Decrying the growing paucity of cars with standard transmissions, Car and Driver magazine recently launched a campaign called Save the Manuals. While applauding the concern, I have the nagging suspicion that all the well-intentioned proactivity in the world can't change the fact that the good ship Stick Shift has already sailed. Indeed, with all the talk lately about autonomous automobiles -- cars that override "faulty" driver inputs and cars that drive themselves -- I fear there's a lot more that's going to need saving in coming years than the dying art of double declutching.
What I mean to say is -- if you think automatics are dull, wait until you don't get to choose which lane you're in, what route you're taking, what line you're carving through the corners, and how fast you're running, to name just a few decisions that your car -- in league with a central computer bank -- will be making for you in the not-so-distant future. I hate to sound like a paranoid crazy and a mind-control Chicken Little, but the self-driving car is more than dystopian prophesy. It's a certitude.
Like so many mixed blessings, this one had its start with the Department of Defense, specifically the Defense Advanced Research Projects Agency (DARPA), founded as a technology incubator in 1958 in response to the Soviet Sputnik program. Since the late 1960s, DARPA has encouraged construction of unmanned vehicles. By the 1980s, Daimler-Benz was happily downloading a billion dollars (in today's money) from the European Commission on the Eureka Prometheus project, which netted a pair of robot-guided civilian vehicles that briefly hit the streets of Paris in 1994. In 2001, Congress mandated that by 2015 a third of armed forces ground vehicles be robotic. DARPA then sponsored a series of vehicle events with large cash prizes, the first in 2004, in the Mojave Desert. One hundred teams entered but none finished. By 2007, DARPA had shifted the competition's focus to urban driving, and despite a worrisomely high DNF rate -- only three out of eighty-nine teams that entered successfully completed the challenge -- the die was cast. Recently, Volvo announced that a "road train" with cars following a lead vehicle filled with sensing equipment would be tested on Swedish roads this year, predicting that such convoys will be routine in Europe by 2020. A Volkswagen/Stanford University team, one of the few to complete the 2007 DARPA contest, built their third-generation "Junior" autonomous car not long ago, and General Motors plans to start field testing its EN-V concept, a two-seater that looks like the scuttlebug robot pod of your worst nightmare, in the next few years.
Bringing it all home to fill this enthusiast with dread was the story a couple months back of a fender bender involving a self-driving Toyota Prius. The property of Google -- the powerful search-engine company with an annual budget about the size of the old Soviet Union's and with a decidedly higher market valuation -- this project is further along than many of us realized, with 140,000 miles of testing on California roads so far. And it reminded us that the benefits of driverless cars -- to Google, to investors in driverless technology, and to other corporate interests -- will undoubtedly be significant enough to overlook the hazards, enormous cost, and inevitable loss of one of life's simple pleasures: piloting your own car.
Dr. Sebastian Thrun, a German-born professor of robotics and artificial intelligence and the co-inventor of Google's paranoia-inducing Street View, splits his time between Stanford and the tech company's luxe campus, where he fronts Google's autonomous-car effort. Thrun has explained that when a friend died in a car crash, his teenage self committed to saving a million lives annually by eliminating car accidents. Pretty ambitious territory, messianic lunacy even, you might say, if he wasn't the brilliant employee of a company that is not only richer than God but also prides itself on doing well by doing good, whether it's actually doing good or not.
To ensure that it gets to live out its autonomous-car dreams, Google recently laid down some heavy baksheesh to persuade Nevada legislators to legalize driverless cars. The state did so in June, barely six weeks before one of Google's self-driving Toyotas crashed in California, where no explicit prohibition is on the books and where the company's fleet of six autonomous Priuses and one Audi TT are working it out, as I write. Who says you need a law permitting driverless cars?
True enough, as Thrun posits, most fatal accidents occur due to driver error rather than mechanical failure. So he reasons that if only we could take the human being out of the equation, there would be fewer fatalities. He's probably right. As countries like China, India, and Russia mint drivers by the hundred-million load, roads will become more congested and dangerous, and each gallon of fuel wasted as the world's population sits idling in traffic will seem that much more foolish. Autonomous cars, he promises, could double or triple highway utilization while lowering consumption and emissions.
But that's not the end of the story. As roads are more densely packed, miles traveled will presumably skyrocket, offsetting fuel-economy gains. And it's not like road fatalities will become a thing of the past; there'll surely be some doozies when, for instance, mainframe glitches simultaneously send 4500 cars into the same Chuck E. Cheese's parking lot headed for the same space at 100 mph.
There'll be plenty of other issues, too, when operators are riding in the back seats of their own cars, drunk, high, and shagging; when the electric grid fails; when terrorists hijack the systems; and when Grandma heads to her local CVS in Ho-Ho-Kus, New Jersey, for some bunion pads and winds up at the grindcore convention in Worcester, Massachusetts, because her 2020 Ford Taurus thought that was where she belonged.
Of course, if we were really concerned about safety, we might try fixing the roads we already have and teaching people how to drive before committing ourselves to spending trillions to make the world's cars drive themselves. And trains work pretty well, too.
Except all these old-tech solutions miss the essential point of self-driving cars. Aside from any profit to be realized in infrastructure investment, the corporate world has every interest in freeing us of our driving chores so we can spend more time chalking up incremental charges shopping, blabbing, and social networking. As the ill-advised march of in-car telematics reminds us, that's what corporations have decided we're here on earth to do.
By and large, carmakers don't care. Planned obsolescence can take any form, so long as it works. They're selling units, not driving thrills, and they're not emotional about it -- as the death of stick shifts attests.
Never thought that an automatic Nissan Versa would seem like wild good fun? It's time to save the car (and driver).