Naturally certain people care about some issues more than others. If your democracy works by electing representatives based on their positions on various issues this means that for most voters not all the positions taken by a candidate will matter equally. At the extreme some voters care about one issue to the exclusion of all others, which means that the candidate they vote for is determined solely by their position on a single issue, assuming the candidates take different positions on the issue. Such voters are called single-issue voters, and they exist in great abundance (a fair number, for example, can be found among those who are anti-gay marriage).
The existence of single-issue voters is perhaps unavoidable, but it has disturbing implications for democracy. Let me illustrate with an example. Suppose that 10% of the population is single-issue voters who want all stores to be closed on Friday (for religious reasons, let us suppose). Additionally in the upcoming election there are two candidates, each of whom supports a different range of policies. One of them is focused on economic expansion while the other is focused on improving the general quality of life. Between these two candidates the population is divided 50-50, with voters unlikely to be switch from one to the other. Both candidates of course want to win, but neither can be assured of victory as things stand. Closing stores on Friday doesn’t seem like a big deal, so one of the candidates may choose to add that to their platform, as a sacrifice they are willing to make in order to win. This would immediately give them victory, with a 10% margin, due to the single-issue voters who now will vote for them (we assume they are divided evenly among the candidates). Naturally their opponent will respond by also promising to close stores on Friday in order to restore the balance. And at this point no matter which of the two candidates wins then the single-issue voters win as well. And there is something wrong with that, since here we have a minority exercising control over the majority, the very thing democracy was supposed to prevent.
Of course in real life things are more complex. There are single-issue voters on both sides, and often the majority of single-issue voters are already leaning towards one candidate or the other. But still, it is a flaw in the system, a flaw that arises because issues and candidates come as a package deal (and thus a flaw which the alternative form of democracy I discussed yesterday does not suffer from). But is it really a flaw? Perhaps because the single-issue voters care so much it really is fair to let them have their way, despite the fact that they are a minority. Some might say that closing stores on Friday really is the democratic solution. But suppose that 30% of the population organized and decided to declare themselves a new aristocracy. If they were sufficiently motivated they could agree only to vote for candidates who promised them some special benefits (thus turning themselves into single-issue voters). Obviously no candidate would ever give them too much at once for fear of alienating the other voters, but over time the advantages would accrue until they really were treated like aristocracy. And obviously this isn’t right, because democracies were created expressly to get rid of privileged classes of people. And essentially that is what a group of single-issue voters is, a privileged group of people whose opinion counts for more.
Of course pointing out problems with democracy is a bit like kicking someone when they are down (Arrow’s impossibility theorem anyone?). But I think the existence of single-issue voters also show that there are real problems with a cousin of democracy, utilitarianism. The existence of utility monsters, people who become extremely unhappy when they don’t get their way (such that their unhappiness outweighs the happiness of everyone else) are counterexamples to utilitarianism. A world that contains utility monsters is a world in which it is wrong to try to maximize total happiness. One utilitarian response to utility monsters is to deny their possibility or, if their possibility cannot be denied, claim that since there aren’t real utility monsters then utilitarianism is still a good ethical system. But in single-issue voters we have an example of real utility monsters. Of course the happiness or unhappiness of a lone single-issue voter doesn’t outweigh the happiness of everyone else. But, as a group, the opinions of the single-issue voters matter more than the majority opinion (and thus the opinions of the other voters), because they care so much about their issue, just as the wishes of the classic individual utility monster outweigh the wishes of everyone else because they care so much about everything. And, as demonstrated, the ability of a minority to outweigh the majority just because they care more is not something we find acceptable. And so here we have a real life counterexample to utilitarianism.