[FRIAM] Narrating Complexity

[Prof David West]

In my opinion, Glen is right to be worried, and more than a little worried about the idea that "narrative" is fundamental to cognition.

Simultaneously, I am a huge fan of "story" and do believe that "story" is fundamental to cognition.

My antipathy towards narrative theory, and the resolution of the partial contradiction above, is that I see narrative theorists as a group of academics that are trying to "formalize" something that is fundamentally aformal. 

That vast majority of what we know, individually and collectively was acquired via story (70-75%); while 20-25% was acquired via direct experience (but shared with others via story); and only 2-5% of what we know was acquired formally as 'book learning'.

Ever since we started talking to each other, and telling stories around the fire pit, we have had no difficulty telling intelligible, meaningful, and useful stories about complex systems - after all we have lived in one and interacted with one, pretty successfully, for tens of thousands of years.

If narrative theory, as presently formalized, cannot deal with complexity, then the issue is with the formalism - it is fatally flawed.

In my opinion, this is the same issue with AI, both classical and the current resurgence of "AI in everything" fad - intelligence is defined as what the machine can do, what the formalisms that drive the machine can handle, and not what goes on in the embodied minds of human beings. When the singularity occurs and the machines take over, it is not because they are 'are cognitively superior and think better than us" it is because we allowed them to define away, as "not cognition," that which is unique to human minds.

Narrative theorists, again only in my opinion, do the same thing with story: anything not congruent with the theory is excluded as not cognition.

davew



On Mon, Dec 18, 2017, at 3:51 PM, uǝlƃ ☣ wrote:
> 
> This was helpful:
> 
> https://www.academia.edu/4163381/
> Narrating_Complexity_The_Antipathy_of_Stories_and_Systems
> 
> Ultimately, I'm a little worried about the idea that narrative is 
> somehow fundamental to thought/understanding. I'm completely ignorant of 
> any official domain called "Narrative".  But I tend to associate it with 
> stories, most importantly *including* stories that include concrete 
> detail.  And concreteness is antithetic to the type of "conceptual 
> model" I'm inferring from both Stepney's and Walsh's (from the above 
> presentation) sense/use of the term.  This is my core concern.
> 
> But first, I'd like to broach a more peripheral concern.  Walsh's defn: 
> "The semiotic articulation of linear temporal sequence" very clearly 
> lays out the sequentiality.  But, at least with computation, isn't it 
> relatively accepted that any parallel process can be simulated by a 
> sequential process?  And if so, couldn't we claim that this 
> sequentiality isn't that much of a limiting factor?  I mean, sure, it 
> may be merely approximate, or inaccurate, or whatever.  But, basically, 
> this would mean their research program would become showing which, in 
> particular, types of complex adaptive systems cannot be approximated by 
> narrative and the naive claim that *no* CAS can ever be adequately 
> approximated would be unjustified or too extreme.
> 
> The more important concern goes back to the accusation of idealism I 
> often lob.  I don't believe narrative is core to *my* cognition.  Yes, 
> when interacting mostly in symbols (like books, email, etc.), narrative 
> seems dominant.  But I would claim that's an artifact, a side effect, of 
> the underlying process(es). A quick way to my point is the accusation of 
> "book learning" vs. doing.  It's notoriously difficult to *tell* someone 
> how to, ride a bike or hit a baseball with a bat.  Such telling does 
> tend to be narrative, a kind of recipe, I suppose.  But is the narrative 
> necessary for *understanding* how to hit a baseball?  Or for 
> understanding what it *means* to ride a bicycle?  I would say "no."  
> 
> Understanding is a concrete thing, done with the *body*, not with some 
> abstraction in the *mind*. To go a step further, the objects pointed to 
> by the signs in a narrative, are fully complex processes. The narratives 
> really are just lossy models of the complex experiences. So, one person 
> compresses her experience(s) into a word salad.  Another person reads 
> that word salad and the signs are interpreted to map to concretely 
> *different* complex processes in the receiver.  I like the quote from 
> Herman below.  But even that doesn't go far enough.  Rather than say it 
> the way Herman does, I'd prefer to say narrative is a justificationist 
> technique for formulating hypotheses (whether testable or not).  Then 
> the concrete experiences of executed *tests* of those hypotheses are 
> understanding (like when a child actually slides a block down an incline 
> as opposed to hearing descriptions of blocks sliding down inclines). The 
> experiment is the understanding.  The narratives are only related to 
> reality to the extent they cause repeatable experiences (through both 
> inter- and intra-agent repetition).
> 
> Anyway, none of what I say is probably new to Walsh or Stepney and they 
> probably have answers.  But it's curious.  I'm intrigued by one of the 
> text boxes in the "Cognition" box of slide 60: "Narrative cognition 
> (relation to other modes of cognition)".  It'll be interesting to see if 
> the upcoming book talks about that.
> 
> 
> On 12/17/2017 09:03 AM, Steven A Smith wrote:
> > 
> >       Narrative Theory and the Cognitive Sciences
> > 
> > [...]
> >     David Herman, ed.
> >     /Narrative Theory and the Cognitive Sciences/.
> >     CSLI. 2003
> > [...]
> > Herman
> > 
> >     /Narrative can have many cognitive functions. It is a system for structuring patterns of events progressing through time: for structuring processes. It can be used to “chunk” experiences into “frames” of stereotypical experiences, then used to compare this typical against the actual. This helps us to understand the world more, and therefore have to memorise less. It allows us to generate and evaluate what-if scenarios. It allows us to draw coherent system boundaries: to extract and bound a relevant collection of participants, events, and structures from the overall stream of events we experience. /
> 
> 
> -- 
> ☣ uǝlƃ
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove