On Birds and Boeings:
Thoughts on the surprises we can expect from Genetic Engineering

by Roger Bourke White Jr., January 2007


Whenever a powerful new technology is developed, its adaptation to human needs goes through some common phases. This is a discussion of those phases, and how they are likely to apply to an up-and-coming powerful technology: genetic engineering.

The basic phases are:

1) Being inspired by a dream until an application of the technology shows its first commercial feasibility.

2) This first feasibility shows one way that the new technology can replace an existing technology in an existing application. This inspires more people to pay serious attention to the new technology – serious meaning: willing to invest time, money and attention.

3) Discovering what other existing applications the new technology can do well.

4) Discovering the surprise benefits that the new technology is capable of supporting.

5) Discovering which parts of the original dream are not likely to be realized.

The first phase of implementing a new technology happens when it is seen as a solution to a dream. The dream may be to replace an existing, but tedious, task, such as getting to work on a cold, winter day. Or, it may be the dream to gain a wonderful, but currently unattainable, ability, such as flying like a bird. Either way, the first implementation of a new technology is almost always big, clunky and dangerous, and it rarely comes out in a way that will allow the inspiring dream to come true.

The first successful implementation of a new technology -- successful in this case meaning use that is widespread enough that other people besides the first adapter consider using the technology -- is almost always to replace an existing, but hard-pressed, technology. The first use of digital computers was to replace thousands of human beings doing the tedious arithmetic it takes to calculate artillery trajectory tables. The first use of cars was to replace horses who didn't like to make house calls in the middle of the night (the doctors riding them didn't mind as much).

The third phase of implementing a new technology is the phase where it gets smaller, more elegant and cheaper to produce. This trend opens the way for the fourth phase, when "surprises" appear -- ways to use the technology that don't relate much at all to the original dream, or the clunky first implementation. The surprise uses are almost always much more interesting than the original use, because they are much better fitted to what the technology can do well. The surprise uses are discovered as mankind comes to understand better what a technology can and can't do well. They are discovered by yet more tinkering done by people who no longer think of this technology as wondrous or mysterious – it’s now just another part of their technology universe, and their thinking about this technology is, “So… why not use it this new way? It’s a neat use!”

Finally, I will talk about those times and places where a technology fails to live up to its dream. These are important, because even though these failed implementations tend to be looked upon as curious and ridiculous thinking by following generations, they were an important part of the drive to get the technology established in the first place, and without these dreams that became failures, the technology would not have been implemented. These dreams opened the wallets of investors and philanthropists, so that inventors could experiment with the new technology.

I will center my discussion on four technologies: the automobile, the airplane, nuclear power, and genetic engineering.

Phase One: Emerging technology plus a dream

In phase one, an emerging technology is mixed with a dream, and something useful emerges.

The Automobile

In the late 1800's, industrialization was producing a lot of newly wealthy people. These newly wealthy dreamed of moving around like old rich people did -- with a horse pulling a carriage. The problem was horses are expensive. They are expensive to buy, very expensive to maintain, and they are unreliable. But several technologies emerging from industrialization had potential to help make this dream of more and cheaper carriages become a reality. The most noticeable of those technologies were: internal combustion engines, better road pavement, and rubber tires. The key feature for cars -- the one that distinguished it from bicycles -- was the engine. In fact, bicycles boomed just ahead of cars, and the demand for better bicycle conditions laid the groundwork for many of the infrastructure improvements we now attribute to being done for cars. Think of China in the 1980's as a near-modern example of a place where a road network was bicycle-dominated rather than car-dominated.

Since they were first developed in the late 1700's, engines have been getting steadily smaller and cheaper and their use more widespread. By the late 1800's, some were small enough to mount on carriages, and the first "horseless carriages" were developed. By modern standards, these first implementations were clunky and expensive, and a lot of key features, such as gear shifts and electric starters, were not available. As a result, the cars of the late 1800's were strange and not very useful, and not many were sold. But these early horseless carriages powered the dream, and they showed inventors of the 1900's what was easy, what was hard, and what was necessary to make cars that were popular for car buyers.

Cars finally became cheap and widespread in America in the 1930's and many surprises came with successful, cheap cars. Many of the differences between American culture of the 1900's and American culture of the 1950's can be attributed to the spread of car ownership through all strata of American society.

Phase Two: The Surprises

Once a technology is modestly established, and a few people are using it, many other people look at it and say, “Hey, I can think of something to do with that!" The more those other people are right, the more the technology spreads widely through society and gets used for many things beyond what the original dream called for. A classic example of this is the audio recording device. Thomas Edison, its inventor, thought it would be useful for business, and having people say their wills, rather than writing them. Using recordings for music playing was way, way, down on his list of possible commercial uses. It was other people who looked at Edison's voice recording technology and said, “Hey, I think that would be great for listening to music!" that started the music industry as we know it today.

This is a common evolution in every powerful technology: It is originally implemented for one purpose, but it becomes valuable to society for totally different purposes. This is the most fun, and the most productive, part of any new technology. This is where people find out what a new technology can do well and easily.

Another good example of this is the airplane.

The Airplane

The airplane was inspired by the bird. Human inventors watched birds flying and thought, “I wish I could do that!"

The realities of bird flying and man flying turned out very differently.

Once again, man flying (of which there are many kinds ranging from sport paragliding through combat fighter flying, I will use commercial jet flying as my example) turned out to be much faster than bird flying. Man flying uses a different propulsion system, different materials, and different physical dynamics. It takes place in a different size range, a different speed range, and a different performance range. Birds can land in trees while airplanes must land at airports. In fact, about the only thing man flying and bird flying have in common is the flying part.

The Incredible Shrinking Technology

If a technology becomes successfully established, it will be tinkered with to produce smaller, cheaper, and more efficient versions. Engines have gone from hulking affairs the size of big rooms (the steam engines that powered water pumps in the mines of England) to so small they can power model airplanes. Computing power is the most spectacular example of this shrinking trend -- the artillery calculators mentioned earlier filled big rooms and had a billionth of the computing power of a contemporary tablet.

This tinkering goes on and on, and it takes a whole lot of effort on the part of engineers and inventors. But each advance in the smaller, cheaper, more efficient trend opens new possibilities in how the technology can be exploited. Electric motors went from suitable for moving trolley cars in the 1880’s to suitable for moving hands in electronic wristwatches in the 1970’s.

One sign that a technology has failed is that it stops shrinking. Nuclear energy is an example of a technology that has failed in this way. When the public thinks of nuclear energy, it thinks of atomic bombs and nuclear power plants -- big devices. The public doesn't think of AA battery replacements. But small nuclear engines are just as possible to design as small gas or small electric engines. More about this in the “Technology Failures” section.

Technology Failures

Who of you reading this article is flying your own personal airplane? The one that you can take off from your front lawn and land on the parking lot at work? Who can take their personal airplane and land it in a tree?

Mankind's flying skill has produced incredible flying machines, and those machines have transformed human society, but it still hasn't produced a flying machine that lets a person fly like a bird, in the literal sense. This was the original dream, but it remains unfulfilled.

Likewise, the original dream that powered work on robots -- the dream of each person having a "robot butler" -- remains unfulfilled.

These are examples of technology failures, places where trying to implement the dream has produced something expensive and clunky, and, as a result, the technology hasn't grown or spread any further in that direction. (Every so often, you will see an old news clip of someone demonstrating a personal rocket pack or personal airplane. Notice how clunky those attempts are.) Cars pretty much fulfilled the dream of making a horseless carriage, and then went on to make other surprises for our society, but cars are an exception. Most of the time, a new technology produces a lot of surprises before it can fulfill its original dream, if it ever can.

And cars have their problems. They have done a great job of replacing a horse and carriage, but problems such as deadly accidents, pollution, traffic jams and hard-to-find parking show places where they have failed as providers of perfect personal transport. Anyone who's been in an airport or a traffic jam, or spent too much time looking for parking, agonized over insurance and car payments, held their breath as they watched a teenager take the wheel ... still dreams of having better transportation.

The Curse of Being Important

Nuclear energy has suffered from a different problem than those experienced by the dreamers of personal airplanes and robot butlers. It suffers from The Curse of Being Important. Some technologies look important from the day they are first dreamed about, and nuclear energy certainly fits into that category. These same technologies often look dangerous, too. If an aspiring technology looks important, the early promoters will ask for government help in developing it. If it looks important and dangerous, the community will wholeheartedly agree and demand that the government get involved in developing the technology in a safe way.

