Content uploaded by Frederic Seraphine
Author content
All content in this area was uploaded by Frederic Seraphine on Mar 13, 2017
Content may be subject to copyright.
日本デジタルゲーム学会 2016 年 年次大会 予稿集
Digital Games Research Association JAPAN Proceedings of 2016 Annual Conference
67
The Ludic Framework
-A Theory of Meaningful Gameplay-
Frederic SERAPHINE
Ph.D. Candidate, The Graduate School of Interdisciplinary Information Studies, ITASIA, The University of Tokyo.
Hongo 7-3-1 Bunkyo-ku 113-0033 Japan
E-mail: seraphine「at」g.ecc.u-tokyo.ac.jp
Abstract Using concepts from semiotics, aesthetics, and ludology, this paper is shaping a framework opening new
perspectives on meaning and emotion evocation in videogames. While building upon the MDA Framework, it poses the
problem of meaning production in game design when the traditional building block, the game mechanics, is a complex
rule-based semiotic compound belonging to the Peircean thirdness. It proposes an alternative building block called the
ludic at a higher level of abstraction, allowing for wider combinatory possibilities and deeper meaning production through
the gameplay itself. It introduces an interpretation layer opening new possibilities for both narrative and emergent
gameplay.
Keywords Semiotics, Aesthetics, Storytelling, Interpretation, Framework
1. Introduction
When in 1987 the French philosopher Gilles Deleuze was
asking his students at Paris 8 University “What is having an
idea in filmmaking?”, Deleuze came up with the concept of
blocks: image-time blocks and image-movement blocks in the
case of cinematography. [1] Using Peircean semiotics, he
created from those conceptual blocks, a taxonomy of
cinematographic images that is still taught in French film
schools to this day. We can think that 30 years later, it is time
to ask ourselves similarly “What is having an idea in
videogame design?” The ludic framework is an interpretative
aesthetic model for games. It places interaction-signs that we
will call here the ludics as the main building block for
videogames. It aims at expanding the possibilities for
meaning production of the well-known MDA(Mechanics-
Dynamics-Aesthetics) Framework. (2008, Le Blanc, Hunicke
& Zubek) [2] The hypothesis that lead me to theorize this new
aesthetic framework was that videogame, despite being
multimodal in nature possesses a language of its own lying in
its interactivity. A language bearing potentialities for
conveying meaning and emotions without breaking the play-
flow with storytelling methods borrowed from other medias.
In videogames, is it possible to reach a level of meaning-
production and emotion suggestion equivalent to that of older
mediums like literature or filmmaking, without breaking the
play-flow? How is meaning produced within gameplay? How
can we manipulate meaning to create narratives and
emotions?
2. Defining Our Building Blocks
First, we need to define the building blocks we are working
with. I defined earlier interaction blocks that I would call the
ludics. Ludics are semiotically different from Aarseth’s
textons and scriptons. In the Aarsethian lexicon, they would
be building blocks of the traversal function instead of signs
that we may discover in the game world. [3] Why not taking
the game mechanics instead, as it is the first step of the MDA
Framework and it is generally thought to be the base element
of any game design. If we take the definition given by Miguel
Sicart, we have indeed our base interaction block. In the
online Journal Game Studies, he describes game mechanics
as “methods invoked by agents, designed for interaction with
the game state.” (2008) [4] While very tainted by the
computer science lexicon, this definition could comply with
our needs for a base building block. However, in “mechanics,”
we hear “mechanism.” And in the way most designers think
of game mechanics, we are probably closer to the definition
given by Ralph Koster (2004) [5]:
Game mechanics are rule based systems /
simulations that facilitate and encourage a user to
explore and learn the properties of their possibility
space through the use of feedback mechanisms.
In this definition, mechanics could be perceived as complex
semiotic compounds. Indeed, mechanics are often perceived
as interactions associated with a rule. If like Deleuze did with
cinematographic images, we look at game interactions
through the lens of Peircean semiotics; we would realize
quickly that mechanics belong to what Pierce calls the
“Thirdness.” [6] As they represent the possibilities of
representations by an interpretant by virtue of a rule or a law.
As a result, we will define our most simple ludics here as signs
belonging to the Peircean “Firstness.” Signs that are unary,
monadic in nature. Therefore, the constitutive elements of a
game mechanics: the rule and the interaction can both be
perceived as signs. A ludic can be an idea, an interaction, a
68
rule or any compound of those.
