Sunday, May 31, 2009

Political Allsorts #3

Meanwhile, my friend Oli has, for a while, been giving me some tough push-back on the decentralist, anarchist and green directions that I've been moving in for a while. And I need to flag that I don't have good answers to him yet.

Oli's arguments can be summarized, roughly, as

a) decentralism is unproven, whereas centralism, and strong power-centres, when wielded in the right cause, have demonstrable achievements.

b) in particular, centralism is necessary to promote egalitarian ends. Examples include the EU which is helping spread wealth more equably through Europe, and counter-examples include "opt-out schools" or the "local democracy" which has failed to bring the US school system to an acceptable state.

So, you need centralized concentrations of power to set and ensure standards. And encouraging decentralism is as likely to allow the wealthy and powerful communities to secede from wider social responsibilities or shared planning.

Like I say, I don't have good answers to this. Though I am looking.


Justin Pickard said...

(b) might be valid, but (a) is a normative & deeply conservative thesis. There's nothing inherantly good about things that have been proven.

I would also counter with the notions of Mill's "experiments in living". Without room for a diversity or pluralism of societal models, we could easily be missing out on something much better than the tried and tested.

zby said...

On the other hand you can view capitalist as a decentralized system - (as opposed to feudalism). I think the decentralized/centralized play is wrong - it is not about being more centralized or more decentralized - it is about establishing centralized structures that allow decentralized action. Capitalism is decentralized in the economic decision making - but it relies on centralized low and low enforcement and military and ... It is all complex and fractal - the real point is to remember that centralized is difficult it requires communication and coordination and costs of that grows much more rapidly than our intuition suggests us. So when planning to centralize we need to be careful about what we can achive and what is really out of reach.

Justin Pickard said...

If you do view capitalism as a decentralised system, you should be incredibly wary of public-private partnerships, monopolies, and market distortions.

Oh, and which capitalism? American capitalism, European capitalism, Southeast Asian capitalism, or Chinese communism? None of them are what I'd call particularly "resiliant".

Oli said...

Hey Phil, you've oversimplified my position here. You're making me sound like a Stalinist!

In reality I agree with what zby is saying: you need a strong centre on some key things in order to enable other things to be decentralised in a fair way.

I raised this both as a criticism of anarchism and as tension with local direct democracy.

For example, in order to remove tax havens and protect the global environment you need global agreements and actions that aren't ignored at a local level. We don't want to let local fishing communities unilaterally opt out of fishing quotas even if they all vote in favour of such an opt out at their local town hall.

Subsidiarity, as practised in theory by the EU, appears to me to be the best way to attempt to solve this tension.

So, IMHO, it's a mistake to assume that all centralised authorities are a bad thing. Sure, there will always be significant room for improvement, but I think we should recognise the important role that central authorities have to play in enabling the kind of social, liberal democracy that I wish to live in.

If I had the time to explore this area more formally my hunch is that a mixture between game theory, human psychology and history would provide good supporting evidence for why you need central authorities to set and police the rules we play by as we try to build a fair, liberal society.

Then we need democratic politics so that the people can control these central authorities and to make sure that we continually improve the rules as we find their inevitable weak points.

As zby points out, the best rules will often be the simplest rules that we can successfully implement.

Composing said...

Ouch! Oli, Certainly wasn't trying to accuse you of Stalinism, sorry. :-(

And agree, it was way too simple a summary.

However, I do think that the basic outline of the argument between centralism and decentralism does end up fairly simple.

The decentralist owes an argument for how global agreement (standards) can be reached and policed. And the centralist owes an argument for how abuse and injustice can be avoided at power-centres.

And undoubtedly, any practical answer is going to be in what I'd call "middle-space" (that zone between the extremes of centralization and decentralization) and I guess you'd call "subsidiarity" (ie. devolve to as local as possible but no more local).

I still think there's an active discussion to be had about the "shape" of middle-space (ie. is it "fractal" / "scale free"? If not, what structures are appropriate at different scales? Is it more like a bunch of parallel systems who's power centres act as checks and balances on each other? Etc, etc.)

I also feel that we're in a historical moment where people are coming up with a lot of new discoveries and inventions to solve some global co-ordination questions. The video I linked from gives some sense of that.

I don't feel the same sense of excitement that the practice of managing the centre, of reducing abuse or injustice, is making similar progress. Though may be you could point out some examples where it is.

Oli said...

Interesting. Surely the 'decentralist' also owes an argument for how abuse and injustice can be avoided at power-centres?

Indeed, I think this is the precise point on which different people’s intuitions diverge. Anarchists think that emergent power structures are always more benign than formal power structures. I disagree with this intuition.

To put it simply: I would rather have a formal police under the control of democratically authorised institutions than a local militia under the control of a local warlord. For me it is so clear that we have to choose roughly one or the other. There simply isn’t a ‘no power structures’ option.

