Merits and Demerits of Compulsory Voting
-
@clay how are jurors to be selected? In what sense are they representative—relative to what? That seems like a whole political system, where we are now essentially wondering what sort of body should elect our candidates. We could then ask, why don’t we also elect jurors?
Anyway, this should be a different topic.
-
@cfrank if i thought there was a way to summarize the manifesto without losing crucial details, i certainly would have done it. i spent weeks writing and rewriting it. indeed, just here in his reply, he demonstrated why the details are important, because he showed a major misunderstanding that reading the manifesto would have avoided.
how are jurors to be selected? In what sense are they representative—relative to what?
wow, maybe...read the manifesto that explains all of this? or put it into an LLM and ask your questions?
it's a random sample of the electorate. it's LITERALLY REPRESENTATIVE.
-
@clay I was aware of random sampling. That ignores district and state boundaries that currently exist.
-
@cfrank no it doesn't. You're doing a random sample of whatever "district" the election pertains to.
-
@clay ok, but for federal or state elections, how would that work? It seems to beg a question about representation for localities of different sizes with different local concerns.
-
@cfrank what are you talking about? a sample is a sample. the only thing you need to change is the jury size.
-
@clay sampling can be stratified. Most large scale political systems have negotiated systems of stratified representation, because population doesn’t always reflect importance of political concerns.
I feel this should be a separate discussion. But my point is, if you admit stratified representation as almost all political systems do, I would wonder why we should randomly sample jurors from fine-grained constituencies rather than letting those constituencies elect their jurors.
-
@cfrank stratification is an extremely bad substitute for statistically random sampling. maybe read the manifesto.
Random and Compulsory vs. Stratified and Voluntary
Many modern sortition implementations rely on voluntary participation with stratified sampling, which compromises true representativeness. This approach faces two significant problems:
First, stratification only accounts for characteristics that designers choose to measure and can measure with precision. Sexual orientation, for example, exists on a spectrum rather than as simple categories. Political affiliation is similarly complex—a "Republican" might be moderately conservative or deeply right-wing, while an "Independent" might be more conservative than many Republicans or more progressive than many Democrats. These labels become largely arbitrary, yet stratification treats them as discrete, meaningful categories. The East Belgian Citizens' Council and Michigan Redistricting Commission both illustrate these limitations of stratified selection.
Second, some stratified selection approaches explicitly mandate equal representation for different groups regardless of their actual population distribution. For instance, the Michigan Redistricting Commission requires four Democrats, four Republicans, and five independents, which distorts representation if the actual population has different proportions of these groups.
Election by Jury advocates for pure random selection with mandatory participation, achieving mathematically perfect representation without these distortions.
-
@clay I won’t inquire about your latest arguments here yet but only indicate that this is a forum for discussion and dialogue, so my preference is that we should try to keep our responses measured in volume and pace so that questions can be addressed. I will read your response and perhaps manifesto soon, but my first impression is that we should build these ideas communally and incrementally, one principle at a time.
I hope you know I am not dismissing or rejecting your ideas, but just engaging with them with immediately available bandwidth and the surface-level questions and concerns that come to mind under those constraints.
For example, I don’t agree that stratified representation is intrinsically bad. When implemented well, it can provide representation for minorities and other constituencies that are politically important and have unique functional and territorial needs despite small population, such as farmers or rural areas.
-
@clay “ which compromises true representativeness.” This is perhaps close to my point—representativeness of what, exactly? Individual people, yes. Stratified, complex political concerns of varying urgency or importance, I would say no.
You’re right that some strata are arbitrary. But what about strata that are not arbitrary, but are functionally, territorially or even historically meaningful?
-
@cfrank said in Merits and Demerits of Compulsory Voting:
I don’t agree that stratified representation is intrinsically bad.
but i proved it.
- statistically random sampling is already optimal.
- stratification is demonstrably vulnerable to manipulation/incompetence. there's no benefit it it, only downside.
-
@clay but “statistically random” does not define the constituencies from which the samples are being drawn. That is a political question, not a mathematical one.
As a thought experiment, suppose a jury itself were ideally representative in whatever sense is considered, and that they elected representatives. What stops those representatives from ultimately adjusting the manner in which future jurors are sampled? Especially, for example, if the question of stratification were raised as a political issue for jurors to consider, as it likely would be.
I think you should raise this as a separate topic here so we can all engage in a more organized way.