UK Constitutional Law Association

affiliated to the International Association of Constitutional Law

Alex Green: Why the EU Referendum Might Be Morally Binding – A Partial Response to Yossi Nehushtan

Alex GreenIn an erudite and sophisticated post on this blog, Dr Yossi Nehushtan argues that the United Kingdom’s recent referendum on membership in the European Union is not morally binding on the British government. Whilst there is plenty there that I agree with, in a number of ways I think his conclusion was too quick. In this brief reply, I provide what I take to be the strongest case for the pro tanto morally binding nature of the referendum.

The Nature of the Debate

On the 23rd of June 2016, I voted to remain because I believed that leaving would likely spell economic and political trouble for the UK, if not outright disaster. Hopefully I will be proven wrong. Whatever the truth might be, it is important for us to separate: a) the question of whether the referendum is pro tanto morally binding; from b) the distinct problem of whether or not leaving the EU is all-things-considered advisable.

Consider the moral nature of promising. All other things being equal, the fact that I promised is a good reason to consider myself bound to do what I said I would. However, other things are not always equal. A promise to meet my brother at a certain place and time may be overridden by more weighty considerations. Perhaps complying with my promise would prevent me from offering aid to an injured person I met along the way, or from doing my part to reunite a lost child with its parents. In these fictional cases my promise was pro tanto binding but outweighed by competing moral demands. I nonetheless have cause to regret breaking my word. The same is true of referendums.

Assume that it would be morally wrong for the British government to activate Article 50. My position is that the EU referendum should nonetheless be seen as presumptively binding for democratic reasons. Such reasons remain morally defeasible if the consequences of leaving would be particularly severe. Not being an economist or political scientist, I have nothing much to say about the likely fallout from activating Article 50. In any event, I take Nehushtan’s blog post to be directed at the pro tanto non-binding nature of the referendum, not the all-things-considered position on whether we should secede from the EU.

Referendums, Minimal Decisiveness and Equal Respect

Throughout his post and discussion in the comments, Nehushtan makes reference to ‘the people’, both in terms of their ‘will’ and their ‘consent’. But this misconstrues the moral case for direct democracy. Majority rule over the use of political power is not appropriate because it captures the unified desires of a political community. Rather, it is apposite because, in circumstances of disagreement, it treats each individual as morally equal by giving them an equal impact upon the ends to which power is put. Bruce Ackerman (Social Justice and the Liberal State (New Haven: Yale Univesity Press, 1980), 283) calls this our ‘minimal decisiveness’ and explains it as follows:

If, say, there are 99 people in an Assembly, then majority-rule gives me a decisive voice when the rest of you are split 49-49; and the same is true of your decision as well. When confronted with the prospect of a tied vote, the majoritarian does not appeal to some unresponsive decision procedure, but instead recognizes each citizen’s right to have his considered judgement determine the social outcome.

Majority rule is not always the all-things-considered best way to decide controversial matters. However, it is almost certainly the most respectful means of doing so. Specifically, when we disagree over the best course of action, minimal decisiveness allows us to reach decisions whilst respecting the fact that we disagree. When governments comply with majority rule by popular vote they treat each voter’s opinion as worthy of equal concern and respect. This provides the pro tanto case for the binding nature of referendums.

Like most modern states, the United Kingdom subjects its residents to coercive rule. Coercion is morally troublesome because it gives rulers dominance over the ruled, creating a relationship of inequality between them. But when it is employed in accordance with majority rule, coercion becomes a function of an egalitarian decision-making procedure, which, all other things being equal, dissolves its inegalitarian nature.

Since all residents of the United Kingdom are subject to coercion by its government, they are all presumptively entitled to minimal decisiveness over how political power is employed. True, it is impractical to always govern via direct democracy. What is more, any referendum is morally flawed (as this one was) where participation is restricted by something other than de facto residence. But equal concern nonetheless implies that majority rule is entitled to respect wherever it is employed in a largely inclusive fashion. That seems to have been the case here.

Now, coercive rule in the United Kingdom is clearly affected by whether or not it continues to comply with EU law and EU institutions. As such, any referendum on whether or not the United Kingdom should be subject to such things is one that relates to how its residents should be coerced. Where minimal decisiveness yields a clear majority view, equal concern requires the British government to accord it presumptive respect. The most obvious way to show such respect is to comply with the decision the referendum provides. With this groundwork in mind, let us turn to the detail of some of Nehushtan’s arguments.

The Narrow Margin

The leave vote won the day by somewhere in the region of 4%. Nehushtan claims that this casts doubt upon the outcome because ‘[w]eightier moral reasons should be provided for refusing to follow the will of a vast majority than refusing to follow the will of a marginal majority’. Given what we have seen, I am not sure how this can be maintained. The essence of minimal decisiveness is that respect is shown precisely when the outcome is marginal, so that the value of respecting disagreement is most poignant where disagreement looms large.

The Turnout

Next, Nehushtan contends that ‘the referendum’s results should be read as follows: 28% of all eligible voters could not care less. 35% voted ‘remain’ and 37% voted ‘leave’. It does not look like a decisive decision to leave, does it?’ Given the plurality of reasons that may have played a part in individual decisions to abstain, there is no reliable way to ascertain the preferences of that 28%. That aside, Nehushtan’s argument here is essentially the same as before, in that it appeals to the size of the majority. But if 37% want to leave, 35% do not and 28% don’t care, we have clear reasons of equal concern for considering the ‘leave’ vote to be decisive. On Nehushtan’s own assumption the 28% would accept either outcome, so the issue becomes whether to prefer the 37% or the 35%. If every vote counts equally, the solution should be clear.

The Subject-Matter of the Referendum

Nehushtan’s most promising argument is that it would be ‘anti-democratic to allow a marginal majority in an isolated point in time to prevent a future majority from having its preferences realised’. That argument is no doubt familiar to most readers of this blog as a justification for why a sovereign parliament cannot bind itself. I think we can accept it for now. Nonetheless, we can at least question whether once we leave there will be no turning back. I think it is probably true that we will not get back in on the same or as favourable terms, but that is not the point. Majorities make stupid decisions all the time and they nonetheless deserve respect where they are made in accordance with minimal decisiveness. Since up until relatively recently it was almost inconceivable that we would ever leave the EU, the notion that our exit would be ‘once and forever’ is uncertain. Politics is complex and unpredictable past the short to medium term. We will simply have to wait and see.

The Age Issue

Perhaps the most controversial position Nehushtan takes is to deny that each voter in the referendum should be entitled to equal concern and respect. Specifically, he claims that older voters should not enjoy as much political impact as younger citizens because they do not have to endure the outcome for as long. This objection has a certain appeal. Under normal circumstances, those living elsewhere in the world are not affected as much by British politics as residents of the United Kingdom. The reduced impact our politics has upon them is a plausible justification for their lack of participation in our decision-making processes. Why not also discriminate on the basis of age?

If I were feeling churlish, I would ask why we should stop at age if life expectancy is what really matters? According to recent studies of the United Kingdom, women tend to live longer than men and those in higher socio-economic classes tend to live longer than those in lower ones. Does that mean that poor male votes should count for less than rich female ones? No doubt there are ethnic and racial discriminations we could make on the same basis. Thankfully we need not go down this road because more charitable objections are available.

First, we need to look at Nehushtan’s claim that this referendum is a special case because ‘it affects numerous aspects of our lives, it has long-term and meaningful implications and it is practically irreversible’. We have already raised doubts about the practical irreversibility of leaving, so I will place that to one side. But is this referendum unique in affecting various aspects of our lives and having long-term and meaningful implications? Every major change in legislation, every significant court ruling and every impactful exercise of prerogative powers have these features to some extent. As long as individuals are subject to the coercive rule of state institutions they have an important interest in all such things. Can we truly make generalisations on the basis of age in this case and no other? When we voted on the alternative vote in 2011, was my vote more important than my mother’s because I would likely live to see more elections? Perhaps, but if that was true then, why does it not hold in relation to the periodic election of governments? Should my grandfather’s vote count for less than mine next time around because it is unlikely that he will live to see the next parliament? The expressive value of minimal decisiveness is such that thoughts like this should at least give us pause.

The second point is connected. Nehushtan’s objection on the basis of age focuses on the immediate economic consequences of leaving. For instance, he mentions the ‘relatively generous and protected pensions’ of the older members of society who voted out. However, by assuming that economics is what really matters, Nehushtan discounts the possibility that some older leave voters acted out of farsighted concerns for national identity, local community or the (un)democratic nature of the EU itself. To be clear, I am not saying that voting for these reasons would have been sensible. Indeed, I think that there is very little to be said for any such position. But that is not the point. In circumstances of political disagreement, why something should be done is just as much up for grabs as what that something should be. Preserving and safeguarding domestic democracy or national identity (however wrongheaded) is not something that obviously affects younger people more than older ones. In fact, they are transgenerational issues. The character of a political community is an ongoing project to which each generation contributes and with which each successive generation has to live. Taking a stand on principle for the common good in perpetuity is both a conceivable and reasonable position, even if the stand taken is wrong. Given that we do not know why all those who voted to leave did so, we owe them the benefit of the doubt.


There are several arguments Nehushtan makes that I have not examined here. The quality of the public debate was atrocious and he is quite correct to bring it up. Similarly, he is right to argue that having a referendum on EU membership in the first place was asking for political disaster. As Leah Trueblood rightly argues in her recent post on this blog, the whole process was carelessly designed and poorly conceived. Most likely, the fundamental nature of this contemplated constitutional change and the complexity of its likely consequences are such that it would have been safer in the hands of the experts that the British public has (apparently) outgrown.

Personally, I would not lose much sleep over a governmental refusal to activate Article 50. I just think that it is important for us to be clear about what such a refusal would entail, morally speaking. The government engaged a process that offered minimal decisiveness to each member of our voting public. Collectively, we made the wrong call. What has to be decided now is whether the danger of the likely consequences outweighs the respect that such a procedure merits. If the powers that be decide that this is so, the result may be for the better. Nonetheless, going against the outcome of the referendum would give us a moral reason for regret.

Mr Alex Green, Teaching Fellow, Faculty of Laws, University College London

(Suggested citation: A. Green, ‘Why the EU Referendum Might Be Morally Binding – A Partial Response to Yossi Nehushtan’, U.K. Const. L. Blog (14th Jul 2016) (available at

2 comments on “Alex Green: Why the EU Referendum Might Be Morally Binding – A Partial Response to Yossi Nehushtan

  1. Yossi
    July 15, 2016

    Thank you Alex for this brilliant response.

    A very quick response to the response.

    As to the margin and the turnout:
    In order to demonstrate the moral importance of the margin and the turnout, it is helpful to use extreme examples. Let’s assume two possible cases. Case A: 100% turnout and a majority of 90% for doing X. Case B: 10% turnout and a 51% majority for doing X. I can’t see how an argument that both results are equally morally binding can be defended. Precisely because the majority rule means treating “each voter’s opinion as worthy of equal concern and respect”, the strength of the moral reason to follow the will of majority depends on the margin and the turnout.
    If you think that the will of the majority should always be followed – no matter what – then ignoring the margin and the turnout makes sense. However, if you think that there can be cases in which Parliament/government should act against the will of the majority, and against its promise to treat the decision as morally binding, then (all other things being equal) ‘weightier moral reasons should be provided for refusing to follow the will of a vast majority than refusing to follow the will of a marginal majority’.

    As to my campaign against old people:
    I don’t think we should worry too much about having a slippery slope here.
    As you rightly said, my argument is limited to a non-legally binding referendum, and to a decision that ‘affects numerous aspects of our lives, has long-term and meaningful implications and is practically irreversible’. Even if the decision is reversible, as you assume, it’s a decision that is unlikely to be reconsidered and changed in the foreseen future. It is hard to think of similar cases – and I don’t think you provided any. If we find similar cases, my argument will apply in a similar way.
    Ant yet again, an extreme example may help. Think of a group of 10 people who are stuck on a desert island. They take a vote for either staying on the island or building a raft and trying to escape (and let’s assume that they can only survive if they stay in a group of at least 8 – so all agree to follow the vote’s result). 4 vote for ‘stay’. 6 vote for ‘leave’. Of these 6, two are terminally ill and their days are numbered. Should their votes count?

    The quality of the public debate:
    In my post I should have made it clear that this is the decisive, sufficient argument to accord very little weight to the referendum’s result. All the other arguments are important, but they merely strengthen an already decisive reason.
    This should be painfully clear and it is frustrating that (for now) both Parliament and government refuse to acknowledge that (and here I repeat my self): many voted for ‘leave’ because of promises that were given by the ‘Brexit’ campaign leaders. But if these promises will not and can’t be kept – then ‘leave’ does not really mean ‘leave’. It meant ‘leave’ under certain assumptions (more money for the NHS; staying in the single market but having our own policy about immigration, etc.) If we now know that these are false assumptions (as admitted by Brexit leaders!) then perceiving the referendum as binding means acting against the true will of the majority.
    The law says the referendum is merely advisory. Parliament, however, made an assurance that it will treat it as morally binding. The advisory body was misled and acted upon false assumptions. All agree to that. What moral principle compels anyone – including Parliament – to treat the advice of an ill-informed and misled majority as morally authoritative?
    A contract or a will that were made by a misled person could be invalidated. A final conviction in court which relied on false assumptions/evidence can be cancelled in a retrial. A referendum is different in some ways, but not that different.


  2. Alex Green
    July 19, 2016

    Hi Yossi,

    Thanks for the rejoinder! I will attempt to reply to the reply to the reply…or something like that.

    As I said in the post, I largely agree with you about the poorly informed votes. However, I think we can inject some doubt about two points you make in relation to them. First, although it is likely that some people voted ‘leave’ out of the expectation that, for example, £350 million a week would go into the NHS, we don’t know (at least as far as I am aware): a) how many leave voters were primarily motivated by statements like this; or b) how many that were would have voted to leave anyway. In light of our ignorance, it we lack an empirical basis for attributing collective intentions to such a large and diverse group of people.

    Second, I agree with you that radically poor public debate may rob a referendum of its pro tanto binding force, if it means that people en masse were genuinely not voting for what they thought they were voting for. However, the differences between this and the invalidity of contracts or wills are important. I assume here (if I am wrong, apologies) that in speaking of contracts and wills ‘made by a misled person’, what concerns you is deception and dishonesty. But misinformation and deception are not always morally equivalent. As far as I can tell the NHS funding statement was intentionally deceptive. However, I’m also willing to bet that quite a bit of the misinformation on both sides owed at least as much to the cognitive limitations of the debaters and the complexity of the issues. Whilst deception bites particularly hard in relation to legitimacy, I think it is at least questionable just how much misunderstanding there must be on my part before my view is no longer worthy of ‘recognition respect’ in political circumstances. I am tempted to think that the threshold must be very high because the issue is whether or not I am entitled to an equal say, not whether I am epistemically well placed to resolve the question.

    Furthermore, the binding force of a contract or will is primarily related to the importance of moral autonomy, whereas the argument I advance about minimal decisiveness and coercion is concerned with political equality. Take contracts. What makes our contract bind is that: a) we have elected to bind ourselves as an expression of our liberty; and b) in so doing, we have encouraged reliance by someone else. Now, it is arguable that the leave campaign made certain promises that bind at least for reason b). If that is so, then they commit a moral wrong by reneging upon them now. However, that is a separate question.

    The issue of whether or not my vote should be accorded equal weight to that of others under a system of majority rule invokes neither of these reasons, although it may in the final analysis be coherent with them. Instead, my vote should count equally because I am subject to coercion, which ‘prima facie’ places me in subordination. Granting me minimal decisiveness over how that coercion is employed at least partly dissolves this problem. As I said above, extreme misapprehension on my part, particularly if it results from deception, can negate this dissolution. However, the reasons at play are different from those that govern autonomy-centric institutions (like contract) and we should not move by direct analogy from one to the other.

    The above points are really by way of clarification. Now I want to respond to what you say about margin, turnout and age. First, I am not sure how your ‘extreme examples’ advance your argument so much as repeat your intuitions, which were quite clear (and attractive) in your original post. With this in mind, I think we have to examine our disagreement more closely. I think we differ over ‘why’ votes matter. Specifically, if I may be forgiven for attempting to repeat your own position back to you, I read you as claiming something like the following: a) a vote represents an individual preference; b) all individual preferences are equally valuable; c) satisfying as many preferences as possible is good; d) majority views are important because they represent many coinciding preferences; e) (therefore) larger majorities are more valuable than marginal ones. If I have mischaracterised your view, I apologise. However, if I am even roughly in the ballpark here, I want to be clear that I understand the moral basis of majority rule quite differently.

    On my approach, it is not about aggregating preferences to maximise satisfaction but resolving controversial issues in a way that is respectful ‘qua process’. In circumstances where we disagree about how coercion should be employed, some preferences will necessarily be frustrated if a conclusion is to be reached in the absence of unanimous consensus. This is always something to regret. Nonetheless, we need to make political decisions. This being so, the question becomes how we can treat potential ‘losers’ as respectfully as possible. Minimal decisiveness accomplishes this because it expresses that, even if someone’s preference was not satisfied, their view mattered as much as those who won the day. To put it another way, the satisfaction of preferences as such does not matter, it is the agents that hold those preferences who do. Respecting opinions is ‘just’ a way of expressing respect for moral equality. On this model there is no reason to respect larger majorities more than marginal ones because the need to demonstrate equal respect is important whenever we disagree, no matter how we are numerically divided.

    Based on this characterisation of our disagreement, I want to address two points you bring up when discussing margin and turnout. First, there is nothing that connects my view to the idea that majorities must be complied with ‘no matter what’. For one thing, it would be incoherent to comply with a majority out of equal respect where the majority in question opted to violate that same value. Imagine that a majority voted to institute slavery. We could hardly rely upon equality to justify enforcing that outcome without self-contradiction. Furthermore, as I say in the post, it remains possible that the importance of demonstrating equal respect can be ‘outweighed’ by other and more pressing moral concerns. For instance, imagine a majority that votes to do something consistent with equality but is nonetheless likely to result in some manner of cataclysm. In either situation we can consider majority rule to be presumptively binding but not all-things-considered authoritative. Second, on my view there is no need to say that some instances of majority rule are weightier than others. Minimal decisiveness as a manifestation of equal respect may have the same value wherever it exists and still have relative importance in relation to other things.

    As to your comments about age, you once more cite the uniqueness of the decision to leave the EU, this time as something that is unlikely to be changed in the foreseeable future. I suppose that it is clear why ‘impossible to change’ is morally distinct from ‘unlikely to be changed’, so I will leave that point alone. As to examples, I think I provided a great many, albeit not by name. Ordinary legislation, in many instances, causes changes in the law that present themselves as (if you’ll excuse the seeming paradox) ‘permanent for the time being’. So if by ‘unlikely to be changed in the foreseeable future’ we mean to say that, at time T, it is unknown how long it will be until some fundamental further change will occur, then I think this is true of many areas of law and policy beyond our membership of the European Union. At the risk of repeating an earlier intuition of my own, we all remember a time when it seemed inconceivable that we would ever leave the EU. There was plenty of commentary then to the effect that the sovereignty of parliament had been practically limited for the foreseeable future by our initial act of accession, even though we could theoretically secede.

    Finally, I turn to your desert island example. With respect, I think that the mistake here is exactly the same one as that which I pointed out in the post itself, when discussing your comments on the economic consequences of leaving the EU. You seem to assume that what matters here is how, whether and where these people live out some ‘average’ lifespan, so that those who will definitely not do not merit a say, or at least merit less of one. But I am not sure we can assume that. What if the two terminally ill individuals vote to leave because they want the chance to die with their families at home? That they will live shorter lives than the others is precisely why this potential desire is both conceivable and reasonable. In these circumstances, and even more so on political questions amongst much larger and diverse collectives, ‘why’ we choose to do something is just as much up for debate as ‘what’ we choose to do. Perhaps these two individuals will definitely die before the raft has any chance of successfully escaping. But even then my argument still applies. Maybe the two terminally ill people voted to leave because they value the chance of the four ‘normal leavers’ reaching home. Surely that is not such an incoherent position that it would not show disrespect to exclude them from the decision making process.

    Maybe this misconstrues your point, so let us consider a third possibility. Say the four ‘remainers’ voted for justifiable motivations, as did the four ‘normal leavers’. However, the two ‘doomed leavers’ voted for utterly capricious reasons. Why should they determine the outcome ‘pro tanto’? The answer is implicit in my above justification for majority rule. The content of the preference does not matter, what matters is that all the voters are recognised as morally equal. Since there is no default answer here (i.e. neither leaving nor remaining is obviously correct) allowing them to determine the outcome is the only respectful thing to do. Of course, if we alter the example so that leaving would be manifestly crazy then everything changes: it would be one of those circumstances where the four ‘remainers’ would be morally justified in resisting the majority decision because their interest in self-preservation would outweigh the importance of equal respect. But I have never argued that this cannot be so.

    I hope this is constructive!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Ordinary membership

UKCLA yearly membership (ordinary)


Student membership

UKCLA yearly membership (student)


Associate membership

UKCLA yearly membership (associate)


%d bloggers like this: