[Wittrs] Re: On the Mechanism of Understanding
- From: "Stuart W. Mirsky" <SWMirsky@xxxxxxx>
- To: Wittrs@xxxxxxxxxxxxxxx
- Date: Thu, 13 Aug 2009 14:34:27 -0000
I actually did a response to this the other day but for some reason it never
showed up here. So I'll try again (and maybe some day both will appear):
--- In Wittrs@xxxxxxxxxxxxxxx, kirby urner <kirby.urner@...> wrote:
>
> On Wed, Aug 12, 2009 at 6:09 AM, Stuart W. Mirsky<SWMirsky@...> wrote:
> >
> >
> > Kirby wrote:
> >
> > . . . someone has a flash of insight, says "aha! now I get it" and yet
> > manages to not follow the rule, didn't really get it after all. Observers
> > have to say this someone doesn't get it.
> >
> > This notion some authors cited here have shared, that "understanding"
> > is some "hidden process [in the brain]" i.e. points to some interior
> > process, is simply not consistent with our manifest and evident use of
> > that word. People understand all the time without pausing to reflect
> > or have any specific experience.
> >
> > The tool "understand" is definitely caught up in the machinery with,
> > is associated with "aha!" experiences, sub-vocalized reading etc., but
> > to claim these time-bound experiences are "the meaning" of what it
> > means to understand, is to misperceive what "meaning" means.
> >
> > My response:
> >
> > I think there's a misunderstanding here, at least insofar as I am among the
> > commenters of whom Kirby is speaking, above. For instance, he suggests that
> > "This notion that some authors cited here have shared, that 'understanding'
> > is some 'hidden process [in the brain]' i.e. points to some interior
> > process, is simply not consistent with our manifest and evident use of that
> > word."
> >
>
> Hey thanks for responding. I think it was actually Hawkins I was
> thinking of, but in a long quote from you (wasn't it?). That thing
> about how one knows, reading a novel, that one understands it, e.g.
> Robinson Crusoe doesn't have to wait until he gets of the island to
> know he understood the twelve books washed ashore with him -- or at
> least he maybe understood eleven of them, this 'Philosophical
> Investigations' thing was maybe a stumper. He'll check in with some
> others if he's ever rescued, maybe find out if he understood it at
> all. He's especially intrigued by the "private language" stuff!
>
Yes, it was Hawkins and I did post the text. I wasn't sure who you meant
though. It could have been my own comments since, from time to time, I've been
accused of something like that, i.e., equating a thought with an image. In
fact, of course, I don't and neither as far as I can see does Hawkins.
What Hawkins does do, which I'm inclined to agree with, is recognize the role
mental images play in our thinking processes. I would just cite, again, my
anecdote about seeing the sign, not getting it and then realizing what it meant
on the drive up through the Carolinas. What happened there? Was it a behavioral
response as Gerardo or Glen might want to say? Was it something that simply
doesn't fit our grammar, as you might say?
In fact, of course, I can describe what happened pretty well (suggesting that
grammar is perfectly functional here): I saw a sign, wondered about it (the
wording had initially stumped me), then, in an instant, it made sense and I had
a whole slew of mental images run past me, images that related to following the
words on the sign or to not following it. There was no change in my behavior. I
just kept driving (indeed I was already doing what the sign required) and I
didn't turn to my wife and say, ah, now I understand that sign. The only
changes involved the mental images I suddenly had, the connections I made to my
stored memories.
But I am not suggesting that any of those images, or even the complete panoply
of connected images, were the meaning per se (that they were the same images as
the writer of the sign had had in mind). Indeed, I would doubt that any two of
us ever have the same images much less the same network of connected images. If
you saw the sign and understood it, I expect you would have other mental images
reflecting your comprehension. It's not the images but something else at work.
A memory or an instant of understanding does not consist of particular images
but of a complex of images, held together in a certain way, and shared
understanding OF THIS TYPE occurs when two or more of us have roughly similar
complex networks. This is in keeping with Edelman's interesting point that
human memory is not like a computer's. (A point Hawkins makes, too.) Computers
must call up a precise replication of what is to be recalled each time or they
are deficient in their operation. As Hawkins notes, if a picture is stored in a
computer's memory, it must be precisely the same each time it's summoned onto
the screen. But human memory doesn't work like that. Edelman points out that
human memory is rough, fluid and always different with each instance (think
Rashomon). Each time it is called up new associations are added and maybe some
fall away or lose significance. So human memory involves constantly
reconstructing complex pictures and that process, that of construction, is
different than simple replication.
Hawkins actually explains it this way: The human brain, unlike a computer, he
says, doesn't store every bit of data it receives. It works by storing the
general structure, an adumbration of the pictures it is holding onto but not a
"pixel" by "pixel" deposit in the storage bank. When called back to active
duty, it's that structure that the human brain pulls up. Other parts of the
brain, with other stored images, are relied on to plug in details.
