A more complex public opinion scale would be pretty straightforward to simulate. You could do it by storing an array of 4 floats or integers for each issue. Each number represents the percentage of people who think the current law is "liberal enough" at a certain tier of the issue. For example:
Free Speech
| C+ | C | M | L |
| 10 | 24 | 53 | 80 |
This means that:
10% want C+
14% want C
29% want M
27% want L
20% want L+
A simpler system would be to store numbers for the relative proportions in each bucket, then scale those values to percentages or absolute population numbers as needed. Comparable with the POP_RATIO system DF uses.
Edit: If the CCS springs out later, they should also spring out stronger. Maybe they don't need the big ramp up of varying strength levels, where they're relatively impotent early on and scarier later. I had them coming out earlier and weaker. But that just seems to make them easier to kill and less threatening for more experienced players. Perhaps making them really quite scary but an endgame feature would be better.
Some ideas for CCS etc:
I was thinking that as specific institutions became more liberal, conservative members of those organizations would either resign, be fired, or converted to liberalism, all leading to a higher percentage of 'liberal' entities within those institutions. A flow of disgruntled ex-members could form the nucleus of the CCS, which would emerge once it had sufficient members (Ex-police, ex-agents etc would be cool, but only in great numbers once police / intelligence laws are liberal).
This would lead to a natural growth of the CCS, tie their appearance and strength to the success of the LCS, and it would be different every time, depending on the strategy and targets you go for. And high ranked/skilled characters could form the 'chain of command' for the CCS to give them some more characterization, story line.
LCS operations against 'liberalized' institutions should have lower, and negative effects once those targets are at about L or L+. Like was said before busting prisoners from a L+ jail should be a negative PR effect, no effect at L, etc. Also for example, publishing documents from an L+ prison should only reveal to the public how well the prison is being run. This could be (very basically) simulated by not generating secret documents when raiding highly liberalized sites (and having out-of-date documents being published only making you looked bad, i.e "get with the times bro")
Also, CCS should do more things that the LCS can do. i.e. CCS should also employ more sleepers, covert ops, training etc. Snooping LCS sleepers should have a chance to detect CCS sleepers. Just thinking about how to implement this idea ... and I got this: once an enemy CCS sleeper is identified, then they appear on raids as a special encounter, based on the CEO kidnap code, an LCS agent can then assassinate, try and convert, kidnap or date them. Realistically sleepers shouldn't lead you straight to the safehouse, but give you lead you can follow up on (i.e. they reveal a location with increased chance to encounter real CCS members)
And how about a conservative paper appear somewhere once enough laws are liberal? the "Conservative Clarion" would be a good name, to rival the Liberal Guardian. Maybe as a right-wing student paper at the Uni, like the RL Dartmouth Review.