An Idealized Design for Carfree Cities and Its Application in the Real World

{jcomments on}

J.H. Crawford


At a conference entitled “Making Cities Livable,” one can reasonably expect general agreement that cars are the most difficult design problem facing in cities today. While most people here probably accept that fewer cars are better and that no cars would be ideal, many may have trouble with the notion that it would actually be possible and practical to remove all cars and trucks from a large urban area while simultaneously improving its economic competitiveness.

I have proposed an idealized case for car- and truck-free cities, what I call a “reference design.” It is based entirely on rail systems for heavy transport, with local transport and light freight delivery provided by walking and cycling. I will describe in some detail the elements of and rationale for this reference design, and then consider its application to both new and existing urban sites. Variations from the reference design will be required in the real world, but the design can still be guided by the reference design. The resulting compact cities would be highly sustainable, offer a quality of life second only to Venice, and can compete with auto-centric cities in the global economy.




I have visited many parts of the world and seen first hand the cities other cultures have built for themselves. During my travels, it became clear that cars were antithetical to good urban environments. About 15 years ago, I worked out the basic form of an idealized carfree city of one million people. Today, I would like first to present this idealized design and then to consider how the principles it establishes can be applied in existing cities.


While cars are useful in rural areas, their presence in urban areas is always disruptive, and I believe that this view is shared by most people here today. It is worth mentioning, however, that I do not foresee the return of draft animals to our cities-the problems they caused were very serious indeed. What is needed instead is a transport system that does not intrude upon the city. I believe that, with minor modifications, rail systems are entirely suited to this task.


Today, the only large carfree cities are Venice, with a population of about 70,000, and Fes-al-Bali in Morocco, with a population of 150,000. Both Fes and Venice were built with very narrow streets that could never have been adapted to automobile use. Fes is probably not a useful model for Western cities, as it depends almost entirely on mules and donkeys for heavy freight transport. Venice offers a better model for modern carfree cities in spite of its unique lagoon setting. Boats provide slow and expensive transport of passengers and freight, but the important point is that the canals are grade separated from the pedestrian environment. Grade separation is the principle that guides the design of the transport network in carfree cities generally. I regard it as economically infeasible to place roadways underground, as they require too much expensive construction, and I do not believe that any urban transport system should ever be elevated above ground, if only for aesthetic reasons.

Venice can guide us in other respects as well. I believe that the medieval city, with its narrow, irregular streets, is the best urban model ever developed. However, the carfree approach can be applied to almost any type of city, with the caveat that density must be high enough to support good public transport.


For this audience, it is perhaps not necessary to say a great deal about the effects of motorized road transport on the quality of urban life. I regard the damage to social systems as the most serious problem with auto-centric cities, and one that cannot be solved by technical adjustments short of burying the entire road network.


As we all know, the arrangement of a city exerts a large influence on its transport requirements and vice-versa. Cars make such bad neighbors that people in auto-centric cities want to live at low densities, which reduces the impact of cars. Unfortunately, this leads to a dispersion of destinations and still more driving. This pattern of habitation is the most energy-intensive in history and almost certainly cannot be sustained. This can be dramatically illustrated by comparing Houston and Hong Kong. Residents of Houston use 86,000 megajoules of energy annually compared to just 6500 for Hong Kong residents. Houston spends 14% of its GDP on transport, Hong Kong just 5%. Public transport service density in Hong Kong is 370 times higher than in Houston, which makes these benefits possible.

We must build sustainable transport systems characterized by:

  • Efficient use of renewable energy sources
  • Efficient use of land
  • Rapid transport of both passengers and freight
  • Economical service

While rail systems are the most energy- and land-efficient of all transport choices, they are of limited usefulness in small cities because the capital costs are high. In larger cities, however, the high capital costs are offset by large operating economies. At the same time, modern passenger rail systems provide a level of comfort that is never reached by buses, no matter how advanced. If street-running trams are adopted in existing cities, which is a reasonable compromise in many circumstances, it is imperative that conflicts with other traffic be eliminated, which can only be achieved by removing cars and trucks from the streets on which they run.


Cars can only be replaced by public transport if the quality of service is high. There is no intrinsic barrier to high-quality public transport in any reasonably dense urban area. The means are well known, if seldom applied. Where high-quality service is provided, people use it willingly. For example, in Zurich, 90% of commuters to downtown use the excellent public transport service.


Let us now consider an idealized design for a carfree city of million people, built without any of the usual constraints of geography or existing development. Such a design could actually be constructed on a number of sites. This “reference design” can serve as a guide when converting larger cities to the carfree model; small cities may require different means.


  • Priority for pedestrians
  • Economical construction
  • High sustainability
  • High quality of life Vital Statistics
  • Population 1,000,000
  • Site 16 kilometers on a side
  • Less than 20% of the site developed
  • 81 inhabited districts
  • 18 “utility areas,” with parking garages, freight infrastructure, and heavy industry
  • Buildings average four stories high
  • Both the transport halt and green space lie within a 5-minute walk of every doorstep
  • Longest door-to-door journey takes 35 minutes


Slide: General plan of the entire city.

The general plan of the city:

  • Provides optimum public transport with no more than one transfer
  • Leaves open space close to all parts of the city Slide: Lobe drawing
  • Portions of 12 districts are shown and 3 utility areas
  • Farms, forests, and parks surround the districts
  • Utility areas are shown at the outer edge of the lobe

Many variations are possible while still following the basic concept, as long as reasonably high density is achieved.


The reference district is largely based on the medieval city form, except that the streets are somewhat wider because I think that very narrow streets are a difficult sell today, even though I prefer them myself. A highly regular street plan is of course possible, but I think that irregular plans are easier to navigate because they provide a ready supply of reference points for the pedestrian, and I believe that they are much more interesting.

Slide: District drawing

District Attributes

  • 12,000 residents and 8,000 workplaces per district
  • F.A.R. of about 1.5
  • Radial design minimizes walking times to the center
  • District radius is 380 meters, or a five-minute walk
  • Mixed-use districts, with housing, shops, offices, and light manufacturing
  • No automobiles, motorbikes, or trucks inside the district
  • A natural district focus at the central transport halt
  • Markets, shops, services, and a large square near the transport halt
  • Public transport running on or under the wide central boulevard
  • Green space surrounding the district


Slide: Block

Attributes of the sample block

  • Short blocks provide a choice of routes to most destinations
  • Narrow streets of varying width, with frequent slight turns, create an interesting streetscape

Slide: Begijnhof

  • Interior courtyard of 40 x 60 meters


  • Building heights are limited to four stories (possibly five or six in very large cities)
  • A wide variety of building types and sizes is desirable
  • Ground-floor commercial space is provided in many buildings


Freight is the most difficult issue for the carfree city. I propose to deliver freight in standard shipping containers, now in wide use throughout the world. A dedicated rail system delivers containers to businesses located along a freight line running parallel to the central boulevard. Local deliveries can be accomplished using a variety of small vehicles: Slides: Freight bicycles Hand carts Small, very slow, electric delivery vehicles


I do not expect that the reference design for carfree cities will ever be built as proposed. Instead, I hope that the principles established in this design will be applied in the creation of large carfree areas in both new and existing cities.

As always, the most challenging problem in making an area carfree is the distribution of freight. In smaller projects, it is clearly not economically feasible to install dedicated freight distribution systems. In smaller projects we must accept some compromises, probably including the use of trucks to make local deliveries. This should be tightly controlled on several points:

  • Hours of access
  • Speed (limited to 20 km/hr) and rigorously enforced.
  • High entry fees, to encourage load consolidation

Clearly, in smaller projects, those businesses that have heavy freight requirements or make extensive use of motor vehicles should be encouraged to relocate to the outskirts of the city.

The conversion of existing urban areas will free up a great deal of street space. We should not neglect to reap the benefits that accrue. All of the signs, pavement striping, speed bumps, traffic lights, large illuminated signs, and other ugly trappings of urban automobility should be removed at the outset. Once the cars are gone, a great deal of street space becomes available for a variety of uses, including sidewalk cafés, benches, flower pots, and fountains. In cities with broad streets, it will be possible to use some of the street space for new buildings. Curbs can be eliminated to facilitate the use of handcarts and to provide unimpeded access for those with mobility limitations. Once the need for brilliant street lighting is gone, we can replace ugly sodium-vapor lights with much softer, full-spectrum illumination.

Carfree districts should be mixed-use areas because they usually assure a high quality of life and minimize the distance that must be traveled to obtain daily necessities.

Carfree areas will not be without their problems, but they can be expected to yield a dramatic improvement in the quality of life while also delivering a large environmental and sustainability benefit. It is time that we availed ourselves of the opportunity.


During the past 30 years, most cities in Europe have made portions of their old downtown districts carfree. These areas have generally won the support of merchants and the public alike, although sometimes only following strong initial opposition.

An example is Zermatt, a carfree alpine resort with a population of about 20,000. Private cars are not allowed into the city. Visitors park a few kilometers away and take the train the rest of the way into town. Residents may park in underground garages at the north edge of town. Zermatt is small enough to cross on foot in 10 or 15 minutes. Small, slow, battery-powered taxis carry passengers and deliver freight, which arrives either by train or in a utility area near the garages. A few specialized vehicles, such as concrete delivery trucks, are allowed to use the streets. Zermatt appears to be popular with residents and visitors alike, and offers an active street life, attractive neighborhoods, and clean air. It is a very prosperous town, no doubt largely as a result of the very high quality of life that it offers.

The success of existing European carfree areas is now almost beyond question, and there is no reason not to expand upon this success. Improved public transport is, of course, essential to practically any carfree conversion.


In Carfree Cities, I included several illustrative proposals for the conversion of existing cities to the carfree model. These examples will serve to show how a variety of challenges could be met in the conversion of existing cities.


Lyon This was the first proposal generated. It is based on the four existing metro lines plus the addition of a number of new tram lines. Nearly all of the existing core urban areas would be retained in urban uses, but much of the outlying area would be condensed and a considerable amount of land freed up for use as parks and productive open space.


Manhattan is already substantially carfree, at least according to the numbers-only about one-third of daily commuters to Manhattan are in private automobiles; the remainder use New York’s relatively good public transport network. However, the city is dogged by intense traffic and all its accompanying problems. The proposal assumes the completion of the Second Avenue subway and the replacement of crosstown bus service by trams, as is indeed already seriously proposed for 42nd Street. Minor cross streets would see the installation of freight trams to deliver containers from terminals on the rivers. Barges would deliver containers from existing harbors in New Jersey to the Manhattan terminals.

Los Angeles

Los Angeles is the icon of the auto-centric city, so I thought it would be instructive to show a conversion of one of its more central areas. Subsequent field inspection revealed densities somewhat higher than I had assumed, and, in reality, somewhat more of the area would have to remain in urban uses, with a resultant reduction in the amount of park land that would eventually be created, but this does not greatly affect the proposal. Los Angeles’s small metro system would be extended, as has already been proposed, and new tram lines would link to metro stations. Utility areas would be installed at freeway interchanges, which would include garages where visitors arriving at the carfree area would park. Rush hour travel should be appreciably quicker than today and competitive with the car even in quiet periods, assuming that there is enough demand to run reasonably frequent service.

Brownfield Redevelopment in the USA

Other candidates for carfree redevelopment are large brownfield sites in US cities. Few cities are without these sites, and most are already served by heavy infrastructure and often also reasonable public transport service. Given that the cost of redeveloping these sites as fairly dense, mixed-use carfree areas should be lower than new construction or other kinds of redevelopment, this offers a means to solve the increasingly pressing need for affordable housing in the USA.


The outlook for carfree districts is considerably more favorable than is usually assumed. People everywhere are sick and tired of traffic, and those who do not own cars obtain virtually no benefit from them while bearing a large burden from their use by others. In many cities, one-third or more of households have no car, and these families are a ready market for carfree districts. With the addition of car-sharing facilities and limited parking at the edge of the carfree district, these areas should find rapid uptake in the market.


It is my dream that the European Commission will see the wisdom of this approach to arranging our cities and decide to fund a new, completely carfree city somewhere in Europe. This city could become the home of those directorates general concerned with transport, the environment, and urban development. Such a city would become a showcase of the best practices for sustainable cities.



Crawford, J.H., Carfree Cities (Utrecht: International Books, 2000) Appleyard, Donald, Livable Streets (Berkeley: Univ. of California Press, 1981)

INTERNET LINKS (includes hundreds of links to related sites):


This article was published in:

International Making Cities Livable conference 34Alpbach and Salzburg, Austria15-19 September 2002.

Leave a Reply

Your email address will not be published. Required fields are marked *