This is an excellent system; in fact, I can well suspect that this could be an ideal system. I suspect that it could find its first real-world trial, not in staffing a normal legislature, but in calling a single-purpose body such as a constitutional convention.<div>
<br></div><div>I suggest that you might want to avoid representatives who got exactly one vote before sortition. Electing such a representative would be too much of a fluke; it would be impossible to make sure beforehand that a vote was valid, for instance in questions of age or citizenship.</div>
<div><br></div><div>Thus, I'd suggest that a candidate should need to pre-register by obtaining some minimal number of nominating signatures. I'd set that number somewhere between 5 and 100 - small enough for any committed individual to get by themselves, but large enough to show that there is at least some level of commitment.</div>
<div><br></div><div>I also think that it would actually be productive to run the procedure at multiple levels, so that reduction in "represented base" was no more than 100:1 at each level. So my full procedure would be:</div>
<div><br></div><div>1. Nominations are submitted, you need 100 unique nominations to be valid.</div><div>2. Nomination validity is checked and announced.</div><div>3. A year after step 6, local representatives are chosen from the valid nominations through delegated sortition.</div>
<div>4. A year after step 3, state/regional representatives are chosen from the local ones through delegated sortition. Local representatives "promoted" to the next level could pass their local voting power to some representative not promoted. Cross-local delegation would be permitted.</div>
<div>5. A year after step 4, national representatives are chosen from the state/regional ones through delegated sortition, as in step 4.</div><div>6. A year after step 5, executive offices are filled through some deterministic single-winner system.</div>
<div><br></div><div>Thus, steps 3,4,5, and 6 would rotate in different years, so that the delegations in steps 4 and 5 would be informed by personal familiarity.</div><div><br></div><div>...Anyway, it's fun to dream. I think this would be a pretty ideal system, but I recognize that there are major constitutional hurdles to getting it adopted anywhere real any time soon. I'd love to hear more ideas for how to start using and prove this system without constitutional change; perhaps for intraparty matters?</div>
<div><br></div><div>Jameson</div><div><br><div class="gmail_quote">2012/1/24 Raph Frank <span dir="ltr"><<a href="mailto:raphfrk@gmail.com">raphfrk@gmail.com</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
You could also solve the narrow-preferences problem by having ranked<br>
ballots. If a candidate gets more than double the average number of<br>
votes, then any other votes received go to the next highest<br>
candidates.<br>
<br>
This might break strategy free-ness. You could have the candidate has<br>
a list of alternatives, who get any surplus votes.<br>
<br>
Under certain assumptions, the power of a voting bloc is equal to the<br>
square of the number of votes held, so, another option would be to<br>
give votes equal to the square root the number of vote received.<br>
<br>
Another issue is that the minority representation would change from<br>
election to election. Depending on how concentrated the voters were,<br>
some elections might distribute 1000 votes, but in others, they might<br>
only distribute 500.<br>
<br>
One thing is that 51% of the resulting legislature still has 100% of<br>
the power. Sharing power more evenly would be another improvement.<br>
<div class="HOEnZb"><div class="h5">----<br>
Election-Methods mailing list - see <a href="http://electorama.com/em" target="_blank">http://electorama.com/em</a> for list info<br>
</div></div></blockquote></div><br></div>