There has been a lot of discussion recently about constitutional reform in New Zealand, and how it should occur. But a lot of the conversations I have been involved in have actually raised a slightly different question – when should a country like New Zealand engage in constitutional reform?

I think this second issue raises two distinct but related questions that, as a nation, we don’t have a very good collective grip on. The first is, what is the standard we should be trying to reach (or maintain) when it comes to having a “good” constitution?

Finding an answer to this question is actually not as easy as it might first appear. Constitutions are what I would call “ideal regarding” institutions – they envisage and in many cases promote an ideal standard for law and government. But failure to meet that ideal standard in practice is not by itself evidence of a constitutional failing. Real-world constitutions are human institutions, and as a result they are necessarily imperfect. Every single constitution will fail to meet the ideal standards it sets for itself at least some of the time.

So the standard we should be striving for is not an ideal standard of perfection. That would simply be impossible. We need a more realistic measure of constitutional performance. There isn’t agreement on what that more realistic measure should look like, but I think the standard ought to be something like “achieving good constitutional outcomes at least most of the time”. This standard allows some leeway for outcomes which are not great from a constitutional perspective without treating those exceptions as compromising what is otherwise a robust and principled system of checks and balances.

But if we accept this real-world standard for judging constitutions, we might seem bound to accept at least some less than desirable outcomes. Should we really tolerate breaches of human rights, or abrogation of Treaty principles, simply because perfection isn’t possible? This is a valid concern, and it leads into the second question I want to address – what evidence should we accept as demonstrating a need for constitutional reform? How do we know when a bad constitutional outcome is just the price for having a constitution in the first place, rather than a sign that reform is needed?

Again there aren’t easy answers to this question. But it’s important to remember that constitutions aren’t immutable – they evolve all the time in small and subtle ways in response to changing circumstances and new challenges (this, by the way, is true of both written and unwritten constitutions). So where bad (or at least debatable) outcomes are reached, the constitution can right itself to some extent.

Think of it like the body’s immune system. If you have a cold, your body will respond to restore you to reasonably good health (after a time to adjust). That’s sort of how a constitution works too. If there are symptoms that something is wrong, various parts of the system will respond and come out the other side more robust than before.

In these examples, we probably don’t need to adopt a conscious reform effort. Things should take care of themselves. But if the symptoms persist, and we get one bad outcome after another, it’s likely that something more serious is wrong. The constitution itself might need to be fixed.

So the upshot is that some problems may be worse than others, in a constitutional sense. And we don’t need to shout “the sky is falling” each time something goes wrong. How do we know when we should panic? It takes an understanding of the constitutional system as a whole and its ability to right itself. And there is room for disagreement on this. One of the reasons I am not overly concerned about the Taylor line of cases, which deal with the constitutional implications of prisoners having the right to vote, is that I think the constitutional system is addressing those concerns in a robust way and over time our system will improve. I don’t hold the view that the fact that the rights of prisoners being violated – as they almost certainly have – is not a big deal. But I do think it is within the scope of the system as it current works to address this problem, and address it well, over time.

I know some informed commentators will disagree with me on this. That’s great, because debate on where the system does and doesn’t work helps us figure out where the truth of the matter lies. Maybe over time I will be shown to be wrong, and systematic constitutional failings will be identified. But in the context of an imperfect constitutional system we can’t just assume that there is a fundamental flaw because there is a less than desirable outcome. We need to figure out why that less than desirable outcome occurred, and whether or not it is something we can, and should, seek to fix.