Semiotics, Character Design, and the ‘Tank’ Problem

Image 1: Merciless Wanderer, Bloodborne

It is important for many games to convey a lot of information quickly. In Bloodborne, if a player runs into this fellow (see above), they’re going to be making a lot of decisions in a very brief span of time. Will they be able to block this creature’s attack? Likely not, its weapon is both huge and blunt. Will they be able to quickly kill it? Again, no – it is hulking and heavy. But will they be able to outrun this creature? Definitely – it is big, lumbering and covered with chains, which means it must be slow.

This… all sounds a bit obvious though, doesn’t it? Of course this big creature is slow, of course it will take some effort to defeat – but why does that feel obvious? How can a still image carry so much mechanical, functional knowledge? Enter semiology, the study of signs – when properly utilised, by understanding the myths and meaning behind signs, a character’s design can convey so much information at just a glance.

The trouble is, when a sign is used for long enough, it can grow… complicated.

Signs, Signifiers, and Social Constructs

Before we dive into another piece about how video games hate women (fear not, we’ll get there), it’s important to establish what is meant by the word sign. Saussure wrote that signs are made up of two parts: the signifier, which is the thing that conveys information, and the signified, which is the information being conveyed (Saussure, 2011). A big red octagon is the ‘signifier’, and the request that an approaching car stops is the ‘signified’.

Barthes argued that there was a third part, one that must not be lost in the analysis – the sign itself. It is not merely that a signifier expresses the signified, but that there is “a correlation that unites them” (Barthes, 2014, pg. 263). The sign, the “associative total of the first two terms,” (ibid.) is inseparable from its component parts, and vital to Barthes’ concept of “myth”.

“Myth” here refers to a way in which signs are gathered together to convey a message, a thing “constructed from a semiological chain which existed before it” (ibid.). A sign (signifier and signified) is “caught” by a myth, folded into it and built upon – what was once a complete sign becomes only a signifier, a building block, in a greater myth. 

He uses a bouquet of roses as an example; the signifier is the roses, the signified is an expression of romantic affection, together forming the sign of romantic flowers. However, when caught by the myth of traditional romance, the bouquet sign is reduced to a signifier, its paired signified being a grand romantic gesture, likely coinciding with an anniversary or important date. This grand romantic gesture can then become a signifier, denoting a relationship becoming serious, and so on.

In this way, the myth is both collecting and generating meaning. While some myths come and go, the prevailing ones are immortalised in culture; their component signs stripped of original meaning, their legitimacy unquestioned. Why, again, are roses romantic?

Putting the Sign in Character Design

Okay, so what are the signs and myths we find in games? There are plenty that predate games completely: big muscles mean strength, long legs mean speed, and so on. There are myths more firmly centered in games, however – take the triumvirate of DPS, Healer, and Tank.

Image 4: ‘Role Playing Game roles’

The DPS character – that is, damage-per-second – is aptly named; they are there to hit things hard, and are normally quite fragile. The Healers (easy guess) heal and protect their teammates, serving a support role and usually doing so from the safety of the back-line. Lastly the Tank, a wall of defences that welcomes and withstands the damage that would otherwise kill their comrades. 

The aesthetic trimmings will vary between games and genres, but the profiles are immediately recognizable: the DPS is tall, gaunt, dangerous, and sharp. The Healer is soft, bright, warm, delicate. The Tank is big, heavy, cumbersome and blunt.

The value of this is immediate and evident once experienced as a player – being able to distinguish the general function of a character in a crowd from their silhouette alone is what allows for the complexity and scale of a game like Overwatch – a dozen characters fighting all at once, but a seasoned player can keep track of the chaos thanks to this thoughtful design. However, as Barthes would be quick to point out, the “myth” is greedy, and oblivious to the damage it can cause…

The Tank Problem

Alright, pop quiz. Below are a few pairs of characters selected from a few massively popular video games. Which of the pair do you think is the most resilient to damage, the most “tanky”?

The answers are B, A, and A. Huh – that last one is strange, isn’t it? Where is the bulk, the hulk, the lumbering size? Let’s investigate further. These are all of the “tank” characters in Overwatch:

Of the three women present, one is a robot, one is a skinny woman in a robot, and the last is (refreshingly) a big woman. A look into the staggering 140+ characters in League of Legends will yield similar results; here are the tankiest men:

And the tankiest women:

It is here, at the intersection of semiotically-informed character design and market-research that the priorities of a game, or of its producers, are made clear. Men are allowed to be grotesque, deformed, rugged, messy and monstrous – but the women are on the box art, the women sell merchandise, and the women perform well in marketing.

Stuart Hall tells us that “meaning is produced by the practice, the ‘work’, of representation” (Hall, 1997, pg. 14) – that meaning is made, not inherent. When women consistently look like this, even at the functional detriment of the games they exist in, what meaning is being made? And what is being lost among the myth?

Bibliography

Barthes, R. (2014). 9 Myth Today. Ideology.

De Saussure, F. (2011). Course in general linguistics. Columbia University Press.

Hall, S. (1997). The work of representation. Representation: Cultural representations and signifying practices2, 13-74.