This helps explain why, when I had blacked out after sitting for a prolonged
period at my computer watching an image on the screen, the first thing I "saw"
as I was coming to was that same image. But, when I tried to focus on the
details, they weren't there and the image began to fade too! It was like an
afterimage, just a rough picture of the overall structure, the look of the
actual screen "page" I had been working on but not the page with all the
details on it, itself. Since it wasn't a real image, there were no details
present and I had no details in storage to plug in. But it looked perfectly
real, just as true to me in my half-wakened state anyhow as the real thing,
until I began to look for the details.
This may also explain the phenomenon of photographic memory. Most of us don't
have it but perhaps the brains of some of us do have the ability to capture
detail to a far greater extent than most of us (or a better ability to use
other brain areas to rebuild lost detail when structure is recalled and
associations summoned up to plug details in).
I know that I've always had a pretty good memory though it is hardly
photographic. I was always great at meetings, never taking notes. Colleagues
used to assiduously scribble down the comments being made, the details of
responses, etc. I would occasionally note something interesting but otherwise
just sit quietly listening (or join in). Yet, afterwards, I could reconstruct
the meetings in great detail. Not verbatim to be sure, but in a way that
resulted in my eventually doing away with notetakers at meetings I ran (staff
assigned to take down the proceedings) since I could do it better than they
could and found them coming to me for details after the meetings anyway.
My ability to do this worked in a way consistent with Hawkins' proposal, i.e.,
I would simply start reconstructing the meeting in my mind and who said what
would start to come back to me. Remember I wasn't producing a verbatim record
but mine was so thorough that it tended to replace most efforts to be verbatim.
It was better than a tape recorder which captured all the chaff as well -- mine
was already edited and since I was pretty good on the details and always tried
to be focused and fair, people came to trust and rely on my recaps. (It also
gave me the ability to shape the record, of course, an added benefit in any
business or bureaucratic environment, especially when I thought important
matters had been left out or inadequately stated).
What I wrote tended to become what everyone later referred back to. It also
became the basis for my own expectations from participants (both of staff and
colleagues). The phenomenon I'm describing was possible NOT because I could or
did recall every detail but because I recalled enough of the general meeting
structure which would trigger my plugging in enough details to create a
convincing and satisfying record of the proceeding. I suspect that Hawkins'
comment on structural recall as opposed to computational recall is as good an
explanation of how this aspect of our thinking works as any as I've seen.
> > > Although I've made it a point to note that understanding in my own
> > experience is not always accompanied by actions and is often solely
> > internal, characterized, say, by a feeling of "getting it" which is also
> > usually accompanied by various images, thoughts, etc., I have also been
> > quite clear to note that I am not arguing that any PARTICULAR image,
> > thought, etc., IS that understanding. What I've suggested is that knowing
> > anything (and this includes understanding anything) seems to consist of a
> > complex set of relationships reflected in various mental mappings by which
> > we hold together pictures of different aspects of the world we experience.
> > (See my moment of insight on the road through the Carolinas.)
> >
>
> I'd go further and saw "no picture at all" is often just as well, i.e.
> "I understand the traffic laws of the state of Oregon" is sort of
> tacitly assumed by the fact that I have a drivers license. I don't
> only understand these laws when I'm thinking about them or driving my
> car. People can say "Kirby understands the rules of the road in
> Oregon" without pausing to consider what I might be reflecting on at
> that moment. The grammar of "understand" is a lot more like that of
> having a badge or award of some kind.
>
> To take another example, I'll walk up to an elevator and simply use
> it, punching buttons while yakking about something else. I understand
> how to use an elevator, but there's nothing whatsoever I feel I need
> to report about mental states or mental pictures or anything. I'm
> actually thinking about a meeting I'm about to have.
>
> There's really know time to reflect on all the stuff I understand as I
> go through my day, yet understand I do.
>
Yes, not all of memory involves images. Many times we remember things
physically (my fingers typing away on this keyboard, hitting mostly the right
keys for instance).
Your point about understanding even when you're sleeping or not engaged in
whatever it is you are said to understand is a good one but not quite to the
point here. The issue I was raising re: understanding had to do with the
instant or condition of getting something, of grasping a meaning, etc. While we
use "understand" to also refer to longer term dispositions to act in certain
ways, etc., that use isn't at issue here. The fact that we can use words in
different ways (as we manifestly can with "understand" here) doesn't eliminate
the interest we may have in a particular use.
> Speaking of elevators, there's a new kind catching on with all the
> buttons on the outside i.e. you commit to a floor before you get on.
> Once on the elevator, no buttons inside. Wild eh?
