Author’s note: The format for this blog is changing. It will contain personal thoughts and considerations around the research I am conducting as part of my PhD into self-determination & secession (independence) from peaceful, democratic states. I will include links where possible to open-source documents that highlight the issues raised. However, the content here is to be considered as ‘op-ed’, that is commentary by the author based on his opinion and not to be considered ‘peer-reviewed’ material. Please bear this in mind while reading and commenting.
The appeal of a referendum in democratic states is self-evident. The average citizen of a democratic state thinks of their political system as existing to serve their interests as a collective, and for their representatives to heed the instructions of the population when deciding on policy. A referendum on a specific policy, on its face, presents an opportunity for the population to unambiguously and directly instruct politicians as to the wishes of population on that given subject.
However, when one considers both the content and process of conducting a referendum, certain obstacles arise that make the certainty associated with the outcome troublesome. There are many reasons for this, but I will focus on three main points:
- Referendum outcomes are typically misrepresented by politicians as evidence of a consensus amongst all participants in the exercise, essentially leading to political positions that only represent the majority of those that voted in the referendum.
- The real world rarely comes down to yes or no. Although the ballot may contain a straightforward question with (typically) two possible responses, the reality of a given policy position may not lend itself to such a neatly bifurcated solution.
- If the voter in a referendum lacks access to evidence-based, verifiable facts about the implications of the potential outcomes, it fundamentally undermines the legitimacy of those outcomes to be put into practice.
The Illusion of Consensus
It must first be acknowledged that no public vote is ever going to result in 100% of the population voting a particular way. In political elections, we are rightfully skeptical when any politician secures higher than 80% of the vote. In referendums, it is even rarer to see such ubiquity amongst the franchise.
There are exceptions: In March 2013, a referendum was held regarding the Falkland Islands where the people of that territory were asked if they wished to remain an Overseas Territory of the United Kingdom. The turnout was 91.94% of the eligible population, and the result was 99.8% in favour of retaining this political status, with only 3 votes against. International observers from six other countries certified the referendum as “free and fair, reflecting the democratic will of eligible Falkland Island voters“. (emphasis added)
However, most referendum results are less decisive, often ranging between as much as a 30% gap to as little as 0.1%. There are a myriad of factors that contribute to this, not the least of which is the content of the question itself. After all, if public consensus was so clear, there would not be a reason to hold a referendum in the first place. As such, most referendums expose political divisions amongst the population of a country, and they result in change that a constitutionally significant number of citizens rejected.
This wouldn’t typically be an issue if normal political means were employed in bringing about change. Here in the UK, legislation is subject to multiple levels of scrutiny and analysis, and often compromises are made in order to generate support for any given policy. In a referendum, by contrast, the outcome is presented as a concrete ‘mandate’ for a specific policy, and the concept of reaching a compromise to unify the electorate has become increasingly anathema to ‘respecting the will of the people’, even if it really means ‘the will of slightly more people on one side than the other’.
This conundrum is compounded when you take into consideration voter turnout. Taking the UK’s 2016 EU referendum as an example:
- voter turnout was 72.21% – or in terms of people voting, 33.577 million out of a potential 46.5 million;
- The outcome of the vote was 51.89% voting to leave the EU (a total of 17.41 million votes), and 48.11% voting remain (16.14 million votes); and
- The UK has a population of 66.685 million people.
At first glance, the outcome of 51.89% is sufficient to say that ‘more voted to leave than remain’. However, in contrast to the amount of eligible voters, that 51.89% becomes 37.44% – or just 1/3 of those that were allowed to vote. Against the entire population – or put another way, against the total number of people that will be affected by the outcome – 51.89% drops even further to 26.1%. When considering whether democracy equates with a true understanding of the ‘will of the people’, the perception of that legitimacy is subject to debate.
It is a dirty secret of democracy that it only works if its citizens participate. It cannot be denied the importance of the electorate doing their civic duty by attending a polling station on the day of a vote, and equally that democratic systems can only rely on those that make their voices heard when called to do so. However, presenting the outcome of an election or referendum as the consensus of the people to pursue a given policy is to overlook the entirety of the population governed. This may appear to be a numbers game, but it should be remembered that each of those numbers are equal citizens in a democracy. Imposing an unforgivingly harsh policy that ultimately commanded support from a minority and representing it as a ‘clear majority’ is in itself damaging to the unity of a state.
The Façade of ‘Yes’ or ‘No’
Rarely, if ever, are most of life’s decisions as simple as a ‘yes’ or ‘no’.
This is even more so when considering potentially controversial policy decisions. Yet, when putting a referendum question to the people, the challenge is presenting these complexities in a manner that lends itself to choosing between two options. Referendums are only of utility when there is a ‘clear’ result – that is, one option is preferred over the other. Typically, this is achieved by carefully crafting the question as to ensure that it communicates clearly the subject at hand, and that the response clearly indicates a policy direction.
However, regardless of how ingenius the crafting of the question, it is inevitable that a ballot question will lack the ability to present the full scope of the choices on offer without the question becoming cumbersome and difficult for voters to understand. By way of an example, the question on the ballot for Québec sovereignty in 1980 read:
The Government of Quebec has made public its proposal to negotiate a new agreement with the rest of Canada, based on the equality of nations; this agreement would enable Quebec to acquire the exclusive power to make its laws, levy its taxes and establish relations abroad — in other words, sovereignty — and at the same time to maintain with Canada an economic association including a common currency; any change in political status resulting from these negotiations will only be implemented with popular approval through another referendum; on these terms, do you give the Government of Quebec the mandate to negotiate the proposed agreement between Quebec and Canada?
« Le Gouvernement du Québec a fait connaître sa proposition d’en arriver, avec le reste du Canada, à une nouvelle entente fondée sur le principe de l’égalité des peuples ; cette entente permettrait au Québec d’acquérir le pouvoir exclusif de faire ses lois, de percevoir ses impôts et d’établir ses relations extérieures, ce qui est la souveraineté, et, en même temps, de maintenir avec le Canada une association économique comportant l’utilisation de la même monnaie ; aucun changement de statut politique résultant de ces négociations ne sera réalisé sans l’accord de la population lors d’un autre référendum ; en conséquence, accordez-vous au Gouvernement du Québec le mandat de négocier l’entente proposée entre le Québec et le Canada ? »
This question ran 106 words long (113 in French), and included a number of proposals that would have been subject to extensive negotiations with the Canadian Government only after the referendum had been decided. The manner in which it was presented to the people of Québec was open to being misinterpreted as fait accompli, requiring only a yes vote to implement this agreement. The complicated and confusing nature of the question certainly contributed to a ‘No’ vote (59.56% No / 40.44% Yes, 85.61% turnout of 4,367,584 eligible voters). The subsequent 1995 independence referendum reflected this by reducing its content to a relatively sveldt 41 words (35 in French):
Do you agree that Quebec should become sovereign, after having made a formal offer to Canada for a new economic and political partnership, within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12, 1995?
« Acceptez-vous que le Québec devienne souverain, après avoir offert formellement au Canada un nouveau partenariat économique et politique, dans le cadre du projet de loi sur l’avenir du Québec et de l’entente signée le 12 juin 1995? »
It should be noted that the “agreement signed on June 12, 1995” wrongfully suggested that some partnership had been agreed between Québec and Canada prior to the referendum. No such agreement existed. Despite this glaring misrepresentation, the result of the 1995 referendum was a ‘No’ vote (50.58% No / 49.42% Yes, 93.52% turnout of 5,087,009 eligible voters). The difference between Yes & No (54,288 votes) was less than the number of spoiled ballots (86,501).
In each ballot, the understanding was that a ‘Yes’ majority would provide a mandate to negotiate for the secession of Québec, but in no way guarantee the specific terms that were suggested in the verbiage of the question. It was also understood that a second referendum would likely be necessary once the actual terms of secession were negotiated as the mandate may not have supported all versions of secession.
What is clear is that ‘Yes’ or ‘No’ as a response could not legitimately convey nuanced support for the ultimate implementation of what would have been an irreversible action. Similarly, the 2014 Scottish independence referendum was presented as a choice of endorsing the SNP (Scottish National Party) terms of leaving the United Kingdom, terms which were entirely subject to negotiation and would almost have certainly changed substantially by the point of departure. The Scottish referendum did, however, manage to reduce their ballot question down to six words:
Should Scotland be an independent country?
The result: ‘No’ – 55.3% / ‘Yes’ 44.7%, 84.59% turnout of 4,283,392 eligible voters.
Though a question may be crafted to illicit a simple ‘yes’ or ‘no’, it does not mean that such a response is sufficient to endorse any one outcome beyond scrutiny. At best, a referendum may indicate the direction of travel for a given policy, whether the public would support further pursuits in any given endeavour. It has been suggested by multiple participants in multiple countries that any referendums that require negotiation with other governments (inter-governmental) or within the governmental structure of a single state (intra-governmental), a second referendum would be necessary to maintain legitimacy for a course of action given the degree of change that may occur during any negotiations. However, the suggestion that every person who votes for the ‘change’ option understands and is supporting an unambiguous, unquestionably-precise outcome is contrived. In any referendum for change, it is only the ‘status quo’ option that can command such a position – it is support for the way things already are, under the conditions that are already known.
In any event, it is a nice pretence to think that life breaks down into a series of ‘this or that’ decisions. In reality, however, there are usually multiple choices and many different factors that go into how we decide to express our desires.
The Need to be Informed for Consent to be Legitimate
There is an infamous response to a journalist in 2002 given by then-US Secretary of Defense Donald Rumsfeld that states:
…As we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns — the ones we don’t know we don’t know. And if one looks throughout the history of our country and other free countries, it is the latter category that tend to be the difficult ones.
Though said in a completely different context, Secretary Rumsfeld’s comments highlight the difficulties associated with policy-by-referendums – the unknown unknowns and the perils they entail. Although a referendum appears to empower citizens to have direct impact on policy, what impact that might be is altogether a different matter. In order for a supposed mandate from the people to have legitimacy, the subsequent governmental action needs to reflect the actual wishes of the people – not just a convenient interpretation.
As has been noted in the post-Brexit discourse in the UK, the departure from the EU is not as self-evident as has been made out by some on the Leave side of the conversation. During the woefully inadequate referendum campaign period, multiple groups were formed to advocate for the UK exiting the European Union, and each offered up their vision of what a post-Brexit UK-EU relationship would look like. These options ranged from a relatively close arrangement with the UK maintaining access to the single market and customs union, through to the so-called ‘hard Brexit’ of no relationship beyond trading on the World Trade Organization (WTO) standard tariff schedules. The consequences economically, politically, and socially for the multiplicity of potential outcomes varies significantly, and it should be remembered that at no point during the referendum was a ‘hard Brexit’ supported by any of the Leave advocate groups.
However, as the process of departure from the EU has unfolded (the ‘Article 50’ withdrawal agreement negotiations), the complications of leaving have become readily apparent. At present, the major deadlock arises over the UK maintaining its commitments under the Belfast (‘Good Friday’) peace agreement by guaranteeing an open border between Northern Ireland and the Republic of Ireland. (This is unsurprising given the original intent of the EU as a means of maintaining peace in Europe.) Furthermore, there are implications in over 80 different sectors of business in the UK that will have to make major changes to their operations in order to accommodate all possibilities that still remain on the table. A number of UK car manufacturers are either closing their operations temporarily or exploring relocating their facilities elsewhere in the EU to maintain frictionless production. There are pension considerations for UK citizens resident in other EU states, as well as questions over the status of non-UK EU citizens resident in the UK (legally) after departure. For all the countless days and weeks of campaigning during the pre-referendum period, only a fraction of the real implications were ever fully laid out. As such, it cannot be conceived of that these ‘unknown unknowns’ would have had a serious and consequential effects over peoples’ decision to vote for the UK to leave the EU.
