A quote from Ed Glaeser:

"The internal combustion engine was more than just another, faster commuting technology. The car is a point-to-point transit method while all of the older technologies were essentially hub-and-spoke, where people walked to a stop and then took the bus or train. The car made it possible to live at lower densities since it was unnecessary to walk to any bus stop or to walk to get groceries or perform any other function. The second impact of the internal combustion engine was to break down the traditional monocentric city. These cities were generally built around a port or a rail yard but that transportation infrastructure became increasingly irrelevant over the twentieth century. Trucks don't require central stops and as a result, factories could be decentralized. The decentralization of employment has led to much flatter cities that look and feel completely different from the monocentric cities of the past." (Glaeser, `Cities, Agglomeration & Spatial Equilibrium' 2008 p.46)

Certain American cities often find themselves hauled up as examples of the awful mess that results from letting cars shape places, but this quote nicely captures that the transformation has been much more endemic. One of the questions I asked in a previous post and the models in the wild talk: is a post-carbon world more likely to be the radical relocalisation outcome talked about by transitionistas or, at the other extreme, can we pull off the infrastructure equivalent of the tablecloth trick - whipping out carbon from under the whole system, leaving it essentially intact? That's the appeal of things like biofuel: being able to just slot in replacements for fossil fuels without disturbing anyone's morning commute. (Seeing how angry people get at motorway roadworks is a sobering indication of just how happy they'd be at larger-scale disturbances.)

Neither extreme is likely, but history tells us that technology largely determines the shape of settlements (even if those settlements' origins lie elsewhere). What jumped out at me from that Glaeser quote: hub and spoke movement is likely the only way to make public-transport-heavy versions of future settlements work. So imagine if (a) Glaeser is right and (b) a post-carbon world must be predominantly public-transport-based. The resulting spatial economic forces will be radically different to those that have produced the places we live and work in now - both for where people live and where production takes place.

So that's a nicely content-free academic statement: "things will change". Uh huh. The interesting part of that, for me, is that in my view we don't have a complete understanding of how those forces work. With good reason - it's a fiendishly complex problem. So most of the time, that old, reliable fallback is used: ignore it. (A point Krugman has made repeatedly; here's a recent-ish take.)

So let's pretend (and it is, currently, still just pretend) we will actually get serious about transitioning towards a no-carbon infrastructure. In the medium term that may mean pretty major demographic changes as the economy rewires itself. Looking at how Liverpool shrunk in the last 80 years is instructive - and maybe a nice illustration of why the economic equilibrium metaphor is useful. Liverpool was already losing its role as a pivotal global port (including for the slave trade) by the time containerisation arrived. Its population is only now starting to stabilise. It seems to me it's useful to think about that as a tension attempting to shift to a new equilibrium, even if that equilibrium is never reached (the reality of economics is much more like a weather system than the static physical models the discipline was built on). That metaphor implies, though, that you can perhaps identify those tensions and develop some useful guesses about which way the wind is blowing.

I'm rather stuck in a Groundhog Day with the planning vs 'shit just happens / spontaneous order' argument, though it's obviously crucial: no city changed through `pure' economic forces, any more than they were purely built by Platonic engineer-planner-gods (can I be one of those!) from a blueprint. I've spent several years sat next to massive pile of Stan Openshaw's old books (thanks to Andy Turner for making sure we still have them!) - many on the history of planning. Getting any kind of overview is a daunting task. I seem to default to the traditional modeller's preference for sitting in a room making pretend worlds. But consider Ostrom's work on managing common-pool resources: based on around five thousand case studies, many years spent sifting and synthesizing - there are clearly alternatives to defaulting to either trite soundbites about markets vs planning or, um, staying in a room making pretend worlds.

Anyway: there's going to have to be a lot of Zapatista-style "walking while asking" going on, isn't there? At the very least, Glaeser's quote has made me think about the role of cars a little differently. The costs and benefits of any carbon-free transport technology lie in what they mean for spatial morphology as much as anything else. Do we need cars to avoid having to re-centralise back to purely hub and spoke models? Are there other transport technologies that could slot into that decentralising role? Or is there just a more mundane shifting-of-balance needed that will be more centralised, but not overly so? In this scenario, over time places will change shape, but at a manageable rate not so distinct to their rate of change now.