"I think we have run a strong campaign. We have explained the issues in considerable detail and all our TDs have been out campaigning personally for a yes vote".
It was the press officer for one of the major yes-parties who presented us with this apology at the count at Dublin Castle yesterday. We choose not to mention his name – though he was happy to speak on the record. For what he presented was a textbook example of how to lose a referendum. They – the yes side – did all the things that you should not do in a referendum.
There are a number of cardinal errors one should never commit in a referendum campaign. This month we have seen them all in Ireland, from electoral greed, complacency and complexity to mixed messages and inappropriate messengers.
All too often political parties – in Ireland as well as elsewhere – treat referendum campaigns as if they were general elections. They are not! In a Dáil election you can smear the opponent. A referendum is about an issue – and you cannot smear an issue.
But more importantly, in a short referendum campaign – such as this one – it is impossible to explain the issues fully. Indeed, even to attempt this is a recipe for failure and defeat. EU treaties are by their very nature complex legal documents. They deal with complicated matters. This means that the more you try to explain the arcane detail, the more mystified the people will become. And when they are befuddled by doubt, they tend to revert to caution and suspicion.
And if voters do not understand an issue, they are more likely to smell a rat – even though the proverbial rodent is absent. Therefore it is essential in a successful referendum that the yes-side has a narrow focus on one – and only one – symbolic issue. And this token issue's salience must have been tested in polls and focus groups to ensure it is the right one; politicians should never choose it alone. That is how referendums are won. By campaigning on one simple, salient, but popular issue! A classic example of this was provided in the referendum in Northern Ireland in 1998, which was billed as a poll on progress – and not as a plebiscite on the 20,000 word intricate details of the so-called 'consociational system of power-sharing', that it in reality was.
It is much easier to scatter 'No' landmines, to scare voters on abortion, neutrality, taxation or unelected presidents. That's why most referendums are actually lost, not won. This is how the anti-abortion campaign almost caught up with the yes-side in the 1995 referendum in Ireland.
The second law of successful referendum campaigns is that personalities do not sell! The fact that the parties campaigned by putting up posters with pictures of endless local wannabe politicians was the second cardinal mistake in the campaign. Voters do look to their political leaders for guidance, but since the vote is on an issue, not a candidate, they do not feel as bound by their party's view. They want to associate with the messenger, but not if they are merely positioning themselves for the next election – clientilism of the worst sort.
To be sure, General Charles de Gaulle effectively used his own popularity to win referendums on Algerian independence and a new constitution in the late 1950s and early 1960s. But few people can match le général when it comes to charisma, and Brian Cowen and Enda Kenny are not among them. De Gaulle also reminds us: 'the trouble with referendums is that the people answer the wrong question…' Especially mid-term, the worst statistical moment to push through a poll.
Referendums are about trust. It is about allowing people to feel comfortable about an issue.
Politicians are generally seen as untrustworthy. Indeed, alongside car – salesmen and lobbyists, politicians are the least trusted profession. Especially at present; especially in Tribunal-heavy Ireland, especially on European matters!
In order to instil this feeling in the voters, it is important that the messenger is trusted; someone with whom the voters can identify. Ordinary voters are more likely to identify with people like themselves. The classic example of this was a referendum in Denmark in 1972, when the yes-camp printed shopping receipts, comparing the prices of groceries in case of a victory. This had the desired effect. In issue-based votes, trade unions, businesses, celebrities and sports people can participate – but not if they squabble and contradict each other -mixed messages do not sell well.
To be fair, the yes-side started doing a few things right in the last week of the campaign, but too little and much too late. We predicted the loss from a month ago. It's almost impossible to come back from that sort of poll negativity – indeed Yes needs to be 2:1 ahead with less than 30% don't knows a month out. They weren't.
In the past it has been a successful strategy to question the motives of the no-side's front people. Asking the question 'who's behind it?' is uncomfortable. This is what the government in Austria did in 1994, when they effectively derailed the campaign of right-wing politician Jorg Heider. That's how the Cyprus referendum of 2004 failed – the Greek Cypriots undermined both the UN and their Turkish compatriots by asking who was backing the yes-campaign and would you trust them?
In campaigning terms, the yes-side was correct in questioning the motives of Libertas and their enigmatic front Declan Ganley. But the campaign was effectively over when they finally got round to this. In 1994, the Democrats effectively used this tactic to win support for a number of measures opposed by the Republicans in Oregon. The offensive in Ireland came too late – and again their message was not coordinated and still less focused.
As a general rule, 'don't knows' split down two to one against – two thirds of those who 'don't know' end up voting no. Voters are cautious – they tend to opt for the status quo, in case of uncertainty; like the rest of us, they are apprehensive about change, which often moves to resistance, if given the ammunition, and the no-side certainly succeeded in that. This pattern also proved to be correct in the Irish referendum yesterday.
All that can go wrong will go wrong. So runs Murphy's First Law.
Given the way the yes-side conducted the referendum campaign it is probably more appropriate to devise Murphy's Second Law: 'all that can't go wrong will also go wrong'. For example the cross-party alliance in favour was portrayed as 'establishment and boring' rather than 'safe and solid – encouraging'
A new referendum is necessary if the Lisbon Treaty is to become law. The yes-side will need quite a lot of tuition before they embark on this campaign. They will need to carefully analyse their mistakes in order not to repeat them – again. Don't they remember Nice I and Nice II? So give us a ring – you've got the number!