2.1 Actum, Tactum, Factum: A Perspective on Ludics
In my previous essay on the semiotics of videogames, I
realized that I needed a grasp on the ludics: I needed a
perspective. And the most natural point of view that one can
get on a videogame is, of course, the player’s standpoint. To
be able to observe and manipulate the ludics, I had to separate
them into three types of signs. What I called the actums, the
tactums, and the factums. [7]
First, let’s consider a category of interaction that is resulting
from a direct input from the player. We will call this type of
interaction an actum. An actum is an interaction-sign
triggered directly by an input of the player. The act of moving
a character around with the analog stick of the controller is an
easy example of a basic actum.
Then if we consider the opposite category of interaction, we
would have a factum. A tactum is an interaction controlled
by an input from the game state itself. If we consider the
player’s perspective on the game, factums are all interactions
which are not related to the player input. Witnessing an NPC
(Non-Playable Character) waving a hand at your character
could be a handy example of a factum.
In between those two opposing categories, we would have a
third category of ludics, a type of sign that would appear from
the contact between two game objects: a tactum. Any in-
game interactions like grabbing, pushing or hitting fall under
this category. Any interaction involving two or more game
objects (may it include the playable object or not) are tactums.
What is implied by considering the actions of the game, the
factums, as sign, is that the player shall be able to interpret
those actions in order to deduce meaning or intentions.
Interestingly, the fact we also have actums, that are in-game
representations of the player’s intent also implies that the
game may be able to interpret and deduce the meaning of the
player’s actions and his intentions.
2.2 The Ludics And Peircean Semiotics
The three previous categories of ludics are mainly tools.
They give us a perspective on the different kind of actions.
Nonetheless, they can still combine with the taxonomy of
Peircean semiotics to produce different layers of meaning.
2.2.1 The Three Trichotomies of Peirce
In Peircean semiotics, there are three trichotomies of signs.
The firstness is the trichotomy of the object, in videogames,
we could see it as the layer of the code. It is a trichotomy of
monadic signs. In this first category, the qualisigns are mere
concepts or ideas, the sinsigns are existent signs, and the
legisigns are rules or conventions. The secondness is the
trichotomy of the signs. In game design, we may see it as the
layer of the game, for it is the meat of what the experience is
made of. It is constituted of dyadic signs, for they are objects
that stand for something else. In this trichotomy, you will
have icons sharing characteristics with an object, indices
denoting an object by virtue of an existing relation, and finally,
symbols standing for an object by the intercession of a law or
a convention. Finally, the thirdness is the trichotomy of the
interpretant. In our field, we may see it as the layer of play, as
it is at this level that the ludic experience is subject to
interpretation. If the previous level was a structure, this stage
would be the path through it.
In this category, the rhemes represent the decontextualized
possibilities of interpretations of a given sign, the dicisigns
are asserting something as true or false, and finally, the
arguments are giving possibilities to retrieve meaning by
abduction, induction, or deduction.
2.2.3 An Example of Ludics At Play: The Jump
Let us take a very common example of game action: the
jump. We will see what different signification it can take as
ludics in the different Peircean trichotomies.
At first, within the trichotomy of the object, in the layer of
the code, we would have the concept of jump, the jump itself,
and finally any rule or convention associated with the jump.
The mere possibility of a jump controlled by the player
would be an actum-qualisign. This would simply be the
concept, the monadic idea of a player’s jump.
A qualisign doesn’t inform us on the nature of the jump, its
height nor its direction; those are other unary concepts to be
associated with it in order to make it an existent object. Such
an association of qualisign is a sinsign. An existent jump of
an NPC, with a given direction, height, and parameters
defining it precisely, would be a factum-sinsign. This time,
given we are dealing with an action controlled by the game
state, this action would be a factum. Also, all the parameters
defining this existing jump are as many qualisigns that
together will constitute our existing dicisign: our sinsign.
Still, in the firstness, any rule or convention associated with
the concept of jumping would be a legisign. For instance, a
convention stating that touching a poisoned object cancels
your jump ability would be a tactum-legisign. Just like any
other simpler rule or convention would be a legisign.
In the trichotomy of the icon, at the layer of the game, we
have the actual signs: Objects that are standing for other
objects. A nice first example would be a player-triggered
jump animation. This sign would be here considered as an
actum-icon. Actum because it is triggered by the player, and
icon because a jump animation shares characteristics with the
object “jump.” In animation, you could have a realistic
representation of a jump sharing most characteristics with an
existing jump. Or you could have a cartoonish jump
animation sharing only broad strokes of the characteristics
that are defining a “real jump.”
