Bob catches his wife cheating on him. He murders her right then, in heat of the moment. This is of course unjust.

But suppose he instead plots revenge over the course of a year, and then murders her later. Bob would likely be charged with first degree murder instead. He’d be charged with a more severe crime and receive a more severe sentence. Committing injustice in the heat of the moment is bad; committing the same injustice after cool deliberation is usually worse.

Everyone accepts that. But when you read democratic theory, you tend to find the opposite intuition. If democratic leaders do unjust act X in the heat of the moment, that’s bad. But if we have a lengthy democratic deliberative process which then decides to X, that’s not as bad. Indeed, that might confer justification upon X, or so most democratic theorists say.

But on it’s face, that’s odd. Suppose Chancellor Merkel decides to hold a surprise referendum tomorrow on the question, should we carpet bomb Tuvalu? Germans get no time to think about it, but they end up voting in favor, say because they are mistaken about just war theory, or mistaken about the facts relevant to applying just war theory to the case of Tuvalu. The bombs fly. This would be terribly evil. (See this related post about American support for bombing the fictional city of Agrabah.)

But now suppose they instead spend a year deliberating about whether to bomb Tuvalu. There is full, fair, and equal participation. Finally, they vote, and they decide to bomb Tuvalu. Again, in doing so, suppose they have either misapplied just war theory, or they have gotten the relevant facts wrong. But they made these mistakes during a process of careful deliberation. In this case, many democratic theorists would have to say that the decision to bomb Tuvalu is less unjust; some would even have to say the decision is legitimate and justified, if regrettable. But that seems odd. Shouldn’t we instead say that the Germans in the second example are even worse, much as pre-mediated, cold murder is worse than passionate murder in the heat in the moment?

Now, perhaps it depends on what actual happens during deliberation.

It wouldn’t be enough that deliberation is equal and fair, or that it follows Habermas’s rules of deliberation. Those norms ensure equality, but they don’t specify anything about the actual epistemic quality of the information people use, or of the epistemic reliability of the thought processes they use to assess that information. If, e.g., in a jury trial, every juror gets an equal say, but they all decide to find they defendant guilty because they believe the defendant is a reptilian, we wouldn’t say that was good deliberation. We’d say the jurors are idiots and the decision is unjust.

If we knew that over the course of the year, Germans A) scientifically reasoned using B) the best available evidence and C) the best just war theory, and then concluded they should bomb Tuvalu, perhaps that would be better than just bombing Tuvalu after a rash decision. But for that to hold, we’d have to imagine that the Germans had reliably obtained misleading information. We’d have to imagine they reasoned in a good way, but made an innocent mistake.

Sometimes, that sort of thing happens. Sometimes the best available evidence points to X, but the evidence is misleading. So, at least in principle, you can imagine a situation where Germans would be epistemically justified in believing they should bomb Tuvalu, even though they should not–where they would have justified but false beliefs. In that case, the Germans would be “innocent aggressors” in just war theory. (If you’re having a hard time with that, imagine a sci-fi scenario, where the Germans justifiably but mistakenly believe that Tuvalu has been infected with the zombie apocalypse virus and the only way to stop the spread is to level the island. Imagine the equivalent of a court case where the evidence strongly implicates that the defendant is guilty, but the defendant is in fact innocent.)

So, in principle, sure, one can imagine democratic deliberation could reduce the injustice of a bad decision. But this does little to justify or excuse actual pre-mediated democratic decisions. Most voters are ignorant and misinformed, and use unreliable thought processes to  arrive at what few political beliefs they have. Voters support parties largely for non-cognitive reasons. More opinionated voters generally post-hoc rationalize that they agree with their party’s platform. Politicians seek re-election and so cater to voters’ biases, especially near election years. Politicians tend to do what will look good in the short term rather than what’s actually good in the long term. Self-interested lobbying corrupts the process after the election. And so on.

Actual democratic “deliberation,” if we even want to call it that, doesn’t look like an example of careful, reliable, scientific deliberation. So, arguing, as many democratic theorist do, for the principle “If democracy engages in a smart, reliable deliberative process but then makes an innocent mistake, it would be justified in acting on that mistake” does little to justify the decisions actual democracies make. Real-life bad democratic decisions are usually more like the pre-mediated murder cases.

In the real world, if your city council decides spur of the moment to impose rent control, that’s bad. If they take a year to make such a stupid decisions, that’s usually even worse.