Prometheus and Mega Lists, May 1999 (Part One)
From: "David H Slater" <dave.slater@telinco.co.uk>
To: "THEFT OF FIRE" <theft_of_fire@egroups.com>, <fire@wwwh.com>
Date: Fri, 14 May 1999 16:47:07 +0100
Subject: [theft_of_fire] Intelligence & thinking
Things have gone rather quiet on these lists lately. I am including below an
article I have offered to James for possible inclusion in the next GoF. Your
comments would be appreciated, and I am sure James would appreciate anything
you might like to write in response.
I am sending this to both lists and apologise to those of you who will get
it twice.
Very best wishes
Dave Slater <dave.slater@telinco.co.uk> or <dave.slater@connectfree.co.uk>
web site: <http://www.fortunecity.com/meltingpot/neches/212/>
Intelligence and thinking – an uneasy partnership
What characteristic of humans is it that gives them the power to change the
world, for better or worse, and know they are doing so? Their intelligence?
No. It is their ability to think; to use their minds to plan, organise and
control change. Of particular importance is their ability to create new
ideas and concepts; to generate novel solutions to both old and new
problems. Intelligence plays an important part in this, but it is primarily
the ability to think that enables us to advance.
Intelligence is the engine that drives thought and one would hence expect
highly intelligent individuals to be powerful thinkers, changing the world
for the better with their powerful minds. Some highly intelligent people
are, without doubt, also powerful thinkers, but the correlation between
intelligence and powerful thinking ability is well below unity, as anyone
who has spent time around the various high-IQ societies will be able to tell
you. Intelligence and powerful (or even capable) thinking do not always live
together happily.
A discussion of intelligence and thinking would be facilitated by a
reasonable understanding of the mechanisms that operate when we use our
minds. Therein lies a serious problem. Our knowledge of the mechanisms of
mind is scant; the inner space of the human mind presents at least as great
a challenge to science as our exploration of outer space. The exploration of
inner space is probably the more important of the two endeavours.
Although we know far too little of the 'physical' mechanisms that operate
when we think, some progress has been made in reaching an understanding of
the 'logical' processes that constitute thinking. This understanding has
been reached mainly by analogy, and is pragmatic in that it is shown to have
value because it works. Edward de Bono has done a lot of work in this area
and I can recommend his many books, but particularly 'The Mechanism of
Mind', to all readers of this essay. I have developed my own analogy to the
mind for use here, but it owes much to de Bono's ideas and I acknowledge my
debt to him.
Before we look at my analogy to the mind I would like to briefly discuss
another analogy, which is common, wrong and misleading. This analogy
compares the mind to an electronic computer and sees it as a mechanism for
processing sensory data algorithmically or procedurally. To those of us who
work closely with computers and understand what goes on 'under the bonnet',
the analogy appears ridiculous. Why? Because the very things the mind is
good at doing computers do very poorly and vice versa. For example,
computers can process huge numbers of calculations quickly, repetitively and
without error, a skill that is way beyond humans (except for a few who
appear to have their brains wired in an unusual way, and tend to be
seriously deficient in other areas). Conversely, humans can distinguish
faces with ease, a skill that eludes even the most powerful computers.
It has been argued that computers only have problems with the things humans
do well because they are, as yet, too slow. Certainly it is true that as
processors have become faster computers have been able to simulate more and
more human mental activities. The key word in the last sentence was
'simulate'. Electronic devices can process many millions on instructions
each second. It is unlikely that biological devices, no matter how advanced,
could do the same. Computers use brute force to simulate rather than emulate
human thought.
Of course, the human brain is not a single processor, but billions of
processors working together in parallel and with huge inter-connectivity,
and this may explain why the mind has capabilities denied to computers. We
are beginning to develop computers based upon the massive parallel
processing model and it may be that future computers will provide a better
analogy for the mind. Present day computers provide a seriously flawed
analogy, and one that has led to mistakes. A Pentium III processor may
enable a computer to do more challenging things, but speed of thought does
not necessary imply a more powerful mind.
The human mind is not a data-processing machine in the same sense as an
electronic computer, but an experience association device (or a
self-organising memory surface). That is, it is concerned not with
processing information using predefined rules, but with linking our
experiences together in a complex pattern, which then controls the things we
think, say and do. Each experience we have changes the pattern, and our
minds therefore have their abilities to respond to further stimuli modified.
Now for my analogy to the mind. Imagine a newly formed heavenly body (a
planet). It is perfectly spherical with a smooth, perfectly consistent
surface. At the surface the temperature is quite high so that ice will melt
and water will evaporate, but not too quickly. An ice meteor hits the
planet, forming a crater from which the water evaporates. A second meteor
also impacts the planet, forming another crater. Eventually the surface is
covered in impact craters. The ice meteors still arrive, but now they are
received slightly differently. They do not make perfectly formed craters, as
did the original meteors. The craters formed by these later meteors will be
influenced by those that already exist, and the water from the melting ice
will flow from one crater to another, opening and widening channels. Thus,
over a long period of time, the surface of the planet becomes a reactive
surface, 'remembering' its previous experiences (impacts) and 'interpreting'
each new experience in a way that is heavily influenced by those previous
experiences.
I cannot prove that the mind works like this idealised planet, but the idea
is not original to me and it make a lot of sense (far more sense that the
idea of the mind as a computer). The concepts of thought and intelligence
fit well with this analogy; thinking is akin to the flow of water and
materials along the channels and intelligence is like the inter-connectivity
between the craters - the greater the inter-connectivity the more complex
will be the responses to new impacts.
However, the analogy also highlights one of the weaknesses of the mind, a
factor which limits our ability to think powerfully. New impacts
('experiences') will be processed by an increasing complex memory surface,
but in a totally deterministic way. Intelligence may provide a more complex
(and perhaps faster) response to an experience, but that response remains
deterministic and has no potential for novelty.
Thus the mind as an experience association device will very efficiently
process new experiences and find solutions to problems according to learnt
rules. A highly intelligent mind will have a wider range of available
processes with which to formulate its response. [Going back to the planet
analogy, an 'intelligent planet' may be one that has a surface material that
is better able to record impacts and the channels that form between the
craters.] The processes will still be deterministic, though they may not
appear to be so.
If (and it is a big IF) the analogy usefully reflects the nature of the
mechanism of mind, then we have identified a serious limitation that needs
to be addressed. The effective thinker (as opposed to the simply
intelligent) needs to recognise that there are limiting factors upon his/her
mental capabilities and find ways of overcoming them. Let us examine a few
of the ways in which an experience association device may limit our ability
to solve problems.
1. The Dead End. To reach a solution using the experience association device
we need to follow channels from one crater to another, until we find a
desired destination. However, we do not have a map and the first route we
take may lead us to a dead-end; a crater from which there is no other exit.
Most of us have experienced thinking dead-ends from time-to-time. The
problem is exacerbated if we do not know which crater we are heading for.
2. The Partial Solution. Assuming that the solution we are looking for is
out there somewhere in mental space, then our task is to find that solution
and no other. There may be several acceptable solutions and we then need to
find only one of them. But after much searching we are fatigued and find a
'solution' that almost fits the bill and our search stops. We settle for a
slightly inadequate resolution to our problem.
3. The Obvious Solution. This is the easy channel out of the first crater,
and we are pleased to have found the going light. But very soon the going
may get tough, and we may not be able to go back. In contrast, a harder
start may lead to easier going soon afterward.
This list is by no means exhaustive.
* * * * *
If this analogy to the mind is valid (in the sense of being useful, rather
than 'true') then it is the very nature of mind itself which limits thought.
All our thoughts are naturally directed in particular directions, and
towards predefined solutions, by the patterns that have been formed by our
previous experiences. There is no escaping the fact that we are not
naturally creative thinkers; our minds require that we process new
information in ways determined by the things that have happened to us
previously. Intelligence, even very high intelligence, does not change this
fact.
Are there ways to overcome this limitation? Edward de Bono believes there
are ways and has written a whole library of books on the subject. His
techniques rely upon finding methods to navigate the pre-formed patterns in
our minds without having to follow the existing channels. On the analogous
planet this would be a little like using a helicopter to get a wider view of
the terrain and then moving directly to the desired location. The most
widely known techniques for this are 'lateral thinking' and 'brainstorming'.
For this short essay I do not have the time to discuss these techniques, but
refer interested readers to de Bono's work (a short list of some key titles
is included as an appendix). My concern here is to stress the importance of
not relying upon intelligence alone to provide us with the mental tools we
need. We in the high-IQ societies are proud of our intelligence (and why
not?) and this carries with it the risk of complacency. It is good to have a
powerful engine under the bonnet, but a powerful engine is wasted, and
possibly dangerous, if we do not know how to drive the vehicle properly.
With effort, we can learn. Then, perhaps, we shall begin to make a real
contribution to the progress of humanity.
Dave Slater
10 May 1999
APPENDIX
I am listing here five of my favourite books by Edward de Bono. The first is
the best, and the one that is most relevant to the above article. They are
all published (in the U.K.) by Penguin Books.
The Mechanism of Mind
Lateral Thinking
Po: Beyond Yes and No
The 5-Day Course in Thinking
Practical Thinking
Date: Fri, 14 May 1999 16:44:33 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: David H Slater <dave.slater@telinco.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com
Subject: [theft_of_fire] Re: Intelligence & thinking
Interesting analogy. However, it may be limited by the fact that it's
deterministic from the start. The brain is not, after all, a planet being
impacted by comets - that's a relatively simple physical system, even with
mounting topographic and hydrological complexity - but a complex dynamical
system that functions on a microscopic scale. Other theories incorporate
quantum indeterminacy and coherence at a variety of interfaces, including
cytoskeletal microtubules and neural synapses (to whose transmissive
characteristics certain quantum processes are said to be critical,
although I'm unclear on the proposed mechanisms).