A good second example in this trichotomy would be an
animation of dust elevating from the ground, triggered by a
giant monster landing a jump. This sign would here be
considered as a tactum-index. Tactum because it is resulting
from the interaction between the monster and the ground.
Index because this dust animation can indicate the fact a
monster just landed a jump even if we don’t see this action
directly.
A handy last example for this category would be a very
specific jump animation that would be associated with a guild
or a group in an RPG. This association being made by virtue
of a convention, it would naturally make it a symbol.
So, a non-playable character making a very special jump to
show his allegiance to a certain group would be a factum-
symbol. Factum because it is an NPC’s action. And symbol
69
because this peculiar jump bounds the character executing it
to a group by virtue of a convention.
In the thirdness, at the layer of play, our last category, signs
are denoting objects through the possible interpretations of an
interpretant. This is where belong complex ludics, what I
called the “Ludophrases” in a previous work. (Seraphine,
2014) What’s interesting with actums and tactums, is that this
interpretant could be the player interpreting the game state’s
factums. However, it could also be the game state interpreting
the player’s intents through his actums. But we will come
back to that later.
A first example we could take here would be the actual jump
mechanics of an NPC and all the possible interpretations
attached to it. Decontextualized, a game mechanics represents
every possible intent, every possible meaning. At the level of
the thirdness, we talk here about the complete jump
mechanics made of several interwoven ludics. This
decontextualized game mechanics and the palate of meanings
that it may convey is a factum-rheme. Because we are
dealing here with a mechanics associated with an NPC, this is
a factum. And given we are looking at this mechanics out of
context with a magnifying glass, this is a rheme that
represents all the gamut of intents and meaning that this
mechanics may convey. For instance, jumping toward the
player or jumping away from them would convey a very
different interpretation. A rheme doesn’t choose between
those possible interpretations; it embraces them all.
In this trichotomy, a good second example would be a
cracking noise, and a hurting animation triggered when
landing a failed jump. This would be a tactum-dicisign.
Tactum because it would directly result from the interaction
between the ground and the landing object. And dicisign
because this sound and this special animation are asserting
that the player was hurt during this failed landing. It does not
give any reason for the failure, though.
Finally, our last example could certainly be the limping
animation triggered by the player after the failed jump. This
would be an actum-argument. Because from the observation
of this limping animation, the player could make the
deduction that his left leg got broken. Arguments are complex
compounds of ludics that may allow us to make deductions
about the result of an action. This result may induce
something about a rule or a convention. And finally, by
abduction, the interpretant may emit a hypothesis about the
meaning of this sign. It is within this last category of ludic
argument that we may become able to construct complex
narratives in gameplay.
Now that we have presented examples of what are the ludics,
we may look at how they take part in an aesthetic framework.
What seem to emerge from this structure, is the notion of
interpretation. Usually, we see the player, or the spectator, as
the only interpretant of any cultural product. Here what is
implied by the opposing relation between actums and tactums
is the dialog that may exist (or that should be designed)
between the player and the game. If the player may
understand the rules and the intents of a game. Conversely,
the game state may also interpret the player’s intents.
3. The Ludic Framework
The ludic framework is characterized by a structure in two
parts: the interpretation layer and the aesthetic layer. (see
Figure 1.) In the ludic framework, ludics do not necessarily
precede dynamics. Ludics and dynamics are interwoven.
Ludics influence the dynamics, but the dynamics can change
the very nature of the ludics. For example, if a game interprets
a player’s “playstyle” as friendly, enemies’ behavior may
change, and the possibility space of basic actions to take
(ludics) may switch from fighting to communication. The
aesthetic layer on the other side is different from the aesthetics
described in the MDA framework. For instance, in the MDA
Framework, narrative play or challenge are considered as
game aesthetics. I disagree on including those in the definition
of aesthetics I’ll use in this framework. In Peircean aesthetics
according to Anderson (1987, p. 55-57.), the goal of creativity
is to make monadic concepts belonging to the firstness (such
as emotions) intelligible through reasoning at the level of the
thirdness. To simplify, the goal of an artist or a designer is to
lead a user to a certain monadic emotional state. The artist
will do so by creating semiotic systems that will guide the
user to this emotional state. That’s why I would define
challenge, narrative play, or fantasy, to cite cite only a few
from the MDA Framework, as higher dynamics instead of
aesthetics. For instance, narrative play could lead us to
happiness, sadness or anger. Or challenge could make us feel
frustrated, proud or entertained. Those emotional states are
the actual aesthetics that, as game designers, we will be
aiming to evoke to the player. An author doesn’t craft a story
for the sake of crafting one; the story is a system made to lead
the reader to specific emotional states. The same applies to
game creators.
