These episodes are the currency of our past and the storyboards we arrange to make sense of the things that have happened to us. We line them up like dominoes that lead to where we stand now. That we do this imperfectly has been written about many times. But I am more interested in the invisible threads running from one episode to the next, the forces that hold our stories together. Some have names, like love, or courage, or fear. Others are harder to pin down.
According to psychologist Dan McAdams, the episodes in our memory are not only the material for anecdotes to amuse our friends. They are also the building blocks of our “life story”—our own version of how we came to be the person we are.
Unless we write a memoir, or visit a therapist, we may never even tell anyone this life story, but that doesn’t make it any less important. McAdams and others argue that the ability to see one’s life as a story is at the heart of identity. In fact, our ability to “narrate” our life’s events may even be the defining mark of consciousness.
Building a life story is a process that begins around the time we turn two years old. That’s when we develop what McAdams calls a “primitive autobiographical self.” As we move into adolescence, we start to emphasize different memories we feel were important—events in which we learned something or changed. Then, during our late teens we start to develop a more complicated “personal fable,” in which we dream of the people we could become, like astronauts and presidents. McAdams calls this a “first draft” of our identity. We choose episodes based not only on who we think we are, but also on who we hope we can become.
As we move into young adulthood (between seventeen and twenty-five), things become a little more urgent as we try to compose a “full life story” that explains not only how we got wherever we are, but also what we believe, and who we will in fact be.
But our own past is not the only place from which our life story comes. The memories are our own, but what they mean and how we put them together come from the lives we see around us, from the stories we read and hear, and from whatever possibilities we can imagine.
For most of us, that full life story is never really finished, and is always subject to revision. Even so, it determines much of how our life unfolds. It’s like a road map through the chaos, with arrows pointing one way or another at turning points like failure and success, death and birth, love and loss. That is what our daughter was really asking: How do you live in a world with sadness and fear? And how should I?
After the door closed behind me at Paul Gruchow’s house, I went back to campus. I graduated and my career went slowly on. Yet even as I wrote story after story—hundreds of them—and even as I became a better writer, I still didn’t quite know what a story was, not exactly. Instead I wrote by feel. A story was something I knew if I saw or felt it, but when I tried to put a definition into words, the meaning would slip through my fingers.
You can find this same problem running through much of the discussion about stories, or narrative, these days—and there’s more of it than ever. Narrative neuroscience and narrative psychology are both growing fields. The Defense Advanced Research Projects Agency, a part of the U.S. Department of Defense, is even researching the use of narrative for defense purposes. Evolutionary biologist and author E. O. Wilson has repeatedly called us the storytelling species, and last April Houghton Mifflin Harcourt published a book by Jonathan Gottschall titled The Storytelling Animal: How Stories Make Us Human. But like other authors who have tackled the subject, Gottschall never quite articulates what he means when he talks about a story, and the book remains a disappointing collection of platitudes.
So what makes stories so important? What makes them stories at all? I finally stumbled across a kind of answer in a field about as far from the English department as you can get: artificial intelligence.
It turns out one of the biggest problems with making a computer intelligent is getting it to do something that we do naturally, something called “commonsense causal reasoning,” which means understanding instantly when one thing causes another to happen.
“It’s very simple things,” says Andrew Gordon, a researcher at the University of Southern California’s Institute for Creative Technologies. “Like if you tell the computer you dropped an egg, you want the computer to know that it broke, not bounced.”
Gordon and his fellow researchers have been working on this problem for some time. They tried to instill this ability into a computer program by collecting millions of stories from blogs and using them to teach it how to deduce that A causes B.
After they had collected these stories, they designed a test in which they asked the computer a question, such as: “The man lost his balance on the ladder. What happened as a result? 1: He fell off the ladder. 2: He climbed up the ladder.” Or this one: “The man fell unconscious. What was the cause of this? 1: The assailant struck the man in the head. 2: The assailant took the man’s wallet.”
“Computers are horrible at this test,” says Gordon. Humans get the answer right 99 percent of the time—more or less perfectly. The best result they could get from the computer was 65 percent correct, or just 15 percent better than chance.