
Third, there’s more to whether a polity (or even a vassalage-based polity in specific) of X composition can be stable than how many people of different ranks it contains. Balance of power, both between royalty and nobility and between different nobles; bureaucratic/administrative capacity available to the rulers; how rulers manage disputes; cultural norms and values; external threats; etc.
And probably more problems that an actual expert could point out.
No way. The things I disclaimed, including the fact that this wasn’t realistic, weren’t included?
That’s wild, man.
I hope you spend some time talking about the Byzantine Empire in-game vs. in real life — much of the feudal game structures you mention make it so the Byzantines are never internally weakened to the point they are an attractive option for invasion, so usually end up as one of the world’s strongest players well into the endgame.
I wish the steppe nomad and/or republic DLCs came out before this series. Partly because it would be neat to see Devereaux break down The Republic 2.0 and Horse Lords 2.0, mostly because playing all sorts of different government types is neat and fun.
An interesting bit of complexity that would be neat to dive into (a proposal for the Patreon Senate to consider?), but tough to model in a way that the player could easily parse. Or that Paradox’s Clausewitz Engine (drink?) would know how to display.
Like, should the parts of France run by the Kings of England be displayed as part of France or part of England? It would be ridiculous to just paint England blue, but you need to display the Angevin Empire somehow—not just their peacetime borders, but also indicating when (for instance) the king of France occupies Auvergne (meaningfully distinct from places that France just owns), or for that matter when Spaniards occupy Gascony. Even if you can make players grok that every one of their vassals could owe fealty to other lords, even enemies, and help them understand the potential consequences of this, creating a system to display all the relevant information clearly for any mechanically possible combination of homage, fealty, and occupation would be a UX designer’s nightmare.
> creating a system to display all the relevant information clearly for any mechanically
> possible combination of homage, fealty, and occupation would be a UX designer’s
> nightmare.
I think that at some point you also run into the problem that real-life kings spent a life-time learning the details of their vassals, while most players would like to be able to grok the world and their place in it in a few minutes or at most hours.
Even my day-dreams run aground on the rocks of reality 🙁
The game actually DID model it in a very parsable way in CK2, at least for a bit – you just had different law and inheritance screens for different titles.
Frankly, I felt that never went far enough, or allowed enough weirdness – you ended up having unrealistically little control over places you were not de jure lord of. As in you had less control over Normandy as the King of England and Duke of Normandy than you did if you were just the Duke of Normandy. Bit odd.
That, and one definitely should have been able to have wildly different inheritance laws for titles of the same level, but that was surprisingly underdeveloped.
Which is why I feel those systems don’t really model the sort of thing I’m talking about. They’re really modeling something different, unrelated to the obligations individual middle aristocrats had to higher nobility.
Does anyone know the name of the banner picture along the top of the page, or who the people are in it? It’s quite beautiful 😊
I did an image search and it appears to be a wallpaper specifically created for the game in question.
Its one of the loading screen images. There are a fair amount of them and they are all of similar quality.
It’s one of the images used by the game as a load screen. The comparable images in EUIV are all of identified (and *usually* identifiable) historical rulers, but in CKII the game art was all allegorical depictions of mechanics and I assume that’s what’s happening in this one, but someone else may be able to spot signature Whatsit of the Whatever dynasty in a minute.
This is one of the loading screens from the game – apparently painted by Craig Mullens!
https://www.reddit.com/r/paradoxplaza/comments/snp11o/crusader_kings_3_new_loading_screen_the_court/
“you will raise and supply form the revenues :
from, no doubt.
Is it possible to square the trend towards decentralization in the early middle ages with the neorealist view of how history works? It would seem to me that decentralization would be a disadvantage in interstate anarchy, so one would expect less-centralized states to be conquered by more-centralized ones. However, this doesn’t seem to be what happened. Is it possible that decentralization was less of a disadvantage that I’d thought? Or is this just an area where neorealist theory fails to explain how the world actually works?
Decentralization can be an advantage over ineffective centralization, especially in the man on a horse age of communications. The central authority might well get more troops and revenue out of an area by ceding control to local authorities than trying to manage it from hundreds of miles away, quite possibly without being able to send letters because no one involved can read. It also means that when the Vikings or whoever raid a border region the local authorities can mount a defense with the forces on hand rather than needing to call in the royal army from its bases.
