It was a night of the long knives for my PhD last week. After much internal bickering amongst the various factions, some of them produced evidence that the leader of the powerful 'planned economy' clique were being funded by the French in a secret plot to overthrow the central question. There was some truth to this, it transpired: references can be found to a title stating -

Aims: to re-examine the 20th century 'economic calculation debate' using 21st Century computational methods'.

(Here is a good summary of that debate.)

A very manly and rugged aim, to be sure, involving what Diane Coyle might call 'macho' mathematics: men in itchy shirts, smoking pipes in the basements of ostensible bookshops, mathematically proving or disproving the theoretical possibility of socialism. (Hayek was above all this, of course.)

But in the grey light of the new year, a Gay Mafia consisting of various 'keepin it real' types carried out the final coup in a moment of wheel-of-fortune-spinning randomness. Well, now the die is cast. Or possibly the dice, I'm never sure. The good news is, it means it won't take me 25 years to finish my PhD - so I've hopefully avoided reaching 60, nourished on stray dogs and fagbutts alone, and submitting a final manuscript on radioactive cardboard written in the bushes of a motorway feed island.

No, I shall finish... within a decade. A brief statement of things as they stand:

• Hayek was right about the complexity of society, though he was a little obsessed about price.
• Places - real places, damn it - have unique human / economic / social networks that centralised planning can kill stone dead.
• But so can enforced marketisation. Prime example: any current government plan that involves shipping current residents out (either because they made the tactical error of living in council properties and so, it seems, signed away all rights to being left in peace, or because councils have banged compulsory purchase orders on them) to 'renew the housing market'.
• Putnam says: "the causal arrows among civic engagement, reciprocity, honesty and social trust are as tangled as well-tossed spaghetti... however, we need to recognise that they form a coherent syndrome."156
• It's from here that I nabbed the idea of a 'human network syndrome' - a dynamic tangle of social and exchange networks. These networks live in our specific communities. Certain scales within those communities can have specific properties; a recent example from the Front Porch Forum blog is the 'borrow a ladder' test for deciding the number of houses to have in a functioning local forum. Note how this isn't simply about cost, as it might be when weighing up whether to walk and buy milk or drive to the supermarket.
• So if this human network syndrome - a mutually reinforcing, geographically embedded set of social and economic dynamics - is a spontaneous order, is it possible to:
• Model it - both with simple thought experiments and more in-depth models
• Measure it - take the theory and come up with a way for checking data for proxies
• Theorise whether / how it can be planned for, what things are bound to wreck it - and whether Hayek was right that our only hope is to decentralise and leave well alone. (A position I'm inclined to agree is better than the government's bizarre mix of state action and market promotion.)
• Think about whether theory and the messy reality of one case study - probably the EASEL area of housing renewal in Leeds - can sit in a room and actually talk sense to each other.

Of course, this just makes matters more complicated in a different way. But at least it's, y'know... grounded. Kinda. To sum up, this is a move away from 'can economies be planned?' (As my supervisor said: well, the answer's probably a bit yes, a bit no. Boring.)

Instead, it's thinking about different notions of planning and their interaction with real social and economic dynamics. Two extreme ways of looking at this: 1. planning as a map (a la J.C.Scott155) - a map wielded by the powerful that will change the landscape to fit it. 2. planning as a frame for network dynamics to grow on, as plants grow on frames provided for them. (The gardening metaphor is originally Hayek's.)

A literal example of map-planning would be Firvale in Sheffield. Well-meaning consultants drew a line around some houses to be knocked down and replaced with bigger ones, in order to create a more 'mixed community'. The people living there objected, pointing out that... well, they were living there. (Luckily, the area has very high levels of owner occupancy. Other areas were not so lucky - such as Woodside, an ideal spot with great views over the city, but council-owned. Currently grassland awaiting a developer.)

The consultants' Firvale plan therefore isn't going ahead, because the (PFI) money was only available for demolition and private rebuild. (Private contractors won't get out of bed for small plots.) Firvale residents have since met in their project group, and I recall one person memorably saying they 'discussed plans for hanging baskets.'

Firvale gets close to the visualisation I currently have in my head. Its similar to what one might picture when thinking of a plan view of a successful cohousing development. That plan view could be exactly the same for an estate where half the houses were boarded up and the survivors lived in fear.

Which is to say, the plan would not show all the most important things about that area. Or maybe physical layout is a necessary but not sufficient condition for certain dynamics. It may well have attempts to 'design community in', but what makes cohousing work is the commitment of the residents from the start, and the development of social networks as a result. Read the second story down here: a stranger is shocked that a cohousing resident doesn't know where their kids are. The resident knows it doesn't matter because of the dynamic that exists there.

Well, there's a fairy tale of community in there, isn't there? There are plenty of dark sides to all kinds of human network dynamics. But cohousing is a nice example of the 'frame' concept of planning - one that's based on a tested theorisation of one good scale for human interaction and builds a frame for it to grow on. In this case, the three vital ingredients are: not so small that one or two families leaving will fatally wound the existing dynamic; not so large that it becomes impersonal; people involved in its development so they're forced to develop a dynamic together. (There's appropriate scale appearing again.)

That's the jist.

One new aspect that I'm a little over-excited about, and mustn't spend three years on, is that I have agreement to develop some actual visualisations using flash / actionscript 3. Geek-point: actionscript 3 has a new language structure that's not dissimilar to Java, and much more powerful than previous versions. Yup, it's proprietary - we had to buy a copy of the Flash 9 developer. But I've produced some java models that spew out numbers, and they're never going to excite anyone. Now, making a visualisation of a community - that's perhaps running live, with little people borrowing ladders off each other and playing bridge - and then seeing what happens to the dynamic when you demolish one whole block of 300 houses? And speed it up, so the dynamic connections move from sporadic to obvious web? That'll be cool. And PhDs are all about cool. I think that's all they ask about in the viva, if I recall. It won't be quantitatively accurate or even academic, but it'll illustrate a point about maps and dynamics to a wider audience, and I reckon that's worthwhile. Hmm... ask me if I still think that in a year's time.

While we're on pointless graphical cool, take a look at the new prefuse flare, originally written in Java and now in Actionscript 3. Pointless, pretty graphs. Mmm. I'm so gonna try and visualise nodes in my PhD with that.