In Sub Terra, players assume roles chosen from a team of eight unique characters.  Here are five of them:

From left: Ishiko (climber), Jai (bodyguard), Amirah (leader), Daniel (geologist), Jack (Diver)

I believe that diverse communities are stronger, and that inclusive representation in media can nudge communities towards diversity.  ITB shares this view, and this is why we worked hard to ensure that the cast of Sub Terra covers a broad cross-section of humanity.

But why do I think this?  Part of it is definitely due to my mixed-heritage upbringing and the progressive filter bubble I happily live inside.  But as I don’t like taking things on faith, I’ve also found compelling arguments to support these statements.  I’ll try to explain these in this post.


Diversity makes groups more efficient.  This appears to be for a number of reasons:

  • Larger Talent Pool
    Because we’re all different, we’ll each contribute different amounts to the success of our group.  Let’s assume that the benefit of an individual to a group isn’t correlated at all with some feature they possess (gender, age, ethnicity…).  If we exclude people based on that feature, we’re shrinking the available pool of potential recruits in such a way that the quality of the group will be reduced.

    Say you’re rolling five red dice and five blue dice (the talent pool), with the aim to pick the three dice that sum to the highest score (your group).  If you throw away most or all of the red dice before you roll, your group is likely to get a lower score than you would if you rolled all ten.

  • Productive Conflict
    Diversity increases the amount of conflict in a group due to the wider variety of opinions, but if managed properly this is a strong asset.  Potential solutions are critiqued more thoroughly and end up more robust and comprehensive.

Diversity also makes groups happier:

  • Reduced Toxicity
    Let’s assume we can divide a group into two parts, Reds and Blues.  Let’s also assume that 10% of the whole group are assholes who like to harass members of the opposite subgroup (i.e. 10% of Reds like to harass Blues, and 10% of Blues like to harass Reds).

    Suppose we have a group consisting of 90 Reds and 10 Blues.  There are then 9 asshole Reds targeting the 10 Blues, but only 1 asshole Blue targeting the 90 Reds.  Being a Blue is disproportionately unpleasant (0.9 harassers-per-target), as they basically have a dedicated harasser each.  Being a Red is still pretty great, as there’s only one harasser dividing their attention between a huge number of them (0.01 hpt).

    Suppose instead the ratio is more even, at 50 Reds and 50 Blues.  The harassment suffered by the Blues is greatly diluted, and not raised that much for the Reds (0.1 hpt for both).  Putting both sides in the same situation also makes it easier for the once-dominant Reds to empathise with the previously-victimised Blues.  Diversity helps.

    This model of “supply and demand of assholery” is known as the Petrie Multiplier, and is a plausible explanation for why women receive considerably more harassment in male-dominated domains.  You might have seen it in action at local gaming tournaments, and you’re probably aware of something similar happening on Twitter when someone raises their head above the parapet.  (There’s also no reason why this is limited to just gender – any targeted minority group can suffer from this)

  • Increased Tolerance
    If we don’t know much about a certain group of people, then we tend to fall back on stereotypical impressions of that group gleaned from what little information we have.  If the only contact with Islam you have is the constant threat of terrorism on the nightly news, you’re going to start thinking of Muslims as terrorists, because statistics and base rates don’t register with our primitive monkey brains.  

    This makes sense – doing the math is hard, and relying on gut feelings is easy.  We can only keep around 150 social connections active at any one time, and if they’re all very similar to us then we have to fill in the outgroup blanks with stereotypes and prejudice obtained from our social echo chamber.  (Note that stereotypes and prejudice don’t have to be negative – you can have a favourable opinion of a group without personally knowing anyone who belongs to that group)

    If, instead, a community is more diverse, then suddenly you’re going to interact with people from minorities on a regular basis.  Because these people are most likely similar to you and not the extreme that appear on the news, your mental model will start to shift.  This is known as the Contact Hypothesis.

Unfortunately, there’s a problem here.  A lot of existing communities aren’t particularly diverse, in that the ratio of people in them don’t match that of the wider population.  For example, the ratio of male to female Magic players is around 2:1, yet in tournaments this plummets to around 20:1 or less.  If diversity is so great, then shouldn’t we expect groups to naturally mix instead of remaining divided?


There are a number of factors resisting diversity and enforcing the status quo.  These include:

  • Intolerance
    As mentioned above, people with low exposure to minority groups are more likely to view them unfavourably based on extreme examples in the media (terrorism, crime) or community opinion (sexism, political views, anti-semitism).  The majority therefore doesn’t want to include minorities in their community, and minorities don’t want to join a community that’s hostile towards them.  Victimised minorities are more likely to leave a community than to stay and change it for the better.
  • Unconscious Bias
    This is related to the previous point, but is less obvious.  People’s expectations are shaped by their environment.  If a group is already mostly men, then people (of all genders) will unconsciously assume that men are better suited to that group than women.  This is of particular importance when hiring staff, where candidates with equivalent background and skills will be unconsciously discriminated against based on their gender or race unless a deliberate effort is made to hide this information.  It also affects minority perceptions of themselves (e.g. “It’s not feminine to be competitive”).