Also, not all else is equal. If internal forces make a realm decentralize while another doesn’t, that won’t compensate for the fact that the Holy Roman Empire is very big. And the big centralized states of the era in the east were often occupied with each other and couldn’t exactly spare the forces or attention to launch a sustained western campaign
One of the problems the Emperors had do deal with continually was that they needed to maintain control in Italy – and be in a position to display force to often-hostile Popes; but the bulk of their powerful vassals were Germans who had little interest in affairs south of the Alps and even less in leading their troops there. A refusal to do so was the reason for the fall of the mighty Welf double duke Henry the Lion- as he got the blame for the Emperor getting soundly thrashed by the Lombards
This is one of the places where neo-realism breaks down.
When, in reality, having a king far away means you get conquered, you break off and raise your own forces. Such internal pressures show the model’s state does not actually exist.
All models are false. Some are useful.
Usually, you don’t break off but instead, the far-away ruler recognises that you need to raise your own forces. In fact, they most likely will tell you to do that. (E.g. British Imperial Government directing the creation of Canadian and Australian forces.) The local government will keep appealing to its formal allegiance to the far-flung imperial center, as this improves their legitimacy. (E.g. Rhodesia did claim fealty to the British Queen even after her government had renounced them.) Simultaneously, the formal suzerainity may act as a deterrent. The potential aggressor cannot really know whether the local adversary will get significant help from the imperial center. (E.g. Estonia vis a vis Russia today or Canada vs. USA in the Crimson war plan between world wars.)
The point where the border region breaks off comes only after the formal allegiance becomes a burden and the local government has no longer need for the added legitimacy given by the imperial recognition.
Depends on how bad communication is, and how formal the break up is.
There were places in medieval Europe where you could only get there by waterways.
One element that neo-realist theory struggles with is the underlying expectations about legitimacy. A bunch of tribal war-bands (eg early Anglo-Saxon England) or other warrior elites (as across much of North India) in competition have little investment in any particular patch of land – so once some ruler gets an edge it can snowball into a kingdom quite quickly. China alway re-unifies, because all the rulers of the divisions agree that it ought to be one – they just disagree on who (same with the Roman Empire – the endless civil wars do not lead to permanent division). The fusion of Germanic/steppe traditions of rule over a people (King of the Franks/German/Poles/English) with Roman notions of rule over territory in Europe made empire very difficult, as each element derived legitimacy from its own particular traditions (so the Angevins end up running a lot, but under about six different hats, with different rules; and Charles V has so many that he gets a headache and retires). The empires break down into their constituent parts. In this sense, even in the dark days of the Capets, there was still a firm sense that France was a kingdom and needed a king.
I would think that, all else being equal, a centralised polity would have a military advantage over a less centralised one. But a larger polity is harder to run centrally than a small one. There are more people further off to centrally command. So they are harder to centrally command.
So you could quite easily get a mix of large decentralised polities, small centralised ones, and large decentralised polities comprised of small centralised ones.
I have three serious objections to CK3’s current set up: the de jure system (especially for the early start date, it implies there were pre-existing beliefs about how polities should be organized), the succession system (which is too legalistic and orderly most of the time), and the representation of armies (which completely fails to model the retinue of retinues construction of a lot of armies).
Though a lot of this was predictable when it was announced that counties couldn’t be split.
A subtle typo:
> (from the German Markgraf or ‘March-Lord,’ with ‘march’ here meaning ‘a boarder military district’)
Boarder, meaning one who pays a stipulated sum in return for regular meals or for meals and lodging, has been substituted for border, meaning edge.
All vassals can be metaphorically called boarders, but the only border ones are edge lords.
In the previous comments I suggested that a player in Crusader Kings 3 is something like the Bene Gesserit in Dune. Developing this, my suggestion for a future Paradox game is Dune Imperium AKA Crusader Kings in Space!
Setting is the universe of Dune before Paul Atreides, ie the Corrino Empire. All the CK 3 game mechanics should work just fine. Feudal society with competing noble houses and manorial style economics. Military forces are the personal retainers of the aristocrats rather than some modern industrial era army. Kinship politics dominate, especially with Bene Gesserit meddling. The tech level is higher than medieval Europe, but also deliberately limited and there isn’t really any kind of tech tree progression going on. (Maybe one option is to play as Ix, where you can innovate but also risk getting Butlerian Jihaded out of existence for doing so.)
I’d use the start of the Corrino Empire, long before inconvenient Fremen obliterate everyone. With ten thousand years to choose from there’s plenty of scope for, er, creative additions without necessarily conflicting with the books. (Books written by Frank Herbert. Maybe Villeneuve can decree all the others non-canon?)
And for the real hardcore players who think they know how to win at CK 3, let’s see you survive from the first Corrino ruler all the way up to 10193 years later when Muad’dib finishes everyone and everything off with an awesome cut scene.
If you are ignoring the other non-FH books, you might check out The Dune Encyclopedia by Willis E. McNelly et al. Not really canon at any point, but full of a lot of fun stuff about the historical and current Imperium. The conceit is that it is an encyclopedia written about 2000 years after the death of Leto II.
(Especially fun is the timeline where Alexander the Great is the first Emperor.)
There’s a game called “Star Dynasties” that at least attempts to do the Space Feudalism thing to a similar degree as CK. It has even been directly called “Crusader Kings in Space” by some review outlets like Rock, Paper, Shotgun. Might be worth a look.
I actually made a mod for CK2 that was based on a similar idea, Crisis of the Confederation. It’s more about exploring the transformation of a liberal democratic republic into vassalage, though. It’s also, frankly, kind of an unfinished mess. But it exists!
I’m curious, how does CKIII provide for this? If two of your vassals are squabbling, what powers do you have to jerk their chains? Similarly, if you’re, say, a Count who is expanding his holdings at the expense of your fellow vassals of the same liege lord or lords, what moves will the AI feudal overlords take to make you knock it off?
Because that’s a pretty major part of this sort of thing, isn’t it? Resolving disputes that have escalated to private wars?
The AI is not a sophisticated player. In the earliest portions of the game, the only real tools for controlling ambitious vassals are forming marriage-alliances with them and murdering them. In the hands of a human player, ‘both’ works pretty well. You can also promote them to your retinue and try to get them killed in battle, but the key point is that they decentralize upon death, with their realm divided among their children in as many ways as the local traditions allow.
Eventually you reach the point in the game where people know how to outlaw internal wars, but that law doesn’t prevent a mass revolt to overturn a liege’s outlawing of internal wars; generally AI players don’t enact restrictions on much, and are much more willing to surrender to their vassals than human players are. The AI mostly solves problems with assassinations or not at all
Technically Royal Court represents this, but the player only gets three court decisions every five years at maximum and only a few of them involve vassal squabbles. Also, if you’re a vassal you can petition your liege for various things, which can include getting a claimed title or de jure vassal transferred or enforcing peace in an internal war. But you’re pretty light on tools for managing inter-vassal relations, except that having high crown authority bans unapproved warring.
There’s a number of tools for trying to deal with vassals threatening your position, though. You can modify their contract to require them to use partition even when other forms eventually become available, you can retract vassals or revoke titles, you can imprison them and execute them, you can straight-up murder them, you can blackmail them, you can cow them by executing other people, you can lessen their obligations, and that’s not getting into the pile of random events.
At least in the tribal stage, you do not even get notified when internal wars start. (You do get notified about external wars, some of which arguably matter much less to you.)
You also do not get notified of peasant revolts against your vassals, which is a pity because you would be allowed to help your vassals with them.
The game allows you to have up to 64 direct vassals, so you ideally you would want to split the realm (apart from the desme and the player heir territories) in 64 equal parts and assign it to 64 different dynasties. (Over time successful vassals will still grow their territories, but it will be a long time indeed before any one vassal becomes threatening. I am not sure how I should imagine a tribal level administration accounting for the taxes and levies owed by 64 different parties, and why any system that could accomplish that would break down beyond that.)
My point here is that while you could click through all your vassals to stay up to date with their warfare efforts, doing so would be extremely time-consuming.
Outside the game, is there a generally distinction among historians between tribe and clan? What is the difference between the two?
My (unqualified) understanding is that a clan is a large extended group with shared (perceived) ancestry. So what CK2 actually models as a dynasty or in the cases of very big dynasties a “house” within that dynasty. A tribe as a form of government (ignoring that it’s been used to describe basically anything modern-era European writers thought was less civilised than them) is then a “clan of clans” where many of those kinship groups come together to govern territory.
My understanding is that “clan” implies that the members of the group are related by blood, while “tribe” doesn’t. I could be wrong though.
I’m not sure if there’s a different distinction in pure history, but from an archaeological perspective I don’t think I’ve ever seen “clan” used as anything other than a alternate or regional/culture-specific term for a dynasty or family.
“Tribe” in archaeology/anthropology has a much more specific usage as a kind of socio-political organisation, involving a group of a few hundred to a few thousand people with a very limited, or completely absent, formal government/administration, and where the most important ties people have are based around kinship. The tribal governments in CK3 don’t really match this definition, since they have a formalised political hierarchy with paramount chiefs/leaders and are organised into much larger polities, and would probably be described as either chiefdoms or early states if you’re following this particular typology.
