hi AdL, I don't think we'll ever know the 'real' answer without looking at the source code, but I've been thinking about this a lot recently, and here are my current theories:
Theory #1: ideology drives issues.
In my previous essays
, I argued that issues drive party affiliation (which party a POP votes for is based on issues, not ideology per se), and that issues do not drive ideology - ideology is not a 'summary' of a POPs issues.My current theory is that it's the other way around: ideology drives issues, but does so via a many-many mapping. This can't be as simple as a 1-1 mapping, as otherwise (a) in a two-party system, issues would become irrelevant: ideological liberals would always vote liberal, ideological conservatives would always vote conservative, and (b) it doesn't explain the 'reactionary example' - if 19% of a voting segment has reactionary ideology, why will only 12% of that same segment vote for the single reactionary party? There are other examples like this, but this is the canonical example.
Well, here's one way it could work:
Suppose that there are a 'set' of Reactionary issues, a 'set' of Conservative issues, and a 'set' of Liberal issues. A given issue may be in more than one set (for example, if you look at the 5 Mexican parties, all of them have 'Moralism' as one of their issues). Each party chooses a subset of the possible issues of its ideology, (note that no two parties have the same party platform, which shows there are more issues than ideologies). Furthermore, you might have a party which does a 'hybrid' - picks, either intentionally or because of random perturbations, some issues from another 'set' (for example, in Mexico the Conservative 'moderado liberal' party might be called this because it chose 50% 'conservative' issues, and 50% 'liberal', for its party platform). A POP then chooses the party that dot-products best with its ideologies values.
If it works this way, we get three nice behaviors:
1) most of the time, a voter will vote for a party of their ideology, as you would expect, but....
2) a voter might sometimes "cross the aisle" and vote for a party which isn't of their ideology, because their 'natural' parties choose a unlucky subsets, while the 'un-natural' party did the opposite.
For example, let's suppose that Reactionary voters really really really value 'jingoism'. Suppose that the particular Mexican reactionary party, which picks all reactionary issues, didn't happen to pick jingoism - it picked other, less juicy reactionary issues. Now suppose that another party, let's say the Liberal party, did pick 'jingoism', as an issue can span ideologies. This might cause the POP to vote Liberal, even with a Reactionary ideology.
3) There's a reason for the "ideology" pie chart
Theory #2: demographics drive ideology.
OK, if party is driven by issues, and issues are driven by idelogy, what drives ideology? Is it turtles all the way down?
I'm too lazy to do the statistical analysis required to validate this thesis, and too poor to bribe Paradox into releasing the source code
, but my intuition is that ideology is driven by demographic 'facts on the ground', such as:- profession
- militancy
- consciousness
- citizenship
Putting this all together, we get the following Unified Grand Theory
:- the economy drives profession
- profession, nationality, and militancy drive ideology
- ideology drives issues
- issues drive party affiliation
- party affiliation drives election results
- election results drive government policy
- government policy drives professions chosen by POPs
and the circle of life is complete - hakuna matata
