The Concepts Mimesis and Diegesis, applied to Interactive Literature

13 views
Skip to first unread message

Steve Breslin

unread,
Sep 16, 2004, 2:01:31 AM9/16/04
to
[The following is a *very* preliminary attempt at the question. You'll
find a great number of important points skipped through, and a greater
number of points entirely elided.]

In order to establish a concept of mimesis and diegesis in any
literary work, we have first to answer two questions: (1) who is the
speaker; and (2) what is presented.

The question of the speaker in interactive fiction is not simple,
first of all because we have a number of speakers, and second because
none match conventional types of literary speakers. There is no clear
answer to the identity of a fundamental speaker.

It is possible, though not necessary, that there is a projected
narrator in some sense (a pseudo-narrator). This depends on the
specific framing of the work. But the presence or absence of this
projected pseudo-narrator does not typically impact the progression of
the text.

At the very beginning, we notice that there is no proper narrator, and
where there is, the projected narrator quickly recedes from view.
Instead, we find one speaker deeply concealed (the one who projected
the pseudo-narrator in that case), much closer in essence to the
dramatist: there is descriptive text of an environment, very much like
the stage setting in a drama.

Upon interaction (and this is implied from the beginning), we find at
least two more speakers, the interactor, and the (implicit) addressee.
In normal cases, the interactor (the reader) is the secondmost
important speaker in the work, as it is he who determines the course
or development of the text. Depending on how the work is framed and
interpreted, the addressee is either (1) the parser; (2) the object
currently in the interactor's control; (3) the work itself; or (4) the
illusive dramatist we've yet to identify very precisely.

Except in the most trivial case, there are other speakers within the
work. The simplest case is the "object," and the most complicated case
is the "NPC" (non-player character) – the object simulating a robust
personality (but which is not in the interactor's direct control). In
these cases, the speaker is represented by the work, and by current
convention does not transcend its own representation.

These complications we announce for the sake of a certain line of
thought: let us take as a working hypothesis that the fundamental
speaker is the machine generating the text (based on (and enabling)
the interactor's input).

Having hypothetically identified the fundamental speaker as the
machine, let us turn to the next question, which we have already
worked out to some extent:

What is presented is a manifold of speakers, including the assumed
interactor, his addressee, and the text in its various characters.

Let us work out a bit more carefully the identity of our illusive
dramatist. What can we tell about him? Certainly, he is a figment of
the genre. He is one outward face of the machine, and he is not the
machine itself. He is responsible for not only the description, but
for reporting the behavior of the work, the interaction. He is not to
be confused with the author, but he rather runs the show. He's a
master, and a masterfully slippery subject.

He is a dramatist, of this we're certain; and as such, his
presentation is mimetic and not diegetic. He cannot describe, for the
simple fact that he does never show his face; he presents, from behind
the scene. His language is diegetic at times, as when he describes a
location for example; his language is mimetic at times, as when he
gives us a piece of dialogue. His performance shifts between diegesis
and mimesis, but the language itself is entirely mimetic, entirely
performative.

Nick Montfort

unread,
Sep 16, 2004, 2:16:39 PM9/16/04
to
ver...@hotmail.com (Steve Breslin) wrote in message news:<f407dc2b.04091...@posting.google.com>...

> In order to establish a concept of mimesis and diegesis in any
> literary work, we have first to answer two questions: (1) who is the

> speaker ... we have a number of speakers ...


> none match conventional types of literary speakers. There is no clear
> answer to the identity of a fundamental speaker.

I did try to give a clear answer in the last two paragraphs of

http://nickm.com/if/toward.html#sec8

the "Diegesis, Hypodiegesis, and Extradiegesis" section of "Toward a
Theory of Interactive Fiction," which you kindly mentioned earlier. I
also cite Graham Nelson and Espen Aarseth there, who have both written
about this topic.

While I think my answer is clear, it does depend on diegetic levels in
narratology and how they can be mapped to levels of simulated world in
interactive fiction, so it's not something that is easily discussed in
posts on the newsgroups. I did try to discuss this work in some other
forums, though, presenting it last semester at Narrative: An
International Conference in Vermont, for instance. People are also
welcome to come discuss it in an old, but still comment-enabled, nook
of the blog Grand Text Auto:

http://grandtextauto.gatech.edu/2003/12/19/toward-a-theory-of-interactive-fiction/

-Nick Montfort <mail> nickm <web> http://nickm.com

Steve Breslin

unread,
Sep 17, 2004, 12:23:57 AM9/17/04
to
Nick writes, quoting me:

> > There is no clear
> > answer to the identity of a fundamental speaker.
>
> I did try to give a clear answer in the last two

> paragraphs of ["Toward..."].

While writing this post, I was thinking about your essay (and about
the argument from the inform designer's manual -- I haven't yet read
Aarseth's work). I don't think anyone gives an answer to the
*identity* of a fundamental speaker. Instead we remark that there
isn't *one*. "The dramatist" I figure is a speculative projection of
an identity behind the other speakers, but at a remove from the
machine itself.

This needs some more thought, of course; it's only meant to be a
preliminary shot. Meanwhile, I'll hope its interest makes up for the
lack of rigor. (Most of all, I badly wanted to avoid copping out with
"sometimes it's mimetic and sometimes it's diegetic, and it just
depends," which would have blunted the point considerably -- though it
would be neat to discuss on what it depends.)

> [I]t's not something that is easily discussed in
> posts on the newsgroups.

You might be right. I took the theme to demonstrate what one might do
with the term "mimesis" (and "diegesis") in an critical context.
(Mostly I did this for the benefit of that other thread about mimesis
and consistency, in which I raised this question of whether IF is
fundamentally mimetic or diegetic, and what mimesis in IF might
properly mean.)

Nick Montfort

unread,
Sep 17, 2004, 9:06:29 PM9/17/04
to
ver...@hotmail.com (Steve Breslin) wrote in message news:<f407dc2b.04091...@posting.google.com>...
> Nick writes, quoting me:
>
> > > There is no clear
> > > answer to the identity of a fundamental speaker.
> >
> > I did try to give a clear answer in the last two
> > paragraphs of [[[section 8 of]]] ["Toward..."].

>
> While writing this post, I was thinking about your essay (and about
> the argument from the inform designer's manual -- I haven't yet read
> Aarseth's work). I don't think anyone gives an answer to the
> *identity* of a fundamental speaker. Instead we remark that there
> isn't *one*.

Yes, it's true that I didn't name a single speaker as fundamental. I
thought that identifying the roles of different narrators was a clear
and appropriate answer to this question, though.

You could ask who the fundamental speaker of Heart of Darkness is, for
instance. While Marlowe narrates the major events, another narrator
tells of him telling the story. The other (unnamed) narrator is
"fundamental" in one sense, because there is no narrator "underneath"
that one. But the novel is only interesting because Marlowe tells what
he does, so Marlowe is "fundamental" in another sense.

IF is not the same as a novel, of course, but the two are analogous in
that IF can have many narrators, each of which are important in
different ways.

Adam Thornton

unread,
Sep 18, 2004, 11:41:27 PM9/18/04
to
In article <f407dc2b.04091...@posting.google.com>,

Steve Breslin <ver...@hotmail.com> wrote:
>While writing this post, I was thinking about your essay (and about
>the argument from the inform designer's manual -- I haven't yet read
>Aarseth's work)

At the risk of sounding snotty, if you're going to be reworking the
concept of diegesis in IF, you'd *better* have read Aarseth's work.
Everyone who responds to your article will have.

Adam

Steve Breslin

unread,
Sep 19, 2004, 8:39:18 PM9/19/04
to
Adam T writes:

> [I]f you're going to be reworking the


> concept of diegesis in IF, you'd *better* have read Aarseth's work.
> Everyone who responds to your article will have.

Assuredly. I'm happy to hear that I'm reworking the concept of
diegesis in IF! High honors indeed for my paltry post -- not really
what I was up to, though. I agree with you: it appears to be necessary
reading for the subject.

I assume from your post that you have read Aarseth, presumably
_Cybertexts_? Or that you're familiar enough with the discourse that
you know at least what the necessary reading is? In which case, if you
have anything to contribute, I'd be first among happy listeners.

I don't think I'll develop my idea into an article for the time being.
I was just putting forward a tentative point or two about mimesis, for
the sake of clarifying the term a smidgen. I'm almost positive that I
haven't said anything about diegesis in IF which has not already been
said, except perhaps that it should be understood in relation to
mimesis. -- But this would require some research.

Steve Breslin

unread,
Sep 19, 2004, 9:03:21 PM9/19/04
to
Nick and I wrote back and forth:

> > > > There is no clear
> > > > answer to the identity of a fundamental speaker.
> > >
> > > I did try to give a clear answer in the last two
> > > paragraphs of [[[section 8 of]]] ["Toward..."].
> >

> > I don't think anyone gives an answer to the
> > *identity* of a fundamental speaker. Instead we remark that there
> > isn't *one*.
>
> Yes, it's true that I didn't name a single speaker as fundamental. I
> thought that identifying the roles of different narrators was a clear
> and appropriate answer to this question, though.

I agree. I *really* like your approach: nothing speculative or
hypothetical; purely describing precisely what we have before us.

> You could ask who the fundamental speaker of Heart of Darkness is, for
> instance. While Marlowe narrates the major events, another narrator
> tells of him telling the story. The other (unnamed) narrator is
> "fundamental" in one sense, because there is no narrator "underneath"
> that one. But the novel is only interesting because Marlowe tells what
> he does, so Marlowe is "fundamental" in another sense.

Yes, this is all quite clear; you're certainly correct.
(_Frankenstein_ is my favorite example of narrative stacking; I'd make
some sophomoric points, but I'd rather save you the trouble.)

> IF is not the same as a novel, of course, but the two are analogous in
> that IF can have many narrators, each of which are important in
> different ways.

Yes. The anology is useful, though in IF there's not so much narrative
stacking as there are concurrent narrators. The narrators all acting
like players on stage, I think the hypothesis of a unifying voice
(analogous to the dramatist) might have some currency. I've been
reading Peter Szondi _Theory of Modern Drama_ recently; I might come
up with some further ideas, but this requires much more thought.

Reply all
Reply to author
Forward
0 new messages