A persona who becomes a kannagi can see the supernatural at any time and in any place. The other option, kamikakushi, lets anyone see the supernatural, but only sometimes, and only in particular places. “Kamikakushi” means “hidden by the kami”, and could be translated “Spirited Away”. Indeed, the Japanese title of the Miyazaki anime called “Spirited Away” in English is “Sen to Chihiro no Kamikakushi”: “Sen and Chihiro’s Kamikakushi”.

In kamikakushi, a kami takes a number of people out of the mundane world, and into a separate world inhabited by the kami. In this world, any human being can see kami and communicate with them directly, and the kami can, and do, use obviously supernatural abilities.

Only kami can initiate kamikakushi, at least to start with, and it takes a significant amount of effort. The more people a kami tries to spirit away at once, the more effort it takes, so small groups are more common. It is not normally possible for the kami to keep someone in kamikakushi indefinitely, and it is not possible to take anything from kamikakushi to the mundane world that didn’t get there from the mundane world first. Time normally passes more quickly in kamikakushi than in the mundane world, so that even someone taken for a long time is only missing for a matter of days, at most. This, of course, makes it very hard to convince anyone else of what happened.

In addition, there are many kamikakushi, all of them different. Most of them look superficially like Japan at different periods of history, from primeval woodland to the present day, but the inhabitants and events of each world are different. This means that the accounts of people who have been spirited away do not agree.

The first session of a kannagara game, then, would typically involve kamikakushi, to get the players involved in the supernatural side of the setting, and give them a chance to talk directly to the kami of their jinja. After that, the personae can only return to kamikakushi if they can give the kami enough strength to take them, and convince him that it is a good idea. These goals play into the themes of Kannagara.

Kamikakushi also provides an opportunity to have pseudo-historical game sessions, when the personae visit a kamikakushi that resembles a period of Japanese history. The kamikakushi is not actually Japan in that historical period, so not all details are the same, but the overall culture and atmosphere is preserved. This is important, because it allows people to run game sessions without doing an inordinate amount of research.

Not all game sessions take place in kamikakushi, so gaming groups must choose how often they go there. If it is very rare, the game will feel more mundane, but the wonder of kamikakushi will be easier to preserve. If it is very common, there will be more supernatural events, but they will become commonplace. At the moment, I think the default will be that kamikakushi is rare to begin with, but becomes more common as the story progresses. Once the characters have become kannagi they can, of course, see the supernatural in the mundane world as well, and this may well make it easier to enter kamikakushi.

Kamikakushi may be linked to one another, and may also be linked to the actual past, as opposed to supernatural re-creations thereof. That, however, is something for the future development of the game. Initially, each kamikakushi will be self-contained, and fairly small. Links to other kamikakushi, other places, and other times can be discovered or created in play.


A fundamental problem faced by all “modern world with the supernatural” games is how to account for the fact that there is no clear evidence of the supernatural in a world where it definitely exists. That is, for the game world to look like the modern world, there must be no clear evidence for the supernatural generally available, as that is what the modern world is like. However, the supernatural does exist in the game world, so one would expect there to be evidence. This is a problem. Some games just handwave it, while others create justifications that work as long as you don’t look at them too closely, and as long as the players are not involved. Kannagara will fall into the second category.

The supernatural creatures in Kannagara are limited in that they cannot do anything that is obviously supernatural where a human being can observe it. This is not a rule or social convention; supernatural abilities fail to function if a human being is watching. Supernatural creatures can appear as humans or animals and tell people (if they appear human) that they are really kitsune or kami, but they cannot, in general, do anything to directly prove it. They can also intervene in human affairs through their powers, but only in ways that do not necessarily appear to be supernatural. Believers believe, but they cannot provide convincing evidence to sceptics.

There are two exceptions to this rule. The first is people who are also kannagi. “Kannagi” (巫) is the old Japanese word for a medium, someone who communicates directly with the kami, and “miko” is written in Japanese with the characters for “female kannagi” (巫女). In game terms, a kannagi does not count as an observing human being; kami and other supernatural creatures can do clearly supernatural things where a kannagi can see. Kannagi do not have any other supernatural abilities, but this one makes a big difference.