A law podcaster, UK Law Weekly, put it succinctly recently regarding the need for informed consent, not mere consent (at 7m48s):
It’s hard to see how any consent can be real and voluntary if it is not informed consent. You might volunteer to help an old lady cross the road, but if I then told you the road was the M25 [motorway], you might think twice.
The same holds in relation to the consent (or ‘will’) of the people as to how they wish to be governed. If the people of Scotland had voted ‘Yes’ to the question “Should Scotland be an independent country?”, only to find out that its ‘independence’ entailed extreme isolationism in closing its borders to people and trade, many of the pro-independence supporters would likely denounce that course of action as contrary to their wishes as expressed in the referendum.
One may reply by saying that the referendum did not specify what independence meant, and therefore isolation was a possible interpretation of that outcome. That suggestion would, however, be unreasonable and completely at odds with the information and education of the public prior to the referendum. Equally unreasonable would be suggesting that the unforeseen or unknown circumstances arising after the referendum in no way could result in the public changing their mind – thinking twice – about their decision. (The best example of the public regularly changing its collective mind is general elections. Nobody suggests that holding an election every 4-5 years is a “gross betrayal of our democracy“.)
The idea of holding a subsequent referendum (not a ‘second referendum’) whereby the people – now in full command of the consequences and implications of known knowns, known unknowns, or otherwise – can confirm their support for the reality of their initial decision is very much in line with the form of democracy that is associated with the holding of referendums. If the people can be entrusted to make the initial decision to instruct a government to begin a process of change, then certainly that trust must extend to the ability to confirm their wishes have been followed, or to change course if the circumstances have changed.
The concept of a follow-up referendum isn’t particularly controversial either. In a recent report (2018 July) on referendums in the UK, the Independent Commission on Referendums recommended that:
Sometimes a referendum may be needed earlier: for example, to initiate intergovernmental negotiations. In such cases, the government initiating the referendum should set out precise plans for what will be done in the event of a vote for change; the enabling legislation would set out a two-referendum process, for use in the event that the settlement does not deliver what was promised.
This commission involved politicians and individuals of all variety, including some who advocated for Brexit as part of the official campaign.
It is understandable that in the event of a relatively weak or ‘fluke’ chance that a particular side of a referendum wins despite their expectations, the concept of having a subsequent poll to confirm support can appear to be an attempt to reverse the direction of the initial referendum. However, it is argued that the greater the implications of the outcome – particularly where it can have dire consequences in practice, or where the action may be perceived as irreversible (e.g., secession/independence) – the greater the need exists to ensure that the public remains engaged with developments and given the opportunity to ‘ratify’ the final destination. It should also serve as a cautionary tale to those that play fast and loose with facts and truth: if a follow-up referendum is involved on the final terms of any course of action, it is very likely that any misrepresentation or lies will become apparent and have the potential impact of turning a victory into an ignominious defeat.
So what you’re saying is…
In the end, what matters most is that those actions taken by any government subject to the outcome of a referendum critically require the continuity of legitimacy. It is not sufficient to simply shut down any notion of confirming those decisions with the very people that constitute the powers with which elected representatives are entrusted.
It is also advisable that prior to any drastic course of action (e.g., constitutional, independence, etc.), the government of the day should seek first to build a consensus amongst all citizens. They should not blatantly set the interests of one half of the electorate directly against the other half, that is if the government wishes to continue to be the government of that country rather than party nepotists. It divides societies, and it risks a fundamental chasm opening up between the unity of the citizens of any State.
Referendums typically involve controversial subjects that are potentially divisive. The absence of these risks would make a referendum pointless. Holding referendums also offers the most direct manner of deciding State policy to the citizens exercising their collective right of self-determination. The decision to hold a referendum should not be taken lightly, and the process and contingency planning should involve a clear set of agreed information being provided to those citizens going to vote. Only where there is a clear informed majority supporting a clearly explained outcome to a clear question can any government invest the finances, resources, and efforts in pursuing a legitimate action.
To paraphrase a piece of common sense advice: Measure twice, cut once.