In the future we will understand how evolution works and we will see into the structure of life. This will be the basis for a powerful, or at least dangerous, new engineering capabilities.

We are making inevitable progress towards this understanding. However, right now, we can’t see it. When we have this understanding, I think it will come in the form of basic principles that apply to any dynamic system, biological or not. These principles will explain the party tricks listed below, and more.

Here are four things life can do that engineers can’t:

1) Make something big by starting with something small and expanding each part

That is a great way to start a business.

Growth is, I think, the most dramatic difference between living and nonliving engineering. Life always builds this way. At least in the realm of material objects, engineers never build this way. Imagine going into a car dealership, getting a tiny car with tiny little metal parts, and somehow expanding it later. It’s hard to imagine how that would work. On the other hand, I have seen enough science fiction movies to know that if something swells up, extrudes appendages, and starts chasing you, it must be alive.

2) Grow all of the differentiated parts of an organism, starting from a single cell, using only local cell-to-cell signaling.


3) Make 100,000 proteins from 30,000 genes

The human genome project found about 30,000 genes. We suspect (but do not know) that there are more than 100,000 proteins in a human body. Each set of genetic instructions can be used in multiple ways “alternative splicing” to make a protein machine. Our feeble human brains may never understand how this encoding works, but I think we will figure out why so much overlap is used.

Imagine if you made a machine where each little piece knows where it should be and what it should become, and where this knowledge changes synchronously over time. Hmmm…

4) Make a brain with 100 billion neurons using 10,000 or so genes

The brain is very complicated, with specific organs, intricately connected, and very subtle and specific instincts that are hardwired in. It controls thousands of types of motion and detects innumerable patterns. What organ helps us learn to talk? What about Chomsky’s universal grammar? We keep discovering new wiring. Heard about the experiment that showed men will spend more money after they look at pictures of pretty girls? If you look through the psychology literature you find thousands of these built-in behaviors. And they all come from a small number of genes. It seems that the genes start the brain off, and then it picks up structural information from its environment as it goes, in some very deterministic way.

There is hope. If we understood 1 and 3, we could probably figure out 2. The answer to 4 is probably the same as the answer to 2.

I’m going to make some guesses about the principles behind trick number one – making something big by starting with something small and expanding each part. I think that a key principle of a device that grows is that it is cellular. Each part is composed of a number of cells, and it can expand by increasing the number of cells per part. Furthermore, I would guess that a second principle is that the cells are substitutable – they can be replaced by other similar cells and the part will still be useful.

These are certainly characteristics of the plants and animals that we know. I think they are general principles because they also apply to something else that grows by expanding its parts – an organization of people. The organization grows by increasing the number of people (cells) in each function, and it survives turnover.

Principles like these can be applied to build software and systems that have smooth upgrade paths. Maybe someday we will buy baby cars, too. I hope they don’t bite.