One of the most worrying things about North Korea's military threat to the South lies in the fact that Seoul--the historical capital of Korea, and a metropolitan area home to half of South Korea's 50 million people--is within range of North Korean artillery. Although--as Karlin notes--growing military superiority may allow for successful preemptive strikes, and despite ongoing efforts to build sufficient shelters for Seoul's population, the city is obviously at risk. Close to the 38th parallel that inspired the post-1953 DMZ, Seoul's prosperity is fragile.
Over at Lawyers, Guns and Money, Charli Carpenter put forward a proposal by one Robert Kelly to diminish South Korea's vulnerability to the North by decentralizing the country, moving population from the northwest of the country to the southern areas.
The various criticisms of this plan--that it would be very expensive, that it would take a long time to make any noticeable shift in the distribution of South Korean population, that it would be illiberal, that it would be a waste of money once reunification/regime change came, that it would be illiberal, that expecting South Korea to move its capital from Seoul would be as plausible as expecting France to move its capital to Lille or Lyon, and that given South Korea's small size it's not obvious that even a partially successful decentralization would make much difference--all stand. The first comment is the one I like best.
Cities are built on geography and human inertia. What starts as trade routes and resource-rich regions result in the financial, government and service structures to support those primary industries. That’s what causes the influx of immigrants, the concentration of wealth, and eventually the self-sustaining nature of the city.
Cities don’t die unless that fundamental geographic economic advantage disappears. While there might be ways of encouraging growth in the south, there won’t be a fundamental shift of population without a regional economic incentive.
Moving the government buildings from one spot to another might shift a population, but only a small portion of it. Only 16 US state capitals are located in the largest city of the state, after all.
The commenters, it should be noted, did agree that inasmuch as state policies discouraged investment and development outside Seoul, these policies should be changed to favour the growth of the second tier of South Korean cities.
This sort of sentiment isn't new. The idea of decentralizing population and industry in a centralized country was most prominent in France, where geographer Jean-François Gravier coined the phrase "Paris and the French desert" to describe the dominance of Paris over the rest of France. Owing to early declines in birth rates, and perhaps also the concentration of immigrants in Paris (and other cities), many regions of France saw their populations stagnate and decline, while Paris become ever-more powerful in a centralized republic. After the Second World War, systematic government planning did aimed to promote decentralization.
Industrialisation in France was based, as in other countries, on coalfields. The black countries in the North, Lorraine and Massif Central were the first centres of the steel, chemical and textile industries. But the second phase of industrialisation was of greater advantage to Paris, as major industries, such as cars, aircraft, engineering and electrical goods, began operations in and near the city. In one hundred years, the population of the capital, which was already 2 million at the end of the 19th century, grew fivefold.
[. . .]
The planning body, DATAR, was set up in the early years of the Fifth Republic (1963). Its work is centralising by nature, but its effects have been contradictory. At first, the division between Paris and the “provinces” (a condescending term, now replaced by “regions”) was accentuated. Paris, the centre of politics, the economy, research and culture is also the hub of infrastructure networks. The web of roads and railways was strengthened by new forms of transport: motorways, high-speed TGV trains and airports.
Furthermore, industrial policy in the Gaullist period focused on aerospace and the nuclear and electronic sectors for reasons both military and civilian. This planned industrial policy, based on nationalised industries, was the origin of what are now called new technologies. But the new technologies were located in the Paris region, where they had all the elements required for their development: grandes écoles, universities, CNRS and the military-industrial complex.
This process of concentrating highly qualified employment in metropolitan areas was extended to other cities. Those that already had an industrial, university and technological basis, such as Grenoble, Toulouse and Bordeaux, benefited from the establishment of aerospace industry, nuclear and electronic research centres and became science cities. Other regional cities created science parks, such as Montpellier, Lyon, Nancy, Metz, Rennes, Nantes, Lille, Nice and Marseille. Publicly funded science research was more evenly spread across the country and privately funded research set up closer to universities.
The net effect may have been to decentralize France, to allow second tier cities to emerge as niche competitors to Paris. French efforts to decentralize the country, however, had two negative effects.
The same problems would apply to Seoul and South Korea. Indeed, Paris and Seoul are classified by one author as "macrocephalic" cities, places where geography and governance and economics and population have concentrated to produce one urban centre that completely dominates the rest of the country. (Vienna and Budapest, former imperial capitals now the metropolises of much smaller rump states, and a Bangkok more developed by far than the Thai countryside, also fall into this category.) Sociology Danny Dorling's 2008 paper "London and the English Desert: The grain of truth in a stereotype" argues that Greater London is starting to acquire a similar position of dominance in England. Other cities--Tokyo-Yokohama in Japan, Buenos Aires in Argentina, perhaps, Johannesburg in South Africa, or Baghdad in Iraq--might be in similar positions.
Is it in the interest of the South Koreans to decentralize their population so radically? Military vulnerability aside, it doesn't seem to be the case. Regardless of what policies have encouraged Seoul to become so dominant in South Korea's urban hierarchy, and the legitimacy of these policies (the decisions of military dictators to concentrate everything in the national capital comes to mind), Seoul is now what it is. Trying to take the metropolis apart--as opposed to trying to promote growth in other major urban centres, and perhaps using high-speed commuter connections to functionally fuse more cities in South Korea with the capital--would involve massive and expensive population shifts, to say nothing of the strong possibility that there might not be many places to hide in a compact South Korea. It would certainly hurt Seoulites efforts to promote Seoul as a world city.
South Korea's population is caught in an unenviable situation, living in a thriving city that's at risk of devastation. In this respect, Seoul might not be unlike the cities of the Cold War world, which regardless of their ideological affiliations were vulnerable to annihilation in the space of a half-hour at most. I can only hope that the experts are right when they say that an escalation to war is unlikely, and that the Koreas--even North Korea, however unlikely it may be--will be as lucky as the rest of the world was.