Inkwell: Authors and Artists

    
We're very pleased to Welcome Gary Marcus to the Inkwell.

Gary Marcus is a scientist and a writer. As a scientist, he is
director of the NYU Child Language Center, and a Professor of
Psychology at New York University. As a writer, he is author of three
books, including The Birth of the Mind, a book about genes and the
origin of the brain, and Kluge, a book about evolution and the
clumsiness of the human mind.


Leading the conversation with Gary is our own Bruce Umbaugh.

Bruce Umbaugh is a philosopher at Webster University, where he teaches
in St. Louis, MO, and online. His work nowadays is mostly about the
ethical and social consequences of technological choices, but he wrote
a book about the British empiricist philosopher, George Berkeley, too.
Back in the day, he wrote a dissertation about rationality that focused
especially on the clumsiness of the human mind.


Welcome, Gary and Bruce!
  

    
Gary, it's great to have you here with us. Thanks for coming by.

Your book, *Kluge*, takes us into an altogether fascinating domain. The
subject of these human frailties first came to my attention in the
literature on framing and choice, and on the limits of memory. It really
changed my thinking about how we should understand human reason (and
understanding).

How did you find your way into the subject?
  

    
Thanks for having me!

I came to Kluge, and the ideas within it, from a longstanding interest
in the development of the human mind. In a nutshell, my dissertation
was on how children acquire language -- itself one of the most
interesting mysteries in modern science -- and that led me to think
about the Chomskian notion of innateness. What is it that is built into
the human mind? And for that matter, how is *anything* built into the
mind? 

Questions like that in turn lead me to a great deal of reading (and
writing) about genetics and evolution, including a book called the
Birth of the Mind, about how a relatively small number of genes gives
rise to the complexity of the human mind.

One of the things I realized in the course of writing that book is
that evolution is (if you'll forgive the anthropomorphic metaphor)
incredibly stingy: once it finds something that works, it tends to
stick with solution, through thick and thin -- even if some other
solution might work a lot better.

So evolution inevitably winds up building what engineers call
"kluges", clumsy or haphazard patchworks that get things done but
really could be a lot better.

Realizing this, it occurred to me that the basic idea of evolutionary
psychology -- that the mind was superlatively well-adapted to the
environments of our stone-age ancestors -- was probably a little bit
off the mark.

Kluge is, in essence, my attempt to answer the question of "if
evolution is so good, how come the brain is so clumsy?"
  

    
Clumsy, indeed.

That evolutionary argument has always amazed me -- and some otherwise very
smart folks have endorsed it quite explicitly. Can you explain why people
find the argument so attractive? And what are some of the "clumsy" examples
that put pressure on it?
  

    
I'll save the examples of human clumsiness for later -- I'm in haste
now -- but take your first question first. Why *does* anybody think
that the mind is optimal?

In my view, there are at least three separate reasons why different
sets of people might at least want to believe that the mind is optimal:

1. Some people might be religious, and find it comforting to believe
that we are built"perfect in His image". A similar inclination
underlies many people's "belief in a just world". 

2. Some people might gravitate towards a notion of mankind as optimal
because they are mathematically-inclined, and find it appealing to
envision humans as neatly captured by a set of elegant equations.
(Hence Homo Economicus, or the "rational man" model popular in
economics, and also something more recent, the idea of humans as
"Bayesian" optimizers")

3. Others might be fans of biology that are so impressed with what
evolution does well that they forgot that evolution has downs along
with its ups...
  

    
> Homo Economicus, or the "rational man" model popular in
economics

Gosh, this seems entirely correct, given the week's events...
  

    

> a relatively small number of genes gives rise to the complexity of 
> the human mind.

That's fascinating.  I've heard many times that humans are genetically
similar to other primates, but I haven't heard what is almost the same
information expressed this way, that only a few genes make all that
difference. How can that be?

Welcome! 

(By the way, for those reading along out on the net, you're invited 
to email any questions or comments to inkwell[at]well.com with the
author's name in the subject line, and we'll pass them along to answer 
here. Or you can just sign up for The WELL and post it directly)
  

    
In response to Gail, who writes

"That's fascinating.  I've heard many times that humans are
genetically similar to other primates, but I haven't heard what is
almost the same
information expressed this way, that only a few genes make all that
difference. How can that be?"

Before I give you answer, let me start with a clarification:  when
people tell you that the human genome and the chimp genome are 98.5%
similar, what it really means is that if you take a hundred human DNA
nucleotides (A's, C's, G's, and T's) and compare them with their
chimpanzee counterparts, on average 98.5% will be similar, not that
98.5% of our genes are identical to chimps genes. In other words, 1.5%
of our 25,000 genes would be 375 genes, but what you find is not that
humans have 375 genes with no chimpanzee counterpart, but rather that
there are 1000's of genes with small differences.

That said, the way I think about the genetic difference between chimps
and humans is this: the genome is like an enormous library of what
computer programmers call subroutines: recipes (or more technically,
algorithms) that can be used and reused for different purposes. Once a
programmer has written code for displaying a window on a screen, he (or
she) can reuse that code many times, to display many different
windows. In a similar way, once evolution stumbles on a set of genes
that govern a basic function, say growing bones, it can reuse those
genes over and over again, to build many different bones.

In that light, even a relatively small number of changes could lead to
hugely different consequences. Think for example about what the
consequences might be for a gene that helped control how many times
one's brain's cells divided. A slight change could conceivably yield a
brain that is twice as big (or half as big). Much as recipe for a big
cake might be virtually identical to a recipe for a small cake, a
genome for a big brain might not need to be that different from the
genome for a small brain. 

The whole power and elegance of the genome comes from the way in which
old mechanisms can get combined in new ways.  Then again, much of the
clumsiness comes from the same place -- when most evolution comes from
the mixing and matching of spare parts, sometimes you wind up with
clumsy patchworks.

But that's a story for a later....

Gary
  

    
I'm enjoying your book very much. This may (or may not) be a tangent,
but it's what's on my mind: The Power of Political Misinformation

<http://www.washingtonpost.com/wp-dyn/content/article/2008/09/14/AR200809140237
5_pf.html>

"In a paper approaching publication, Nyhan, a PhD student at Duke
University, and Reifler, at Georgia State University, suggest that
Republicans might be especially prone to the backfire effect because
conservatives may have more rigid views than liberals: Upon hearing a
refutation, conservatives might "argue back" against the refutation in
their minds, thereby strengthening their belief in the misinformation.
Nyhan and Reifler did not see the same "backfire effect" when liberals
were given misinformation and a refutation about the Bush
administration's stance on stem cell research."

I'd love to hear your perspective on this, especially the difference,
if true, between 'liberal and conservative brains.'
  

    
This stuff makes so much sense...off to buy the book!
  

    
Gary, that's a great explanation.

Like sbmontana, I'm finding this a timely topic, since we're looking at the
sorts of patterns you discuss writ large at the social level. Susceptibility
to narrative over actual evidence in voting decisions, for example. Perhaps
overconfidence in our own smarts leading to missing some obvious things
(bond failures don't conform to the quants models because bond defaults
aren't independent events!) in the case of AIG, for another.

Can you help us understand what's going on with some of these patterns? And
how we could do better? Or are we just out of luck and stuck with our clumsy
endowments?
  

    
Last night I read the 'Language' chapter (I'm a poet) which was a
pleasure -- the next chapter. And today I'm back with another question.


The common wisdom just now is that, given our economic crisis, Obama
is in a stronger position than McCain, since the population wants
change. 

But you observe that, in crisis, folks want the familiar. Does this
mean that the Obama strategy of tying McCain to Bush might backfire, as
Bush is the 'familiar'?

Further, you note (on page 50):

"... all people tend to become more negative toward minority groups in
times of crisis ..."

Still further indication that this situation might put Obama in a less
favorable position?
  

    
Sharon, sorry to interrupt the chain of this conversation, but I
wanted to chime in with a few thoughts and questions to Gary for later
response.

I'm just digging into this excellent book and in the first chapter, on
human memory, I was prompted to notice an interesting feedback loop
that might exist between human and computational memory/logic. If I
understand your theses correctly, and have some dim understanding of
the pursuit of so-called "fuzzy logic" in computing machines, is it
possible that our engineers and scientists are attempting in that now
decades-long pursuit, to reverse engineer the useful but weirder
abilities of the human mind?

Provoking this question was, in part, a recent article in The NYT
<http://www.nytimes.com/2008/09/16/science/16angi.html?_r=1&oref=slogin>
on the innante approximate numeracy of not just mammals, but species
such as bird and reptiles whose brain and behavioural patterns
preceeded ours in their origins. My next questions are the opposite, in
a way of my first. Are we recently attempting to endow our machines
with superior performance in calculating numbers with precision and in
a big hurry that our own human minds cannot acheive? If so, will these
two evolutionary vectors perhaps merge, one at our own direction?

Btw, do take the time to check out the aprroximate number sense test
embedded in the article ref'd above.

http://www.nytimes.com/interactive/2008/09/15/science/20080915_NUMBER_SENSE_GR
APHIC.html
.

It's kind of spooky to have a peek inside of our ancient and
unconscious "lower" mental faculties. Good clean scientific fun!
  

    
Hi, Folks,

Just a brief note to say that while a conjunction of banker's panic
and travel has keep me from making my appointed rounds today, your
questions all look fascinating. 

Keep on posting, as questions occurs to me, and I'll do my best to
cathup by Sunday evening (if not sooner).

Cheers,
Gary
  

    
Another study, recently released: 

"People who startle easily in response to threatening images or loud
sounds seem to have a biological predisposition to adopt conservative
political positions on many hot-button issues, according to unusual new
research published yesterday.

"The finding suggests that people who are particularly sensitive to
signals of visual or auditory threats also tend to adopt a more
defensive stance on political issues, such as immigration, gun control,
defense spending and patriotism. People who are less sensitive to
potential threats, by contrast, seem predisposed to hold more liberal
positions on those issues . . . "

<http://www.washingtonpost.com/wp-dyn/content/article/2008/09/18/AR200809180226
5.html>

I tend to be skeptical of this study and its conclusions, not only
because it uses a very small sample, but because all the 'sensitive'
types I know are liberals. 

But I'd be interested in other folks' reading.

 
  

    
By the way, one reason that the human mind can seem awfully clumsy is
that the environment in which it spent most of its evolutionary history
has (for most of us) changed abruptly in the last 50,000 years or so
after the advent of civilization, and even more abruptly in the past
several hundred years of the industrial age.

(E.g., the tendency to fight with others may have been a good
adaptation in a tribal world with limited resources, but in this era of
international relations and WMD's, it could be, needless to say, very
bad.)
  

    
Hi, Folks,

Sorry for the delay in returning to the Well. 

When I look at the elections -- obviously the topic of greatest
interest here and in the country at large -- what strikes me the most
is how little effect current events seem to have on the polls. 

By any objective measure, the United States is in far worse shape now,
at the end of the Bush administration, than it was in 2000, at the end
of the Clinton administration.
Yet it looks fewer than 5% of voters will change their minds. 

What's going on? Four words: "confirmation bias" and "motivated
reasoning". Confirmation bias means that people notice evidence that
fits with their preconceived notions, "motivated reasoning" means they
spend more energy in critiquing the flaws of their opponents than the
flaws in their own side. Although we human beings possess a lot of
fairly sophisticated mechanisms for "deliberative reasoning", those
mechanisms are often in conflict with older, more reflexive systems,
which work hard to protect our egos but rarely lend themselves to
objectivity.  

In my view, the weakest link in democracy is the human mind. I
absolutely believe in democracy in principle, and see no better
alternative. But leaving things to the people almost guarantees that
many decision will be made on grounds that are far from rational. Only
with improved education -- and a serious recognition of human
limitations -- are voters likely be able to learn to vote with their
heads as well their hearts.

I wish I had better news!

Gary

p.s I think that Sharon is very much right whether the country's
general frightened state will lead people to hang on to the status quo;
this is very much part of what happened in 2004, when fear of
terrorism led many people to want to stick with Bush over Kerry. 
  

    
I'm a cultural anthropologist with a background in the thought of
Gregory Bateson and cybernetics.  I have never believed that humans or
organisms in general are optimally designed or behave to maximize
things.  Our "kluginess" is a major argument for evolution and fits
with the general outlook of cybernetics and systems theory (as opposed
to academic economics -- which is not exactly at one of its triumph
moments -- and most varieties of sociobiology/ evolutionary
psychology).

I do have one politically correct comment to get out of the way.  I
have lived for long stretches of my life among isolated hunter-
horticulturalists in the lowlands of Papua New Guinea.  They impress me
greatly.  Luria's experiments where he found traditional people not to
use syllogistic reasoning don't indicate anything to me except lack of
formal education, i.e. a particular kind of context of knowing that
the other person is testing your individual capacities.  Testing and
grading as we know it in school makes us aware of the kind of thing
Luria was doing, but without that experience we would interpret his
questions differently.

Traditional knowledge of horticulture, hunting, gathering, and complex
social organization is very rich and intellectually challenging, and I
am sure I could find analogues to syllogism in the ways that people
think about kinship, for example.  Living among them is not like living
among people who don't reason.  

OK, now I've said that.  I'll tell a story from Bateson in my next
post that bears on this.  But in general I am excited by this book and
its ideas.  
  

    <hidden>
  

    
Nice, thanks!
  

    
Phillip,

Thanks for your post. In the passage that you're referring to, I
certainly didn't mean to imply that non-literate people can't reason,
only that they don't necessarily reason according to the laws of formal
logic. Most of the time Westerners don't, either. 

That said, I'd be quite interested in hearing of examples among
syllogistic-like reasoning in cultures where there is no formal
education, especially since understanding how intuitive reasoning works
can give a great deal of insight into questions about the origins of
human thought, and what a relatively untutored frontal cortex can
manage to accomplish. So bring on the examples!

Cheers,
Gary
  

    
I appreciate the answer.  I came down a little hard, I know.  

In hopes of rekindling the discussion I'd like to get back to the big
themes of the book.  It seems to me that there are a lot of different
kinds of not-optimum our brain processes can be.

We can have failures of memory or reasoning that anyone can see as
disadvantageous to any purpose we may happen to have.  We can also have
"failures" to live up to cultural standards of rationality that are
not necessarily good ways for people to live their lives.  Some of the
ideas of how to live a rational life that come out of corporate culture
or academic economics might not in fact be models of the "good life"
and some people have argued that some of our "imperfections" are
actually aspects of our humanness that need not be seen as defects.

I suppose "efficiency" is one concept that I tend to suspect when it
is used as a high-level goal rather than an instrumental goal related
to a specific purpose.  Also, there are a lot of goods or desires out
there, which may be incompatible with each other to some extent, and
yet which we may wish to obtain or satisfy to some degree.

Our brains may in many ways be "kludgy" just in general, but sometimes
doesn't the "kludginess" come from evaluating them with respect to
ideals that are inhumane to begin with?  Should we not be trying to
bring our ideals down to human scale and adapt them to human beings,
rather than the opposite?  Very hippie of me I know, but where better
than the WELL to indulge such questions?
  

    
On the subject of bankers and kluges, some of you might enjoy a piece
I posted yesterday at the Huffington Post, a discussion of how the
phenomenon known as anchoring is perhaps distorting the bailout debate.

http://www.huffingtonpost.com/gary-marcus/could-a-quirk-in-human-br_b_128776.h
tml

Onwards to Phillip's message in a bit.
  

    
That's very illuminating.  The false choices and the numeric anchoring,
both.
  

    
Nice work.
  

    
> When Paulson tells us (as he said in yesterday's testimony) that his
plan "is much better than the alternative," the only real alternative
he is considering is having no bailout plan at all.

But Anatole Kaletsky of The Times 
<http://www.timesonline.co.uk/tol/comment/columnists/anatole_kaletsky/article48
20549.ece>
thinks the current Paulson "plan" is still no plan.

> The staggering incompetence of the US Treasury Secretary is now
acknowledged - and is a disaster for George Bush
....
> But as the cross-examination rolled on, and Mr Paulson just waffled
- “we will ask experts to advise us”, “we will get the best and
brightest financiers to suggest ideas” - the terrible truth dawned.
There was no such thing as a Paulson plan. Not only did Mr Paulson not
know what he was doing. He did not know what he was talking about.