Instructions for dancing (1)

I don’t think the Appeal Court’s ruling last week – on whether Labour’s NEC had the power to set a retroactive ‘freeze date’ for eligibility to vote in the leadership election, disenfranchising some 130,000 people who are otherwise members in good standing – was wrong in law, as Jeremy Corbyn suggested. But, by the same token, I don’t think we can say it’s definitely wrong to say it was wrong in law. (Bear with me.) There is no law that can only ever be read one way, no case that could only ever have been decided one way. The fact that the Appeal Court reversed the previous week’s court ruling isn’t a demonstration of corruption or incompetence, but one example of a perfectly normal phenomenon in law: one reading of a legal question being superseded by another reading.

This isn’t to say that judges are free to decide cases, and interpret statute, any way that they please; on the contrary, legal rulings – particularly at Appeal Court level – need to be, and are, justified by closely-reasoned argument. When one court’s decision gets reversed by another, it’s very rarely a matter of Judge B announcing “Judge A was wrong, I’m right”. Rather, the higher court examines the argument in which the first judge’s ruling is embedded and puts forward a ruling grounded in a better argument – better in the sense of greater logical coherence or comprehensiveness, greater appropriateness to the situation at hand, better fit to statute and existing precedent, lesser probability of creating problems in future cases, and so on. Sometimes the greater appropriateness/coherence/etc of the higher court’s ruling is glaring and unarguable; sometimes it’s more debatable, and in these cases the original ruling may eventually be reinstated – either through appeal to a yet higher court or, in the longer term, by the precedent set by the appeal being distinguished (i.e. disregarded) so consistently that it falls into disuse.

The key point here is that the question “is this ruling correct?” both does and doesn’t have an answer. An Appeal Court ruling gives a definitive statement of how the law should be interpreted, together with supporting arguments; the Appeal Court ruling is the law (unless it’s reversed by the Supreme Court), and the answer to the question of whether it’s correct has to be Yes (unless the AC has really screwed up). At the same time, the arguments supporting the AC’s judgment give one particular reading of the body of materials which the court had to work with, together with reasons for adopting that reading. It’s possible for a reasonable person to hold that, although the AC has ruled that reading A applies and consequently the law is X, it would have been preferable for the AC to choose reading B, in which case the law would now have been Y. And, as I’ve noted, it’s even possible for the law-making power of that particular ruling to be, in effect, eroded over time, if the judgments of future Appeal Court hearings concur in preferring reading B and law Y.

All this is by way of saying that the Appeal Court judgment in the Labour Party case (Evangelou v McNicol) doesn’t (in my view) correct anything unproblematically identifiable as an error in the original ruling; what it does is propose a different reading of Labour Party rules and – more importantly – a different way of reading the rules. The key passages are in clause 4.II of the rules, headed Procedural rules for elections for national officers of the Party, and specifically sub-clauses 4.II.1.A and 4.II.2.C.vii (!), which respectively read as follows:

The following procedures provide a rules framework which, unless varied by the consent of the NEC, shall be followed when conducting elections for Party officers. The NEC will also issue procedural guidelines on nominations, timetable, codes of conduct for candidates and other matters relating to the conduct of these elections.

and

The precise eligibility criteria shall be defined by the National Executive Committee and set out in procedural guidelines and in each annual report to conference.

A submission from the NEC to the appeal drew attention to 4.II.1.A and the NEC’s power to ‘vary’ whatever is written down in the rules. The claimants objected on the grounds, roughly speaking, that this had not been brought forward before, and their case might have been different if it had. The AC agreed, but with one significant qualification:

In our view, the only relevance of Chapter 4, clause II(1) is an aid to the construction of other powers and requirements in the Rule Book, which has to be construed as a whole.

This brings us to the two different ways of reading the rules which were put forward in the earlier ruling and the AC ruling. The difference hinges on how much importance is given to what’s not in the text – shared assumptions, common knowledge, established practice and so on. One approach – what you might call a purposive approach – would start from common knowledge about what the rules are for and what kind of association the Labour Party is, and skate generously over lacunae in the text. So, we know that the party is a democratic organisation which elects its leaders, and we know that the party’s stated policy is to engage all members in activity and participation in the party’s structures; does it matter that the rules don’t explicitly say that all members get a vote in leadership elections? Similarly, we know that there needs to be a date beyond which new members can’t join the party and expect to get a vote – even if you give a vote to members who join on the day of the election, you need to specify that – and we know that it’s highly unusual, based on past practice, to set a ‘freeze date’ as much as six months in the past: does it matter that the rules don’t explicitly say that freeze dates shouldn’t be set six months in the past?

A purposive approach would say ‘no’ to both of these questions; interpreting the rules, on this approach, is partly a matter of filling in the blanks by referring to the purposes of the rules, the purpose of the organisation and the way things generally work in practice. This is, broadly speaking, the line taken by the original court ruling. It’s worth saying, incidentally, that although a purposive approach in this case favoured the claimants (and by extension Corbyn), there’s nothing inherently radical about taking an approach like this; it could equally well be argued that the purpose of marriage is to support procreation, or that the purpose of trade unions is to promote industrial harmony, leading potentially to highly conservative readings of the relevant laws.

The alternative approach forswears any of this assuming and skating-over; sticking with the letter of the text, it arrives – where the text allows – at results which are clear, definite and hard to challenge. The Appeal Court took a textualist approach in its ruling, albeit a modified, and arguably incoherent, textualist approach (I’ll come back to this). Where the freeze date is concerned, the textualist approach can close the case by asking and answering two questions: Yes, the rules do provide for the imposition of a freeze date; No, the rules don’t state that a freeze date cannot be six months in the past; The End. On the question of whether all members should presumptively get a vote, the AC is unyielding: the rules don’t say that anywhere, but they do say that the NEC has the responsibility of defining “precise eligibility criteria”. A purposive reading would lean heavily on that word ‘precise’ – doesn’t that imply that broad eligibility criteria already exist and are known, even if they aren’t necessarily written down? The textualist reading – and the AC – says that the use of an adjective to qualify X, when X is named, isn’t nearly enough evidence for inferring the existence of a broader, unnamed form of X; that would be like saying that a shop sign advertising ‘high class menswear’ tells you that you can get lower-grade clothing further down the road.

So is this an open and shut case? Not quite. Remember that legal arguments are justified in part on the basis of their potential for creating clear and appropriate precedents; also, remember that I argued that the AC took a modified textualist approach. The key point here is the use they made of that clause 4.II.1. Discussing the rule book at the time of the challenge to Corbyn’s automatic inclusion on the ballot, Carl Gardner drew attention to this clause, pointing out that it effectively frees the NEC from the restraints laid down by its own rules: “The NEC could vary the procedure however it liked, so long as it was reasonable.” This is the ‘power to vary’ which the NEC brought forward – rather late in the day – in this case. There is – potentially at least – an argument here about whether this power to vary the rules simply gives the NEC the power to decide whatever it wishes and ignore the rules, or whether it only empowers the NEC to vary the rules in an individual case having stated that it is doing so; the former reading would be so broad as to make the rules meaningless, but the latter wouldn’t cover anything the NEC has done in this case.

But the point is moot; as we’ve also seen, the AC ruled that this clause would not form part of its decision – except insofar as it was “an aid to the construction of other powers and requirements in the Rule Book”. Now, that’s quite a big ‘except’. What it says is that the entire Rule Book is read from the starting point that the party has both rules and a rule-making body, and the rule-making body can legitimately step in any time the rules need amending or seem to be giving the wrong result. The gaps in the rules as written – gaps which any textualist reading will inevitably find – are plugged by reading the explicit power to vary as conferring an implicit, general power to vary, as and when necessary. In effect, it’s a textualist approach within an overriding purposive approach, and as such arguably incoherent – after all, do the rules say that 4.II.1 is an aid to the construction of other powers and requirements, or is it just one sub-clause among others? A thorough-going textualist approach would surely choose the latter.

The result, in the words of Corbyn’s campaign, is as “a ‘make it up as you go along’ rule”; I wouldn’t go that far, but this reading would certainly make it very hard to win any case concerning the rules against the NEC. Some will welcome this ruling for precisely that reason – the courts shouldn’t be getting involved in the internal workings of political parties; the NEC is an internal party body, and anything that makes it less likely that members will take it into their heads to drag it through the courts is to be welcomed. But I think they should be careful what they wish for. If the NEC is the rule-making body, and if the rule-making body has the power to vary the rules, what limits are there on the power of the NEC? The AC’s ruling addresses this question in terms of the discretion of decision-making bodies and the limits to such discretion. In the words of a 2008 case (Socimer):

a decision-maker’s discretion will be limited, as a matter of necessary implication, by concepts of honesty, good faith, and genuineness, and the need for the absence of arbitrariness, capriciousness, perversity and irrationality

Wednesbury [un]reasonableness is also invoked, if anyone was worried it wouldn’t get a look-in.

Now, these are very broad limits. Restrictions on eligibility to vote are arbitrary if there’s no good reason for the line to be drawn in one place rather than another; they’re capricious if the line is drawn chaotically or at random; they’re perverse if their disadvantages outweigh their benefits; they’re irrational if they have no rational justification; and they’re Wednesbury unreasonable if they are so unreasonable that no reasonable person acting reasonably could have chosen them. But that still leaves plenty of scope. As we can see, a six-month retrospective freeze date doesn’t qualify under any of these headings; how about twelve months? Or how about defining eligibility to vote in terms of attendance at party meetings? contributions to party funds? membership of an approved party organisation (e.g. Momentum)?

I think it’s a very problematic ruling, in short, and one which – given a Left-dominated NEC – may well come back to bite the very people who are now celebrating it. If they are celebrating it.

Next: what was going to be parts 2 and 3 of this post. Let’s face it, this is quite long enough as it is.

Advertisement