<div>Expressiveness considers voting as being a communicative act. How, then, should it be structured so the society's members best understand each others' preferences? Probably the most expressive ballot would be an unlimited, norm-referenced Range ballot - you could express any level of preference, with some presumably universal reference - "each point is a dollar", or "10 points of difference represents something you would die to acheive". However, unless you actually collect a Clarke tax, such a system would be overwhelmed by strategic effects. That's true even if you're not even dealing with an election, with practical effects on results, but just some kind of poll, whose entire purpose was for expression. By exaggerating my vote, I can effectively supress the expressive power of other voters.</div>
<div><br></div><div>However, in more traditional systems, there are generally few or no strategies to supress others' expressivity, if expressivity is the only goal. Consider Range voting: if, on a simple nonbinding poll, I "strategically exaggerate" to an approval-style response, that can't really be called a dishonest response, as I am simply making a choice to express only the most important distinction I see. Only when the vote is an election, not a poll, does dishonesty come into the picture, as utility conflicts with expressivity.</div>
<div><br></div><div>Thus, dishonest/strategic voting presents two separate problems. It may or may not reduce the overall utility of a voting system's result. (In fact, in some cases it may improve the result.) But, by weighting outcome over expressivity, it cannot help but reduce the effective expressivity of a system. This leads to a certain paradox: systems which seek to increase expressiveness by increasing voter freedom - for instance, Range as compared to a Condorcet system - could increase strategic opportunities, and thus in the end reduce expressiveness - for instance, if Range were to end up as pure Approval in practice.</div>
<div><br></div><div>In my conclusion (email 9 of this series), I'll come back to this paradox, and attempt to start resolving it.</div>