Stop making Sense (“connecting the dots” and the narrative fallacy)

The narrative fallacy has to do with the fact that natural objects and events don’t have meanings.   The signicance we give objects and events does not naturally and neutrally attach.  Objects and events just are.  An object, event or sequence of events just “is what it is”.  Narratives purport to attach significance.  This metaphysical fact lies behind the “narrative fallacy”.   It’s the fallacy of supposing that the stories, narratives, theories, propositions or accounts we associate with an object or event reveal something neutral, objective or inevitable about it.  In point of fact the accounts are inevitably provisional and subject to revision in light of further evidence or consideration.

Nevertheless, narratives, stories or theories usually have the appearance or pretension of fact.   But really, narratives are just hypotheses.  Narratives back-fit what has already happened or what you already know (or think you know).  By the neatness and completeness of that fit, some narratives or theories insinuate they are neutral and canonical renderings of the way things are.  But really, the world is wide open, resistant if not constitutionally impervious to being accurately and fully described or adequately narrated.
This is obvious if you go a little “meta-theoretical”,  or analytical on the problem.  If we step back from any particular narrative or theory we embrace, we have to admit there are a lot as yet undetermined facts and events that could affect the future or our future understanding of something we may think we’ve gotten to the bottom of.  In Donald Rumsfeld’s famous or notorious turn of phrase, there are lots of “known unknowns” and potentially or even very probably more  “unknown unknowns”.  From the perspective of common sense and logic, it’s kind of  remarkable that narratives ever really have worthwhile predictive value[i].  The fact that theories and narratives often have a temporary or provisional validity may say something about the orderliness of the world, or perhaps our inability of conceiving of things-in-themselves absent our narrative accounts.
Its a basic result of modern mathematics that you can always  “back fit” a theory, function or narrative account to your prior observations or known data.  You always know that there are nicely behaved functions and internally consistent, coherent and totally plausible narrative accounts that fit each moment with your observation at that moment.   The problem is that there are infinitely many functions and narratives that will fit the bill.  Objectively, there is no reason to suppose that any one of the functions (or stories) you choose to fit your facts or observations will have worthwhile predictive value going forward. While there are an infinite number of stories that will back-fit your prior information and explain what has happened so far, going forward these stories can and will differ (in every possible way).  You can set up a meta-system that will purport to help you choose among the possible candidates that back-fit properly, but that just re-introduces the problem of narrative again.  Turtles stacked on turtles.

There is no firm ground beneath on which to ultimately ground your account in.
If you’ve gotten this far with me you might be thinking “what on earth does this have to do with art?”  Fair question.  Maybe my interest in this is only idiosyncratic (idiopathic? idiotic?).  In any event I am troubled by the common place in criticism and critical theory among many people who talk about visual art to think, say and act as if it is important and somehow probative to recognize that all art is “narrative” or has a “narrative”.  Newsflash.  Everything has a “narrative” in that sense.   People, so far as we can tell, have always been imposing theories and narratives upon anything that falls under their gaze. (As a group, people who actually make visual art or who can actually wordlessly view a work without feeling any irresistible compulsion to force it into some Procrustean narrative probably tend to attach less importance to the narrative possibilities of art than do less “visual” people.)   Some people find some deep significance in the commonplace that all visual art has a potential narrative.  Which is nonsense.  For as long as they have been talking, people have been attaching narratives, stories, and theories to everything that falls under their gaze.
We are always looking for, imagining and sometimes actually finding causal relationships between separate events, objects, thoughts and feelings.  Humanities’ narratizing, theorizing, pattern-finding, connection-seeking proclivities seem to be more about human psychology than about the things those narratives get attached to.  Consider a distant ancestor of yours or mine who ate a taboo food the day before the wind blew down his hut.  He was almost genetically programmed to conclude the events were related.  He almost could not help himself from devising a theory that posited his culinary infraction had offended the wind gods.  Even though he came up with some such theory, I’m not convinced the particulars of his theory are what is interesting about the situation.  Rather, why did he feel the compulsion to narrate?
People will find a story to fit anything that passes in front of their eyes.  Sometimes, it seems a better use of one’s time to just enjoy the spectacle rather than to make up some story about what it all means.  Having a potential for narrative does not seem like an interesting, probative or even distinguishing aspect of the visual arts.  The unique thing about traditional visual art, as opposed to writing and talking and other essentially verbal or narrative arts and practices (prattleworks) is that visual arts have something more going on than just a “narrative”.  There is depiction, there is facture, the hows and why’s of which seem much more interesting to me than the use of of visual artifacts as springboards for verbal exhibitions.   With visual art, understood as an enterprise involved real objects in the real world, there is a tangible, concrete, separately existing and self-sufficient artifact that stands apart and separate from the narratives attached to it.  It’s true that the verbal arts have artifacts too.  Things like fonts and medium I suppose.  But those artifacts seem to have taken some sort of distant second place to the linguistic meanings attached to them.
No matter what some theories may posit, it is possible to wordlessly enjoy a painting or a sculpture, music, or even the semantically unburdened sound of some voices.  The artifact or physical occurrence or performance has interest just in its unnarrated perception.  Not so much a novel, an essay or a journal article.  It’s true we can attach narrative accounts to visual art, but there are a lot of other things going on besides a potential susceptibility to narrative.  But, of the things we cannot speak, we must remain silent.  And that is unacceptable if your only tool is a narratizing tendency.  The lack of accompanying chatter for some aspect of visual or non-literary arts doesn’t mean there’s nothing there.  Rather, it means only that it does not yield to denaturing and rendering in the prattleworks.   But, if the only tool you have is a hammer, then everything looks like a nail.  And if the only tool you have is a narratizing tendency, then everything looks like a text.  But there is more between heaven and earth than nails and  texts.  You just need the eyes to see them.  The fact that one can attach a narrative to a work of visual art does not impress me as worth noting.  What seems most interesting and worthwhile are precisely those aspects that can’t be rendered  and denatured in the prattleworks.
In principle there are an infinite number of internally consistent and ostensibly coherent narratives that can be attached to anything someone puts their mind on.  So much sound and fury, signifying almost nothing.   Why should the fact that narratives can attach like remora or leeches to anything that lives, breathes or falls under our gaze, make that susceptibility an important or defining trait of an object?  The fact of that occurrence seems almost like an epistemological tautology:  to be perceived is to be narrated.   I think there is a lot more interesting stuff going on with visual art than just the prattle that may potentially attach to it.   Pull up the ladder behind you and enjoy the view.   Sometimes its better to just quiet your mind and look.

[i]  One big result in mathematical analysis is the “Weierstrass Approximation Theorem” which essentially states that for any continuous function or set of data points for an interval of values for a variable there is an infinite number of polynomials that will approximate arbitrarily closely all the values of all the data points in the interval.  I can’t remember how all the trees go together to form the forest view created by this theorem, but the general observation that you can “back fit” any set of discrete data points with an infinite number of different polynomials or other nicely behaved continuous functions (and the resulting corollary that they won’t provide you with a canonical, mathematically and logically neutral way of predicting the future), has colored my way of looking at scientific theories and  other narratives over the years.
By way of illustration the Weierstrass Approximation Theorem would tell you that if you have data about the temperature at different times at a certain place, or the location of a particular sparrow at particular times or the Dow Jones Industrial Average at different times, you can find an infinite number of different relations or “functions” that will fit your observed facts as closely as you would like.

If you’ve followed me this far, the result is kind of unsettling, once you tease it out.   What it says, among other things, is that if you take the data for the temperature at a particular spot in Toledo, Ohio for each minute over the last 50 years, or the value of the Dow Jones Industrial Average each minute the market was open during the last half of the 20th Century, or the location of a particular sparrow each minute last Tuesday afternoon, there are an infinite number of different accounts or “functions” that will give values for different times that will fit your data as close as you may like.  .  And whichever function from the list you choose, it will give you the fact or datum you observed at the time your observation of fact was made.  They all fit your data.    That is essentially the same thing as having as many tidy little theories or stories that “connect the dots” for all the points in time or events you have information for.  No matter how much information you have about those temperatures, values or locations you are guaranteed there are infinitely many nicely behaved functions (read narrative accounts) that will “fit” your known facts as precisely as you would like.

 “If one puts an infinite number of monkeys in front of (strongly built) typewriters, and lets them clap away, there is a certainty that one of them would come out with an exact version of the Iliad.  Upon examination, this may be less interesting a concept than it appears at first:  such probability is very low.  But let us carry the reasoning one step beyond.  Now that we have found that hero among monkeys, would any reader invest his life’s savings on a bet that the monkey would write the Odyssey next?  In this story, it is the second step that is interesting.  How much can past performance (here the typing of the Iliad) be relevant in forecasting future performance?  The same applies to any decision based on past performance, merely relying on the attributes of the past time series.  Think about the monkey showing up at your door with his impressive past performance.  Hey, he wrote the Iliad.  Quickly, sign him up for the sequel.”  Taleb, Nassim, Fooled by Randomness, (Texere, New York, 2001) p.113.

My point here is that without certain metaphysical, as opposed to purely logical, assumptions about how things hold together there is no reason to suppose that any one narrative that back-fits prior events will be any better than any other, and given that there are an infinite number of back-fitting (but prospectively radically divergent) theories or formulae that will fit the limited information you have, the chances of stumbling on one that has predictive value is vanishingly small unless you are willing to make some metaphysical assumptions that give you some meta-criterion for choosing among the proffered back-fit theories.   And that meta-criterion, to the extent it makes a positive assertion about the way things are, will like all other positive, falsifiable theories, eventually fail in practice.  The world is just too rich and complex in its possibilities and interconnections to ever fully and finally yield to our narratives.