RTT/Grand Strategy 2004-2013, developed by Creative Assembly
A general is an organizer of warfare. They take the rough and ready stuff of violence and sort it into ordered aggression, directed at specific points and with (ideally) an objective in mind. In this piece I would like to look at the representation of generals, and how they are conceptualized, by the Total War games. Total War was, when it came out, almost unique (not totally unique, I have fond memories of the pre-total war real time tactical game “Warhammer: Shadow of the Horned Rat”, which was objectively bad, but made a lot of the same design choices as Total War: Shogun). Total War offers the opportunity for a player to take the role of a general, and invites the player to think of their battles in tactical and historical terms. In many ways Total War games represent whistle stop tours of military history of the time period they choose to represent. They actively attempt to get the player to think in terms of the tactical choices open to commanders of the time period. The importance of the General, and their role upon the field, changes from game to game, and the ways in which they do so are fascinating.
In Rome and Medieval 2, Total War, the General is the seat of the character of most of the game. One’s generals are the sons, or adopted sons, of the ruling clan of the faction. When playing as the Romans, they are all competing scions to the power and glory of Rome, and they are a strange bunch. Rome’s generals gain character traits according to how they act on the campaign map. These traits range from zealotry to blood-lust and back all the way to corruption and pacifism. Most brilliantly, these traits are conveyed via the pre-battle speeches that the Generals give—the zealous will speak of great portents of victory, the unhinged will ramble on about nothing much discernible as sense, and all do so in pleasing regional accents reflecting their geographical background. There is so much work that goes into these pre-battle speeches that it amazes one that Creative Assembly ever decided to do them—there are hours of audio here, which speaks of a labor of love (cringe-worthy as some of those speeches may be).
All traits also have some mechanical impact, the most common being to add some modifier to the General’s command score. The command score is a strange thing, because while one gets the sense that it is in some way important—being displayed beside the commander’s little man on the campaign map, the game never really comes out and tells one exactly what all the little gold stars actually mean. The stars do have some mechanical impact—the more command stars a general has, the more they buff the morale of nearby units, and the larger the General’s bodyguard becomes. This is important because the General’s bodyguard are some of the best cavalry in the game, which incentivizes the player to use them as shock troops to break the enemy when they are weak. Rome is always pushing the player to use their general on the front line, as a support for the morale of the troops, as well as a combat machine in his own right. However, there is a risk to using one’s general in this way, they could always die to an errant rock hurled from a catapult, or simply be swarmed by less noble troops and beaten to death. One is given generals almost at random, when one’s sons come of age they are counted as one’s generals. The General therefore is a valuable resource to be fostered and nurtured, a rare (until around 100 turns in, when one’s empire is lousy with listless children) and powerful unit, as well as an interesting character.
Rome wants to be a game that has the player tell stories about great men and the armies they lead, it wants to be a game where general matter and the player invests a great deal of emotion in them. There is much to like in the way in which Rome does generals, but other things to be wary of. The game knows what it wants one to do with generals, it wants you to use them on the field of battle, and it wants the generals to lead from the front. The reason for this is clear in two ways. First, it is mechanically most appropriate to have the general near the troops, and the general’s unit is undoubtedly a powerful force on the field. Second, it is thematically most appropriate that the general be at least somewhere near the troops and the fighting, generals in those times did often take it upon themselves to get into the melee and let themselves be seen fighting. Total War, further, is a game about war, not peace and good governance. What it aims to emulate is some really good, but relatively shallow narrative history. It wants to give you the experience of the most exciting cliff-notes of the battle of Cannae, and it does not mind committing to a great man narrative in which Caeser wins all his battles basically by himself.
The course of a campaign of Rome in many ways approximates a great man narrative told about the player, and that is an enjoyable power fantasy, wrapped in a deep and interesting strategic and tactical puzzle. However, given the power fantasy Rome presents, it has some puzzling design choices. Let us for a moment anachronistically compare Rome with Crusader Kings 2. They bear comparison in the way they model their characters, as shaped by random chance influenced by player choice. In Rome, there are a number of negative traits that one’s generals can gain at random. They might become drunkards, or they might go mad, be infertile or be cuckolded. These happen without any player input and with the smallest of notifications. Not only that, they are often permanent and can only really get worse as the game goes on. In Crusader Kings 2 on the other hand, the player is almost always given a clear choice about their character. “Hey, look” the game seems to say “something awful is going to happen to your ruler, but I am going to give you a choice about what awful thing it is going to be, ok?”. In Rome it does not work like that. In Rome, there is only a couple of ways that the player can communicate to the game who any given character is, via the battles that they fight, and the forces that they take to those battles. Outside of battle the player is given no agency over their characters.It must also be said that for all the random chance involved in the acquisition of traits, there is something of a tendency for all old generals to become essentially the same character. In my experience, and perhaps this is simply a result of how I play, almost all my successful generals end up battle-hardened-old-bastards. These old men are scarred beyond measure, as tough as an elephant and hit like a truck – they are also generally world renowned for their network of assassins. It does not matter if they are Gaullic, Grecian or Roman, their character traits fold together to a single, bloody point. Perhaps there is something of a message in that fact. With their sons there may be great variety, until they become leaders of an army themselves, at which point the cycle repeats.
Medieval 2 goes some way to giving this system some more interest, adding a couple of new stats to the General’s character sheet. In addition to a command, management and influence, leaders in medieval also have “chivalry”, “loyalty” and “dread”. These extra stats give characters new axes along which to grow. Primarily they allow the player to choose whether they want their great generals to be feared or loved. This is facilitated by the fact that one can actually capture prisoners, and ransom them back to the enemy, after a battle—or indeed let them free without ransom. Chivalrous characters who let their prisoners free are loved by the populace, whereas characters who kill all their prisoners cause fear in enemy—and their own people. This is precisely the kind of choice that Rome lacks in shaping the outcome of one’s characters, giving the player some measure of agency in how they wish to approach the game from the perspective of that character. It is clear that the focus in Medieval is even more tight on the characters of the Generals than Rome’s was. This is only thematically appropriate, the narrative history of the middle ages is filled to the brim with the names of Kings with odd character quirks. Further, while Rome can draw from the deep well of Roman history in providing for itself a character and a feel, the history of the middle ages is in many ways far more messy. No longer is there one great empire which comes to represent, from a central governing body, the forces of the majority of Europe. Instead we are placed in an age of squabbling Kings and asked to care about the countries that they lead and the forces that they muster. The natural “in” to such a time and a place, is to get the player to care about the character of the King and their family. Regardless of whether one knows anything about the kingdom one is playing, one can understand the motivations of kings, and one can come to find a character for one’s country by the predilections of the man in charge.
The next title in the Total War series was “Empire: Total War”, which boldly attempted to take total war into the age of gunpowder. Empire is widely regarded as one of the worst Total War games, but it has a special place in my heart. It earned that place by its sheer ambition, which towers over the rest of the series, stepping out of the well established Total War comfort zone, into an entirely new era and style of play. The hubris of attempting to model a significant chunk of the world, populate it with multitudinous factions with a huge variety of units, adding a research tech tree for the first time in Total War history and at the same time completely change the dynamics of the now well worn battle system, is simply breathtaking. There are so many moving parts added to the Total War formula by Empire, it is a marvel that it even works (at launch it often did not). Unfortunately, this step into the unknown, was also something of a step into the wilderness. One of the knock-on effects of adding all this complexity, is that elements of the game that used to be more complicated and nuanced are left by the wayside. The way the game models generals remains mostly the same, along with—to some degree—the family tree. However, generals are now purchased rather than given to the player at random.
The stats of the General are paired down to the absolute minimum—command stars are all they have. Further, the utility of the General’s bodyguard in battle is obviously limited by the ubiquity of the musket and gunpowder artillery. Thus, the General steps back from their important role in characterizing a force—they are re-made in the form of a pure battlefield tool, just like any other of one’s units. This leads to an odd disconnect, the General is still the most important man on the field of battle, if he falls the entire force still takes a huge moral hit. However, this no longer really makes any thematic sense. In Rome and Medieval, the General was the charismatic core of the army, he played a role in battle that could be easily seen, and he could make a difference by himself, leading from the front. The force would, over time, become identified by the General that lead it, and take on the character of that general (even if that character became uniform over time). The General in Empire is not a charismatic leader, they are a buff that one attaches to an army, because without it the army will do worse. Of course the troops should care about the General, but it is hard to see why they do, given that he does so little apart from stand behind them, I imagine them giving a little wave and the odd “good show chaps!”. He is now an organizer of violence, rather than an agent of that violence, but the game never allows the General to do any organizing—all of that is handled by the game’s new tech tree. The removal of the more interesting aspects of characterization lays the nature of the game, as a game, bare.
The follow up to Empire, Napoleon, solved many of the issues of its predecessor. Napoleon has a tighter battle system, one that more fully understands the time period in which it is set. However, it still lacks generals with personality. It is greatly aided by the name recognition of Wellington, Blucher, Napoleon and Kutuzov. However, for a game that centers around the personality of one man, the characterization of that man is sorely lacking. On the battle map, Napoleon is a general with a lot of command stars. Apparently, when he stands behind men, those men feel a good deal better about being shot to pieces. Also, importantly, generals can now give out an “inspiration” buff to a single unit, which makes that unit feel special enough to kill people better. It is all a bit strange, the General is improved as a game piece, now being much more of a force multiplier, but at the same time, does little to none of the business of being a general, such as giving commands or organizing the carnage. Of course, this is because the player is the real general—but it seems strange that there was little thought put into the idea that perhaps the General could be used in order to get units to perform specific functions they would have been incapable of on their own, or indeed to speed up recruitment or weapon development. These are, after all, the things that generals spent their time doing in the Napoleonic wars. As such, Napoleon does not really manage to put the personality back into empire, although it does manage to take Empire and make it a better game—by limiting its scope and tightening its mechanics.
In Shogun 2, Creative assembly decided to upend the formula for what makes a general in a Total War game. Or at least, they decided to completely throw out the Empire formula—generals can, for instance, no longer be purchased like so many apples. Shogun takes the character of the General a very different direction from the games that went before, rolling a good deal of what makes one’s army operate as it does, into a single character. In Shogun 2, generals still gain random traits—although they are rare. For the most part however, they gain experience points by fighting battles and leveling up. When leveled, a general can put points into skills that will make him better at appealing to a troubled populace; allow the army he is with to move faster on the strategic map and use different formations on the tactical map. Or indeed, one can simply spend all one’s points to turn his unit into a terrifying death machine.
What this represents is a move towards player agency. Previously, the general was a character and a playing piece. He had a dual role, one mechanical and one narrative. Although those roles came together to some degree, the general could never really be expressed as a character through playing the tactical game. As I said previously, the general in Rome or Medieval is a playing piece with a defined role—as morale boosting shock cavalry. Different generals will be better or worse at performing this role on the field depending on their overworld stats, but those stats do not change their mechanical interactions with the tactical game. In Shogun 2 different generals express their different command styles on the field by the different roles that their unit comes to have in the armies that they lead. One general will dismount from horseback and sit down in the middle of his troops causing them —for some reason—a to fight like men possessed. Another will ride with the cavalry, while another still will stay back and perform the traditional, Empire style, job of waving and saying nice things to the troops. These different specializations can make all the difference in Shogun 2, whose factions are, by Total War standards, extremely lightly sketched—they share almost the entire unit roster with one another. By the skills one picks for one’s general, one can craft a few, very different kinds of army. So, in Shogun 2, the general becomes less of a character, but at the same time gives their army a distinctive mechanical fingerprint. Thus the focus of characterization moves, to a certain degree, from the individual of the general, to the army that the general leads.
If Total War aims to be a history, Shogun 2 is less the narrative history of great men that Rome and Medieval were, and are a more narrow military history. They are focused on methods and styles of warfare, rather than personalities. The thematic master that that the general then serves, is not that only of their own character, but also of the faction and army that they lead. Here mechanics roll into theme, one can feel immediately the difference between playing a thematically Date army—all powerful charges and brutal melee, with the general leading from the front—to a thematically Chockosabe army—which strikes from afar, then runs away and strikes from afar again, with the general buffing the rate of fire of the archers from the rear. Because every army draws from essentially the same roster, it is easy to pick out the different roles which are emphasized in any army, and the lack of variety underlines the importance in the choice of preponderant unit.
This trend is refined in Shogun 2’s expansion “Fall of the Samurai”, which moves the timeline of the Total War series even further on than Napoleon does—to the 1860s. Once again, in this late period, generals are reduced in fighting effectiveness—however, what that fighting effectiveness is replaced by in FoTS, is a choice of possible skills set of deep choices about army doctrine. How one chooses to tailor the general in FoTS is integral to giving the armies one leads any character at all—drawing as they do from an even more limited pool of possibility than Shogun 2’s, and mostly being a thousand different flavors of rifle armed line infantry. Thus small changes in their performance can lead to interesting differences between forces. The stories that these choices tell are grander narratives that the ones that Rome or medieval attempted to tell with their generals, they are stories about armies, not individual men. Further, one comes to value one’s generals immensely, because it is the general that allows one’s forces to fight in the way one wishes them to fight—their absence is sorely felt on the field, and the death or defection of a great commander is a tragedy not simply because it makes the game harder, but because it means that one’s army loses what made it unique
At least, that is, until the next guy gets enough battles under his belt to replace the former commander. There is an unfortunate tendency in Shogun 2 towards carbon-copy commanders who share exactly the same skill set—because there is one skill set one really wants to play with. Like the “battle-hardened-old-bastard” tendency I mentioned with Rome and Medieval, this is unfortunately unavoidable over the long term. The General, as a human being then, rather than an embodiment of a way-of-war, is less visible in Shogun 2 the longer one plays. One’s first general will be memorable, but eventually they will simply become a cipher for how one wishes to fight.
A focus on doctrine, and building armies to themes is something that is further emphasized most modern Total War games. Rome 2 limits the number of armies that one can have on the field at any one time, and each must have a general. Further, this limited number of armies are named—and can be re-named by the player—and gain experience points, which earn skills on an attendant tree. One is encouraged to think of them as having personalities of their own, at least somewhat independent of their generals. Oddly the generals in Rome 2 are somewhat sidelined, at least as personalities. They do have traits, although they can only have three at a time, and the negative ones can be relatively easily lost by making sure that generals are always fighting. They also have no real personality at all, they are nothing but skill tree. That skill tree is, as well, flattened and less full of interesting choices about differing doctrine. The General in Rome 2 almost fades out of view, simply by the fact that he is always there. I have played games where I did not know the names of any of my generals, and did not really care to. The General is lost in the haze. Much as they do in Empire or Napoleon, they simply fade into the noise of everything else. Rome 2, for me, represents taking the Shogun 2 idea that the general should characterize the army they lead, a step too far—making the general identical with the army. When the general becomes identical with the army that they lead, they lose all independent character, and it is impossible to care about them independently of the troops that they lead. The strange, flattened skill tree which they climb also promises few interesting choices in the terms of the doctrines with which one can lead an army. This is important not only from a gameplay perspective, but also from a thematic perspective. If Total War wants to tell an interesting military history, it must give the player some sense of the turning points in their wars. One way to do this was to have a general learn a new skill, that perhaps the player would use to win the next battle. This would mark a change, something memorable that the player could see made a difference, a military innovation. Without interesting turning points, with only a 3% increase in melee damage or whathaveyou, there is less meat to the story of one’s game. It is simply progression for progression’s sake, mechanics un-moored from narrative, and that is a great tragedy.
The General in the Total War games is, in many of those games, one of the most interesting storytelling tools in a videogame. They are, or can be, the point at which mechanics meet narrative, in such a way that a player can craft their own narrative through the mechanics of play. This is true in Rome, Medieval 2, Shogun 2 and its expansions. However, the least interesting Total War games (or at least the ones I find least interesting) Empire, Napoleon and Rome 2, all get the balance wrong. There is more than one way of affecting this meeting of mechanics and narrative, and we see two methods played out in the transition between Rome and Shogun 2. The first is reactive, random, and predicated upon the player’s actions—without invoking the player’s agency directly. This style can create something surprising and different, as the generals are in Rome, at least before they all turn into the inevitable “battle-hardened-old-bastard” archetype. The second method is more co-operative with the player, it asks them what they want and acquiesces to it. This creates a more interesting connection between mechanics and play, but leads to carbon-copy generals more often than the first method. Both are interesting, I prefer the second, but then again I must admit I could never really get into Crusader Kings 2, and I could see that if I was the kind of person that could, I would probably prefer the first method. In either case, the general has a story to tell, and I do wish more games would listen to that story, and apply it to their own case.