David writes, "However, the analogy also highlights one of the weaknesses
of the mind, a factor which limits our ability to think powerfully. New
impacts ('experiences') will be processed by an increasing complex memory
surface, but in a totally deterministic way. Intelligence may provide a
more complex (and perhaps faster) response to an experience, but that
response remains deterministic and has no potential for novelty." This
follows from the analogy, but again, is deterministic. It also ignores the
prospect of experiential reduction...that mounting experiences provide
cumulative information on how to reduce the associated complexity to a
simpler and more efficient energy surface.
David continues, "Thus the mind as an experience association device will
very efficiently process new experiences and find solutions to problems
according to learnt rules. A highly intelligent mind will have a wider
range of available processes with which to formulate its response. [Going
back to the planet analogy, an 'intelligent planet' may be one that has a
surface material that is better able to record impacts and the channels
that form between the craters.] The processes will still be deterministic,
though they may not appear to be so." Again, we have determinism.
The distinction between intelligence and power is also problematic. The
reason: what David calls "power", or that which enables a mind to think
clearly and effectively to effect desired change in the environment, can
be posed as a problem (what it is and how to do it), and the problem can,
by the definition of intelligence as problem-solving ability, be solved
using intelligence. This higher level of intelligence relates to ordinary
intelligence as a metalanguage relates to a language, i.e., in such a way
as to unbind mental (and experiential) syntax and its real-world
interpretation so that they can be "transcendentally" optimized.
If the mind is deterministic, then so is intelligence, and so is "power".
So any effective scheme to rise above experiential topography also has a
deterministic implementation. Such a scheme is informational, and the
information can be learned. It would follow that intelligence, like
experience and learning, can be acquired...i.e., that people can "learn to
be more intelligent", provided that they possess a critical threshold of
efficiency in certain neural parameters.
In addition to the basic transmissive and switching parameters known as
"g", these parameters would include the learning function...the function
by which a neural network weights its own connections according to a
combination of existing structure and input. Neurologically, the
efficiency of this function would constitute a "higher level of g", call
it g', that depends on g and therefore cannot be factor-analytically
distinguished from g itself. This {g,g'} complex can itself yield a
yet-higher order of efficiency g'', and so on for higher-order complexes
{g,g',g''},..., the nested (convergent) sequence constituting an
intellectual "self-optimization faculty". Higher orders of intelligence
may account (at least in part) for the relative deficiencies sometimes
found to exist in correlations between standard IQ test data and data from
power IEQ tests, which may involve the modification of neural connections
over time.
This idea, which is original as formulated, was previously introduced on
MegaList.
Chris Langan
Date: Sat, 15 May 1999 13:42:08 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: David H Slater <dave.slater@telinco.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com
Subject: [theft_of_fire] Re: Intelligence & thinking
David writes, "My analogy does not relate to the brain, a physical entity,
but to the mind, a conceptual entity. This distinction is important. It
was not my intention to explain anything about the way the brain works,
but simply to explore the way we think. Although thinking necessarily uses
the brain as its physical platform, mind is an emergent property and can
be properly considered independently."
This is very promising in that it recognizes a potential distinction
between mind and brain. There is no a priori reason to identify the two;
we have grounds for isomorphism, but not identity (i.e., we can assume
identity only up to isomorphism, a mathematical fact lost on (e.g.) Dan
Dennet). But at the same time, it's important to remember that
epiphenomenalistic approaches to mind - and this is what is most often
suggested by "emergence" - are doomed by the fact that purely subjective
attributes are effectively excluded from scientific reality by the
(objectivistic) scientific method.
Good point about skill and discipline being essential to powerful
thinking! But if one wished to be argumentative, one could of course say
that those who fail to develop the requisite skill and discipline are "not
intelligent enough" to recognize the necessity. In any case, for skill to
be acquired and discipline to be effective, intelligence is a primary
necessity.
Of course, I'm in 100% agreement about the IQ-IEQ distinction. All in all,
a well-considered response!
Chris
Date: Sun, 16 May 1999 14:58:08 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Heather <hlp@alum.mit.edu>
cc: fire@wwwh.com, THEFT OF FIRE <theft_of_fire@egroups.com>,
IAN <I.M.Hocking@exeter.ac.uk>
Subject: [theft_of_fire] Re: Intelligence & thinking
On Sun, 16 May 1999, Heather wrote:
> > BTW: I agree completely with the comments on powerful thinking and
> > intelligence - that a high IQ does not guarantee clear thinking - and would
> > like to add a reminder that there are many more facets to the mind of an
> > excellent thinker than intelligence, such as: creativity, motivation,
> > determination, knowledge, common sense, humour, wit........ but I won't go
> > on.
>
> If we could just add "compassion" to that, the social IQ would increase
> dramatically as well. It's not required for someone to think excellently about
> abstract topics, but it would certainly help when dealing with other people, or
> topics that are fundamentally social. Which is amusing -- I have always
> disliked the "squishy" subjects and here I am defending compassion.
Probably
> because it tends to mean the difference between a brilliant person and a great
> one.
>
> Heather
First, let me say that as a newcomer, I find the routing here a bit more
complex than expected. For one thing, I've been in TNS for a year, and
just now learned they had an email list...that interfaces with "fire"
through something called "groups". Next, I'm seeing random second-hand
pieces of messages I didn't receive first-hand. Anybody want to clue me in
on the routing, or direct me to a FAQ?