Furthermore, while we may complain about excessive policing tactics (such as the debate around kettling) in my mind the true story of the G20 policing was the new level of public monitoring of police behaviour. Surely this was a great example of the ways that new technologies and ideas are allowing the decentralised people to exert even greater authority over their own centralised power structure.

Emergent power structures favour the physically, economically and socially strong over the weak. How can justice be fair without having a single law of the land under which all are meant to be equal?

zby said...

Phil - you seem to propose a linear scale of decentralized through some 'middle-space' to centralized - but you can have policies that are centralized in one aspect and decentralized in another one. You could say this is one of the middle-space things - but in my opinion this is oversimplification - because it misses distinction between this policy and the opposite one.

Composing said...


I don't think anarchists are necessarily "naive" about this. Or rather, some are, some aren't, just like holders of any other political position. (Eg. plenty of people cite )

The main decentralist answer to the problem of abuse of power-centres is that the scope for abuse is much reduce, as power-centres are just smaller and less powerful. They're less capable of doing harm. And they're more easily checked by rival peers.

Sure, power isn't everything. There's also morality. And a particularly nasty local-warlord or town bully can cause more local suffering than is felt from a more powerful but benign emperor. The decentralist doesn't (or at least, shouldn't) claim that local is a guarantee of moral superiority.

Nevertheless, we'd believe that statistically direct oppression and suffering would be reduced as power is distributed.

I guess that's where evidence starts to become important as you're up against the Hobbesian counter-intuition that suffering is reduced by leviathan.

I think the case of preferring the police under an elected state vs. an unelected warlord is slightly misleading as there are two issues : the scale and the electedness. To compare like-with-like we'd have to compare police under an elected state vs. police under a locally elected police-chief or town-council.

Or police under an unelected president vs. under the warlord.

In both *these* cases, my intuition falls in favour of the localist version. But I suppose you might argue that the leader's responsibility to the electorate is just easier to maintain at the national level than the local. Eg. locally elected councillors are more prey to corruption than nationally elected MPs. Or that national MPs can be more easily frightened - by the sheer size of their electorate - into behaving themselves.

I can see that that might be an interesting argument to make (that responsibility to electorate is easier to protect at certain scales than others).

Composing said...

Zby: I'd call them all "middle-space" strategies, although clearly the details of the structure in each case are crucial.

Oli said...

Oops, OK Phil, so maybe now I'm starting to paint you as taking a straw man position :)

We seem to both agree in the need for various scale levels of government and some form of subsidiarity. We also no doubt agree that this 'fractal' structure of governance will always need improvement.

I guess an interesting argument is whether or not we should be moving towards more local or more global in general. Obviously each issue needs to be assessed on its own merits, but I guess we're talking here about general hunches of which way the 'traffic' should be going.

Maybe this is where we disagree.

I see a world where global communication, commerce and travel have shifted people's allegiances away from their local region towards more global concerns.

While there is a certain romance in the notion of the 'local' I think ultimately this shift towards caring about global issues is ultimately a good one.

I want to feel like a global citizen, not a little Englander. I want a fair playing field for everyone on the planet - and that means that I increasingly care about the larger scale governance issues a little more than the smaller scale issues.

For me the really important issues of our day are largely issues at the global level where the EU and the UK are the actors that actually have influence. For example: environmental issues and global trade issues (such as workers rights and intellectual property). These are not issues that can be meaningfully engaged with at an international level by Brighton and Hove council.

So, to me the question is not how to make these important issues be handled by the more local scale governance arrangements. No, to me the question is how to best ensure that the people of the world have an effective and meaningful way to be democratically involved in the larger scale governance structures that have to handle the global issues.

Sure it's a scandal that the UK politicians have been fiddling their expenses, but the real scandal is that the upcoming EU elections are such a non-event. The UK is drifting away from the EU rather than trying to improve the EU.

I'm not saying that we should get rid of local government. No, local councils, local police authorities and local health authorities all perform a vital role in providing accountable local government (which includes tackling the local dimensions of the global issues).

But, in general I think there are many important issues that have to be moved up the scale structure in order to be effectively handled.

I guess the question is: do you disagree, or are we just going around in circles here :)

John Powers said...

This discussion is really thought provoking. I have been mulling over Phil's: "I still think there's an active discussion to be had about the "shape" of middle-space." I don't think I've ever thought about the issue in terms of the shape of the space.

Tonight in a discussion about BALLE Networks (Business Alliance for Local Living Economies) I was musing about your question and linked to a brief description of Constructal Theory posted by Adrian Bejan.

There are hazards in applying mechanical engineering to social systems, nevertheless I was struck by this bit:

"The main principle of the constructal theory is that every system is destined to remain imperfect."

"According to this, the best that can be done is to optimally distribute the imperfections of the system, and this optimal distribution of imperfection will generate the geometry or shape of the studied system."

"The constructal way of distributing the system’s imperfection is to put the more resistive regime at the smallest scale of the system. The constructal law is the principle that generates the perfect form, which is the least imperfect form possible."