All of these terms and their definitions, of course, are also very contentious in scholarly debates, but most archaeologists would recognise the specificity around terms like band/tribe/state that wouldn’t apply to a term like “clan” (even if they dispute the former exist or are useful concepts outside the narrow set of type societies they were built off of).
It’s an interesting point how scandinavia (the three scandinavian countries developed differently, but in this sense they are similar) never quite developed the vassalage model, but somewhat adopted the *terms* of feudalism, while often meaning a different thing.
Using European terms was good for diplomacy, as it brought legitimacy. My personal favourite is the case of St. Bridget, born Birigitta Birgersdotter, who was a high Swedish noblewoman. In Italy, she went by the title pricinpessa de Nericia. This was not really a lie: Bridget’s husband, Ulv Gudmarsson was a knight, one of the lords of the realm (riksråd, roughly a member of King’s cabinet) and the Judge (lagman) of Närke. A medieval lagman was the high judge of a province who usually had assistants to run the things, but who also sat his own court for appeals and more significant cases. As the thing was not only a judicial but also an administrative body, this made him the strongest nobleman in the province. Being a “knight” was also a big thing: in Sweden, only the very highest nobles were knighted formally, and in the 15th century, there were times when only one or two Swedish noblemen were actual knights. On the other hand, the Swedish noblemen had no titles nor even surnames.
Translating the position as a “prince” for Italians was honest enough, and signified well what kind of social position Ulv Gudmarsson had held in Nericia.
A similar case of translating social differences, from the other end of social scale occurred in the 1830’s. The son of a Finnish shoemaker who had moved to St. Petersburg was trying to get admitted to the local university, but the university declined his application, for he could not present the permission of his owner: as a son of a shoemaker, he was presumed to be a serf. The Finnish Minister-Secretary of State wrote him an affidavit, which has been preserved. This stated, rather simply that “In Finland, it has been the ancient custom that all subjects are free, and without a lord, and thus free to pursue their wish to serve the state via obtaining university education without obtaining a prior permission from anyone.” The practical shorthand for Russian administration was that all Finns should be considered administratively as noblemen (in essence, non-serfs).
Sounds similar to Poland. There was manoralism and peasants working the land, but no hierarchy of vassals. Nobles were equal in principle and proud of that. The king appointed various officials such as wojewoda, but these were not hereditary. Property was divided equally between sons and daughters of a nobleman.
Titles like count or baron were given to nobles of foreign origin. Also, foreign rulers would sometimes give these titles to Polish nobles. They happily accepted that for the extra prestige, but it had no actual power.
While Polish and Lithuanian nobles enjoyed the equality, they were strongly opposed to admitting others into the system. For example Cossacks wanted to be admitted into the Commonwealth, but for Polish nobles it was inconceivable that a mix of commoners, peasants, convicts, nobles and others could be equal to them.
It would be interesting to see someone make a comparison between the fall of Carolingian central authority and the developement between the Heian and Kamakura periods, since Japan is also sometimes considered to have been a “feudal” society
“‘feudalism’ only really arrives in England with William the Conqueror (of Normandy), who after seizing the English throne in 1066 imposed a more ‘French’-style system of governance.”
Does it seem feudalism in England is quite different than in continental Europe, especially France? The lands of France were control by powerful dukes and counts, ruled like autonomous kings, whether in Norman England, only a small amount of land were controlled by the nobles (the earldoms). The rest of land are ran by sheriffs – officials appointed by the monarchs themselves, while their vassals were given scattered pieces of land (manors) across the kingdom. Thus English kings started off with more power than the French ones.
That’s what comes from being a conqueror: You can set up your vassals in the ways that you want rather than having to deal with already existing structures. (CK actually models this fairly well) William could basically arrange things in an “optimal” manner for his purposes (IE: With nobles still scattered across the land so they could help him administrate, but without powerful base to challenge him)
Having played CK3 for a week now, I have to say that the dynastic point of view the player takes gave me a new appreciation of the motivation of some of the more evil Game of Thrones characters (which life in a later world, but still are very much in the vassalage system).
* The Boltons carefully cultivated image of cruelty is not a weird deviancy but actually serves a perfectly rational aim. Any of their bannermen (or peasants) thinking about rising up against them will answer the question “what is the worst thing that can happen” not with “my head ends up on a pike” but with “my entire family is flayed alive”. In CK3, the corresponding stat is called dread and a perfectly rational way to keep your empire together.
* The cruel father of Samwell Tarly, some petty lord who forces his son to take the black, effectively disinheriting him and rather openly threatens him with death should he ever come back is simply playing CK3 and has found that his primary heir has stats which he considers (rightly or wrongly) unlikely to make him an efficient ruler of his barony, so he has to get rid of him.
* The length Tywin Lannister goes to annihilate the marriage of his son Tyrion also make some sense from the dynastic viewpoint: every child of such a marriage would be a contender for Casterly Rock. And besides, marrying a peasant without any claims is pointless.
* Likewise, trying to murder all of king Robert’s bastard offspring is obviously prudent, as every one could become the nucleus of a rebellion. (I doubt that in CK3, it would be effectively accomplishable: murders would take to long to scheme and executing a score of realm-faith subjects will probably make everyone really hate you. Besides, from my experience, landless nobles with a claim to some kingdom or empire are a dime a dozen in CK3.)
More generally, the mechanics of religion make zealotism perfectly rational. At the moment, I am in the process of replacing the aristocracy of some sixty counties with the wrong religion (mutually Evil with regard to the realm faith) with a completely new rulers with the correct faith. This will not immediately change the faith of the population, but it will at least make sure that if they rise up against me, I will face mostly peasants instead of three kingdoms worth of nobility.
Speaking of meta strategies, having a large number of behaviors considered criminal is sound policy. Yes, it might mean that your ruler might be suffer because they practice some of these behaviors, but the leeway you get with selectively enforcing these laws is totally worth it. The ability to imprison and troublesome vassals or heirs (and then retract, in that order, first their titles and their heads) for “Fornication” is great.
That last paragraph sounds remarkably like modern Russia. Apart from “Fornication” no longer being the big crime.
Yes, you can make a religion where Lust is a virtue and all fornication/adultery is a crime; if you also choose ‘no bastardy’ and full gender equality, virtually every noble in your realm will openly admit to an imprisonable offense fairly often.
This reminds me of the time I was playing the original Crusader Kings and plotted to murder my own son because he was mentally ill and wasn’t producing any children with the daughter of the Duke of Oxford and I suddenly realized, “On my god, I’m TYWIN LANNISTER.”
On the other hand, there were options such as becoming a maester or septon that would have removed Sam from the line of succession as well, Randyl Tarly just refused to consider them because he’s an awful person obsessed with martial skill to an absurd degree.
As someone slightly better versed in West African politics, no, feudal doesn’t quite make sense for West Africa either, even for Ghana. Feudal implies a hierarchy over land that simply didn’t exist in West Africa, where most structures of vassaldom were over people. Though obviously this isn’t the same as “tribal,” which seems to be some kind of messy mosaic of anarchic societies and nonliterate societies, and feudalism in the European style did exist in East Africa.
I’m doing a presentation about CK3’s mechanics at ModCon next week, and I do briefly mention that – Jack Goody is the scholar probably best known for attacking the assumptions about African “feudalism.”
Basil Davidson, a British historian, wrote in his book “Black Mother” that there was actually feudalism in Africa! Only, on a scale that Europeans could not comprehend.
It was a feudalism of tribes! Tribe X was a vassal of tribe Y, who could be a a vassal of Z. So perhaps it was a bit similar to India’s caste system.
Well, feudalism of tribes is a feudalism of people, isn’t it?
The Indian caste system is really a separate thing from Indian feudalism. India does sort of work as “Feudal” government in CK3, because it had a similarly low land/people so did opérate as a feudalism of territory.
Besides, I tend to align with the Manchester school interpretation of the caste system – most historical evidence shows that the varna didn’t become anywhere near as rigid or impactful until the British arrived.
I, the Duke of Anglia, rose to power, became King of Frisia, conquered in the north within the Empire, and in just a few decades held one side, the Emperor simply wanted to arrest me, stripped me of my titles, fought guerrilla attacks with inferior troops to break straight through the capital, three times to make a white peace, and the Emperor still wanted to arrest me again without end, so I couldn’t fight and voted, but he simply stripped me of all my titles in prison, and then the game was over!
Sometimes the only way to keep a game going is for your character to die.
Or to be exiled and stripped of titles. I remember one CK2 game I played where my character was excommunicated for reasons I don’t remember, leaving me facing down a massive horde of loyal French Catholics who wanted to knock off his crown, alongside some other wars.
I realized that surrendering to the French would just pass the titles on to my then-current character’s daughter and get the French off my lawn, leaving me with few enough enemies that I could actually win the wars. And I did! Shame the save ended up corrupted within a generation.
About a vassal having two lords, i have heard about it quite frequently in CK community but the only example I’ve seen cited is always William the Bastard, so much that i thought that he’s an exception that prove the rule (that vassal most of them time have only one lord). Can you enlighten me any other relevant example in the era? Thanks.