Regarding content, one of my original points was this. As we examine
progressively higher levels of intelligence - i.e., levels of
"intelligence applied to itself" - it becomes increasingly hard to
separate intelligence from other (involved) factors like motivation.
This is especially obvious with the power IEQ tests, where the
accurate measurement of higher levels of intelligence depends on the
presence of sufficient motivation to permit the measurement. The bottom
line: the higher the ceiling (space and time complexity) of any test of
problem-solving ability, the more dependent the results on the motivation
of the subject. So from a factor-analytic standpoint, one is no longer
dealing with separate factors.
The scheme involving "levels of g" reflects the fact that intelligence
displays a self-optimization imperative whether associated with mind or
brain, and that this imperative is goal-directed. Higher levels of g
reflect the self-optimizative "motivation" of the intellect. This scheme
was presented in terms of neural structure and functionability to
emphasize the fact that mind and brain are isomorphic, and that to show
that whatever we say about intelligence must be expressible in neural
language.
Let's look at it another way. Most people want to quantify intelligence
in proportion to the maximum complexity of problems that can be
"effortlessly" solved by the subject. But as we all know,
"effortlessness" applies only in the bottom of our range, some amount of
effort being required as the problems get harder. So a motivation-free
functional description of intelligence is impractical. This leaves the
prospect of a structural description. But until we can scan the brain's
neural structure and measure the excitative and inhibitive connection
weights, neurotransmitter concentrations, etc. in vivo, no structural
measurement will be possible.
In other words, separating intelligence from motivation might be
semantically convenient for referring to the structural aspects of
intelligence, but remains impractical from a psychometric standpoint.
While intelligence and motivation may be distinct mental factors simply
inasmuch as one can logically distinguish them, the distinction may be
worth little for practical purposes.
I agree emphatically with Heather's comments on compassion and "social
intelligence". The ethical intelligence of individuals - the extent to
which they are able to recognize themselves as part of the identity of a
larger (social, planetary) entity and behave accordingly, or equivalently,
acknowledge higher orders of their own identities - reflects the
involvement of social utility in motivation and rationality. Ultimately,
we arrive at a cosmological formulation of the whole issue...mind
considered as an essential aspect of reality (CTMU territory).
Heather, could you please let me know if this somehow gets routed away
from the TOF list? And Steve, are these lists archived?
Date: Sun, 16 May 1999 23:31:26 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com,
IAN <I.M.Hocking@exeter.ac.uk>
Subject: [theft_of_fire] Re: Intelligence & thinking
On Sun, 16 May 1999, Paul wrote:
> <Artists 160.... Statesmen 165>
>
> >Naturally these numbers are pretty inexact, but nevertheless
> >it seems to be the case that most geniuses have an IQ of
> >around 3-4 sigma or more.
>
> Pretty inexact indeed, although what I would expect (i.e. that geniuses
> would have IQ's of 3-4SD above the mean rather than 6-8SD above it [or 12SD+
> as claimed for Sidis!]).
>
> [Actually, I think I've read some of Cox's book - they have an old, dusty,
> fragile copy at the University library. The "estimates" were just that as I
> recall - with no IQ tests administered. I forget the details, but I seem to
> remember that they were dubious.]
I think we may be missing the point here. Intelligence is what real
geniuses have by definition. If any real-world genius shows what might be
interpreted as a "deficiency" on an IQ test, it means that the test is not
measuring "intelligence". The test is measuring some kind of shallow
facility, conceptual glibness, the quickness and agility of a housefly
zipping randomly from puzzle to puzzle without ever manifesting the higher
intellectual functions necessary to advance human knowledge. To say that
Sidis was eight standard deviations above Einstein in "intelligence" is
nothing but a thoughtless misuse of language fostered by psychometrics run
amok.
Date: Mon, 17 May 1999 01:37:50 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com,
IAN <I.M.Hocking@exeter.ac.uk>
Subject: [theft_of_fire] Re: Intelligence & thinking
On Sun, 16 May 1999, Paul wrote:
> >I think we may be missing the point here. Intelligence is what real
> >geniuses have by definition.
>
> Which definition? What are "real" geniuses?
People who significantly advance human knowledge or artistic expression.
>
> >If any real-world genius shows what might be
> >interpreted as a "deficiency" on an IQ test, it means that the test is
not
> >measuring "intelligence".
>
> Not measuring intelligence at all? Are you sure?
Absolutely. If it were measuring intelligence (as opposed to a mere
component thereof), it would be registering the genius's intellectual
superiority over desultory puzzle aficionados.
>
> >The test is measuring some kind of shallow
> >facility, conceptual glibness, the quickness and agility of a housefly
> >zipping randomly from puzzle to puzzle without ever manifesting the higher
> >intellectual functions necessary to advance human knowledge.
>
> In your opinion, will a 5-year-old child have the same level of intelligence
> in 30 years time? Maybe yes, maybe no: it depends upon your (perhaps
> idiosyncratic) definition of intelligence.
The search for a time-invariant measure of intelligence has barely begun.
