How to Write Characters

For anyone who clicked on this wanting to know how to write ASCII characters to an LCD screen, I apologize in advance for misleading you. This post will discuss some of my recent thoughts about how people should go about writing good characters in visual novels and why it works. Specifically, I'm going to go over a few different things I do in order to write mine, in order to achieve what I believe is a well-written character based on some of my favorite games, books, and TV shows.

So I think it would help if I explained how I came up with the main cast of characters for the original Detective Butler. By definition, we need Butler. Then, the premise revolved around Gilligan and his father, Galvano. After that, we needed suspects: I decided to evenly divide them among genders and among board/crew members. So three were with the company, three were with the ship, three were male, and three were female. The total number of main characters at this point was nine, which was one less than ten, and that seemed suitable for a medium-sized mystery.

After that, I decided which one should be the culprit, and made their personality suitable for that. I gave each character a particular quirk to make them stand out: the best friend, the lover, the stoic, the charismatic captain, the clumsy maid, the wise old man. There was a convenient variance in age groups as well, ranging from 18 to 60. Then, as I wrote the story, I was able to dig deeper into each of their personal characteristics and create their individual backstories, and of course, incorporate them into the mystery.

There are some particular qualities which make for good, life-like characters.

Let's go into each one in detail:


Each character must be different. No person in real life is the same. Therefore, there should be at least one thing which makes a character stand out from all the others. Furthermore, no character should sound the same. Their diction and vocabulary should have variance based on the kind of person they are. A sophisticated person might go on for a while in long sentences using word you'd never heard of in a British accent, while a more down-to-earth character uses odd expressions and maybe has a Southern accent. Their speech patterns should also vary. For example, Gilligan mostly speaks how I would. While Butler talks. Kinda like this. A-and Cecila, s-she talks like this! Each character then has a distinctive voice that the reader will remember.

Additionally, each character should be visually distinctive. Perform the "silhouette test" by simply looking at the silhouettes of your cast. How recognizable is each character from the others? Body language and posture will affect the silhouette, and are also key ways of conveying the personalities of your characters. More dynamic poses than the generic "standing there with arms at the side" do wonders, which is why I included such "secondary poses" in Detective Butler. Also, when making your characters visually distinctive, run the "cosplay test". What would make you recognize this character at an anime convention? What are they wearing that makes them stand out from all these other fictional characters (and the "normies" too)? Consider what kinds of things the character likes or represents and try to incorporate that into their outfit.


Now that we have created different characters, a result of this difference will naturally be that these characters will conflict with each other. They should have their own individual goals and thought processes, and most of the time these will -not- blend together nicely. Good story arises from good conflict, and good character development results from conflict between two characters. Even during a scene where no plot development occurs, you can show sides of each character by having them argue or disagree over a particular issue. This also makes your characters resonate with the reader more, because now your reader can relate to at least one of them. If you only present one side of an argument, you're excluding half your readers; by providing two characters with differing points of view, you can appeal to both. For me, this works out very conveniently with Butler and Gilligan, who are near polar opposites. The former is more logical and cynical while the latter is emotional and optimistic. So under any given situation, they will almost certainly be at odds with each other.

One great example of character conflict is the television show Breaking Bad, where everyone has their own goals which are constantly in conflict with each other. The closer one person gets to achieving their goal, the more it sets back someone else. Walter, Jesse, and Hank have a particularly interesting dynamic: Walter and Jesse need to cooperate to survive while avoiding Hank, who needs to catch them. The closer they are to achieving their goal, the more risky it is and the more likely they'll be caught by Hank; but in addition to that, Walter and Jesse are constantly at odds with each other, and so it's difficult enough for them to even work toward their individual goals.

Conflict is also the way in which we see characters grow. As we watch people overcome conflicts, we cheer for their victory. We want them to win. With each win, we build a connection, realizing that they (or we) have overcome a significant obstacle and are one step closer to their goal. This is something we do in real-life relationships as well; this is how we bond, by sharing our struggles. We're sympathetic to the cause of our characters, and we empathize with each of the many setbacks along their journey. Therefore, creating conflict in the story is a necessary part of crafting good characters, because this conflict is what changes our characters over time (and therefore keeps them interesting).


Indeed, so as characters change over time, so might their beliefs, emotions, and actions. People always have the choice to take a particular action or not; it is only that we assume people will act in ways they have previously acted (which is still a very good indicator of future behavior). However, people might do different things for different reasons. They might change their mind; they might be turning over a new leaf. Or they might just be doing something different "just this once". It is very easy, however, to keep a character trapped inside the bubble in which you have created them. After hand-crafting their personality and characteristics, you might automatically assume there are certain actions that this character would never, ever take. This is wrong.

As I said earlier, undergoing conflict during the story forces characters to change -- it forces characters to evaluate themselves and their actions to see if they should remain the way they are. People continue to do the things that work well -- if something is not working well, they stop doing it and try something else. Eventually, over enough time, this could result in people doing things they never would have done in the past. This is why having a long enough story is important; in all these short stories, there is not enough time for conflict, character development, or any contradiction.

Good characters will contradict themselves. They will not make sense. This seems odd, but it's true -- just look at real life. How many times has someone done something which you cannot possibly rationalize, given their past behavior? People are unpredictable; there is no real way to tell what a person might say or do next. Some might say this is "out of character", but the key difference here is that out of character actions have little to no justification for occurring. When a character does something contradictory, it is not necessarily out of character if there exists a reason for which this person decides to act opposite their usual position. We do not necessarily need to be shown or know this reason right away, but at the time of writing it must exist within the mind of the author.

