Speak Up! Creating Meaningful Characters Through Dialogue

  • 14 May 2019
  • Nick Jones

Perhaps one of the most important aspects of writing for interactive narratives – in particular, video games - is dialogue. This is because dialogue is the primary way we as the player and/or viewer get to understand character. You can have the best graphics, the greatest gameplay loops, and the deepest lore, but if your dialogue comes across hokey, then your narrative will never get off the ground.

So how do we write good dialogue? With the pumping of Hollywood screenwriters into the games industry throughout the past ten years or so, many are convinced that game writing follows the same rules as writing for film, and while there are definitely some good tools we can manipulate from that genre for our game narratives, writing dialogue for interactive content really is its own beast. It requires a different vocabulary to do well.

Having studied dramatic screenwriting at a masters level and worked on commercially and critically successful games, I will in this article outline the things I believe you need to know about writing dialogue to develop dramatic characters for games, giving you some practical tips along the way.


The Different Formats of Dialogue for Games

In his 2012 GDC talk “Choice Architecture, Player Expression, and Narrative Design in Fallout: New Vegas” Game Designer, Josh Sawyer talks about three main approaches to writing dialogue in video games. He refers to them as the Linear/Cinematic Approach, the Keyword/Subject approach (which I tend to refer to as the Lists and Lore Approach), and the Branching Tree Approach. (Sawyer, 2017)

While there are others out there, these three styles give us the major formats used in interactive narrative media. As you will see, these approaches aren’t a situation of “either/or” and often times a narrative can use two - potentially even all three - in tangent with one another. In fact, arguably a good game should switch it up a bit to create some diversity.

Regardless of which approach you use, or how many you use, it’s important to know what they are as well as the advantages or disadvantages they might bring to the table.


Linear/Cinematic Approach

Beginning with the cinematic approach, we can from the get-go state that this is perhaps the most popular format in video games today. It focuses on removing control from the player periodically to play them a short-film, updating them on the narrative state of the world. Once this update has concluded, control is given back to the player to continue onwards in their quest.

This style is seen overwhelmingly in AAA titles because for the many studios that are producing AAA games, being akin to the next blockbuster film is what’s important. A cinematic approach maintains a high level of authorial control, which allows the narrative to be potentially complex and filmic in nature. The other major advantage of this format is the ability to really leverage voice acting and graphics to produce high quality sequences, where characters act and react dramatically towards one another.

However, taking this approach is not necessarily the greatest as it’s all about trying to fit a square peg in a round hole. A cinematic approach is, as the name suggests, a cinematic technique and games are not cinema. In a way, games represent the antithesis of cinema. They represent a space where the audience has the control and power to influence and change the story world around them.

Cut Scenes, as I previously mentioned, try to remove player agency. Player interaction is either rendered null and void, or made to be very limited. In the case of an interactive cut scene, players may be able to move around, but they are unable to do anything else to influence what’s going on. The NPCs will talk at them until they’ve finished their monologue, and that’s all there is to it.

There is also the risk of over-using cut scenes. In fact, a reliance on them is detrimental to player engagement. I call this the “Kingdom Hearts effect.” If you’ve played those classic Disney/Final Fantasy crossover RPGS, you will be very familiar with what I’m talking about. Particularly in the beginning of Kingdom Hearts 2, and well… ALL of Kingdom Hearts 3, you’ll find yourself running forward a few steps, being confronted with a cinematic (losing control), running forward a few more steps and being confronted with another, and so on and so forth. To the point where, in the case of Kingdom Hearts 3, all the cut scenes amount to a whopping ten or eleven hours of content! (Kingdom hearts 3 - All cutscenes in english (full movie), 2019) To put that into perspective, Square Enix believe that a regular playthrough of the entire game will take thirty hours to complete. (Asarch & Square Enix, 2019) That works out as roughly 33% of the game being unplayable!

Kingdom Hearts can get away with this because it’s a beloved property with millions of dollars invested into it, but for most of us, that would be totally unacceptable to our players.


Lists and Lore Approach

Moving on, we have the Lists and Lore approach, found in games such as Diablo III, The Incredible Adventures of Van Helsing and my own Path of Exile. It also features heavily in the MMORPG genre as it is a very quick and easy way to get across expository lore to players in manageable chunks.

How it works is simple. When a player interacts with an NPC, they are presented with a menu of choices. Each choice with be a short sentence or keyword about a topic. Clicking on that sentence or word will prompt the NPC to wax on in a monologue, explaining to the player all about that given subject before returning them to the menu to either ask about something else or leave the conversation.

While this approach works well if you’re trying to get across a vast amount of information, it doesn’t fare too great in terms of dramatic potential. The characters become less like actual people and more like robots or walking, talking encyclopedias of content.

Players are also at risk - especially in the case of sloppy writing - of missing out entirely on important information if they click through or don’t investigate a particular subject. This can cause a lot of confusion later on when they’re trying to place themselves within the framework of the story.


Branching Tree Approach

Lastly, we have the Branching Tree Approach – a format of dialogue that is arguably surpassing the cinematic approach in popularity. This approach is powerful as it is very good in creating a sense of player agency within the story. When done right, Players come to feel like they are directors of the narrative and that their choices within the game actually matter.

For those of you who are unfamiliar with it, branching tree dialogue functions very similarly to cinematic dialogue, however with one key difference. It gives players options on what to say to other characters. Each of these options branch off into different versions of the same conversation and can affect how characters feel about you the player, how they make future decisions, and in theory, ultimately how the story itself concludes.

This approach allows for a lot of diversity, as well as bigger more complex series of stories to be told. And yet, this potentially powerful allowance is also the Branching Tree’s greatest downfall. If you branch out too much, with too big divergences, then you can very quickly have multiple versions of the same game going at once. This creates an insurmountable amount of work for everyone in the development team and often results in certain “versions” of the game being flakier than others.

The other thing to watch out for in this format is being too “loud” with the illusion of choice. I touched on this in my previous article (here) when I wrote “We tend to think of video games as being the one true form of interactive, non-linear narrative, but in truth, what most video games present to us are illusions of interactivity and non-linearity… …a video game might present you with two, three or even four paths to choose from, but this isn’t really freedom to create your own story – you’re working within the confines given to you by the developers.” (Jones, 2019)

Divergences are intended to create the sense that the decisions a player makes will result in different options yet diverging too greatly or too often tends to result in the exact opposite feeling. This is because, unless you are really committed to creating a gigantic epic, those big divergences will take up so much of your time that all your smaller ones will bottleneck back into linear progress, making obvious to the player that their decisions don’t really have much effect at all. On top of that, the bigger the divergences becomes, the harder they are to maintain, so those large changes run the risk of crashing into the ground.

Branching Tree dialogue choices need to be clear about the result a certain option will create. The now infamous “Glass him” option in Telltale’s A Wolf Among us demonstrates perfectly how an unclear choice description can result in a seemingly surprising - and immersion breaking - result.

Player Agency

Once you’ve decided on a format or formats to use in the creation of your games dialogue, you next need to start thinking about player agency and how you can give a sense of control to your player in the acting out of these scenes.

Once again, Josh Sawyer provides us with a very good framework here. He lists four considerations you should make when deciding on how to show to the player that they are affecting the world around them. (Sawyer, 2017)

Firstly, consider Story Agency, that is, how we grant the player agency on where the story ultimately ends up, or how characters ultimately feel about the player. For example, a Player’s decisions on how they present themselves to other characters or in certain situations might open up and/or close down different narrative pathways, leading to different narrative endings.

This obviously works well with the branching choice approach, but can be adjusted to fit other formats as well. Consider with the cinematic approach, how a player enters a cut scene, or where they enter from. Perhaps the player wearing a certain outfit or carrying a certain weapon or item that a character can see, changes how that character feels about them.

An example of how this could work can be found in From Software’s Bloodborne with the outfit “The Black Church Set.” In the game, if the player wears this outfit, then the worlds NPC’s will react differently to them. Certain enemies upon seeing the player will shout “Death to the minister!” (Miyazaki, 2015) Adella the Nun is also inconsolable until the player approaches her wearing the outfit, upon which she will calm herself and speak with them. (Bloodborne Wiki, 2017)

Similar to player agency, Character Agency means to grant the player options on the kind of person they want their character to be. This is an opportunity to let players roleplay, much like in a D&D campaign. Allowing them to play their character as an asshole or a thug through dialogue (and again outfits and equipped items) gives them a greater sense of ownership in the narrative.