There is nothing "idiosyncratic" about the distinction between
time-invariant and time-dependent measures of intelligence.
>
> >To say that Sidis was eight standard deviations above Einstein in
> >"intelligence" is nothing but a thoughtless misuse of language fostered
> >by psychometrics run amok.
>
> I think that this point has already been made.
But perhaps not absorbed. Relax - I don't want to split hairs with you.
The object is to expose and reject the wealth of possible bad assumptions
embedded in the standard (increasingly inadequate) view of intelligence.
Date: Mon, 17 May 1999 13:35:48 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Kerry Williams <kerry@micronet.net>
cc: fire@wwwh.com, THEFT OF FIRE <theft_of_fire@egroups.com>,
IAN <I.M.Hocking@exeter.ac.uk>
Subject: [theft_of_fire] Re: Intelligence & thinking
On Mon, 17 May 1999, Kerry Williams wrote:
> At 9:37 PM -0800 5/16/99, Langan wrote:
>
> >On Sun, 16 May 1999, Paul wrote:
> >
> >> >I think we may be missing the point here. Intelligence is what real
> >> >geniuses have by definition.
> >>
> >> Which definition? What are "real" geniuses?
> >
> >People who significantly advance human knowledge or artistic expression.
>
> This definition may result in the recognition of "geniuses" who are only
> near our level. I've long held the opinion that a high IQ is as much a
> measure of being different as it is a measure of better problem solving
> abilities. The significance of this is that we are better at understanding
> those who are near our level much better than those who are well above *or
> below*.
Good point. However, when I say "advance", I mean just that. If Einstein
had never achieved recognition, he'd still have "advanced human knowledge"
by dint of being human himself. Immediate recognition is not necessarily
implied. This, of course, means that some geniuses - especially the
ones far ahead of their time - are hard to spot. But that's better than
making genius a popularity contest.
> Among other things, this means that it is much easier for us to distinquish
> between Prometheus members and TNS members, for instance, than between
> Mensa members and average people. It also means that it is easier for the
> average person to distinquish between themselves and Mensa members than to
> distinquish between higher IQ group members and anyone else (even himself;
> one reason that geniuses are often seen as just nuts). The same goes for
> the recognition of accomplishment and its communication.
>
> The average man can recognize an act of "genius" coming from a Mensan. But
> he has to be told that relativity is one because it's so far beyond his
> comprehension. The other side of the coin is that I can't tell when an
> average man does something well beyond what he is normally capable of, just
> as he can't tell when I do something significant. We're just too far from
> mutual understanding.
Once again, any relativism in the term "genius" is dispelled by the fact
that we're defining it in terms of absolute intellectual achievement...a
measure that exists relative to the entire species, not just certain
groups.
> I see no reason not to believe that the same would be true of our (lack of)
> recognition of someone several standard deviations above us.
>
> How do we determine what is or is not going to be recognized as genius a
> century or two from now? I strongly suspect that often what we can
> immediately see as a "significant advance in human knowledge" will someday
> be seen as just another incremental advance. And that the true significant
> advances won't be noticed until enough incremental advances have been made
> to notice that someone has already been there, done that, a long time ago...
When this is the case - when a lot of "incremental advances" are necessary
to recognize that someone else has been there and done that - we must
still explain why, if one has really been "beaten to the punch", the
repetitive nature of one's achievement was invisible without the
clarificative increments. Perhaps it took genius just to realize the
modern applicability of an old insight.
> >> >To say that Sidis was eight standard deviations above Einstein in
> >> >"intelligence" is nothing but a thoughtless misuse
> >> > of language fostered by psychometrics run amok.
>
> Perhaps. But maybe he just couldn't find the language, the common ground,
> to communicate his best insights to those down at our level.
I've read the essay on consciousness that Sidis fans have posted on the
web, and agree that there was something there...something that might have
been developed had Sidis not become discouraged. But eight standard
deviations above Albert "Revolution in Space and Time" Einstein? Come on.
I don't care how old somebody is when he learns his second language, he
doesn't get to enjoy an 8 sd "intelligence" advantage over somebody like
Einstein (or in all likelihood, Feynman) as a result.
Chris
Date: Mon, 17 May 1999 16:01:27 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: David H Slater <dave.slater@telinco.co.uk>
cc: fire@wwwh.com, theft_of_fire@egroups.com
Subject: [theft_of_fire] Re: Intelligence & thinking
On Mon, 17 May 1999, David H Slater wrote:
> -----Original Message-----
> From: Langan <clangan@suffolk.lib.ny.us>
>
> >I think we may be missing the point here. Intelligence is what real
> >geniuses have by definition.
>
> Not so. Genius has more to do with the ability for original thought
> ("creativity") than with intelligence. The first definition in my Collins
> Concise English Dictionary is: "a person with exceptional ability, esp of a
> highly original kind". This suggests to me that intelligence is not even a
> prerequisite of genius, and certainly not its defining factor.