I included at least one very contradictory character in Detective Butler -- the culprit -- and people responded to that quite confusingly. Sometimes this character was their favorite, and other times, this confusion was the worst part. But that was my point -- this character's contradictions have caused the reader to emotionally react, and that is exactly what I want out of my story. Sometimes people are contradictory; their past actions do not align with their current ones. But isn't that how people grow? How can you ever grow if you continue to make the same choices as you did yesterday? Therefore, realistic characters must indeed leave their own comfort zones and take actions they may not necessarily would have taken in the past, or would even want to take in the present.


Now, with all that said, this last thing is the most important: keep it a secret! By revealing all your cards at the start, there's no point in writing the story. So many people love to share biographies of their characters on their website or whatever, but that entirely defeats the purpose of reading. If you do such a thing, do not get carried away with it -- or perhaps, deliberately mislead your readers. But certain things need to be kept a secret and a mystery in order to create depth in your story and characters.

Depth, as I am using the word, means the eventual uncovering of information not yet known. As you dig deeper into a story, you learn more about the characters, and it is this process which makes it all worthwhile. If I write you a summary of a story, it's not going to be as engaging as reading the actual thing. It is the slow process of uncovering a mystery that excites people. Therefore, it is not enough that you create characters with meaningful backstories, but you must also reveal those backstories in a clever and meaningful way.

This is where you must integrate your characters with your story. You could for example, take Umineko's cast of characters and dump them into a different story, and we would know an entirely different set of facts than we currently do. Specifically, if there was no murder mystery and was entirely slice of life taking place at a school or something similar, we would not see all the various (mostly dark) sides of each character. People act differently under different circumstances, and it is a fallacy to assume that your characters are being totally honest in front of the camera 100% of the time. They are lying, even to you, and even to themselves, only bringing out what they want the people around them to see. This is why we must follow the characters through a variety of conflicts and situations, because only then can we see these different sides and average them into the whole person.

It is particularly because the characters experience the things they do over the course of the story which allows us to peer into their backstories, and the significance of those events shapes them into the characters they are at the beginning. Use the premise of your story to emphasize the interesting qualities of your characters, and advance the plot in directions which shed light on your characters even more.


Whether or not English visual novels, compared to other forms of media, make particularly good use of these points is probably a matter up for debate. However, I would argue that there is considerable work to be done in that area. There is (on average) very little mystery, minimal conflict, shallow difference and the wrong kind of contradiction (if any). Quite honestly, my standards for EVNs are so abysmally low that I have almost entirely lost interest in playing them. I can't play an EVN without worrying about the game crashing, or being filled with spelling mistakes, or just being cringeworthy to read. Japanese visual novels have significantly higher standards and are still fun for me to play. I would argue that Western developers should take note at what the Japanese are doing and apply it, because currently, I don't think EVNs are headed in a very good direction.

My current prediction is that we're converging toward EVNs which perpetuate bad design. What I mean by this is that I'm noticing some bad trends in design which are not going away any time soon. Firstly, there's the trend to show off and describe the main characters in the opening post. This almost always eliminates depth from the story because the author reveals way too much information before I can even download the game. Also, most descriptions of the actual plot are very lacking, ironically in contrast to the character descriptions, which include all these minor details such as sexuality and gender (which almost always ends up being some snowflake buzzword if it's not immediately obvious from the character art). I'm not quite sure when these factors became important for me to know about a character before reading a story, but clearly there is some assumption that we are supposed to enjoy the game for who is in it rather than the actual merits of the story. Just because your game has a "diverse" cast doesn't mean they're well-written.

In the actual games themselves, characters have a tendency to remain in the box the author puts them into. This is a very key point. There is no conflict or growth, possible for a few reasons: the game is so short you can't have time for them to grow, or the characters themselves avoid conflict at every opportunity. Or rather, the author fails to take various opportunities to create conflict. I can only guess from this that these characters are extensions of the authors, in that they do not want to generate conflict either, preferring not to voice criticism but err on the side of caution.

There is typically a failure to give each character their own distinctive voice. Many character designs are also bland and forgettable (possibly because 90% of EVNs take place in a boring and forgettable setting). Typically, there is very little discussion regarding varying points of view on a particular subject. If any philosophy is discussed, it's through a third person narrator, not a character in the game who can be challenged by another. In other words, it's all totally a soapbox for the author.


I'm going to suggest that there is a very dominant audience for English visual novels -- and this audience mostly consists of fellow creators themselves. As Super Mario Maker has taught me, a level can be fun for you to make but not fun for others to play, and indeed I think this is the case for most English visual novels. The majority of EVN creators are a particular way, and honestly, I'm not that way, so what works well for them doesn't work well for me. And so naturally I'm really not interested in what their stories might be; I am not their target audience, and most likely, they are not mine.

My last bit of advice here is to stop reading other EVNs and just go outside. Observe other people and how they interact. Compare and contrast people of varying viewpoints. Read up on a bit of psychology (just a bit, not too much) to understand why people do things. And of course, use your own brain to analyze situations and form your own ideas of how things work. Everyone has their own original thoughts to contribute, and I get the feeling that 90% of the time what they're doing in EVNs is just rehashing what they've seen someone else do over and over without thinking.

I would like to see more games which make extensive use of the four points I've outlined. I want to see longer stories, with lots of conflict and more complex characters I can get invested in. I want that investment to be worth my time, to have a payoff. I want an EVN to convey a message that was created from intellectual thought, not fulfill some escape fantasy. If there's even one good EVN that can make me have some faith in the medium (you mean genre, right?) then I'll gladly take it. For now, I'll be working on my own game which will absolutely feature all of these things, because I make games that I want to play, and I'm not really getting that from anywhere else.