An Alternative Proposal for Primaries

In a previous post ("Election by Citizen Assembly") I proposed a different way of handling elections, in which a small group of citizens would elect the president after being given focused attention by candidates. The hope is that they would make a much more informed choice, and thereby choose a person who is in fact more representative of what the people want.

It has since come to my attention that James Fishkin and a few others have recently put forward a similar proposal under the name of  'deliberative polling'.

The proposal that I put forward was radical, and I can't see the United States ever adopting it. However, I suspect that some of its advantages can be secured with rather less radical changes.

Presently, primaries are conducted on a state by state basis, starting with Iowa and New Hampshire. Generally, opinion coalesces pretty quickly. One advantage of this is that it allows candidates to focus on presenting themselves to a smaller group of people. Candidates can tour Iowa and New Hampshire, and hope that an early victory there will bring them to greater national prominence.

To some extent, this system mitigates the effect of the media. The way that the news presents candidates can have a huge impact on their prospects. Even knowing who is in the lead and who is falling behind can have a significant effect.

Suppose that instead of starting with one or two states, we started with 5-10 small districts of about 30-300 people each conducted simultaneously. Collectively, the districts could be representative of the whole population. The candidates could be expected to talk to each district extensively, and each district could be compensated by the party for their time. The key would be to provide them with information that wasn't just through soundbites or political ads.

The purpose of these votes wouldn't be to settle on the nominee candidates. The states would then vote in their current fashion. Instead to provide a better baseline for the rest of the primaries. Instead of allowing the media or name recognition to settle the leading nominees, the informed opinion of a representative sample would.

I see two primary advantages to this. First, each voter would have significant power. No voter would need to feel like their vote for a marginal candidate was meaningless. If a voter likes a marginal candidate, their vote could push the candidate to greater prominence. Second, each voter would be much better informed. Their choice would be much more meaningful.

The results of such a system would have a strong impact on the remaining primaries. The results of the first few primaries can greatly upset the trajectory of the race. If everyone knew that the first primaries were both representative and comprised of citizens who were well informed, they might be more willing to defer to their judgment. At the least, it would provide a better baseline for the bulk of the primaries than the present system.