>
> > Any given individual's instance of understanding something will certainly be
> > accompanied by understanding behavior when appropriate but, while it will
> > not be identifiable with any single image or even the same image that others
> > have, I want to say that it looks to me like we make a lot of connections in
>
> ... or any image at all mind you, let's not anchor understanding to
> any ghostly or interior process, *not* because it's somehow illegal or
> irrelevant to mention such things (as I've already said, those "aha!"
> moments enter into it) but because the grammar evidently (on the face
> of it) follows a different pattern.
>
Depends which use of "understand" we're interested in. Understanding as getting
it, as having a moment of insight or realization is not the same as when we
speak of understanding a language or of how to do something even when we're not
doing it.
> If you read a lot of political stuff, you'll see propositions like
> "UNDP failed to grasp the implications of these funding changes and
> continued its spending spree" i.e. we easily involve corporations,
> NGOs, agencies in the grammar of "understanding" ("failing to grasp"
> is a kind of misunderstanding). You might say "oh no, I'm talking
> only about what humans do, not agencies" but that's to simply slice
> away a big piece of the grammar and our goal is not to twist the
> meaning of "understand" but to accept standard usage and investigate
> accordingly.
>
> > our networked map-pictures (which constitute a level of representing -- a
> > nod here to Gerardo) and that it is the rough connections, types of
> > linkings, etc., that are the internal experience of understanding something.
>
> Not everyone is so navel-gazey about this stuff. Introspective types,
> always fascinated by their interior processes as we call them, are
> probably not the most qualified to investigate meanings per
> Wittgenstein's philo. The PI is always taking us out of these first
> person scenarios and reminding us of when *other people* get to say we
> understand or not -- and they *do* get to, i.e. it's grammatically
> correct to say stuff like: "I don't care about your process, I just
> want to know if you understand or not, and for that, I don't need to
> give a fig about your 'mental life' as that's entirely irrelevant to
> my making my assessment" (that might be a driving instructor).
>
> Indeed, you'll get these defensive people who, when you tell them they
> don't understand something, like to fall back on lengthy descriptions
> of some mental process that led them to conclude thus and so. Who
> cares really? They don't understand, nuff said.
>
I think the evidence shows Wittgenstein was a pretty introverted guy! More,
much of his thinking arises from long bouts of solitude or intensive thinking
about how his own mind works. The private language argument, for instance, (an
insight, really, more than an "argument") could only be constructed if one
thought carefully about what one is doing when using a language and compared
that in different scenarios including a private one.
> > In keeping with Edelman's point about memory, I would suggest that when we
> > share understandings with others it isn't because we have the exact same
> > linked representations, the same map connectors in play but, rather that
> > there is enough similarity for each to understand the other (even if
> > particular mental images differ as they very likely to).
> >
>
> I can't think of any mental images that'd be important to "prove" I
> understand the rules of the road in Oregon. I can play a "mental
> YouTube" of me stopping at a red light, putting out a flare when
> broken down, signalling before turning, but these all seem gratuitous
> and beside the point, to you as well I would suppose. Like, who cares
> what I imagine? This is about understanding, not what I'm imagining
> -- distinct concepts with different grammars.
>
> We don't say: she's sleeping now, so for the next few hours she has
> no understanding of politics or driving a car. How peculiar and
> strange that would sound, really goes against the way we use the word
> "understanding". Anyone talking in such a bizarre fashion is likely a
> philosopher or AI guy of some kind (a euphemism for "vaguely retarded"
> in some circles).
>
Same point as I made above. This really refers to a different use of
"understand". I am asking about what it means to get something at the instant
one gets it, not what it means to have demonstrated an ongoing ability to do or
get something.
> > Edelman's point that human memory is rough, approximate, constantly
> > changing, unlike computer memory, and that it thus operates differently than
> > computer memory does, seems to me to be key here. Understanding, which
> > certainly relies on the memory function, is imprecise and fuzzy in the same
> > way. This explains a lot, including our frequent uncertainty about what
> > others mean, the difficulties inherent in translating between languages and
> > our own shifting grasps of things. Understanding is like shooting at a
> > moving target though, as we know in the world outside ourselves, such
> > targets are often successfully hit and some of us are better than others at
> > doing so. Moreover, despite inherent capacities, our skills level can often
> > be improved by practice and dedication.
>
> I don't think Edelman's focus on memory is especially explanatory w/r
> to "understanding". He's drawing us back to thinking the meaning of
> "understanding" is some first person ghostly process we can introspect
> about and point to with our mind's eye. That's so not helpful.
> Introverts who endlessly obsesses about their interior life shouldn't
> be allowed to drive a Wittgensteinian car as they'll just hit the
> first tree they come to. They won't understand.
>
> Kirby
>
> > SWM
> >
> >
>
Wittgenstein was very much an introvert according to all the evidence.
SWM
Other related posts:
