Home

Perpetual Motion Machines

Driverless cars won’t be a new form of transportation but the end of it

The video that introduces Nissan’s IDS automated concept car resembles any other car commercial: vaguely propulsive background music, tracking footage shot from a helicopter sweeping over a city, a handsome man behind the wheel. Then, as the narrator promises that Nissan’s technology will make driving more “enjoyable” by allowing computers to take over during moments of heavy traffic, the car’s manual controls vanish beneath an elaborate folding-panel system. The driver role is replaced with the equally familiar role of passenger, gazing contemplatively at the passing scenery of the same conventional streets and bridges and office buildings that would be visible today.

But new technologies may ultimately evolve far beyond machines “automating” the recognizably human task of driving. Hypotheses about “driverless” cars still presume there will be such a thing as drivers and passengers, trapping us within the current incarnation of our transportation system. Frequently applied terms like “automated” and “driverless” are inadequate in that they continue to posit manually piloted vehicles as the norm from which the new technologies deviate. Rather than robot drivers piloting cars that humans might otherwise be driving, these new technologies may transport us in an entirely different way that dispenses with accommodating human capabilities.

While attempting to describe an upcoming future that we do not yet understand, predictions like those in Nissan’s IDS video remain burdened with obsolete concepts. It is telling that Nissan’s concept car and the vehicles imagined by Volvo and IDEO retain familiar characteristics of gasoline-powered cars. They have a hood and front grille as ornamentation, for instance, even after their electrical propulsion mechanisms have rendered them nonfunctional. The electric drivetrain of Tesla’s Model S makes the front hood vestigial; the company’s nickname for that anachronistic space where the engine once was — a “frunk,” or front trunk — embodies the awkwardness of adapting new designs to our current expectations.

Our future passenger experience might bear little resemblance to either driving or riding; we’ll inhabit a space that only coincidentally happens to be in motion

Once designers of automated vehicles are no longer bound by the outdated limitations of accommodating either internal combustion technology or human operators, they could move far beyond our present-day intuitions of what a car should look like. Replacing bulky gasoline engines and transmissions with multiple smaller electric motors and slim under-floor battery packs would enable radical new possibilities for the configuration of interior space. As early as 2002, GM’s Hy-Wire concept car separated an interchangeable passenger compartment from its fuel cell and electric motor powertrains, opening up space for an interior that more closely resembled a living room than conventional expectations of passenger-car seating. Where one would expect to see a hood and dashboard, the windshield extended to become a panoramic window framing the road ahead as a scenic view.

The Hy-Wire’s technology suggests that the focus of car design could turn inward, yielding a range of new possibilities for vehicle interiors. Our future passenger experience might bear little resemblance to either driving or riding within a vehicle; we’ll inhabit a space that only coincidentally happens to be in motion.

With a system of automated vehicles, transit passengers will no longer need to pay any attention while distances are being traversed. With the possibility of traffic collisions theoretically eliminated, safety requirements mandating fixed seats, air bags, and seat belts would become obsolete. Passengers who no longer needed to be restrained would be able to move around freely. After ease of handling becomes an irrelevant design consideration for new vehicles steered by computers, designers will be free to stretch wheelbases, raise ceiling heights, and specify softer suspensions to make that movement more natural and comfortable. And since the people inside wouldn’t necessarily need to see where they were going, a growing range of possible wall fixtures — storage cabinets, LCD screens, perhaps a kitchen sink — could substitute passenger convenience over views of the world outside. The elimination of the driver will mean the end of the car as a car.

The social impact could be broader than we expect. When we don’t have to look where we are going, we have to deliberately choose what we want to see. One of IDEO’s more radical visions of how automated vehicles could be used, the WorkOnWheels mobile office, is designed to allow employees to travel to new locations as they work. The pod contains office furniture and pull-down shades over the windows, letting workers choose which aspects of their surrounding environment they want to see, without having to visually process the travel in-between. Cityscapes become optional, consumable on demand rather than by necessity. Meanwhile, the mobile workplace’s controlled internal habitat would remain constant no matter where it was.

Such a vehicle would not have to travel any faster for us to perceive a dramatic reduction in travel time. The time once spent in vehicles inertly waiting to arrive could now be filled with the same sort of activities we’d be doing if we were already there — or had never left.

The time once spent in vehicles inertly waiting to arrive could now be filled with the same activities we’d be doing if we were already there — or had never left

The opportunity to multitask while traveling could make the journey into the destination. Given the expanded possibilities of what one could do inside a vehicle, our existing distinctions between vehicles and buildings, between transit and destination, between static and mobile spaces, may begin to blur. Imagine commuting while sleeping, or socializing at happy hour while the bar transports you home. Imagine if a garage was also the car. If commuting entails being in a space that is functionally equivalent to being at home, one might eventually skip returning home, and commute perpetually. The journey to work could commence as soon we fall asleep. The idea of having a destination becomes as obsolete as drivers and cars. Highways would host listless roaming bedrooms, meandering through the night.

Our understanding of a house as a stable locus of physical and emotional shelter could become diluted. There would be no reason for homes to not also be vehicles. A range of new options for customizing these vehicle-home hybrids would emerge: Homes could be made up of modular docking pods, and specific rooms could be shared, swapped, rented out, or sent away for cleaning or restocking. Modern conveniences that we currently take for granted — such as being able to use a bathroom without needing to arrange for its presence in advance — could become tomorrow’s luxuries. The homeless would be the only people not constantly in motion, the people closest to retaining a fixed physical location called home. Stasis would become homelessness.

If vehicular interiors can accommodate the activities possible at most destinations — if the vehicle becomes a destination in and of itself, and destinations become other vehicles — the mediating experience of a journey between places would be eliminated. There will be no signs to point us anywhere. There would be no need to know directions, and no sense of what being “on the way” to somewhere looks or feels like. There will be no need to know how to get anywhere once we forget the concept of having anywhere to go.


Driverless cars will not be the first transit technology to challenge our conceptions of time and space. The travel speeds of the first railroads were unprecedented, surpassing the contemporary ability to perceive the distance between destinations. Train routes became abstractions, navigated by means of timetables rather than maps. Eventually, transit system diagrams, like the iconic Vignelli New York City subway map, eliminated realistic representations of geography. Mass-market novels grew in popularity as a way for riders to pass the time while their capacity to comprehend or influence the direction of their journey was suspended.

Geographic proximity became less relevant than whether or not the destination was connected to the transportation network. Early transit-oriented developments, such as theme parks and department stores, were built by railroad interests to take advantage of the audiences captive within their systems. Growing suburban commuter towns expanded to the limit of convenient walking distance from a train station; areas beyond that boundary remained rural.

At the same time railroads were offering passengers prescribed choices between linear routes, other technologies were bringing a wider scope of self-directed travel to many consumers. The growing popularity of early bicycles was met with a moral panic over whether they would allow female riders the freedom to travel unsupervised and mingle with members of the opposite sex. While exploratory automotive road trips are now romanticized as integral to American culture, a continuing reminder of the bicycle’s early reception can be seen in Saudi Arabia’s laws prohibiting women from either driving cars or riding bikes.

The user interface for navigation would no longer be a map, but a clock or calendar. Place would be synonymous with occasion, and more closely resemble verbs than nouns

External rules can always be imposed to limit the freedoms that might seem innately afforded by transportation technologies. Driverless cars would seem to retain the automobile’s capability to allow passengers free individualized movement, but their software may introduce new avenues for regulatory control over those movements. Physical impediments like gates and cul-de-sacs would become less relevant compared with restrictions or service fees implemented at the level of code. People and buildings in different service networks might pass each other by without experiencing the slightest hint of one another. And a software error could make certain places impossible to access even as you go right through them. It may require special attention for passengers to know what choices they actually have over their journeys, what potential detours they might be missing. Passengers content to surrender responsibility over their journeys could find themselves back on de facto railroad tracks.

A “driverless car” could become conceptualized as a horizontal elevator. After an elevator’s initial acceleration, the difference in time between reaching higher and lower floors is minimal. Traveling between buildings could become closer to traveling between different floors in the same building, and with no greater awareness of the other numbered floors or buildings blinking past in between. Destinations become equally accessible entries in an arbitrary numeric index, with the differences in access time reminiscent of the slight delays in retrieving digital information from a mechanical hard drive.

It should be no surprise that Google, a technology company focused on information retrieval, has been the first to replace the analog interface of a steering wheel with the binary option of a single push button. Our wider urban environment could become randomly accessible in the same way that Amazon’s “Chaotic Storage” warehouses already organize their contents, independent of any traditional spatial categorization scheme.

Maps would no longer be relevant outside the internal processes of a vehicle’s guidance computer. If one sought, say, the nearest coffee shop, it would not have to be a question of geography. The desire for coffee wouldn’t be a matter of a destination or a journey. Behind the scenes, software would instruct a vehicle to take its passenger to a nearby coffee shop, or it could summon a mobile coffee shop toward the customer. There would be no trip to a fixed location, only trajectories calculated dynamically to unite the various moving parties to facilitate an exchange. The divergent aims and cross-purposes of individual drivers pursuing their goals would be subsumed by a swarm of vehicle-buildings coordinated across a shared network, moving collectively in fluid patterns. Extrapolate this principle, and one can see how dispersed low-rise communities of mobile buildings might replace fixed, vertically oriented cities.

Once physical locations are rendered as abstract coordinates in a user interface, they effectively become arbitrary, as interchangeable as the retail spaces of big-box stores. The experience of inhabiting any particular interior space might become decoupled from its existence within a specific place, free from the baggage of associated historical and geographic context. Real estate would no longer need to be valued according to its location, because proximity would always be subject to change. Travel to visit or inhabit buildings still standing in fixed physical locations might join horses and antique cars as nostalgic hobbies for the wealthy.

Our memories of the spatial processions encountered while traveling through urban architecture — approaching the public facade of a building, the transition between the street and lobby, the awareness of landmark reference points on a skyline, the interstices between buildings — might eventually begin to fade. The experience of passing from one destination to another could become akin to watching the progress bar of a software download. Traveling to a different location, or having that location travel to you, would be more akin to updating an app.

The user interface for navigating space would no longer be a map, but a clock or calendar. Distances once traced on a map would be transmuted into blocks of time plotted on one’s daily schedule. Place would be synonymous with occasion, with movement through time corresponding to automatic movements through space. Frequent destinations such as “home” and “work” might transform into abstract zones differentiated mainly by when rather than where they happen. Our motives and desires would be foregrounded over the experience of traveling, shifting our conception of destinations to more closely resemble verbs rather than nouns. Your workout routine might take place in a different gym than it did the morning before, but you wouldn’t know the difference; they would be identically convenient. As soon as our scheduled time within one destination expired, we would be able to walk through a docking port into the next, like a cinematic cut skipping the passage of mundane events that might otherwise have unfolded between selected scenes.

Driverless passenger cars and delivery vehicles will further accelerate our current move to on-demand services that let us bypass those inconvenient interstitial moments of everyday life — walking to a store, standing in line, cooking a meal, and so on. The logistics of scheduling automated vehicles will ensure that even more of our time becomes consciously programmed and structured, optimized for maximum productivity. With each advance, our surrounding environment will become increasingly hostile to serendipity and chance meetings, known sources of creative breakthroughs.

Contemporary urban-planning guidelines are based on assumptions that the rich pedestrian life of a street or a park emerges from adjacencies with surrounding businesses. Driverless cars posit a possible future without street life and without spaces for spontaneity. As with previous planning mistakes in developing automotive-oriented cities, carmakers and technology companies are moving forward with their ideas without reckoning with the full range of potential social impacts. These futures must be imagined before they can be embraced or resisted. Otherwise driverless cars may steer society into a blind cul-de-sac, and we will discover we have nowhere left to go.

Chenoe Hart is an architectural designer in cyberspace.