    The same applies to gaming groups.  If your group is entirely men, then you’re going to unconsciously assume that women are less interested in games, and you’re probably not going to be aware that you’re making this assumption.  If you’re playing online and your opponent is anonymous, then you’re probably going to assume they’re male by default.  This effect requires conscious effort to overcome.  (Don’t think this applies to you?  Test yourself!)
  • Racist Shapes
    Check out this cute interactive simulation about squares and triangles.  I’ll wait.

    Back?  The Schelling Segregation Model in the simulation takes similar assumptions to ours – namely, that people like being in diverse groups but hate being part of a minority – and shows that this inevitably leads to society-wide segregation as the unhappy people move around.  Additionally, even if people don’t mind being part of a minority, this won’t magically cause segregated societies to diversify.  This is depressing.

These factors conspire to keep things where they are, even if we’d all be better off if we mixed things up a bit.  So how can we do that?


Let’s go back to Schelling’s polygons.  The model shows that if the only pressure is that people will move groups when they find themselves in a local minority, then groups will naturally segregate.  But it also shows that if people decide to move when they’re in a significant local majority, then the society gradually un-segregates itself and overall diversity increases.  This suggests we need forces actively pushing towards diversity to make a difference.

Unfortunately, we’re in the business of making games, not laws.  Can we still help?

Well, maybe.  There are several options available to us:

  • Positive Discrimination (Affirmative Action)
    In addition to not overtly discriminating against minorities, we can also provide direct incentives to reward diversity.  This approach is often criticised as it feels like the ‘fair’ thing to do is not meddle with the system and let people be sorted entirely on ‘merit’.  The problem with that line of reasoning is that discrimination can still occur without the need for obvious bigotry (due to unconscious bias), and being part of a minority in a larger group can still be unpleasant and intimidating.  Extra incentives can help overcome this.

    In the gaming world, the best examples of this are all-female tournaments.  Taken at face value, these may seem sexist and patronising (“aww, girls aren’t good enough to play with the big boys!”).  But the idea behind them is to create a space where female players aren’t in the minority for once, removing the inevitable toxicity of male-dominated environments.  This allows women to be more visible in the tournament scene, which encourages more women to take up the game and feel more confident about attending mixed tournaments due to their greater numbers.  The aim isn’t permanent segregation, but a temporary step to shift the gender ratio towards that of the population at large.

  • Positive Representation
    We’re naturally drawn to groups where we imagine we’d fit in, and put off by groups where we imagine we wouldn’t (for all the reasons given earlier in this post).  Positive diversity in fictional characters suggests that the group that consumes this fiction is tolerant and accepting of this diversity, and increases engagement.  A lack of diversity, or stereotyped and offensive portrayals of minority groups, suggests the opposite.  This is supported by studies on multicultural literature in schools and data from the film industry, and there’s no reason to suppose that games are somehow exempt.

    The message for games is therefore: “make sure your cast of characters is sufficiently diverse, and don’t portray characters in a way that’s offensive”.  Doing so will win over far more people than you’ll put off.  Over the past few years, Magic has been increasing its character diversity and cutting down on objectified bikini-mages, and has seen a dramatic growth in the female player base.
  • Indirect Contact
    The Contact Hypothesis has primarily been studied with regard to ‘direct contact’ – people from different groups socialising, collaborating and realising that maybe the other lot aren’t that bad after all.  This is great, but it requires the groups to mix in the first place, which isn’t something they’re likely to do on their own.  ‘Indirect contact’, on the other hand, shows that people from different groups don’t need to directly interact to get the same benefits, and even if some of them are fictional.  It’s possible that this could apply to characters in a game.

    Firstly, there’s parasocial contact, where people just have to witness positive interactions between members of two groups without the need to be directly involved.  The interaction itself doesn’t have to be real – the subject of the paper was a fictional TV show, but it’s not clear if this still applies to media that doesn’t involve live actors.  We can hope, though.

    Secondly, there’s imagined contact, which requires people to imagine a positive interaction between themselves and a member of another group.  Now, in a board game setting this effect is going to be fairly small – you know that “Brock Goodstuff, attorney-at-law” is actually “Sophie from accounting” no matter how well she’s acting, so it’s probably not going to make you that much more sympathetic towards lawyers.  But it could help a little.

The conclusion of all this is pretty obvious to me – include positive portrayals of minorities in your games, which will increase tolerance and encourage diverse groups to form in real life, which will make people slightly happier overall.


So, bearing all these things in mind, let me finally introduce you to the cast of Sub Terra!

Amirah (Team Leader)


Amirah (Team Leader)

Amirah was deliberately chosen to be a middle-eastern, Muslim woman in a leadership role.  This is an archetype that’s been lacking in Western popular culture for some time (though Marvel comics are doing their part).  Muslims are being unfairly oppressed all over the Western world, at least partially due to the overwhelmingly negative portrayal of them in popular culture, right-wing demagoguery, and legitimate current affairs.

