It’s party conference season, and one of the common stories of that period always used to be of the party leadership (and it didn’t matter which party) facing down the activists in their party. The ‘activists’, we would be informed, would want a policy way out of the mainstream while the leadership was being sensible and moderate. The reason I don’t specify a party there is because it’s a common story based on a common assumption: that the activists within a party are much more radical than the party leadership, and if the party wants to be successful (and appeal to the electorate, which is assumed to be moderate) the activists have to be faced down and/or defeated.

Now, there are two parts to that assumption. First, the difference between party leaderships and activists/members and second, the idea that a party being more moderate will get it more votes. This post is going to look at the just the first one and assume the second as given, but we’ll look at in more depth in a post another time (look out for me talking about spatial models and Downsian theories).

Curvilinear_DisparityThe issue we are talking about is known as curvilinear disparity, or to give it its full name, May’s Special Law Of Curvilinear Disparity. The diagram to the left gives a pictorial representation of it – leaderships and supporters are more moderate, and activists more radical, thus further from the centre. Why is this thought to be the case?

The main explanation is that party leaderships and activists are thought to have different goals and reasons for being involved in politics. The primary goal of leaderships is thought to be office-seeking while activists are said to be policy-seeking. That is, leaderships are more concerned with getting into power (and thus moving towards the centre to get them the votes to do that) while activists are concerned with issues of policy, and more concerned with ideological purity than moving to the centre. Meanwhile, ‘below’ this fight, the less active members and supporters are held to be in roughly the same position as the leadership. So, you can draw a curve from the leaders down to the members that swings out from the centre to represent the position – hence, curvilinear disparity.

So, political science, political journalism and party leaderships all agree on something, which means you won’t be surprised to find out that when researchers have actually looked in detail at party members and leaders and whether their attitudes differ they’ve found little or no evidence to support curvilinear disparity. Indeed in some cases, they’ve found that party leaderships and elites have had more radical ideas than members, who’ve tended to be more centrist.

So why does the idea persist? I’d give two reasons: first, it’s useful for party leadership to be able to send out signals that they’re standing up to the activists to be sensible and moderate. Whether they are or not, they want to send that signal out to the electorate as a whole to show that they’re positioned near the centre, and picking on the activists is a good way to do that.

Second, my personal theory is that there’s a perceptual bias at work. All political parties contain a range of opinions and it’s a rare party that can find a leadership that reflects all strands of opinion within a party. However, I would hold that a party leadership would be more likely to reflect the ‘moderate’ strand of opinion within the party because the ‘moderates’ are more likely to include a majority of the party membership. The ‘radicals’ are thus the members of the party least likely to be represented by the leadership and so are the most likely to complain and be visibly in opposition. However, they are not a majority of the membership, and neither is their position the average of the entire membership, rather it is the average of the non-represented membership – which almost by definition is unlikely to be a majority, as if it were, it could replace the leadership – but it is more visible, and this gives the impression of a ‘leadership vs activists’ battle.

I still need to work out that explanation a bit more, but it feels like a workable explanation for me, but the main point to take is that while many do believe that curvilinear disparity is real, the evidence collected thus far doesn’t support the case for it (for instance, see Pippa Norris in the first issue of Party Politics, if you have access to academic journals – sadly, most of the arguments and evidence about this is in journals). The battle of leaderships and activists is not a fundamentally existing part of party politics, no matter how much people tell you it is.