Urbanism and transportation


A tale of all cities

Cities have existed for as long as civilization has existed, over more than six millennia.  And throughout this vast time period, they have been the hubs of economic activity, in both goods and services.  The reason should be obvious: the city is where all sorts of economic activity is within reach.

In turn, the above factors created enough surplus to spend on culture, making cities ever more inviting places to live.  Thus, for as long as civilization has existed, everybody not directly involved in agriculture tended to move to cities.

This all depends on what is called urban density.  There are many different definitions, but all of them try to formalise the intuitive how many places of interest are there per unit area.

Places and non-places

This is also a good opportunity to introduce the notion of places and non-places.  A place is anywhere normal people might want to hang around, and go there on foot.  Houses, shops, parks, offices and squares are places.  Multi-lane roads, parking lots and mown grass verges along them are not.

Obviously, the benefits of urban density are created by places.  The non-places dilute this, and as such, are antithetical to the very concept of cities.  Interestingly, most non-places are closely related to automobiles and the problems they cause.  Wide roads, parking lots and the "green space" to protect people from their unpleasantness.

A high urban density has benefits for the wider environment, too.  It enables the sustainable modes of transport, because destinations tend to be close by, or because there is sufficient patronage to make public transport systems viable.  In turn, where the primary means of getting around are walking, cycling and taking public transport, their infrastructure takes little space, and is often a place itself (pedestrian streets), allowing higher densities.


We can see that all over the world, regardless of culture, people built fairly similar cities for millennia.  Even when not enclosed by walls, the cities were dense, because everybody wanted to live as close to the centre as possible.  And today many are widely admired tourist destinations, with modern people paying generous sums to spend some time away from suburbia.

You might notice that only in the last few decades—after the world wars, really—was the millennia-old trend of wanting to move to the cities reversed temporarily.  This was a result of badly-planned cities meeting the American Dream—though calling it Hallucination would be more approppriate—of living on "a little house on the prairie".  Indeed this dream was largely fueled by the badness of the American cities.


Throughout the world, we see innumerable examples of small settlements, sometimes just a few hundred people, living in a small high-density area, with a very clear border to the surrounding farmlands.  Even in the case of such small villages, without being forced to, people naturally built this way.

These villages and small rural towns existed for the purpose of providing services to farmers.  As a result, when the per-capita productivity of agriculture increased, or the number of farmers declined, these settlements contracted and became backwaters.

This contrasts sharply with the modern suburb, where people with city jobs reside.  They do their shopping far away, too, in big box—warehouses, essentially.  Often there is no clear edge to the countryside, either, just progressively larger plots.

This recent reversal came about due to American cities being built wrong.  They became inhospitable places, driving people out into the suburbs, a phenomenon which has never existed before.

Back to the top

Patterns of failure

Most cities of the world follow the same millennia-old, proven patterns.  Without any communication between their cultures, they built in much the same fashion.  Today, many such cities still exist, and are great tourist attractions.  Obviously, if they are so livable and walkable they attract tourists, they cannot be bad places to live. 

Now, clearly, most American cities are different.  Nobody in their right mind would go on a vacation to visit American cities like Chicago or the Bronx.  Nobody would go on a tour of modern suburbs, either.  Lots of people know from experience, these places are inhospitable, and they try to get away from them.  Here I would like to shed some light on the reasons for this difference.

Point of reference: traditonal cities

If we take a look at old cities, and even some fairly modern (19th century) non-American ones, we notice a few patterns.  One is the often irregular road layout—this shows a lack of central planning.  Another is the scale of the streets.  By modern standards, the streets are very narrow.  Many of them are just 4-5 metres "wide", and several are even narrower than that.

Even in more modern cities, built in the 19th century, on a planned road network, many streets are barely 6 metres wide.  Most significant European cities have extensive heritage from this period.

Another pattern we find is that the buildings are right up to the property line.  Often the front door opens right onto the street, with just a small step.  Also, they tend to be attached, forming a continuous line along the street.

However, the streets rarely, if ever, feel like canyons.  This is partly a result of the varied architecture, but largely it is because the more major roads have shops all along the streetfront.  Apartment housing (or offices) on the upper floors, small shops facing the street.

A further pattern again concerns the road network.  There are a few major roads, 20 metres or even wider—but only a few.  The great majority of streets is significantly narrower.

There are also quite a few small squares or parks—often in the courtyards of buildings.  Many buildings have one; in other cases, the townhouses around the block share their backyards as a private park.

These all come together to create all the positive characteristics a city can have: high density, high permeability, and generally being a pleasant place to live.

Back to the top

19th century hypertrophism

Hypertrophism is the condition of each element being significantly larger than its proper size.  In the case of cities, this means buildings, blocks and streets being much bigger than what had been normal—humans didn't expand with them.

What I'll call 19th century hypertrophism actually appeared in the late 1700s, and persisted until the interwar period. 


Almost all cities in the United States follow this pattern.  There is no question however, that its absolute poster child is Manhattan.  These cities are characterised by being built on a rectangular grid of wide streets, and often sporting tall, boxy buildings.  The latter is not necessary, indeed that was the last element to appear, and these cities were already bad even without them.

These cities were mostly founded during the 1800s, drawn on a map with rulers.  The grid itself is not actually a problem, but serves as a marker.  The problem was what the grid consisted of: very wide streets.  Too wide.  Unlike narrow streets in all earlier cities, they proved too large to pave before asphalt, and were thus left barren strips of dirt.  To avoid having to tread through a sea of mud when it rained, the sides along buildings were covered with wood—the ancestor of the sidewalk.  The middle of the road was thus, by default, left for wheeled traffic, despite wagons being few and far between.

More importantly perhaps, these wide, unpaved roads everywhere took too much space, and risked decreasing density.  The solution was to make building blocks larger.  Although this might have worked, in practice it only created another problem.  Plots became very long—over 50 metres in some cities—and thus either very large, or very thin.  Very large plots for detached family houses meant low density; very thin plots lead to the creation of the shotgun house, a uniquely American format.  Alternatively, where higher densities were required, terraced housing was built.  But again, the plot was too deep, thus very narrow but rather tall (5-6 floors) townhouses, and strangely spacious backyards were the result.

During the early 20th century, plentiful steel, asphalt and the automobile arrived.  Now the wide streets were paved, motorised traffic started running down their length, and the sides grew highrises, creating the well-known "concrete canyons" from which people flee to the suburbs.

For contrast, the European cities dating from the same period followed more traditional layouts.  Their streets—mostly no more than 6 metres wide—could be paved much earlier, for the use of pedestrians, mostly.  And even where motor cars are allowed, there is only room for one narrow lane of traffic (one-way or shared between directions) and parking on one side.


Perhaps more interesting is what happened to the small settlements.  This is the vaunted "Small Town America" to which designers periodically try to return.  Interestingly, however, we find these places already suffering much the same ills as the cities.  Even villages had vast roads running down their middle, and they again proved unpavable.  More worryingly, perhaps, we find that the houses did not resemble the millennia-old pattern, but were freestanding farmhouses in the middle of enormous plots, while clearly not being farms.

Consequently, even in what should have been small villages, distances were significant, and walking often unpleasant.  Thus even fairly modest residences often had carriage houses, later becoming garages.  We can see that "Small Town America" was already halfway dependent on personal transportation, a hundred years before the invention of the automobile.  As a result, anybody trying to return to it can only fail and create car-dependent communities.

Back to the top

20th century hypertrophism

This pattern is characterised by superlatives: tremendous highrises, enormous highways and gigantic green spaces.  From a helicopter, such a city might look astonishing—but since people experience it from ground level, the result is very different.

First appearing on plans in the interwar years, by Le Corbusier and other modernist architects, they are completely dysfunctional as cities.  Yet inertia is enormous, and more of the same is being built, the world over, somehow expecting a different result.

In my opinion, the architects were overwhelmed by the successes of industry.  Mass production was high-tech, the way of the future, and economies of scale meant big was cool.  Thus the choice of "aesthetics" being giant flat surfaces of concrete and glass, and the complete absence of ornamentation.  To match, the buildings are enormous boxes.  While this may be fine for a "monument to human capability", as a modern Great Pyramid after a fashion, everyday buildings should not look like this.

This is partly to imitate the industrial buildings, which were (and are) built without any care for looks, only functionality.  However, the maxim of "form follows function" does not apply to residential buildings.  Their form is an important part of their function as pleasant places to live.

Add the recent invention of reinforced concrete, allowing such visions to be realized at tolerable costs.  And the automobile as a means of transportation.  The modernists were positively enamored with it!

The industrial way of thinking also meant cities were designed in the manner plumbing is designed.  In a fully functional manner, without any regard to what we call user experience.  Also, they were planned from above, looking at scale models on a table. 

Another result of this thinking is that highrise buildings are "presented" with empty space around them.  This is sometimes a car park, sometimes mowed lawns, sometimes trees.  In any case, these are not useful parts of the city, these are absolute non-places.  It might look nice on a design (viewed from above), but it does not belong in a city.


This pattern was mostly built after the Second World War, and is an even more spectacular failure than its predecessor.  Despite the highrise construction, density is actually very low.  Between the buildings the intrepid pedestrian—a pedestrian! The automobile-age term for a person—encounters vast expanses of greenery (often mowed lawns), parking lots or a freeway.  Obviously he cannot get anywhere on foot in a sensible manner, but needs a car.

And the city certainly does provide for cars.  There are motorways right through the city.  There is an abundance of parking.  Too bad there are no places for people.


When the inhospitable cities met with the idea of living on a fake farm, and using a car to go everywhere, the suburbs—also known as urban sprawl—were created.  Or they developed, step by step, from the 19th century hypertrophic small towns.  You see, when cars were introduced to the latter (replacing carriages), the larger streets of the grid started to see uncomfortable levels of through traffic.  Main Street, previously the place to live, fell from favor as people choose calmer streets.  And the designers were ready to serve: they introduced cul-de-sacs, crescents and other layouts to exclude through traffic from the neighbourhoods.

In parallel, Main Street—the place for shops to be—faced hard times, and for obvious reasons.  Customers came by car, but there were no places to park.  So shops started to look for new locations, where adeqate parking provisons could be built.  These might as well have been in the newly-built residential areas.  However, the residents did not want a shop to draw traffic down their calm street—and the councils readily enacted single-use zoning.  So the stores moved to the edge of town, congregating into the big box warehouses we all know and hate.

And now the people were left with everything far from them.  The jobs still were in the city, far away.  The stores were on the edge of town, no greengrocer on the corner anymore.  And the housing districts were built to much too low a density—and with disconnected street patterns obstructing pedestrians—, so public transport was not financially viable.  Too few people lived within a reasonable walking distance of a stop, quite simply.  Thus the absolute dependence on cars, with almost every adult owning a personal motor vehicle.

Back to the top

The way out

All this might sound a bit depressing, but there is a way out.  Designing things right is no more difficult than designing them wrong, and in many cases there are clear and simple warning signs.  It isn't very difficult to imagine how they will be used, either—whether they are going to make doing the right thing easier, or are going to encourage misuse.

The important thing is to keep in mind what are the results we want to achieve, and why.  The goals are livable, people-friendly, high-density cities, with a high proportion of walking, biking and public transport.  The reasons are all over these two articles.  The tools are listed below.


Back to the top


The modes of transportation we want to encourage are very sensitive to detours.  Thus we should ensure that reasonably direct routes are available, especially for pedestrians.

In the case of grid layouts, this limits the length of blocks.  It is not a problem, though, since most streets should be rather narrow.  Collector and through streets, with separate space dedicated to motor vehicles and to pedestrians (pavement/sidewalk), need to be placed in a similar manner either way.  The additional narrow streets can fit in without causing any problem.

On the other hand, motor vehicle traffic is a lot less sensitive to detours.  Thus through traffic can be excluded from vulnerable areas without any adverse effects.  Indeed such solutions can help make other options more appealing.

An example would be the creation of traffic cells.  This usually means the creation of a cross-shaped pedestrian area in the town centre.  Motor traffic can still reach any part of the town, but it might have to go around one or two arms of the cross.  Usually, a "ring road" is provided, but it is rarely close to circular.  Often it has a rather different shape, both because that is easier to come by, and because that provides a less direct route than a circle.

Where rat-running becomes a notorious problem, it is often possible to convert the street to a dead-end or a crescent with a few planters or bollards.  However, there should be enough place between them for pedestrians and cyclists to pass unhindered.  This is called filtered permeability.

To put it together, motor vehicles should be channeled into their own roads.  This can be easily accomplished with a strict, hierarchical system of roads.  Normally there are three levels:

The hierarchy of roads is already a standard tool of urban planning.  However, its current form is missing a key element.  This dendritic structure is fairly direct by motor vehicle standards, but has unacceptably long detours from a walking or biking point of view.  Thus the latter should have numerous "shortcuts" available to them, while keeping motor traffic out.

Back to the toolbox

Properly narrow streets

The recipe to useful cities is very simple: build on the scale approppriate for humans.  This means building most of the residential streets about 5 metres wide.  This is wide enough to provide access for emergency services, garbage pickup, etc.  This is wide enough to even allow two cars to pass each other carefully.  However, the narrowness itself discourages through traffic.  And if that isn't enough, filtering permeability is always an option.

It should not be a question that pedestrians are the owners of the street, and any motor vehicles are "guests" and are expected to behave.  Some further suggestions to ensure this is the case are:

The result should be that people are comfortable walking down the middle of the street, and there is certainly no sidewalk.  Children should be safe, even without adults accompanying them, as was once the case.

This is similar to the Dutch invention of woonerven, translated as living streets, to emphasize their role as an outdoors living room (woonkamer) of sorts.

When the streets become a calm and livable place, houses no longer need a "front yard" of mown grass as a barrier from the hostile street.  Even without any other change, this permits back yards to be proportionately larger and thus more useful.  Alternatively, this could be used to create more blocks and higher density.

Properly-sized streets also allow block thickness to be reduced.  This prevents plots from becoming excessively deep, and thus a wider variety of sizes becomes available, still of useful proportions.  As a result, comfortable single-family detached houses could be built for much lower land costs.  This could go a long way towards solving the lack of affordable housing.

Back to the toolbox

Squares vs. oversized streets

The recipe to useful cities is very simple: build on the scale approppriate for humans.  There are easily recognisable signs when a human-centric (a.k.a. pedestrian-centric) area is too large.  Specifically, if it isn't a square, and it grows clutter.  Some examples:

This happens for two reasons.  First, the place feels large and barren, thus the need to break it up.  Second, it is not actually used by pedestrians.  If it were, they would complain and have the obstruction removed.

Of course, this does not apply to squares.  They are designed to be somewhat empty spaces, to be filled with this sort of activity.  But they can be told apart from wrongly-used spaces by several key qualities:

A square is most useful when it isn't surrounded by motor traffic.  This way, people can be comfortable on the space in front of the shops, and the businesses can also use it for e.g. outdoor seating.

Back to the toolbox


Already I mentioned front and back yards.  They might be nice, but their size is severely limited, and they mostly compensate for the rest of the city being hostile.  What if instead we made the rest of the city more livable?  One way to do that is to have many small and medium-sized public parks.

For instance, if we devoted 4-5% of potential plots to small parks instead, we could provide a 4-plot park for every 100 housing plots.  That is certainly a rather small park, but it's an order of magnitude larger than the private backyard of even a rich person.  More importantly, one can be placed in every few housing blocks.  Then everybody would live within a two minutes' walk of one.  That very short walk being along the calm street.  Would that not be nice, and worth the cost?

However, small parks are not everything, and a few larger parks are needed, too.  Obviously, block-sized and larger parks cannot be built nearly so often, so people would often arrive by public transport. 

But it is also important to recognise what is a park and what isn't.  Parks—much like squares—have some clear defining properties:

"Green space" and other landscaping that does not fit is a warning flag.  They exist either as a buffer from e.g. motor traffic, or to break up an overly wide street.  They are not useful places however, where people would go to relax.

Since people go to a park to get away from all the noisy traffic, ideally the park would be shielded from larger roads by a row of buildings on all sides to make it the tranquil place is should be. 

Back to the toolbox

What about highrises?

With most residential cities only 5-6 metres wide, and most buildings without front yards, height might appear to be strictly limited.  After all, if buildings were taller than the 4-6 floors practical without elevators, then the streets would be dim, narrow slits, right?

Not necessarily.  This limit only applies to the streetfront of the buildings.  They could have a low-rise "pedestal" with perhaps patio or gardens on top, surrounding the high-rise part of the building.  The latter could also have progressive setbacks, gradually narrowing.  The Empire State Building is a very good example of this style.

The Empire State Building is also a great example of intergating highrises into a street.  Although it is surrounded by 19th century hypertrophic streets, there are restaurants and other features for pedestrians to interact with.  As a result, somebody standing in front of the building might not realise its actual size, both because the "pedestal" makes it look like a normal building from ground level, and because there is something other than a stone (or glass) wall fit for a fortress.

Also, not all streets are quite so narrow.  Shopping streets draw enough people to fill a somewhat wider space, but the signs of overprovision should always be watched out for.  Roads with dedicated motor traffic lanes, squares and parks are even wider.  These locations can tolerate taller streetfronts without becoming dim crevices.

Back to the toolbox

Life in the city

What would life in such a city be like?  First of all, by encouraging the sustainable modes of transport, people would be freed from their dependence on cars.  Thus two-adult households would not require two heavily-used cars, just one lightly used one—or none at all.  Since car ownership and use are expensive, this would tremendously help the financial situation of all but the richest.

Secondly, we are trying to encourage medium-density housing for the benefits it brings.  This spans the range from single-family detached houses (on less deep plots), through attached townhouses to low-rise apartments (with courtyards).  Homes in these types of buildings would not be as expensive as the suburban mansions typical today.  The lower footprint per person and lower construction costs would mean housing that is more affordable.  Middle-class people should be able to buy decent homes without borrowing three times their annual income.

Thirdly, the ability to go to various places without a car would greatly help those not able to drive.  From minors to the elderly or disabled, the opportunity to go out on their own would be greatly empowering.  From a more selfish point of view, the parents or equivalent caretakers would not have to chauffeur them around, giving them more free time.

The increased use of walking and cycling—as well as parks for recreational activities—would also combat the problem of sedentary lifestyles and the resulting obesity epidemic.

Fourthly, these cities would be much more livable places than either the 19th century hypertrophic ones or the suburbs.  With calm narrow residential streets, small parks and many small shops, the cities would be built for the needs of people, not cars.

Back to the toolbox


And hopefully, the architecture of individual buildings will cast off its modernist shackles of large, unbroken surfaces, too.  As far as I can tell, small—residential—buildings have been tending towards having more features over the last decades.  But I must ask, how is this still possible, that we, with the highest productivity in the history of mankind, live in buildings with so little exterior decoration?

I have the good fortune to live in a European city with extensive neoclassical and neorenaissance architectural heritage.  I can see and compare daily that the general shape and layout of modern buildings is not very different to them.  Obviously, both were and are built for use by humans.  The difference is mostly in the decoration, which is incidental to the engineering of the building.

What I mean is that the presence or absence of exterior decoration does not immediately affect the function of the building.  The ledges, pilasters and pediments are not load-bearing parts, nor do they weigh a lot.  Modern balconies and hanging corridors do not need the support of corbels, but they could be added without any difficulty.  A railing does its job of preventing falls, whether it is a sheet of expanded metal, fancy ironmongery or a balustrade.  And it certainly does not make a difference if walls are plastered flat, have some texture to them, or low-relief ornamentation is added.

Except these things do matter.  In day-to-day use, they are much more pleasing to look at than the large negative spaces.  This is not anything new: designers already know to avoid creating large and empty swathes of pavement.  Instead they pepper them with road furniture or bits of greenery.  Then the result is called a promenade or some other ambiguous term for something which is neither a street, a square or a park, and thus is not a place.

And the difference is not that great.  Even in the late 19th century, some types of modular decorations were moderately mass-produced in factories.  It should not be a significant hurdle today.  And since it does not really matter, from an engineering standpoint, proper decoration could be added into new designs almost as an afterthought.

Form fills function.  Where the design calls for a railing, it can be a metal mesh or a balustrade.  Where it calls for a wall, that can be flat or visually broken up by some features.  Aesthetics can be used to fill the gaps of which a functional design does not speak.

Back to the toolbox
Back to the top