The player at the opposite will traverse those systems and
experience, and through his experience of those, will maybe
reach the emotional states intended by the designer. (Or
unexpected states if we go for emergent game design)
What this framework implies is very similar to the model
of interactivity proposed by Brian Upton, as it presupposes a
certain level of understanding of implicit rules and
conventions by the player. [8] The player can predict the
game’s intents through an analysis of the ludics, especially
the factums. But what is at stake here, and what differs from
Upton’s model is that the game state can act similarly to the
player, and use the actums to predict and analyze the player’s
Aesthetic
Layer
(Emotions,
feelings,
mood)
Ludics
Ludophrases,
mechanics …
Dynamics
Interpretation
Layer
Designer’s Intent
Player’s Intent
Figure 1. The Ludic Framework
70
intents. (See Figure 2.)
4. Conclusion
Interpreting human input is something that is currently
entering the field of possible with the recent advances of
machine learning, made possible by higher computing
abilities. When GPUs become more and more powerful, and
the first quantum computers are being built, AIs ability to
interpret data will only enhance. Robotics are still limited by
the cost of technology, but the possibilities of AIs without a
body will soon become almost unlimited. Training AIs and
game systems with the right semiotic labels could probably
help us designing games with a responsiveness never seen
before. That is why understanding the workings of semiotics
in videogames is more important than ever before today.
Thanks to a relevant semiotic labeling of objects and actions
in the game world, we could create games that would interpret
and respond, instead of just reacting as they often do. But this
will probably be the subject of another paper.
References
[1] Deleuze, G. (March 17, 1987). Qu’est-ce que l’acte de
création? [Conference]. France: Les Mardis de la Fondation.
Retrieved January 2, 2017, from
https://www.youtube.com/watch?v=2OyuMJMrCRw
[2] Hunicke, R., LeBlanc, M., & Zubek, R. (2004). MDA: A
Formal Approach to Game Design and Game Research. Retrieved
July 5, 2016, from
http://www.cs.northwestern.edu/~hunicke/MDA.pdf
[3] Aarseth, E. (1997). Cybertext: Perspectives on Ergodic
Literature. John Hopkins University Press, Maryland, United
States.
[4] Sicart, M. (2008). Defining Game Mechanics.Game
Studies: The International Journal of Computer Game Research,
8(2), 1. Retrieved from http://gamestudies.org/0802/articles/sicart
[5] Koster, R. (2004).A Theory of Fun for Game Design.
Paraglyph Press.
[6] Peirce, C., Sanders. (1955).Philosophical Writings of
Peirce. (J. Buchler, Ed.). Dover Publications.
[7] Seraphine, F. (2014). The Intrinsic Semiotics of Video-
Games[Kindle book]. KDP.
[8] Upton, B. (2015). The Aesthetic of Play. MIT Press.
Boston, USA.
ルディックフレームワーク
-意味のあるゲームプレーの理論-
セラフィン・フレデリック
東京大学・学際情報学府・ITASIA 博士課程 〒113-0033 東京都文京区本郷 7-3-1
E-mail: seraphine「at」g.ecc.u-tokyo.ac.jp
概要 この論文では、記号学、美学、ルドロジーの概念を用いて、ビデオゲームにおける意味と感情の誘起に関
する新しい視点を開くフレームワークを形成しています。MDA フレームワークから構築しながら、ゲームデザイ
ンの伝統的なビルディングブロック、ゲームメカニクスが複雑なルールベースの「Peirce の第三性に属する」記号
的な化合物である場合、ゲームデザインにおける意味的な生産の問題を提起します。抽象度のより高いルディック
と呼ばれる別のビルディングブロックを提案しています。ルディックは、ゲームプレイそのものを通じて、より広
い組み合わせ可能性とより深い意味の生産を可能にします。この小論文は、物語とエマージェントのゲームプレー
の両方に新しい可能性を開くインタープレテションレイヤーを導入します。
キーワード 記号論、美学、ストーリーテリング、解釈、フレームワーク
Player
Game
Ludics
Dynamics
Game State
Player’s
Intent
Interpretation Layer
Figure 2. Interactive Model