Several readers took me to task for defining life as a chemical system capable of undergoing Darwinian evolution. This idea comes from NASA. Since the space agency searches for life in the universe, it needs to agree on a working definition of what it seeks.

The readers questioned whether this is the right definition. Is Darwin's theory really the essence of life?

For this and other such questions, biologists directed me to Jack Szostak, a Harvard professor trying to understand life by re-creating it in a test tube.

That quest also made him ideal to answer a question posed in an earlier column: In the absence of an accepted theory for how life began, are scientists acting on faith to believe in abiogenesis - that life spontaneously emerged from nonlife?

Szostak is in demand, having won the 2009 Nobel Prize in Physiology or Medicine. He shared that prize for work on telomeres - structures in our cells associated with aging.

 Last week I finally intercepted him at Pennsylvania State University, where he was delivering two talks.

In the first, "The Origin of Cellular Life," he noted that astronomers recently estimated our galaxy holds about 500 million other Earth-size planets. How many, if any, harbor life depends on whether forming life is easy or hard.

Szostak hopes to answer that question by starting with some of life's chemical building blocks and coaxing them to form living cells. "Ultimately we want to see the spontaneous emergence of Darwinian evolution," he said.

The first life may not have been as complex as life today. That's important because there's a long-standing misconception that the origin of life was a simultaneous aggregation of cellular machinery, as unlikely as a tornado assembling a 747 from scraps in a junkyard, to use astronomer Fred Hoyle's analogy.

But today scientists don't think it happened all at once. All that's needed are two things, Szostak said. One is a DNA-type molecule that can carry a genetic code and copy itself, but imperfectly. The other is some kind of membrane.

Once you have those elements together, Darwinian evolution can proceed.

In his talk, Szostak showed videos of proto-membranes he made with the help of a graduate student. Though just blobs of fat, they looked alive as they formed spheres, then squashed down into filaments that broke and reassembled themselves as spheres.

All they need to divide this way is a little jostling.

Does evolution define life? Indeed it does, Szostak said, but Darwinian evolution requires a few things beyond just change: a way of making copies with variation from which nature can select; and a way for the variations to be propagated into future generations. Darwinian evolution, he said, is the unifying principle of life.

The consensus is that the first living things did not use DNA for their genetic codes, he said, but they might have used a related code-carrier, RNA, which is made up of a single strand and needs fewer external parts to reproduce itself.

RNA is a long chain of individual links. Each of these links, called ribonucleotides, is formed from several pieces that don't like to stick together, making it hard for scientists to envision how RNA could have formed spontaneously.

Two years ago, biologists in England made some progress by changing the order of assembly. Instead of trying to stick together these two building blocks of RNA, they started with precursors of the building blocks. That worked.

I was also curious about Szostak's response to the abiogenesis question that readers keep posing: Did he have faith that this happened?

He looked puzzled for a moment. "Life wasn't here, and now it is," he said. "It had to have come about by a process."

I said I thought the creationists were accusing the scientists of acting on their own faith to assume it happened without God.

He replied that falling back on a supernatural explanation would be like giving up the inquiry. And there's no reason to give up a problem just because it's hard.

"All we can do is break it down into smaller problems," he said.

Later, Szostak spoke to a biology class taught by Penn State biology professor Andrew Read, after which Read asked a more difficult question:

Even if he does create life in a test tube, how will Szostak know whether his method was the way life originally came about?

The hope, Szostak said, is that some insight will come from answering the more accessible question of how life might spontaneously form.

Maybe the scientists will find there are three or four or 10 ways it can happen, he said, and maybe someday they will have ways to choose the most likely.

And really, even for a guy who already won a Nobel Prize, creating life from chemicals would be a big achievement.

Contact staff writer Faye Flam at 215-854-4977 or fflam@phillynews.com. Read her "Planet of the Apes" blog at www.philly.com/evolution.