Again, a good point. What the problem really comes down to is "what are
the defining factors of intelligence and genius?" When we try seriously
to answer this question, we see that it's ultimately as pointless to
distinguish between creativity and intelligence as between motivation and
intelligence. A monkey can get very "creative" with fingerpaints, but
without intelligence, the creativity is empty. Similarly, whenever one
solves a very difficult problem without prior examples, one is necessarily
thinking creatively. It is fashionable to speak of intelligence as somehow
separate from all of the "other factors" of intellectual performance. But
when we look closely, we find that removing these "other factors"
effectively vitiates the concept.
> A 'quickie' definition of a genius is: "a person who has at least two great
> ideas in his/her lifetime". This lacks depth and rigor, but captures the
> feeling of the term.
You're right. Unfortunately, this compels us to specify what constitutes a
"great idea". Sometimes it's not so obvious up front.
Chris
Date: Mon, 17 May 1999 17:24:18 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: theft_of_fire@egroups.com, fire@wwwh.com
Subject: Re: [theft_of_fire] Re: Intelligence & thinking
On Mon, 17 May 1999, Paul wrote:
> >>I think we may be missing the point here. Intelligence is what real
> >>geniuses have by definition.
> >
> >Not so. Genius has more to do with the ability for original thought
> >("creativity") than with intelligence. The first definition in my
Collins
> >Concise English Dictionary is: "a person with exceptional ability, esp of a
> >highly original kind". This suggests to me that intelligence is not even a
> >prerequisite of genius, and certainly not its defining factor.
>
> Indeed. I was going to make a similar comment.
>
> It seems to me that awesome creativity and originality of thought are the
> characteristics at the core of genius. Genuine passion for the subject and
> dedication (hence hard work) are also clearly of central importance. And
> although I've already said that I think intelligence is generally
> over-emphasized, I feel that some minimum level is probably necessary for
> the highest levels of achivement in a given field.
>
> Chris should remember that there are many forms of genius (e.g. the creative
> genius of Mozart and the comic genius of Peter Cook). It is a mistake to
> suppose that the unique capacities of such individuals would or should be
> identified (even "measured") by some hypothetical and idealised
intelligence
> test.
True enough. However, this is equivalent to saying that there are many
ways of *expressing* intelligence or genius, not that there is a different
kind of intelligence for each. Problems of content or design formulated in
some logically-structured mental language, be it conceptual or aesthetic,
are getting solved in every case. That's why all the usual categories of
genius cross-correlate on IQ tests...why a musical or artistic genius is
as likely to have a superior IQ as a mathematical or philosophical genius.
That this has become a largely semantical discussion owes to the fact that
"genius" is already used to refer to two things, extreme general
intelligence and extreme context-specific talent (I was using it in the
first way; Paul is now using it in the second). But these are just two
ends of one similarity mapping between general and specific intellectual
ability. Either way, creativity is ultimately an inseparable aspect of the
concept.
Date: Mon, 17 May 1999 18:42:15 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: "Russell F. Vaughan" <fred@vaughan.cc>
cc: fire@wwwh.com
Subject: Re: [theft_of_fire] Re: Intelligence & thinking
On Mon, 17 May 1999, Russell F. Vaughan wrote:
> A very interesting discussion.
>
> > That this has become a largely semantical discussion owes to the fact that
> > "genius" is already used to refer to two things, extreme general
> > intelligence and extreme context-specific talent (I was using it in the
> > first way; Paul is now using it in the second). But these are just two
> > ends of one similarity mapping between general and specific intellectual
> > ability. Either way, creativity is ultimately an inseparable aspect of the
> > concept.
>
> I am reminded of a grady Towers GoF article from back in the eighties
> called something like, "The Twisted Pear." In that article Grady showed
> that creativity correlates with intelligence up to between the three and
> four sigma level and above that does not correlate at all! I buy it!
> There are those among us who are extremely brilliant in an encyclopedic
> sense who I would doubt could come up with a method of breaking out of a
> wet paper bag if they had not read about it first and others that I
> sense might blow a hole using nuclear fusion through a wall rather than
> use something that has been tainted by normality.
Well, Grady was probably equating intelligence to the standard definition
of "fluid g". But I've already attempted to distinguish higher levels of
g, in effect stratifying the intelligence concept. So in addition to being
statistically coupled to g - which would account for the hypothetical lack
of correlation above the ceilings of g-loaded tests - (higher levels of)
creativity may couple to higher levels of g. One way or the other, I can
challenge any attempt to say that creativity is a meaningful intellectual
performance factor beyond the point where it ceases to correlate with
intelligence. By the nature of its construction, the phrase "unintelligent
creativity" denotes nothing but a waste of time.
We have to be cautious when we make statements about the dependence and
independence of terms that have never been satisfactorily defined. I've
already pointed out that any attempt to independently define the terms
"intelligence" and "creativity" is doomed. This reflects a feature of
the
deep logic governing the assignment of intellectual attributes.
> I definitely side with those who think that "creative genius" (if we
> could be a little more specific) is something that correlates with
> intelligence up to a point, but is definitely *different*.
I disagree. What I say is this: only tentative, contextually-restricted
distinctions between these terms can possibly exist. That is, they remain
semantically coupled at all levels of predicate logic, and unavoidably so
at maximum generality.
> As far as Einstein and Feynman are concerned -- they both seem extremely
> challenged by philosophy to my mind.
Do you mean "challenged by the philosophy of mind"? Because otherwise, I'd
have to point out that technically, Einstein started out as a topnotch
metaphysician rather than a physicist. That is, he was forced to employ a
metalanguage of the language of classical physics (this is a purely
logical distinction). Somewhat the same can be said of Feyman regarding
his path integral approach to quantum mechanics; by utilizing the concept
of advanced waves, he was forced to reject another component of the
language of classical physics, one-way time. That's a metaphysical and
therefore philosophical operation (as Feyman apparently realized, since
he talked extensively about the "equivalence of representations" of
physical reality). Philosophy bagan as the Mother of Science and will
always so remain.
I'll bet your talking about what was really Einstein's "greatest blunder":
not the cosmological constant, but the way quantum mechanics stuck in his
craw!
It's a great pleasure to finally exchange views with the great Fred
Vaughan. I really like some of your contributions to GoF.
Chris
Date: Mon, 17 May 1999 19:12:13 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com,
IAN <I.M.Hocking@exeter.ac.uk>
Paul opines, to some extent correctly, that it is oversimplistic to
wholly equate genius (or creativity) and intelligence as currently
(fuzzily) defined. But it may still be possible to *exactly* define them
so that they possess enough internal variety to accommodate each other as
distributed aspects. Not to dwell too heavily on the logic, but this
amounts to saying that in order to properly define these terms, one is
forced to do so recursively (i.e., so that the definition of each
incorporates the other). This semantical coupling then characterizes all
parts of their interactive relationship, with context-specific
distinctions remaining possible. The reason that no such definition has
yet been presented is that their actual relationship has been only poorly
understood.
Paul then writes,
> Exactly. Personally, I would not equate "extreme general intelligence"
> with "genius", although I would tend to expect the former in the latter.
> However, amongst those with extreme general intelligence, I would also
> expect a large number of "non-geniuses" - individuals who are
> intelligent, yet lack one or several important attributes required to
> create "works of genius". Amongst this group there will be people with
> even higher IQs than some of the geniuses. There is also a group of
> people that have the raw capacity for genius, yet did not use it for one
> reason or another - perhaps bad luck. These could be called
> "potential geniuses".
This is well-thought, but a subtle distinction still needs to be
explicated. If a nongenius has a higher IQ than a genius, it's because
the IQ test he took didn't really measure his intelligence. An ideal
intelligence test would not permit such a discrepancy; the kind of
intelligence it measured would truly be "general" enough to accommodate
real intellectual excellence (genius). Pursuing this line of thought,
two geniuses displaying equivalent facility and power in mental languages
corresponding to distinct talents would end up equal on such an ideal
test, and anyone scoring higher than anyone else would simply be
expressing his higher "intelligence" through a different set of talents or
mental languages.
Chris
Date: Mon, 17 May 1999 22:42:33 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com,
IAN <I.M.Hocking@exeter.ac.uk>
On Mon, 17 May 1999, Paul wrote:
> >This is well-thought, but a subtle distinction still needs to be
> >explicated. If a nongenius has a higher IQ than a genius, it's because
> >the IQ test he took didn't really measure his intelligence. An ideal
> >intelligence test would not permit such a discrepancy; the kind of
> >intelligence it measured would truly be "general" enough to accommodate
> >real intellectual excellence (genius). Pursuing this line of thought,
> >two geniuses displaying equivalent facility and power in mental languages
> >corresponding to distinct talents would end up equal on such an ideal
> >test, and anyone scoring higher than anyone else would simply be
> >expressing his higher "intelligence" through a different set of talents
or
> >mental languages.
>
> I prefer to say that "intelligence" is one narrow faculty of mind that can
> occur (to varying degrees) independently of the other faculties, though I am
> aware of the "interference" that must occur in any one person's brain, in
> practice. I'm not happy about describing mental attributes as "faculties",
> but it helps to explain my position: essentially, I find it unsatisfactory
> to reduce the rich diversity of human mental capacities to a single spectrum
> or hierarchy of underlying intelligence. How useful would it be, for
> instance, to say that Mozart and Einstein are of equal "intelligence" (as
> revealed by the perfect hypothetical test you mention)? It seems more likely
> that Mozart had a brain that lent itself to music, a different *type* of
> brain to Einstein, wheras your concept would seem to suggest that Einstein
> could have been a musical genius and Mozart a nobel preizewinning physicist.
To have interference you need independence; if one thing is dependent on
the other, then the other plays a causal role, and the dependent quantity
can *only* occur in conjunction with some specific configuration of the
independent variable. But if two attributes cannot be independently
defined, then we have dependence on a logical level. Another thing to
remember: there are powerful philosophical objections to a pluralistic
view of reality, and any insistence on qualitative "diversity" runs into
them when pushed beyond a certain level of generality.
Yet another point to consider, especially for a neurosurgeon-in-training:
if an attribute is really "intellectual", then it can be expressed in the
neural language of the brain. Now, fluid g has already been identified
with basic neural efficiency, involving neural and synaptic density,
distribution, transmission and switching. Crystallized g then inheres
in learned synaptic weights and firing thresholds, and higher levels of g
in a dynamic involving both. The question thus becomes: how is creativity
expressed in neural language without being expressible in terms of g and
its variants? It's not a question of interference, but of embedment.
