Hiroshima, Japan. Photo by Banter Snaps

You might think that a city like Tokyo, with a metro population of 14 million, has very little in common with a city like Sydney or Melbourne, with populations of less than 5 million. Or what about the ancient city of Kyoto, or the previously devastated city of Hiroshima?

A recent three week visit to Japan – my first – accompanied by some reading, has changed my perceptions significantly.

A colleague recommended that I read Andre Sorensen’s 400 page The Making of Urban Japan – Cities and Planning from the Edo to the Twenty First Century as I caught trains and walked around Tokyo, Osaka, Kyoto, Hiroshima and Nagasaki – not really holiday reading (thanks to Rod Simpson for that), but it sharpened and more deeply informed my observations, especially the role of both governance and economics in shaping these cities, very much like Australia.

This article makes a number of observations, also informed by Sorensen’s analysis, reflecting on our own situation.

While Japan is perceived as one of the few genuinely “foreign” countries by some, Sorensen sees significant similarities as well as differences with Western countries that share similar levels of wealth and urbanisation.

Even though Japanese cities are no longer growing like their Australian counterparts, one can clearly see the impact of decades of rapid and significant growth.

Hiroshima for example is a harbour port city industrialised along its harbour. Like Sydney it converted large rural/ farming areas to urban uses.

Whether there are lessons for us in Australia is a question explored here.

Sorensen and The Making of Urban Japan

Sorensen’s survey digs down into Japan’s very particular political system, economics and culture, that he argues has remained a top-down national structure from the Edo to the present.

This, coupled with the focus on economic growth over liveability following its devastation after the Second World War through to the 1980s, has deeply influenced the form of its contemporary cities.

Sorensen identifies five key “influencers”:

  1. The dominance of central government throughout Japan: while this has existed from feudal times, it hasn’t actually changed, despite post-war attempts to break down the centralisation of power. In a sense Japan has gone from feudal to free market without changing its top down governancestructure. A tradition of urban neighbourhood self-reliance for day-to-day needs, without meaningful local government authority or representation. The individual had, and still has, responsibilities rather than rights. Citizen groups that emerged in the 1970/80s have only had limited successes. A national focus on economic development: while this increased significantly after the Second World War, economic development was seen as imperative to counter the increasing influence of foreign powers from the late 19th Century.
  2. Central government funding has favoured economic development over social infrastructure, and, Sorensen argues, continues to do so, at the expense of focusing on civil society, acquiring open space as greenfield areas were urbanised. A preference for public infrastructure and projects over regulation of the private domain: the combination of weak local governance and centralised power, resulted in local planning controls being seen as a hinderance to economic development. Most local plans have not been supported by the National Diet over time.

Sorensen highlights a number of impacts on Japanese cities: urbanised land increased from 15 per cent to 80 per cent over the 20th century. This is not dissimilar to Australian cities. Both have rezoned very significant areas of rural lands to relatively low destiny residential at the edges of metropolitan cores; the world’s best and most heavily trafficked rail system; increasing road congestion and a vast network of elevated freeways; lack of public domain.

The lack of controls on many of the major metropolitan cities has led to significant sprawl; while roads were delivered as essential to deliver development, open space was not.

Like Australian cities, the high cost of urbanised land mitigates against the purchase/allocation of sufficient open space for rapidly growing areas that transition from rural to urban.

While Japanese planners introduced measures where 30 per cent of upzoned land was to be contributed for roads and other public domain, central government was not supportive.

Japan’s history of local self-reliance means that at the local level citizens have a “duty” to look after their local affairs rather than “rights” over how their area is developed at the larger scale.

Central government’s strategy has focused on delivering national economic development projects first, postponing social infrastructure.

This is a high-risk strategy for Sorensen, as outer areas rarely receive adequate social and public domain infrastructure, since value uplifts on rezoned land make postponed acquisition unaffordable (something we are all too familiar with in Sydney).

Japanese cities observed

So how is this manifested in contemporary Japanese cities? Are Sorensen’s five influencers evident to the observer?

Let’s take Hiroshima and Tokyo as examples. Tokyo – considerably larger than Sydney – is also a global city. While Hiroshima experienced almost total physical destruction during the war, much of its street pattern remained (see figure 1).

Figure 1:  the “parkway” that cuts through Hiroshima.

