I don’t know shit about the theory of economics, or politics for that matter – but here goes…

I’m currently reading a book(!)[1] – Dr. Strangelove’s Game – that offers a history of the major figures in the development of the science of Economics.
One of these, the Marquis de Condorcet, was a minor French aristocrat and peer of Adam Smith, whose major contribution to the then-nebulous science of economics/statistics/calculus is known as the Condorcet method.

Marquis de Condorcet

The Condorcet method is an approach to elections that attempts to produce the result that is most representative of the wishes of the voting population as a whole. Condorcet developed it in response to the Voting paradox, which he also first identified.

In short, the paradox describes a situation in which the behaviour of individual voters can skew the results of an election away from the majority vote. The full detail of the paradox is worth reading, but there is only one aspect of it that is relevant here – so I will ignore the wider problem statement.

I will now proceed to hang myself with the just-enough rope provided by my tenuous grasp of what-the-fuck-this-dude-is-on-about.

Consider an election where three candidates are standing for election: Bush, Obama and Krazy Pete. And let’s let three classes of voters show a preference for these candidates as follows:

In this scenario there is no clear winner in a simple majority rule; all three candidates have one supporter in a one-choice-only election. And in a rank-your-favourites election the first two candidates draw level while Krazy Pete loses out. The Voting paradox, from here, goes into a lot of detail about how one of the voters could now cause the result to be skewed etc. but what I am interested in is this:

Condorcet’s paradox illustrates that the person who can reduce alternatives can essentially guide the election. For example, if [the Conservatives] and [the Moderates] choose their preferred candidates ([Bush] and [Obama] respectively), and if [the Liberals] were willing to drop their vote for [Krazy Pete] – who is going to lose anyway, then [the Liberals] can choose between either [Bush] or [Obama] – and become the agenda-setter.

To state this more simply, the voter who is willing to give up his first choice can effectively override the choices made by the other two voters.

This brings me to my point – Karl Rove.

The Thinker?

Karl Rove is generally credited with winning the 2000 and (especially) the 2004 elections for GW Bush. He did this by pursuing a relentless ‘divide and conquer’ strategy.
His approach was to slice up the demographic of the electorate into minute groupings based on issues that voters were willing to override their common sense for and then to target them aggresively to swing the vote.
For example, he would find a micro-segment of the electorate who generally preferred John Kerry, but who could be radically mobilized on the issue of the teaching of intelligent design in schools. He then targets them and pushes this one issue so hard that the targeted voters would willingly give up their general preference for John Kerry and vote *sigh* Bush.
If you can, for a given segment of the voting population, find an issue that would cause them to drop their first choice and settle for the alternative that you offer you can use this small segment of voters to override the decision made by the voting population as a whole[2].

It’s brilliant – and a powerful example of the weakness exposed in just about all electoral systems by Condorcet’s paradox.

Of course in the 2006 mid-term elections this strategy just wasn’t strong enough to overcome the horrible bungle that the Bush administration had made of the war in Iraq and Rove had his ass handed to him.

But this doesn’t mean that the tactic won’t work again in 2008 – the question is who will use it and against whom.

At the moment all the candidates[3] are, naturally, keeping to the high ground and not wanting to start slinging mud around. But I’m willing to bet that Giuliani wouldn’t mind finding the but-he’s-a-black-man or but-she’s-a-woman groups and riding their swing votes into the oval office.

*

As an aside, we don’t see these sorts of tactics used in South African politics – why?
Firstly because the South African political landscape is dominated by an overwhelming majority for the ANC – which means that small groups have no chance of influencing the outcome. And secondly because the tools of communicating political messages to small groups isn’t nearly as advanced as in the US where television ads on local stations can target audiences by city, by age group, by class(financial position) – radical.

[1] The first one that I’m not just picking up and dropping after a chapter or two this year.
[2] A similar, but reversed, situation occurred in 2000 when Al Gore couldn’t convince Ralph Nader’s supporters to drop their support for Nader in favour of himself based on their common loathing of Bush.
[3] Whether in primary elections or already posturing for the main event.