Looking at the Wall...

Implications - Set Two

1. The second set of implications from this involve gestalt perception and a related topic--argumentation about patterns.
   

2. We know from perceptual psychology that we perceive patterns, not absolute units. We even have 'hard' neurobiological data that demonstrates that patterns in our visual field ACTUALLY CREATE spatial images in our sensory cortex! [see Tootell, Switkes, Silverman, Hamilton (1988), "Functional anatomy of macaque striate cortex. II. Retinotopic organization", The Journal of Neuroscience, 8:1531-68, cited by DeMasio(above), p. 103-104]. We have long known that the mind sees the whole first, before it sees the parts. In other words, as you read this, on paper or screen, you see the LETTERS first, BEFORE you see any pixel aberrations (malformed serifs, for example). [Print media has mastered the dots-per-inch science of the minimums required for a perceiver to see the 'whole'.]
   

3. Let's take the case of a large, but low-resolution picture of a face. For example, if I printed a copy of the Mona Lisa on an old dot-matrix printer (at low resolution), it would probably look enough like ML for basic recognition. But there are two points about this that I find instructive:
   
   1. The farther I get away from the image, the MORE easily I 'see' Mona; and conversely, I can get so close to the 'dots' that I cannot see the pattern. (Does this sound like my 'push to over-precision' or WHAT?!);
      

   2. I could never argue someone into 'seeing' Mona in that image.
   
   

4. Point #A fits rather well with the general observations above--that the 'closer we get' to the sub-gestalten stuff, the less 'solid/stable/obvious' it becomes. The pattern may still be there, but we cannot see 'the forest for the trees'. (More on this later).
   

5. Point #B has high relevance in the world of social and communicative structures.
   

6. Consider the following argument, between one who 'sees' Mona on the sheet, and one who doesn't:
   
   
   • 
     Gullible: Hey! Somebody printed out a picture of the Mona Lisa on this old printer! That's great!
     
     -- Skeptic: What are you talking about? That looks like a simple printer test. You know, where they print random characters.
     

   • 
     Gullible: Huh? No, look at it from back here. Can't you see her in there? It's HER! See the corner of the lips in that mystical smile?
     
     -- Skeptic: You are seeing things--what an imagination! I see a bunch of dots, colons, semi-colons etc. There's nothing there, bud.
     

   • 
     Gullible: I can't believe it! You really cannot see it? Look, here--if I connect these dots like this you can see her hair, and look with a simple connection here, you can see the eyes...
     
     -- Skeptic: Gotcha! You are seeing connections where NONE are. YOU are adding the connections--it's just random dots to me. You are ADDING STUFF to the picture! No wonder you see stuff 'there'--it is NOT THERE until you imagine it 'into the printout'...There is no "MONA" there; there are only individual "MONAds"...he, he, he...
     

   • 
     Gullible: Wow, this is frustrating...it is as plain as the nose on my face...I am at a loss how to proceed to get you to see her in the pattern...
     
     -- Skeptic: All I can suggest is that you power it down, and run the self test again (heh, heh)...maybe it will connect the dots for you this time!
     
   
   

   What advice could we give the Gullible in this example? Probably none, if the situation is confined to those two participants. The 'proof' was ONLY in the inspection of the image. It was NOT demonstrable in the least sense of the word. It was purely ostensive; purely a 'pointing' kinda proof; purely a kerygmatic event. He could only 'proclaim' that ML was in the image, and hope that the other person would 'see' the gestalt. Without the connected dots, there was no 'evidence'.
   

   
   

7. But really, now...would lines have been evidence anyway? Consider this version:
   

   • 
     Gullible: Look at the fax I got of the Mona Lisa! Someone made a copy of the original and faxed it to me.
     
     -- Skeptic: That's not the Mona Lisa. I've seen it in the Museum and I remember specifically that it does NOT have a fax phone number printed on it!
     

   • 
     Gullible: Huh? Of course it doesn't, but this image is a perfect replica of ML...
     
     -- Skeptic: No it's not--this is black and white (the original is not); this is too small (the original is bigger); this is printed on thermal paper (the original is not)...the nose is too small...Nah, this is too different--in MAJOR ways (color, size, materials)...
     

   • 
     Gullible: Huh? I know all that stuff, but that stuff is incidental to the issue...this face is a scale replica of the original...the proportion is roughly the same (it's a little different because of the fax scanner) but it's close enough...
     
     -- Skeptic: There you go again... why do you suppose it's 'close enough'? It is just too different to call it a picture of ML...even you admit that its proportions CONTRADICT the original! How blind are you going to be to these SERIOUS and MATERIAL differences between the two images? (This is even worse that that printer self-test that you made into a Rorschach ink-blot test the other day!)...
     

   • 
     Gullible: Hmmm...it is so obvious to me!....(you must be a Mona-Lisa-hater, huh!)
     
   
   

8. So, any advice for the Gullible here? Not much more than before. It is a pattern perception problem. Can the gullible accuse the Skeptic of being dishonest? Probably not--there is simply not enough data to make a judgment that the SK actually DOES see it, but is acting in denial.
   

9. There is one approach the G or SK might try to 'convince' the other--call over a third person. So, the G calls someone over to look at the printout (or the fax)...
   "what do YOU see here?"...
   "I see a blurry picture of a Madonna"...
   and we are still at an impasse (or worse!).
   

10. But suppose they said something that agreed with the pattern-perceiver G...what will the Skeptic do? At only one observer, probably nothing, but if the next 15 people that come up ALL see the ML in the picture, he/she would probably start looking VERY intensely to 'try to see' the shape in the picture...And conversely, if the first 15 saw a simple printer-test, what would the G say?
    

11. This is an interesting question, and one that can sometimes be answered from perceptual psychology. IF the pattern was VIVIDLY obvious to the G, or INTEGRATED the 'confusion of dots' SO WELL, there is a very strong probability that the G would NEVER be able to 'un-see' ML! Once our brains 'see' a gestalt, it is nigh on impossible to 'erase' the pattern; we can BARELY get our brains to see an ADDITIONAL pattern (e.g. the familiar "maiden vs. old woman" optical illusion). Many of the classic optical illusions show the power and fixity of gestalt perception. We organize our experienced world around 'wholes'.
    

12. So, the moral so far is that the argument CANNOT be resolved by appealing to the data--the data is being 'seen' differently. And this difference is NOT between "I see pattern A" and "I see pattern B"; but one of "I see a pattern" and "I don't any pattern at all". The two types of disagreements have DIFFERENT RULES OF EVIDENCE, if you will.
    

13. For example, in the first case--Pattern A vs. Pattern B--that the discrete 'pieces' (e.g. dots) DO CONSTITUTE a pattern is agreed upon...this issue is which 'implicit connections' are more weighty or primary or prior or material? In the 2nd argument--Pattern Exists vs No Pattern Exists--the disagreement can perhaps be seen as a METHODOLOGICAL one (you can/cannot connect dots that are NOT explicitly connected) or a TERMINOLOGICAL one (patterns that are incomplete/discontinuous are NOT patterns). [There are alternate ways of understanding this, but these will suffice for the point here.] You cannot demonstrate the existence of a pattern by appealing to discrete/discontinuous 'atoms'--they do NOT individually 'carry' the image--except perhaps in holograms. The 'atoms' have an ambiguity (relative to the pattern), that can only be reduced IN THE CONTEXT OF THE PATTERN. In other words, this dot on this spot on the page is just a dot WITHOUT the pattern, but in the context of the pattern, it is the lowest point of her chin. It carries a hierarchically-higher semantic 'load', much as a word in a sentence. And so, the G may fault the SK for getting 'too close' to the data, pushing it for one-dimensional precision; whereas the SK may fault the G for investing the dot with unwarranted, unproveable, uncontrolled speculative meanings. And, if you are following my argument so far, the SK has jumped into 'hyper precision unreality'...one CANNOT demonstrate a pattern from dots...the dots have all the ambiguity of atoms, particles, waves, fields...it is only at the 'flashlight' level does light become concrete, solid, stable--useable.
    

14. An even WORSE example of this can be given (from personal experience). I recently saw my first 'magic eye' diagram. I had seen boatloads of these posters in the malls, and struggled to relax(!)! to no end in my efforts to see the spaceship, or the rabid rhino or the Ferris wheel or whatever--all to no avail. I even began to wonder if the whole phenomena was a clever commercial version of "the emperor's new clothes"! But one day, I finally 'saw' one! It was bizarre...this seemingly fragmented (but symmetrically patterned) 2D picture 'became' a 3d carousel of zebras, antelope etc...
    

15. I have often thought about HOW one would defend the 'existence' of that image to someone who had never seen it! Think about this for a second. There was not a SINGLE PIECE of data or argument that I could 'point' to or advance, to prove to someone that the image was real! You either saw it or you didn't! No common ground--no way to 'lead' the other to see it! (I began to wonder how much of our worldview falls into this character--a certain strain of Christian thought, of course, maintains that it is the vast majority of faith that falls into this "you either see it or you don't" character.)
    

16. One interesting consequence from this is that the part CANNOT be studied in abstraction from the whole. If one looks at the dots that form the 'smile' in ML, UNLESS THEY ARE CONSCIOUSLY INTERPRETED AS A PART OF THE IMAGE, any study of the dots will be 1) irrelevant to ML; and 2) misleading in terms of its 'meaning'. This is an important point for me. If the various 'levels' of reality in a picture of ML (the picture of ML as a unit; the various color-patterns, the texture of the paint, etc) can ONLY be related through the higher levels of gestalt, then ANY analysis of the lower levels WITHOUT THE CONTEXT of the higher levels--will be incomplete and potentially misleading. If the splash of color that forms the smile is studied in the context of colors INSTEAD of in the context of ML, we will LOSE that smile...it will cease to be a 'carrier' of meaning in a gestalt. Subsequent studies of that isolated (and now reduced, I might add) data WILL NEVER lead us back to the seductive Mona.
    

17. What this means is that the WHOLE MUST BE 'SEEN' and seen continuously for ANY of the 'parts' to make sense. Reductionistic studies can therefore ONLY be of value, when the limitations of that study are duly noted. In other words, a study of chemical behavior in brains during moments of decision making by conscious subjects CAN NEVER 'lead us to' a knowledge of consciousness or decision theory or anything 'higher' than chemistry (or possibly, bio-chemistry). It is only when this study is seen as a 'semantically limited' product--a part which can never lead us to the whole--that such studies will be divested of their character as epistemic idols (in the same way that a mathematical analysis of the pattern of dots on the paper will never lead us to Mona).
    

18. Now, let's move from a visual field argument to a semantic field argument.
    

19. What does the word 'plane' mean? Actually, as just a word, it doesn't have any meaning; in a given culture, it has a set of possible meanings (and rules for stretching or changing those meaning temporarily). But in the sentence "John took a plane from Houston" it means something very specific. Whoever 'launched' that sentence had one meaning in mind (with exceptions: double entendre, etc.). So, lets visit our 'irreconcilable differences' arguers again...
    

    • 
      Gullible: My dad called and said he was bringing me a plane from Houston. I have always wanted a toy airplane.
      
      -- Skeptic: Huh? Why do you think he meant 'toy plane' instead of 'real plane' or even 'woodcrafters plane'?
      

    • 
      Gullible: I just know my dad--he couldn't afford one, and he knows I HATE arts and crafts.
      
      -- Skeptic: But you are being illogical. The word 'plane' can mean so many more things than just 'toy airplane'--what data do you have that it means that in his sentence...I mean 'hard data' not guesses about your Dad's finances or his awareness of your obvious over-reaction to bad Cub Scout experiences as a child? What linguistic data can you advance to show it means 'toy airplane'?
      

    • 
      Gullible: Hmmm...well, he said he was going to 'carry it'... I would assume that couldn't apply to a BIG airplane.
      
      -- Skeptic: But 'carry' can mean different things too..."Carry" in some contexts can be 'carry the debt for' like in 'I don't think I can carry that large of a note'...maybe he borrowed the money for the plane.
      

    • 
      Gullible: Huh? But you could do that to EVERY word in EVERY sentence! Every word has a number of different meanings (according to the dictionary), so any sentence with multiple words COULD theoretically have thousands upon thousands of possible meanings (even this sentence, no?). But I certainly got the impression from Dad that he meant something VERY SPECIFIC in his discourse to me.
      
      -- Skeptic: Yeah, but there you go being 'dogmatic' again...You must realize that you cannot be so sure of yourself, since you cannot 1) identify the specific meaning of any word in those sentences (with certainty) and 2) the context, which theoretically WOULD act as a selector between the possible meanings IS ITSELF MADE UP of those notoriously ambiguous words...In other words, starting from WORDS you cannot determine CONTEXT in which to isolate the meanings of those original WORDS..."You just can't get there from here".
      

    • 
      Gullible: Hmmm...well what if I start with an assumed context of meaning? Then that context WOULD identify the specific meanings of the individual semantic elements (i.e. words) and then the parts would make sense. Would that work?
      
      -- Skeptic: Hmmm...maybe so. It seems that if we START WITH the WHOLE, then the parts can be "disambiguated", but if we START WITH the PARTS, we cannot construct the whole...but I see a new problem in that--the seeming arbitrary character of the 'starting point', but I am too tired to think about that now--plus, we've got to fix this printer...this printer test is still producing replicas of famous paintings (according to you)--we've got to fix it to where you don't see any faces, fish, or fowl in the printouts!...
      
    
    

20. Now, let me give a visual example of this.
    

21. I have an image of unconnected dots...take a quick look at it and see if you can figure out what it is a picture 'of'...click here
    

22. Most of you will discern that the pattern is not 'random' but most will not 'see' the content at first.
    

23. Now, to find out WHAT is in the picture, click here.
    

24. Now that you have seen the answer, you will NEVER BE ABLE to look at the pattern WITHOUT SEEING the 'content'. Try it, click here again. See what I mean? The power of pattern, which unifies and gives meaning to the details, seems so basic and primal in our understanding.