Since it's election time
A recent conversation brought to mind this post Bill made to the BARGE list during the 2004 election:
As I have said before, a two-party system perpetuates this type of equilibrium. If one party gains a clear majority (as it did under Reagan during the Gingrich era or arguably during some of the Clinton years), not only does the other party try to be more inclusive, but the party in power may narrow it's agenda because its what most of the party wants and it no longer has to cater as much to the middle to win. Some moderates "defect" to the other party, and the cycle continues. In fact in the extreme case where one party happens to have a extreme majority for a long time, it's almost a guarantee that there will be a schism in that party and a realignment such as the protestants in the former majority party joining the tattered remains of the minority party.
Of course you can look at each voter as a point in an N-dimensional space, where each axis represents an issue or viewpoint. How far you are on each axis is the product of how strong your opinion is on the issue and how important the issue is to you. Some axes may even be wealth or geographical location.
For example, on the extremes there are people who believe there should be no government social programs at all and others who embrace socialism completely, but one may also have either view but feel that say foriegn policy is much more important, see Ed Koch for instance who after campaigning for Bush immediatly endorsed Hillary Clinton the day after the election. On the other hand the Amish here in Pennsylvania completely oppose Bush's actions in Iraq, but voted for him by a wide margin (with paper ballots of course).
Now suppose we paint all the Kerry voters blue and Bush voters red. As a first approximation, one can find a best-fit plane separating the two types of dots. That is a plane where on the "left" side, the vast majority of the people are blue and on the "right" side the vast majority are red. We can find a plane which maximizes the percentages of blue and red on both sides. Now imagine a line running perpendicular (or normal) to this plane. The direction of this line represents the axis of issues normally associated with each party. For example the "left" direction will be positive on the pro-choice axis.
We can do a better job than a plane probably though, we can fit a more complex surface that better separates the blue and the red though. Also it's not a static situation--new voters are appear all the time, and in most districts now some voters die off. Also, individual voters move in the n-dimensional space. Simply moving towards the middle on any issue may not net more voters. For example, a more "moderate" stance on abortion and same sex marriages but still to the right of Kerry's may lose voters like the Amish for Bush without corresponding gains.
I love the "Of course" in the second paragraph that introduces the N-dimensional topological discussion. (But it's actually not that hard to understand, and pretty neat.)
As I have said before, a two-party system perpetuates this type of equilibrium. If one party gains a clear majority (as it did under Reagan during the Gingrich era or arguably during some of the Clinton years), not only does the other party try to be more inclusive, but the party in power may narrow it's agenda because its what most of the party wants and it no longer has to cater as much to the middle to win. Some moderates "defect" to the other party, and the cycle continues. In fact in the extreme case where one party happens to have a extreme majority for a long time, it's almost a guarantee that there will be a schism in that party and a realignment such as the protestants in the former majority party joining the tattered remains of the minority party.
Of course you can look at each voter as a point in an N-dimensional space, where each axis represents an issue or viewpoint. How far you are on each axis is the product of how strong your opinion is on the issue and how important the issue is to you. Some axes may even be wealth or geographical location.
For example, on the extremes there are people who believe there should be no government social programs at all and others who embrace socialism completely, but one may also have either view but feel that say foriegn policy is much more important, see Ed Koch for instance who after campaigning for Bush immediatly endorsed Hillary Clinton the day after the election. On the other hand the Amish here in Pennsylvania completely oppose Bush's actions in Iraq, but voted for him by a wide margin (with paper ballots of course).
Now suppose we paint all the Kerry voters blue and Bush voters red. As a first approximation, one can find a best-fit plane separating the two types of dots. That is a plane where on the "left" side, the vast majority of the people are blue and on the "right" side the vast majority are red. We can find a plane which maximizes the percentages of blue and red on both sides. Now imagine a line running perpendicular (or normal) to this plane. The direction of this line represents the axis of issues normally associated with each party. For example the "left" direction will be positive on the pro-choice axis.
We can do a better job than a plane probably though, we can fit a more complex surface that better separates the blue and the red though. Also it's not a static situation--new voters are appear all the time, and in most districts now some voters die off. Also, individual voters move in the n-dimensional space. Simply moving towards the middle on any issue may not net more voters. For example, a more "moderate" stance on abortion and same sex marriages but still to the right of Kerry's may lose voters like the Amish for Bush without corresponding gains.
I love the "Of course" in the second paragraph that introduces the N-dimensional topological discussion. (But it's actually not that hard to understand, and pretty neat.)