The Witcher III: Wild Hunt has a brilliant example of this, where it gives the player some truly funny and raw options in dialogue, such as the infamous “I can’t believe we fucked!” approach. (Tomaszkiewicz, 2015)

Consider the players ability to make Strategic Decisions – using foreshadowing and development of relationships between characters and the player to force the player to make strategic decisions about the world around them and how they conduct themselves in it.

Perhaps, in a politically charged GoT style narrative, the player senses a growing distrust occurring in a character that they believe to be their ally. This can be done through the dialogue and how both player and character agency affects it. Build your game and narrative so that they can then make a strategic decision (preferably unprompted) about whether to assassinate that character before they turn on them, win back their loyalty or remove them from the playing field altogether.

If the player responds to an NPC in a certain way, then cut off certain dialogue options later on. Set up your character relationships in a way that gets the player to think about what they’re doing, why they’re doing it and how it will change the future.

It’s important to note here though that there should be no “fail state” if a player chooses a certain direction in how they play. Writing for video games well is a difficult task because it requires multiple version of scenes and endings that feel equal to one another and are written almost as if they’re the only pathway. As we say in screenwriting, the audience needs to look at your ending, be surprised by it, but sit back and think “ah, it was always going to end this way.”

Lastly, and simply, use some good One Liners. Taking a step back from main characters, make use of your “extras” – the NPCs that populate the world who have little to no part in the plot of the game. If the player is rude to the faction leader, then have faction members occasionally shout insults or act coldly towards the player when they’re nearby.

This approach allows the world to show how it is being changed by the player’s choices. Being repeatedly teased and called “Chicken Chaser” in Fable because of kicking a chicken that ONE time is a great example of this in action. 


Developing True Moral Choices

I want to build further on the branching tree path we’ve been discussing now, with the addition of Moral Choice options. Made popular by games like the Fable series, Life is Strange 1 & 2, Telltale Games and Mass Effect, Moral choice options are incredibly popular in today’s gaming and function exactly as they sound: the player is given options designed to reflect and build the moral character of their avatar.

This being said, Moral choices aren’t as simple to develop as you may first have thought. It’s important to understand exactly what a Moral Choice is and what it is not. In a purely “dictionary definition” context, I would define a Moral Choice as being a choice offered to the players that causes them to have to pick an option that is often neither right, nor wrong in the context of the story world. No matter what option they choose, it will have both obvious positives (usually in the form of physical or social gratification, often immediate) and negatives (Always in the form of an immediate or gradual destruction of personal values or belief systems AND SOMETIMES the destruction of physical and/or social gratifications).

However, let’s dig a little deeper.

I would argue that a successfully written moral choice is one that the player is in CONTROL of. Therefore, a moral choice made under the threat of immediate death to the player, or in a heightened situation, or while the player is “drunk,” or under the influence of an externalized force (drugs, booze, possession, mental illness etc.), does not work because we as audiences and players are mostly incapable of empathizing with or - as is the case of with interactive narrative - embodying a victim character. We all like to see ourselves in our favorite protagonists, and we are all, to a degree a little vain. We want to see ourselves in protagonists that are making decisions for themselves and not just having stuff “done to them” or around them.

Within other media contexts, such as a TV show, a character can make a moral choice involving life and death because they are a part of the world. However, in something where players are the characters, this no longer works because of player agency and the suspension of disbelief. Without proper character development, it’s not a moral choice for a player to kill a person infected with a zombie virus for example, because we know on a certain level that this is not real. We are simply making a strategic decision - one without emotion.

A Moral Choice being made about the fate of another character should be presented only if that person has been developed as a character that we empathize with. Only when we empathize with a character, will the choice concerning their fate become a difficult one to make.

Also, I would also argue that any decision which is timed, also is no longer a moral choice, rather it is a choice in resource management - i.e. “she gets to live because she’s better with a gun.” Stress and pressure reduce empathy and increase rational, reactive thinking. That’s not to say that you can’t set up resource management choices as well, but it isn’t what we’re discussing here.

Almost always in video games, the end goal for the player is to take part in an empowerment fantasy of some kind - a story world in which horrible things happen, but that we can act against using our wit, strength, character and morals (or lack thereof.)