For example, not all kami can take on human forms that appear mundane. In order to communicate with a human being, they must use supernatural abilities. This means that they can only clearly communicate with kannagi, although they might be able to send dreams to other people. If the kami use a form of telepathy, so that only the kannagi can hear what they are saying, they can communicate with the kannagi while there are other people present, and the kannagi can pass along what the kami has to say. As far as the other observers are concerned, the kannagi is speaking; it is up to them whether they believe that she is passing along the words of the kami.

While communication is one of the major benefits, kami can also use supernatural abilities to both help and harm a kannagi directly, which opens up more possibilities for a kannagi persona. If she has a good relationship with a kami, she can use supernatural abilities by proxy.

Personae do not normally start as kannagi, and the idea is that the game can be played indefinitely without any kannagi among the personae. Such personae communicate with kami through traditional divination or dreams, building relationships indirectly. However, the assumption is that most personae become kannagi sooner or later, as this makes the game more directly supernatural, and introduces new elements of play. One thing that needs to be determined in development is the default length of play at the pre-kannagi stage. This is something that a group can easily change; indeed, they could decide to start with kannagi personae. However, the rules will make a certain length of time natural, requiring house rules or determined effort in play to spend a significantly shorter or longer time as normal humans.

There is, however, a problem with this. If the personae cannot see anything supernatural to start with, the first sessions of Kannagara will not properly set the tone of the game: they will be much more mundane than the game is supposed to be. The second exception to the rule against seeing the supernatural is designed to cope with this issue.

Persona Transformation

I mentioned early on that, as far as logically possible, all persona options will be open to all personae at all times in Kannagara. That means that a persona who starts as a male human can become a female human, or a male kami, or a yuki-onna (snow maiden). It may be more difficult for a persona who starts with no blood relationship to any shinshoku to become the scion of line going back centuries, but changing the past is something that might come into the game at some point. At first, however, the “logically possible” limit is that personae cannot change their established past. They can certainly discover things about their past, and even discover that they were mistaken about their past, but the past itself cannot be changed.

That still leaves a wide range open. How does a male human become a snow maiden? Magic is obviously part of the answer, and will also be part of the answer as to how a male human becomes a female human, or an old human becomes a young one.

These transformations will not be easy, because I want a setting in which most people do not go through them. I want a setting, in fact, that looks, on the surface, like present-day Japan, and in present-day Japan men do not often change into women, or vice versa, much less into foxes. However, part of what makes the transformations rare will be that most people have no idea that they are possible, much less what to do to achieve them, and personae, of course, will learn that in the course of play.

Fundamentally, a transformation will be a kind of creation. The rules will be the same as those for creating a poem, or a table. The transformation may require high abilities, and will always require access to certain elements. Without those elements, the persona simply does not know how to achieve the transformation. The quest for the elements necessary to achieve a desired transformation is an obvious source of stories, requiring investigation and the solution of mysteries. Some transformations will also require other creations, as part of the process of transformation, and those creations also inspire stories. A transformation could also require the persona to form particular relationships.

Transformation will always be self-driven. That is, the kami cannot transform a human being into a kitsune, although they might be able to temporarily turn a person into a fox, under certain conditions. Other personae may help a persona with her transformation, and in some (maybe most) cases it will be necessary to get help, but the persona who will be transformed must be an active participant, and the driving force, if the transformation is to succeed.

Some transformations may have other transformations as prerequisites. For example, it may be impossible for a male human to transform directly into a yuki-onna. First, he must become a female human, and then he can undergo further transformation. I think I will want to make it impossible to reach “kami” quickly, as a matter of narrative pacing as much as anything else. That is, a human will probably have to pass through at least two other states before becoming a kami. You can make a case for this based on historical Shinto belief, but really this would be a modification to make things work better in a game.

In fact, at the moment I plan to have one transformation, from ordinary human to kannagi, be the prerequisite for all transformations. That will be the topic for the next post.

Teachers and Students

In Kannagara, teaching is an important part of the game. It is a way for personae to develop, and a way for personae to help characters to develop. In addition, the relationship between teacher and student is an important one, and relationships between people are another important feature of the game. That means that Kannagara needs good rules for teaching.

There are two ways that a persona can be involved in teaching: as the teacher, and as the student. In theory, we could use the same rules for both, but at the moment I would prefer not to. Having someone keep track of experience points for all characters is likely to be too big a burden, and characters cannot use the same advancement rules as personae anyway, as they do not have much opportunity to do things that are important to the personae’s narrative. So, at least to start with, I will create different rules for the two situations, and in this post I want to consider what happens when the persona is the student.

First, a teacher is a source of study, and as such has the same statistics as a book: experience point bonuses, elements he can teach, details of content, and features. However, a teacher is also a human being, and so the persona should be building a relationship with her teacher.

This could, of course, be a central part of the story, involving all the personae. The rules for the effects of building a relationship should certainly allow for that possibility. However, the default assumption has to be that the relationship with the teacher is part of the background of a single persona, and that while she might build a good relationship with her teacher, the other personae might know him as little more than a name.

This can work in two ways. First, the group could organise one-on-one sessions to deal with the relationship between the persona and her teacher. For certain groups, this might be a very useful option, while for others it might be incredibly disruptive. Therefore, this needs to be clearly possible, but also very clearly optional.

The other alternative is to have building the relationship be part of the persona’s downtime activities, but with an influence on play. For example, the persona discovers that, in order to improve her relationship with her teacher, she must do something, or learn something, and then that becomes an individual goal for that persona, one that she tries to achieve during the story. In general, there need be no secret about why she is trying to achieve something, nor need she do it alone. Indeed, it is often better if she is open and gets help, because that makes her teacher an important part of the background for everyone.

This is, you may have noticed, still very vague. This is because I have not yet discussed the rules for building and improving relationships, so I do not yet know what kind of rules go into this framework. Those rules are due for discussion soon, but there is one last point about advancement to consider first. In Kannagara, personae do not just become better at what they do, they can also completely transform themselves. How does that work?

Knowledge Advancement

Knowledges do not get better as you use them. Remembering what you know about Japanese mythology does not teach you more about Japanese mythology. Knowledges improve when you read books or attend lectures. Some knowledges can also improve if you just sit quietly and think about them, putting information you already have together and drawing new conclusions. Personae should be encouraged to pursue these kinds of activities.

The problem is that they are not obviously exciting to play out. “I sit in a comfy chair in a nice warm room, with a cup of coffee, and read a book for a couple of hours.” There is no conflict, no interaction, and nothing more to say. Certainly, the GM should not offer a lecture on the contents of the book. The same problem applies to practising a skill. “I sit in a comfy chair in a nice warm room, with a cup of coffee, and write norito for a couple of hours.”

So, the question is how to incorporate these activities into the game.

The first step is to award experience points if the player says that her persona is doing such things. Even at this level, the activity becomes a part of the game, as it is something that the persona does. It takes time, and so the persona should not get experience points immediately, but this is also a good thing. It establishes an ongoing task in the persona’s background, and that helps to flesh out the world.

This is not really enough, however. The next step is to define sources of experience, such as books or practice exercises. The personae gain access to these sources in play, and a source either grants bonus experience points, or access to an element. Realistically, sources should have a level, and grant more experience points the closer the persona’s current ability is to that level, but that might be too much book-keeping.

At this point, sources of study would have about the same level of detail as they do in Ars Magica. Years of experience have shown that this is a reasonable level of detail, and does make study quite important in the game. However, I would like to make it even more important in Kannagara.

One possibility is to define the subject of a source of study in some detail, and grant a bonus experience point if a persona, in the course of play, connects that subject explicitly to events that are important in the narrative. “I read about it in Hogwarts: A History” would be the simplest possible form (although not that exactly, of course), but it should be possible to do it a little more elaborately. In effect, this would allow a persona to gain experience points for using knowledges, as long as the player explicitly tied the use of the knowledge to whatever the persona was currently studying. This might work best if invoking the study allows the player to claim experience as if the roll were for a skill: one point on any failure, and a point on success if the difficulty is greater than four times the number of dice kept.

Another possibility is to give the source of study specific features, which could be as simple as a physical description of a book, or of an exercise, and give the player bonus dice to roll if she makes those features important to the narrative. That is, she uses the features to make the fact that the persona is studying from this source into part of the story.

This may be enough to add to the Ars Magica model. Study should be an important part of characters’ lives in Kannagara, but it should not dominate. However, there is still one important source of study that we have not considered: teachers.

Skill Advancement

Personae in roleplaying games have advanced, become better at what they do, since the hobby began. There is room for debate over whether this is necessary for roleplaying in general, but as one of the main themes of Kannagara is personal development, it is clearly necessary for this game. Indeed, advancement should be rather fast, because players should feel that their personae are developing.

