but I've really learned to appreciate them.

I've learned that they're really the essence

of what makes for progress,

even when they seem to be terrible.

And I'd like to review

just how unintended consequences

play the part that they do.

Let's go to 40,000 years before the present,

to the time of the cultural explosion,

when music, art, technology,

so many of the things that we're enjoying today,

so many of the things that are being demonstrated at TED

were born.

And the anthropologist Randall White

has made a very interesting observation:

that if our ancestors

40,000 years ago

had been able to see

what they had done,

they wouldn't have really understood it.

They were responding

to immediate concerns.

They were making it possible for us

to do what they do,

and yet, they didn't really understand

how they did it.

Now let's advance to 10,000 years before the present.

And this is when it really gets interesting.

What about the domestication of grains?

What about the origins of agriculture?

What would our ancestors 10,000 years ago

have said

if they really had technology assessment?

And I could just imagine the committees

reporting back to them

on where agriculture was going to take humanity,

at least in the next few hundred years.

It was really bad news.

First of all, worse nutrition,

maybe shorter life spans.

It was simply awful for women.

The skeletal remains from that period

have shown that they were grinding grain morning, noon and night.

And politically, it was awful.

It was the beginning of a much higher degree

of inequality among people.

If there had been rational technology assessment then,

I think they very well might have said,

"Let's call the whole thing off."

Even now, our choices are having unintended effects.

Historically, for example,

chopsticks -- according to one Japanese anthropologist

who wrote a dissertation about it

at the University of Michigan --

resulted in long-term changes

in the dentition, in the teeth,

of the Japanese public.

And we are also changing our teeth right now.

There is evidence

that the human mouth and teeth

are growing smaller all the time.

That's not necessarily a bad unintended consequence.

But I think from the point of view of a Neanderthal,

there would have been a lot of disapproval

of the wimpish choppers that we now have.

So these things are kind of relative

to where you or your ancestors happen to stand.

In the ancient world

there was a lot of respect for unintended consequences,

and there was a very healthy sense of caution,

reflected in the Tree of Knowledge,

in Pandora's Box,

and especially in the myth of Prometheus

that's been so important

in recent metaphors about technology.

And that's all very true.

The physicians of the ancient world --

especially the Egyptians,

who started medicine as we know it --

were very conscious

of what they could and couldn't treat.

And the translations of the surviving texts say,

"This I will not treat. This I cannot treat."

They were very conscious.

So were the followers of Hippocrates.

The Hippocratic manuscripts also --

repeatedly, according to recent studies --

show how important it is not to do harm.

More recently,

Harvey Cushing,

who really developed neurosurgery as we know it,

who changed it from a field of medicine

that had a majority of deaths resulting from surgery

to one in which there was a hopeful outlook,

he was very conscious

that he was not always going to do the right thing.

But he did his best,

and he kept meticulous records

that let him transform that branch of medicine.

Now if we look forward a bit

to the 19th century,

we find a new style of technology.

What we find is,

no longer simple tools,

but systems.

We find more and more

complex arrangements of machines

that make it harder and harder

to diagnose what's going on.

And the first people who saw that

were the telegraphers of the mid-19th century,

who were the original hackers.

Thomas Edison would have been very, very comfortable

in the atmosphere of a software firm today.

And these hackers had a word

for those mysterious bugs in telegraph systems

that they called bugs.

That was the origin of the word "bug."

This consciousness, though,

was a little slow to seep through the general population,

even people who were very, very well informed.

Samuel Clemens, Mark Twain,

was a big investor

in the most complex machine of all times --

at least until 1918 --

registered with the U.S. Patent Office.

That was the Paige typesetter.

The Paige typesetter

had 18,000 parts.

The patent had 64 pages of text

and 271 figures.

It was such a beautiful machine

because it did everything that a human being did

in setting type --

including returning the type to its place,

which was a very difficult thing.

And Mark Twain, who knew all about typesetting,

really was smitten by this machine.

Unfortunately, he was smitten in more ways than one,

because it made him bankrupt,

and he had to tour the world speaking

to recoup his money.

And this was an important thing

about 19th century technology,

that all these relationships among parts

could make the most brilliant idea fall apart,

even when judged by the most expert people.

Now there is something else, though, in the early 20th century

that made things even more complicated.

And that was that safety technology itself

could be a source of danger.

The lesson of the Titanic, for a lot of the contemporaries,

was that you must have enough lifeboats

for everyone on the ship.

And this was the result

of the tragic loss of lives

of people who could not get into them.

However, there was another case, the Eastland,

a ship that capsized in Chicago Harbor in 1915,

and it killed 841 people --

that was 14 more

than the passenger toll of the Titanic.

The reason for it, in part, was

the extra life boats that were added

that made this already unstable ship

even more unstable.

And that again proves

that when you're talking about unintended consequences,

it's not that easy to know

the right lessons to draw.

It's really a question of the system, how the ship was loaded,

the ballast and many other things.

So the 20th century, then,

saw how much more complex reality was,

but it also saw a positive side.

It saw that invention

could actually benefit from emergencies.

It could benefit

from tragedies.

And my favorite example of that --

which is not really widely known

as a technological miracle,

but it may be one of the greatest of all times,

was the scaling up of penicillin in the Second World War.

Penicillin was discovered in 1928,

but even by 1940,

no commercially and medically useful quantities of it

were being produced.

A number of pharmaceutical companies were working on it.

They were working on it independently,

and they weren't getting anywhere.

And the Government Research Bureau

brought representatives together

and told them that this is something

that has to be done.

And not only did they do it,

but within two years,

they scaled up penicillin

from preparation in one-liter flasks

to 10,000-gallon vats.

That was how quickly penicillin was produced

and became one of the greatest medical advances of all time.

In the Second World War, too,

the existence

of solar radiation

was demonstrated by studies of interference

that was detected by the radar stations of Great Britain.

So there were benefits in calamities --

benefits to pure science,

as well as to applied science

and medicine.

Now when we come to the period after the Second World War,

unintended consequences get even more interesting.

And my favorite example of that

occurred beginning in 1976,

when it was discovered

that the bacteria causing Legionnaires disease

had always been present in natural waters,

but it was the precise temperature of the water

in heating, ventilating and air conditioning systems

that raised the right temperature

for the maximum reproduction

of Legionella bacillus.

Well, technology to the rescue.

So chemists got to work,

and they developed a bactericide

that became widely used in those systems.

But something else happened in the early 1980s,

and that was that there was a mysterious epidemic

of failures of tape drives

all over the United States.

And IBM, which made them,

just didn't know what to do.

They commissioned a group of their best scientists

to investigate,

and what they found was

that all these tape drives

were located near ventilation ducts.

What happened was the bactericide was formulated

with minute traces of tin.

And these tin particles were deposited on the tape heads

and were crashing the tape heads.

So they reformulated the bactericide.

But what's interesting to me

is that this was the first case

of a mechanical device

suffering, at least indirectly, from a human disease.

So it shows that we're really all in this together.

(Laughter)

In fact, it also shows something interesting,

that although our capabilities and technology

have been expanding geometrically,

unfortunately, our ability to model their long-term behavior,

which has also been increasing,

has been increasing only arithmetically.

So one of the characteristic problems of our time

is how to close this gap

between capabilities and foresight.

One other very positive consequence

of 20th century technology, though,

was the way in which other kinds of calamities

could lead to positive advances.

There are two historians of business

at the University of Maryland,