As the backstory of the characters isn’t part of the main game, we gave Amirah a hijab as this was the easiest way to visually indicate her culture.  Not all Muslim women wear the hijab, or follow a conservative dress code; it’s actually fairly controversial.  As discussed at this excellent GDC panel, it’s hard to represent such a diverse culture through a single character, so there are definitely aspects to Islam that we don’t have space to portray.  But as many non-oppressed Muslim women wear the hijab as a cultural choice, this felt like a suitable statement against recent cultural discrimination.

Amirah is a capable, commanding, fearless leader who will stop at nothing to keep her team safe, and we tried to portray this in her illustration.  If anyone told her what to wear, she’d punch them.

Jai (Bodyguard)

Jai (Bodyguard)

Jai (Bodyguard)

Maybe this is due to them comprising only 1% of the US population, but Indian-Asians haven’t been well represented in Western media at all.  As this demographic comprises 2.3% of the UK population, and 18% globally, we felt it was about time they got some love.
Jai is a quiet, brooding mercenary with a mysterious past who hails from the Indian subcontinent.  We based his look on Bollywood star Salman Khan, and he is every bit the tough tank character his illustration suggests.

Zoe (Engineer)

Zoe (Engineer)

Zoe (Engineer)

Minority representation in STEM fields is a big deal, so it was pretty clear from the start that the Engineer shouldn’t be just another white dude.  Zoe is calm, competent and rational, while possibly not being so great with people.  She’s good at blowing stuff up, though.

Louis (Medic)

Louis (Medic)

Louis (Medic)

Louis is a straight nod to the video game Left 4 Dead, as Sub Terra takes a lot of mechanical inspiration from this cooperative zombie shooter.  We even tried to keep his optimistic personality intact.  Stay positive, guys!  He’s got a good feeling about this!

Ishiko (Climber)

Ishiko (Climber)

Ishiko (Climber)

Ishiko is our upbeat Japanese climber.  We needed an optimistic female character to balance out the stern Leader, the clinical Engineer and the aggressive Scout, and this exploration role seemed like a good fit.  She has the most stereotypically feminine personality of our characters (warm, sensitive, kind, caring) while still being independent and inspiring.
Daniel (Geologist)

Daniel (Geologist)

Daniel (Geologist)

Daniel is our nerdy professor from Spain.  The game mechanics needed a character who had some specialist knowledge of caves, and a scientist seemed like a good choice.  We decided that ‘geologist’ was more recognisable a profession than ‘speleologist’, so we doubled down on this and gave him a cool rock to look at.  Shiny!

Kate (Scout)

Kate (Scout)

Kate (Scout)

The Scout’s abilities push her towards a reckless style of play, pushing forward at all costs and to hell with the consequences.  It therefore made sense to mirror this in her personality.  Kate is an arrogant overconfident loner, who outwardly doesn’t care what anyone thinks about her.  She’s passionate, quick to anger, and usually thinks she’s right.  Inwardly, however, she still cares strongly about her team, and would risk her life to protect them without hesitation.

Jack (Diver)

Jack (Diver)

Jack (Diver)


The Diver was the last character to be finalised – before the story and setting had been properly fleshed out, this slot belonged to the fearsome Chaplain.  Jack is our only white male character and the newest member of the team.  He’s insecure, awkward and intimidated by the others, despite being highly skilled.  The graphic novel is told primarily from his point of view, as his low status and inexperience made him a useful recipient for exposition.


We only had eight characters available, so many other ethnicities/gender/role combinations had to be discarded (we’re not Overwatch).  For example, we only had room for one real tank, which turned into the male Bodyguard (although the Leader and Engineer can do their fair share of soaking up damage too).  Any future expansions will add more combinations rather than retread old ground.

The backstory of each character was kept deliberately light in the main game, so representation had to be visual.  This was ideal for gender, ethnicity and personality, which we could try to communicate through pose.  However, this meant that referencing intangible characteristics like sexual orientation weren’t possible – it’s also irrelevant, as this is not a game or story that wants to involve sexual attraction.  Trans* characters were also hard to portray for this reason (how/why should you be able to tell?).

Finally, the cave setting meant that we couldn’t go to extremes on age, weight or disability.  All our characters needed to be fit and fully mobile in order to believably traverse the environment.  This is an area that doesn’t get as much attention as race, gender and LGBTQ rights so I wanted to deliberately highlight it here.  Future games may not be quite so restrictive.


So that’s the why and how of diversity in Sub Terra!  This is a topic where feedback is incredibly useful, so if there’s anything you’d like to comment about or any characters you think we should have handled differently, please let me know!

In my next post, we’ll finally be able to show you the Sub Terra rulebook!  I’ll discuss why the rules ended up as they did and shed some light on the many paths not taken.  See you then!

Tim Pinder

Tim Pinder

Game Designer, Sub Terra

Indie gamedev robot. All opinions, fatalities and apocalypses are the responsibility of my creators, who should have worked harder on the control problem