So if you want my two cents, I would suggest that all moral choices in a piece of interactive narrative should be constructed based off the following template:


  1. The player should be presented with a choice between holding on to the values that either they, or their character have held prior to the conflict of the story, or taking one step closer to becoming worse than the bad guy/s or monsters that exist within the story world.
  2. If the player chooses to take that step towards eventual “monsterdom,” they are then rewarded in the immediate, but in the long term will be made to suffer (physically, psychologically, socially or morally.)
  3. Or: If the player chooses to hold onto their values, or their characters authored values, they will be granted the moral satisfaction of staying true to the goodness inside themselves, and suffer in the long term (and in some cases, the short term as well.) This suffering will manifest physically, psychologically or socially, but not morally.
  4. Lastly, a good moral choice should have no obviously “right” resolution to it. There is no happy ending.


To make things easier, all you need to do when writing this kind of dialogue is remember the three laws of moral choice:

  1. The player is in complete CONTROL of the decision.
  2. The player has TIME to weigh the pros and cons.
  3. If the choice regards the fate of another character, the player must EMPATHIZE with that character.


Writing Good Dialogue

With all that being said, how do we write good dialogue in general? Our friend, Josh Sawyer suggests that we should always a) consider the choices that a player can make, b) define the player/character expression (namely, how can the character be played as? How would a character respond to this situation?), c) Establish the Narrative Goals of your story (what are you trying to say with your game? Are there any overarching themes or morals?), d) Gain perspective on the interaction between the previous three steps, and finally, e) write your dialogue. (Sawyer, 2017)

I would add to this by stating the following: I’ve found that video game dialogue is at its very core, all about figuring out how to convey as much narrative information as possible to the player, in as little amount words as you can. You want to showcase all that lore and deep lore that you’ve created via believable dialogue. Your characters are not versions of your story bible – they should never “tell” this information to the player. The player must interpret it for themselves.

A good way to do this is through the use of things like one liners, sayings, proverbs and mantras. Think of “Praise the Sun!” from Dark Souls or “Fear the old Blood!” from Bloodborne. These lines are short, but they work because they suggest deeper meaning and are repeated often throughout the game (almost like a chorus), allowing these concepts to become ingrained in the minds of players.

Lastly, I would suggest taking a page from traditional screenwriting – weight your dialogue lines like a scorpion – put the stinger at the end. Writing dramatic dialogue is not only about trying to figure out what to say, but how it should be said. Instead of “I hate you! Leave me alone!” you write “Leave me alone! I hate you.” If you weight the most powerful words at the end, it creates a kind of mini-suspense line that gets the audience or player asking – what happens next?

The last piece of advice I can give is with conversations between characters, treat dialogue like actions. Characters in dramatic stories use dialogue to get what they want. So whenever someone is saying something to someone else, ask yourself – what is this person trying to get the other to do? What is their agenda for saying what they’re saying?

In the end, your narratives will not be remembered for their stunning set pieces, their adrenaline fuelled gameplay or even their intricate lore. No. Your narrative will stick in the hearts and minds of players for one reason, and one reason alone – the relationships they had with the characters, and at the end of the day, that’s what dialogue is all about.



Asarch, S., & Square Enix. (2019, January 30). How long does it take to beat 'kingdom hearts 3?? Retrieved from https://www.newsweek.com/kingdom-hearts-3-play-time-kh3-how-long-1311834

Bloodborne Wiki. (2017, May 23). Black church set - Bloodborne. Retrieved from https://bloodborne.wiki.fextralife.com/Black+Church+Set

Jones, N. (2019, May 10). A visual feast: Learning the movements of environmental storytelling. Retrieved from https://www.thepunkwriter.com/article/a-visual-feast-learning-the-movements-of-environmental-storytelling-for-video-games

Kingdom hearts 3 - All cutscenes in english (full movie) [Video file]. (2019, January 30). Retrieved from https://www.youtube.com/watch?v=dzh8W03FkpE&t=9893s

Miyazaki, H. (Director). (2015). Bloodborne [Video game]. Japan: FromSoftware.

Sawyer, J. (2017, April 6). Choice architecture, player expression, and narrative design in fallout: New vegas [Video file]. Retrieved from https://www.youtube.com/watch?v=LR4OxNfzTvU&t=212s

Tomaszkiewicz, K., Kanik, M., & Stępień, S. (Directors). (2015). The witcher 3: Wild hunt[Video game]. Poland: CDProjekt RED.



Share this post