The problem with early government involvement in an aspiring technology is that it brings with it the demand for huge amounts of time and paperwork to get the funds to develop it. This bureaucratic overhead strongly favors supporting big projects and existing implementations of the technology. It is death to small-scale projects, and "backyard experimentation", which are how the most innovative ways to use and produce the technology are discovered.

Compare the early history of personal computers with the early history of nuclear power to see the effect of The Curse of Being Important. In the 1980's, personal computers were considered toys by mainstream computer users and designers and the government had no interest in them. In the 1940’s, developing nuclear power was a top-secret, wartime project that produced the world's biggest bombs. The government was into nuclear power with both feet as quickly as science fiction writers were.

Nuclear power is an example that shows that an aspiring technology can fail. It can fail to become as cheap, small, and efficient as was originally dreamed, and, if this happens, it will not spread as widely as was originally dreamed. If it doesn't spread, it's not going to produce many surprise uses, either. Today, we don’t seriously consider the possibility of nuclear energy powering cars or artificial hearts, but there is no technological reason why this cannot be.

Genetic Engineering

Now let’s look at the future: genetic engineering.

Genetic engineering is inspired by the dream of building better people than natural and artificial selection have. Mankind has watched plants and animals change with time, and has learned how to have some control over those changes. Controlling how plants and animals change with time is called breeding. Strawberries are one example of what can be accomplished with breeding. Wild strawberries still exist today. A wild strawberry is sweet and tasty, but about the size of a peanut. If you want to eat wild strawberries, you spend a whole lot of time and energy picking them, so, while they are good, it sure would be nice if they were bigger! Strawberry breeders have been successful at increasing strawberry size -- the supermarket strawberry of the 2010's has about twenty times the volume of a wild strawberry, and it's just as sweet and tasty. This is a good example of what breeding can accomplish.

Breeding is "bird" technology in the field of genetic manipulation. Genetic Engineering is "airplane" technology. What can we expect to see come from Genetic Engineering?

What this means is that making an existing plant or animal larger or smaller is likely to be difficult to do with genetic engineering. Changing size is something that breeding will do easily, not genetic engineering.

From what we have seen of genetic engineering to date, what will be easy to do with genetic engineering is add or subtract proteins from what grows inside a cell. This will be the "fast" part of genetic engineering. Genetic engineering is more likely to resemble "organic nanotechnology" than it does breeding -- it will make molecules very quickly, not whole organisms. (Nanotechnology is the up-and-coming way of making things using molecular-sized tools. What goes on inside a cell is Mother Nature’s version of nanotechnology.) 

The powerful dream that will not be realized through genetic engineering is building better babies. Building good living organisms involves so many tradeoffs and complexities that it will remain in the province of breeding -- but in the future, the breeding will be assisted by genetic engineering tools.

Likewise, the science fiction fear-theme of thousands of genetically identical human clones marching around in threatening ways is one that will look curious in less than twenty years (because it's so unbelievable).

Genetic engineering suffers strongly from The Curse of Being Important; witness the continuing uproar over genetically modified (GM) foods. This means that genetic engineering will be biased in favor of big, expensive projects and that "backyard genetic engineering" experimentation will be stunted.

The best thing that could happen to genetic engineering is for a "toy" version of it to somehow appear and become popular -- a version that doesn't look important or threatening. If that appears, that will be the part that leads the field in innovation and world-changing surprises.

Update: This 19 Aug 12 Huffington Post article, DNA Storage Advance: Entire Genetics Textbook Encoded In Less Than One Trillionth Of A Gram by John Bohannon, is an example of a surprise use of genetic engineering: As a memory storage device.


Whenever a powerful technology is implemented, we can expect surprises. We can expect that a technology will be driven to usefulness by a dream, but that fulfilling that dream will be only a modestly good fit with the technology's capabilities. Once the technology has proven itself, it will become popular for some surprise applications, because it can do those surprise applications much cheaper and better than it can do the dream that originally drove its development.

In the case of genetic engineering, the driving dream is making our babies better. We will find that genetic engineering is only modestly good at that. Like an airplane versus a bird, it will be much, much better at doing other kinds of genetic manipulation, and it will have a much bigger impact elsewhere in our society.

Genetic engineering can fail -- fail in the same sense that nuclear power has failed. It can fail to have many world-changing impacts on our society. If it fails, we can lay blame on The Curse of Being Important.


-- The End --