The “parkway” that cuts through Hiroshima

In both cases, very significant economic growth and infrastructure transformed these cities.

With respect to infrastructure, major motorways cut though both cities, especially Tokyo. In the case of Hiroshima, it was both transport infrastructure and the filling in of the harbour to create large areas of industrial uses, separating the city from the harbour in a similar way to Darling Harbour in the early 20th century. (See figure 2).

Figure 2:  the “parkway” that cuts through Hiroshima.

The difference is that the industrial infill coming almost a century later was on a significantly larger scale and has not become redundant, there being nowhere else to relocate to.

There is very little amenity on the harbour (see the Hiroshima city plan below).

Hiroshima city plan

As Sorensen noted, the push for productivity resulted in industrial and transport infrastructure at the expense of civic infrastructure.

This is evident in the lack of public domain, apart from temple, palace, and castle lands converted to parks.

Notwithstanding this, both cities have developed an amazing and diverse jumble of uses, places, and built form, due in part to the lack of private domain regulation (see figures 4 and 5).


There are high levels of street activity where the original street grain (not built form) is preserved, despite no real footpaths, just painted line markings and a jumble of overhead powerlines in the central area (see figure 6).

As Sorensen notes: “the best places escaped “comprehensive planning”, the worst places were not so fortunate”.

Cite and Ville

The reality is that we cannot plan our growing cities without some form of comprehensive planning or “city shaping” at the metropolitan scale.

How then are we to plan the “best places” within our growing metropolis? From these very cursory observations of Japanese cities, the best cities appear to combine what Richard Sennett refers to as the “Cite and the Ville” (see Building and Dwelling; Ethics for the City, Richard Sennett, Allen Lane, 2018).

For Sennett, the Cite is the accumulation of fine-grain human places that evolve slowly over time, tinkered with and adapted from the bottom up at the local level where we “dwell”; the messy vibrant places described above, and, famously, by Jane Jacobs.

The Ville is the comprehensively planned city of top-down plans and policies; in the case of Japan this has built one of the best metro systems in the world in Tokyo. Sennett associates Jane Jacobs with the Cite and Lewis Mumford with the Ville.  The question is, how can we have both?

So what insights can we gain from these cities that have, like many others, embraced comprehensive economic development as the dominant planning model?

What is particular about Tokyo’s vibrant street life that has survived “comprehensive planning”?

Could Tokyo have been even “better” as well as “bigger” if more had been invested in public domain as it grew at breakneck speed?

While there are no simple answers, some reflections below are offered to provoke discussion.

The Australian context

Australia is similar to Japan in that central government has significant control over the spending of state funds (refer to Marcus Spiller’s argument that a “vertical fiscal imbalance” favours federal funding of large state projects – see Australias Metropolitan Governance Challenge, Spiller 2018).

In Japan this extends to local spending also. So in that regard, Australia may be slightly less top down.

Nevertheless in both countries there is a tendency to regard large federally funded state infrastructure projects as the primary ‘”city shaper”.

While this may deliver much needed improvements with respect to accessibility to jobs and housing, it does not by itself create the sort of urban grain, fabric and open space that create the places we value so highly.

As seen in both Tokyo and Hiroshima, it is the fine, walkable-scale grid that pre-dated “comprehensive planning” that creates the valued places.

While this combination of new infrastructure and fine-grain places may work in existing urban fabric structured prior to car dominance, the real challenge is how to achieve such places in greenfield areas where there is no fine grain and a need for significant infrastructure delivered relatively quickly?

In the case of the outer greenfield areas of Japan this does not seem to have been successful, especially with respect to open space or place making.

So how do we meet this challenge where we are only now catching up with infrastructure and bringing it to very large areas of formerly greenfield or rural? Can we comprehensively plan the Ville and “make” Cite places at the same time?

Are Australian cities with stronger local government authority and community voice better able to respond to citizens’ demands for open space and social infrastructure, especially in the more established areas where such amenity is strenuously guarded?

While there is currently significant concern that the pace of development is not matched by sufficient schools and open space, these concerns are reflected in the ballot box at local, state, and increasingly at federal levels.

Can we create a greenfield “Cite”? What should we do?

In the Australian context the challenge is greater where significant renewal is proposed for a future community that has no voice yet.

While this may be the case in both brownfield and greenfield areas, the challenge in the greenfield is greater.

In the cases of Pyrmont and Green Square, major infrastructure and intensification has overlaid 19th and early 20th century urban fabric, much of which has been retained.

In both cases multiple levels of government combined to develop master plans that included significant public open space as well as affordable housing, through the Better Cities Program, and  much of the existing urban fabric was retained even though the built form is significantly more intense..

With Green Square, the South Sydney Development Corporation brought together state and local government.

As discussed above, the task is much greater in large upzoned greenfield areas where major urban infrastructure projects are planned and delivered before a fine grain pattern is established.

Like Japan, the quantum of transport, social, and green infrastructure is far greater than brownfield.

The Pyrmont core precinct was some 26 hectares (note Green Square). The Western Sydney precincts are each over 500 hectares.

As noted above by Sorensen, prioritising “economic infrastructure” at the expense of social, risks inadequate delivery of the latter, especially as land values climb as a result of government investment in major economic infrastructures.

Like Japanese cities, we also require strong connectivity to jobs and housing; such infrastructure itself, however, is not sufficient to deliver a liveable city.

There is an opportunity to learn from this and consider how both the major (both green and transport) infrastructures and “fine grain” fabric can be delivered to create productive and liveable new places for communities over time, to develop both a Cite and a Ville.

Richard Sennett once remarked to Jane Jacobs that she was better at Cite and Mumford at Ville. To which Jacobs replied, “What would YOU do?”. Indeed, what should we do? Can we do both?

Philip Graus FAIA MPIA is an adjunct professor at the University of Technology Sydney’s Design Architecture and Building.

Spinifex is an opinion column open to all, so called because it’s at the “spiky” end of sustainability. If you would like to contribute, we require 700+ words. For a more detailed brief and style guide please email editorial@thefifthestate.com.au

(Visited 1 times, 1 visits today)

Join the Conversation


Your email address will not be published.

  1. Great article.

    Full disclosure, I had the opportunity to have coffee with the author while we were both in Tokyo for the first time as part of 3 week trips to Japan.

    My reflection on Tokyo’s development mirrors Phils. The key thing that stood out for me, traversing Tokyo with a 2 year old, was that the inner-city is poorly designed for children. The lack of investment in public realm really stands out when you’re maneuvering a pram around in all aspects, but a couple of really poor aspects stand out:
    – The use of lines on the sides of many streets to denote pedestrian areas (as opposed to footpaths) meant venturing into traffic regularly. The 1 metre zone for pedestrians was shared with telephone poles, parked cars and numerous other blockages, meaning you have to regularly step into the street to get down it.
    – There’s also next to no green space. Plenty of public or civic space, but most utterly devoid of interest to a child.

    I really felt like Tokyo had been developed well for commuters and industry, but there appeared to be an expectation that children wouldn’t be sharing the city. If we’re to avoid some of these poorer outcomes in Sydney, ensuring our city works for everyone is a must.

  2. I like this. I was astounded by Tokyo’s public transport and particularly rail network. Almost too much stuff but really an inspiration for modal shift. So yes there are some big roadways but mostly people in the metro area use the rail network. If we grow be densifying we can plan for such a future. If we grow by dispersing we cannot. We continue to grow by dispersing despite some counter trends and policy. What would you do?

  3. Hi Phil,

    I enjoyed your piece. In my limited reading of the Japanese city, I believe that cultural acceptance of density and much smaller-scale statutory measures are the twin reasons for the success of Japan’s urbanism.

    These statutory pieces are:

    -‘inclusive’ zoning – where the most difficult or polluting zone allows every other use, the second most excludes only the most polluting, and so on

    -limited local opposition: as I understand it, if a building is more than 500mm from its neighbour, it cannot be blocked on grounds of opposition. In general, verticality is accepted.

    -complete lack of on-street parking allows quality streets

    These controls (and public transport networks) simply allow mixing of people and uses in a way that is largely unseen elsewhere. The vertical mixed uses of Tokyo in particular is unlike anything I’ve ever witnessed.

    I encourage you to read ‘Learning from the Japanese City’ by Barrie Shelton, and ‘Tokyo: A Spatial Anthropology’ by Hidenobu Jinnai – both are very insightful.

    I believe Tokyo, much more than the usual citations such as London, Paris and New York, offers a blueprint for the future of Australian cities.