In contrast, apparent distinctions between talent and intelligence may in
some cases devolve to anatomical and physiological variation...structural
and dynamical peculiarities in specific parts of the brain. This would
imply that in people with specific talents, there is "more intelligence
and creativity" in one part of the brain than another. But is not a brain
whose intelligence and creativity are better-distributed - i.e., a brain
"more of which" is intelligent - not in fact "more intelligent"? If we
want to move towards a more logical view of genius and intelligence, the
answer must be affirmative. After all, who would most of us say is more
intelligent: a genius whose ability is limited to one talent, or a
"universal genius" who can range freely among expressive languages?
Extreme cases of the latter are da Vinci and von Goethe; extreme cases of
the former are called "idiot savants". Perhaps the main point here is that
our analysis of intelligence and its meaningful quantification will
ultimately depend on what we learn about brain structure and dynamics as
the statistical approach to psychometrics gradually yields to a more
exact, neurologically-based science.
One last point: Paul talks about the important distinction between genetic
and environmental aspects of intelligence, and between potential and
actual genius. Beyond the point that environmental stimuli play a role in
the brain's structural development, no time-independent (input-
independent) description of intelligence is possible; even the
developmental potential may change nonlinearly over time. But a structural
description, as opposed to one based on actual performance trials, should
still be possible, given adequate technology. Brave New World, here we
come.
Date: Tue, 18 May 1999 19:56:34 -0400 (EDT)
From: Langan <clangan@suffolk.lib.ny.us>
To: Paul <paul@jitterbug.prestel.co.uk>
cc: THEFT OF FIRE <theft_of_fire@egroups.com>, fire@wwwh.com,
IAN <I.M.Hocking@exeter.ac.uk>
Subject: Re: Intelligence & thinking
On Tue, 18 May 1999, Paul wrote:
> This has been an interesting discussion, which I will continue to follow,
> but it's expanding rapidly and I can't personally devote any more time to it
> just now (especially since I have to write an essay on osteoporosis by this
> Friday). However, I would just like to ask Chris the following question
> again, because I'm not quite clear on his opinion:-
>
> How useful would it be to say that Mozart and Einstein were of equal
> "intelligence", as revealed by the perfect hypothetical test you mention?
It
> seems likely to me (bearing in mind that the brain is a complex,
> heterogeneous structure and not a simple, undifferentiated neural network)
> that Mozart had a brain that, by virtue of its physical structure, lent
> itself to music in some way - a different *type* of brain to Einstein.
> However, your concept would seem to suggest that Einstein could have been a
> musical genius and Mozart a nobel prizewinning physicist - as they were both
> simply of "genius level intelligece" [by your definition].
The utility would be less likely to show up in comparative evaluations of
the brains of Mozart and Einstein than it would be in designing an
artificial construct able to simulate both. Here, it is valuable to know
that the same essential faculty is being expressed in the neural languages
of music and mathematical physics.
>
> I would like to write something for GoF in the future [assuming that there
> would be some interest] on basic brain structure and neuroembryology, with
> particular emphasis on the differences between the structure and function of
> the brain and a connectionist neural network or computer. The reason is that
> I suspect some people have in mind a slightly over-simplified model of the
> brain and that this may be causing problems in relation to discussions such
> as these [for instance - Chris refers to the working brain as a "neural
> network" with shifting "synaptic weightings", presumably seen to
optimise
> behaviour (through learning) by the dynamic modification of input-output
> mapping].
I very much look forward to Paul's article. However, the brain remains a
neural network, and it necessarily exhibits synaptic weightings. What Paul
apparently means is that it also exhibits internal structural (and even
dynamical) distinctions. But in no way do my comments contradict this
(obvious) fact. The distinctions in question represent variations in the
distribution of neural subsystems and processes.
> > I'd also like to comment briefly on the following, from Chris:- >
> >Yet another point to consider, especially for a neurosurgeon-in-training:
> >if an attribute is really "intellectual", then it can be expressed in
the
> >neural language of the brain. Now, fluid g has already been identified
> >with basic neural efficiency, involving neural and synaptic density,
> >distribution, transmission and switching. Crystallized g then inheres
> >in learned synaptic weights and firing thresholds, and higher levels of g
> >in a dynamic involving both. The question thus becomes: how is creativity
> >expressed in neural language without being expressible in terms of g and
> >its variants? It's not a question of interference, but of embedment.
>
> Interesting, but I think that this low "level of analysis" of brain
> operation could be misleading in the same way that is it misleading to
> reduce such diverse objects as a child, a sandwich and a kettle to "systems
> of interacting atoms".
Paul is right - crucial distinctions can be missed if we strictly confine
ourselves to a low (i.e., distributed) level of neurological analysis. So
let's try not to do that! And if we must, in the present forum, do it
anyway, let's try not to do so in a way that will lead to contradictions
when we progress to higher levels of analysis.
Chris Langan