Because development is one of the themes of the game, development should happen in play. This means that skills and knowledges must be handled differently, because personae do different things to improve them. Basically, one improves a skill by using it, but one improves a knowledge by studying.

This makes skill improvement easier to handle. The improvement process involves the persona doing things with the skill, and “doing things with skills” is what we play out in the game. So, the first point is that using a skill in the game will improve that skill. People learn from their mistakes as much as their successes, but do not learn much from succeeding at things that are easy for them. Failing at something that should be easy for you can be a great learning experience, of course, and you can learn a lot from both success and failure at a difficult task.

This is a place where the precise numbers will be determined by playtest, but as a first pass a persona gets an experience point when she attempts an important task with a difficulty that is greater than four times the number of dice she gets to keep. It does not matter whether she succeeds or fails. The chance of getting greater than 4 with one die is one in three, while the chance of getting 9 or greater on two dice is less than a quarter, so these are tasks at which failure is likely. Of course, the number of dice being rolled will also affect this probability, so personae with high knowledge scores might well advance while succeeding, but that is not necessarily a problem; as mentioned above, advancement should probably be fast.

Similarly, a persona gains an experience point if she fails an important task with a difficulty that is less than or equal to four times the number of dice she gets to keep. Succeeding at this level does not teach much, but failure certainly can. To put this another way, a persona gets an experience point whenever she fails at a task, and when she succeeds at a task with a difficulty greater than four times the number of dice she gets to keep.

Experience points go with a particular skill, and when a persona has a certain number of experience points, the skill increases by one die. I think I want to use a flat number of experience points per increase, so that it might always cost 10 experience points to raise a skill by one die, whether the persona has a skill of 1 or a skill of 10. However, the actual number involved will have to be decided as the game is developed and playtested. It might even be something that players are invited to change, to set an advancement speed that their group likes.

I noted above that these rules apply to “important tasks”. That means a task that is part of the narrative of the game. Players cannot just have the characters go off and fail consistently at hard tasks in order to acquire lots and lots of experience points. However, it is obvious that personae can go off and practise their skills in downtime, and that they should improve if they do that. This is a mode of improvement that also applies to knowledges, and so I would like to discuss it in its own post.

Ability Types

At the moment, I think that Kannagara will involve two kinds of ability: knowledges and skills. The distinction is straightforward: knowledges represent what a persona knows, while skills represent what she can do.

When creating something, a player will normally roll the persona’s knowledge, and keep the skill. The more you know about a field, the more likely you are to create something good, and the less likely you are to create something truly appalling. However, if you are lacking in skill, you will never do very well, while someone with great skill but poor knowledge might produce something great.

Knowledges are also used when checking to see whether the persona knows something, or understands the significance of something. In this case, the player rolls the knowledge, and keeps a number of dice based on the environment, and the hints that it gives towards the answer. A question with no context gives one die, and the difficulty to recall a fact is never higher than six. If a persona has studied a topic, there is a chance that she knows any fact, no matter how obscure. The difficulty to understand a situation may be higher, but as the situation itself provides more cues, the player also gets to keep more dice.

In principle, both knowledges and skills can be rolled or kept. Thus, there might be situations where a player rolls one knowledge and keeps another, or rolls one skill and keeps another. I suspect that rolling a skill and keeping a knowledge will be rare, but I don’t want to rule it out as a possibility. For example, there might be a skill “exam technique”, which personae roll when taking an examination, keeping the relevant knowledge. Given the topics of Kannagara, I think that rolling a knowledge and keeping a skill will be the most common situation.

In many cases, a knowledge and a skill might share a name. Thus, knowledge of norito and skill in composing norito would both be called “norito”. However, they should be written in different places on the persona sheet.

The issue here is ease of use. Having the same name for two abilities might be confusing, but calling them norito knowledge and norito skill is clumsy, and trying to come up with completely different names for every pair of abilities will become difficult, and lead to some strained terminology.

Abilities represent general ability, not particular facts a persona knows, nor particular techniques. Those would be represented by elements, at least in the creation process, and elements might well apply to single actions as well. Although someone with only one die in a knowledge has a one in six chance to know any general fact, some facts might be represented by elements, and unknown to any persona who does not have access to that element. Similarly, some techniques might be simply impossible if a persona does not have the relevant element to go with the skill.

The fundamental difference between the two types of ability not only affects how they are used in play, but also how they are improved. I suspect that I will not want to distinguish them during persona creation, but for later persona development the mechanisms will be different.

Jinja Personnel

One important feature of personae is their relationship to the jinja. Kannagara assumes that all personae have a close relationship with a particular jinja, just as Ars Magica assumes that all characters have a close relationship with a particular covenant, and the stories develop around that jinja.

There are several different close relationships that a persona can have to a jinja. The first is to be a shinshoku at the jinja. In this case, “priest” is actually a perfectly good translation, but as there are no good English terms for the other relationships, it would be odd to avoid the Japanese for just this one. Shinshoku are specially trained, which normally takes a few years, and there are several levels of training, levels of job, and levels of status that they might hold. Those will be important in the game, but I will deal with them later. Shinshoku lead the matsuri, typically performing the harae, offering the mikë, and reading the norito. Sometimes they also perform the kagura, but not always. In most jinja, the shinshoku also do most of the management tasks, and even in the largest the office is typically headed by a shinshoku.

For local jinja, the post of shinshoku is very often hereditary in practice. There are no rules requiring the son or daughter of the old shinshoku to take over, but it is made particularly easy if they want to do so, and many jinja want shinshoku who were born into a jinja family, even if that family is responsible for a different jinja. This means that shinshoku often have a very close relationship with the jinja, and with the surrounding area. However, there is currently a shortage of people who want to be shinshoku, which means that jinja cannot always get such a shinshoku, so some are new arrivals.

The next group of people are the ujiko. These are lay people, with no special training, but they live within a defined area around the jinja, and help to manage it. They normally provide financial support, and assist at large matsuri. In most cases, the ujiko’s family have been ujiko for generations, possibly for as long as the shinshoku family have been there, and maybe even longer. At some jinja, the ujiko managed all the ceremonies until the late nineteenth century, so that shinshoku are something of an innovation. Ujiko have a long-standing relationship with the jinja, but it is not always very close.

Sūkeisha are lay people who strongly support the jinja, but are not ujiko. Someone who does not live within the defined area around the jinja cannot be an ujiko, no matter how strongly she supports the jinja, and so will remain a sūkeisha. In many cases, someone who is not from an ujiko family cannot easily become an ujiko, even if she lives in the right area. There are no general rules restricting this, however, so the jinja can change this if the people in charge want to. In Kannagara, a sūkeisha might want to get the rules changed so that she can become an ujiko, or so that she can take on a particular role despite being a sūkeisha, if she does not want to move into the jinja’s area.

The final group of people are the miko. All miko are female, and they are almost without exception young, under 25. Historically, they communicated directly with the kami, but in the present day they assist the shinshoku. They wear a distinctive outfit, of bright red hakama (trouser skirts) over a white kimono. Miko need have no formal qualifications, but they often perform the kagura, so many of them learn to dance. Some miko, particularly at larger shrines, are actually fully qualified shinshoku, but have not yet been able to find a job as such. In Kannagara, of course, miko are likely to find themselves taking on their traditional roles.

Each relationship creates certain assumptions. A shinshoku knows a lot about Shinto and matsuri. An ujiko has a family connection to the shrine. A sūkeisha is deeply committed to the shrine. These relationships will be developed as the game progresses, and the jinja develops under the personae’s direction.

Persona Creation

Character creation is an important part of any roleplaying game. While it is sometimes thought of as part of preparing to play, it is part of the game in many cases, most notoriously in the original Traveller, where your character could die before character creation finished. For new players, however, character creation often does not feel like part of the game. It feels like study, and if you don’t fully understand the whole game you might make stupid decisions that land you with a broken character. This is why so many games provide pre-generated characters for new players. The problem with that, of course, is that the player cannot create her own character, and so might not have the character she really wants to play.

In Kannagara, I want players to create their personae during play.

At the beginning of the first session, the players know that their personae are human beings who live in Japan and have some association with a particular jinja, but that is all. All further details, such as sex, age, and specific relationship to the jinja, and all mechanical details, will be determined in play.

For background features, a player will define a feature of her persona when it matters, and when the session has made it clear why and how it matters. It does make a difference whether your persona is old or young, male or female, but the significance of each decision in the game is not clear until you start playing.

Mechanical features work on the same basis, in that players will choose them when they becomes important. In this case, however, the rules need to ensure that one persona does not become omnicompetent, leaving other players with nothing to do. The standard mechanic for this in contemporary roleplaying games is a pool of points, which must be spent on mechanical abilities. I intend to use something similar.

If the pool is limited, however, spending the points on one ability means that the player cannot buy others, and the player might regret her decision when a new ability is introduced. It seems unlikely that it will be sensible to introduce and explain all the available abilities before defining the first one, so I suspect that I will want to divide the pool of points. That way, players can spend all their available points on the abilities offered without jeopardising persona development, or their concept. This does mean that a group of abilities will need to be introduced before the first one is chosen.

A similar point can be made about the level of abilities. For simple tasks, where the options are success and failure, a persona who succeeded might actually have more dice available to roll or keep. In that case, the abilities can be raised later. If the persona failed, however, that ability is set. If the persona had more dice, she might well have succeeded, so we can say that her ability is capped. Similarly, once an ability has been used to create something, it is fixed, because we know what score the the persona contributed to the creation. However, the rules, and the structure of the initial scenario, should avoid fixing the ability scores before it is truly necessary,

Because Kannagara places a great deal of importance on building and improving relationships, personae should start with problematic relationships with other personae or characters. These are features that we want players to take, so rather than costing points, they should probably grant them. However, the number of relationships per persona should be limited, so there may be a second pool of points for relationships, and taking the relationship converts those points into points that can be spent on abilities.

This strategy means that introductory scenarios need to be carefully designed, to introduce abilities properly and allow players to make sensible choices. Of course, Kannagara makes it possible to change your mind about where you want your persona to go at any point, so mistakes at this point are not fatal, but it is still better to start with the persona you want to play. Introductory scenarios may cover more than one session, and ordinary scenarios may include opportunities to reveal more about the past of a persona. The first bit of actual game material for Kannagara will be an introductory scenario, to allow persona generation.

Experienced players can, of course, create personae by simply spending the pool of points on the abilities that they want. For an experienced player, creating a character in that way, carefully weighing up options and balancing things to get as close as possible to the concept you have, can be a lot of fun. It is one way of playing the game, and I want Kannagara to support it.

Persona Development

The development of personae is a very important part of Kannagara. Personae get better at what they do, and gain new abilities. Sometimes, they even transform into something other than what they were.

Because this is such an important part of the game, there is an important principle that I want to respect. Any persona, at any point, can work towards becoming anything that the game makes available as a possibility to any persona. The change might not be possible immediately, and might be difficult, but it is always possible. Furthermore, this possibility is part of the standard game; it does not need a special exception of any kind.

Most existing games break this requirement. Ars Magica, for example, provides no standard way for a character with no ability to work magic to get The Gift. Similarly, a character who starts off as a faerie cannot become human. The reverse is possible, but far from easy. In Pathfinder, a character cannot easily change race. Even changing between classes is difficult, as the first level of a new class counts as a higher level, and the new rules for retraining, which recognise the appeal of this requirement, are outside the normal methods of advancement — retraining takes time, not experience points.

Of course, it makes perfectly good sense to break this requirement. If you have fantasy races, it makes sense that characters will not be able to change their race. Real people can’t do that, after all. That means that Kannagara will be a little artificial. Fortunately, I have supernatural powers available, and those will allow personae to make the necessary changes. Sex changes will be possible, but so will more important changes: humans can become kitsune, kitsune can become kami, and so on. Further, no change is irreversible.

That does not mean that all changes will be equally appealing to all personae. Some changes will build naturally on each other, so that once a persona starts down a path, continuing down that path will seem like a better use of time and effort. Similarly, it may not be possible for a persona to be a particular combination of possibilities at the same time. Kitsune are an obvious example. A kitsune is a fox who can take human form, so it is not possible to be a kitsune and a human at the same time.

New persona options will be introduced along with the ways in which personae can discover that they are possible. In that way, players will never feel that their personae would have made a different decision in the past. A player might wish that her persona had known about a certain possibility earlier, as might the persona, but it will never be the case that the persona did know about that possibility even though the player did not, and it will also never be the case that the possibility is no longer open to the persona.

There is a second important principle. Development should be part of play. That does not necessarily mean that the group will play out all parts of development, although they will do so for some. It does mean that development will not be abstracted, and that personae and players will make the same sort of decisions. Development in Ars Magica does follow this principle; characters make the same developmental choices as players, deciding what to study or research in a given season. Pathfinder does not; the details of development are abstracted. Kannagara will be closer to Ars Magica, but with even more emphasis on playing out character development.