The proposal, passed by a large majority in the country’s parliament in October, is designed to redefine the respective powers of parliament and the president.

Under the present constitution, it will be adopted if more than half of the votes cast in the referendum support it, provided that this number represents at least a quarter of those on the electoral roll (just over 2.5m citizens, according to police figures). So theoretically the change should require about 650,000 ‘yes’ votes.

However, although official reports of the last parliamentary and presidential elections show a turnout of 1.6m voters, in fact generally only a million or so Armenians take part in elections, and the extra supposed support for the ruling Republican Party is usually the result of rigging and the votes of ‘dead souls’ – the half million Armenians working abroad, mostly in Russia. Billboard in Yerevan's Baghramyan Avenue advocating a yes vote. Photo CC: Yerevantsi, 2015Rigging is also facilitated by the Electoral Code. In the past this allowed people to vote only on production of a passport. But on the eve of the referendum, Parliament voted through an amendment to allow voting on an ID card, without any system of checks on whether its holder had voted at more than one polling station.

In other words, the law will allow about 180,000 ID card holders to vote as often as they like. All attempts by the opposition to modify the amendment were voted down by the supporters of the new constitution.

The law will allow about 180,000 ID card holders to vote as often as they like.

The players

The pre-referendum campaigns by the supporters and opponents of constitutional change punch at very different weights, with the government using massive administrative resources to promote its side of the argument.

To take one example: heading the ‘Yes’ campaign is none other than Prime Minister Hovik Abrahamyan. He has six senior officials as his deputies, including the head of the presidential administration Vigen Sarkisyan, the minister for Territorial Administration and Emergencies Armen Yeritsyan, the head of the presidential Control Service Hovik Hovsepyan and other august figures wielding considerable power.

The ‘Yes’ campaigns in the regions are run by local governors, and a pre-referendum chat show on Armenia’s Public Service TV channel is hosted by Minister of Education and Science Armen Ashotyan, although the Electoral Code specifically forbids their involvement in an electoral campaign. In fact, of course, the entire ‘Yes’ campaign pyramid consists of members of Armenia’s power vertical.

A single example will illustrate the difference in the financial resources of the two campaigns: according to Boris Navasardyan, head of Yerevan’s Press Club, which has been monitoring the media during the campaign, the supporters of reform paid out for 26 times more TV air time than their opponents.

The voters

The referendum, unlike all previous voting processes, has been marked by an almost complete absence of public opinion polls. The fact is that very few polls in Armenia can be trusted. There is nothing surprising in this: to paraphrase the old saying, ‘he who pays the agency conducting the poll calls the results’.

Protesters in Armenia give the government the red card. (c) PHOTOLURE News Agency/demotixIt is, however, worth looking at two polls. One of them was conducted between 10 and 21 September by Gallup International Association - GIA (an organisation that has merely borrowed the name Gallup and is not to be confused with the famous US polling company), which has been operating in Armenia since 2005-6 and whose results are no more trustworthy than those of local pollsters. Many experts in Yerevan also believe that the company has close links with the Armenian government and simply provides the results they need.

Despite all this, the GIA poll reported that only 42 per cent of Armenians were planning to vote in the referendum. 47 per cent out of a poll of 1105 respondents said they would not vote, and 11 per cent were ‘don’t knows’. Furthermore, 29 per cent said they would vote for constitutional reform, 26 per cent that they would vote against and the remaining 45 per cent of respondents ‘didn’t know’.

The other poll was carried out across the country between 11 and 20 November by the Advanced Public Research Group (APR Group). It is unclear who commissioned and financed this poll, but its results seem relevant in the context of a previous similar poll carried out in summer 2014.

The new poll involved 1,300 respondents from all Armenia’s regions and was of a semi-standardised type, containing a mix of standard and non-standard questions and giving a 95 per cent trustworthy result. To a question on how Armenia’s current problems could be solved, 32 per cent of respondents replied that the country needed a change of government; 22 per cent argued for a consistent interpretation and application of its laws; 14 per cent supported a change from a presidential to a parliamentary government model and 12 per cent believed that the constitution and laws needed to be reformed.

When asked whether they were aware of the proposed constitutional changes, 83 per cent of respondents replied that they were and just 17 per cent that they were not. By contrast, only 47 per cent of respondents to the 2014 poll knew about the forthcoming constitutional changes, but overall both polls showed that there were three times as many opponents of constitutional change as supporters, and that over the past year support for the change rose by a mere one per cent.

In other words, despite an overwhelming advantage in terms of administrative, media and financial resources, the ‘Yes’ campaign is losing to its opponents: three out of four voters are against constitutional reform.

Very few opinion polls in Armenia can be trusted: ‘he who pays the agency conducting the poll calls the results’.

The obvious question arises: how did this situation come about? There are a number of reasons, but two are particularly significant. The first is that Armenians are calling President Sargsyan’s bluff. They know that there is no real need for constitutional change, and that the proposed reforms have only one goal – to leave Sargsyan in power, but in a new guise. They came to this conclusion in response to the ‘No’ campaign, which played down the details of the proposed amendments and focused on a clear and succinct message for the voters: ‘Serzh Sargsyan wants to stay in power!’

The second reason is that the ‘Noes’ have run an informative and creative social media and general internet campaign (many specialist bodies believe that Armenia has the most free internet of all the CIS states). It has also made active use of mobile campaign groups and cheaply produced promotional videos. The number of ‘Noes’ is also still growing, and they include not only the usual awkward squad but a good many young people, members of the middle classes and small business owners. The ‘Yes’ campaign, on the other hand, has been very unprofessional and inept.

Standing in the way of ‘no’

If we could believe in Armenia as a democratic country, we could confidently predict that the supporters of change have no chance of winning the referendum. But that is not the case. So all we can say is that the opponents of constitutional reform are in the lead, but that this by no means guarantees anything, given the massive opportunities for vote rigging available to the government.

Clashes between protesters and police in Yerevan, 23 November, 2015. (c) PHOTOLURE News Agency/demotixWe can, however, note some of the problems and ways in which the ‘Noes’ might overcome them:

• by making sure they are in control of polling stations during both the voting and the counts, working with members of electoral commissions, election agents and observers to ensure this happens. There are about 2,000 polling stations, 10% of which the ‘Noes’ cannot control and where they cannot rely on electoral commission members. They plan instead to use election agents and observers here,

• by making sure there is a high turnout. This is one of main problems for the ‘No’ campaign: three quarters of the electorate may be against the changes, but that doesn’t mean they will turn out to vote. There is a certain apathy among voters: they assume that the results will be rigged, whatever happens,

• given that only 1m of the electorate actually cast their votes, the ‘Noes’ have to be sure of at least 500,000 votes – a perfectly attainable figure if they can exercise the necessary control over the polling stations,

• only nine local NGOs have offered to provide observers, and their numbers add up to not much more than half the number of polling stations. Moreover, the largest, most professional and formerly US funded organisation, ‘It’s Your Choice’, which could singlehandedly provide enough volunteers to control almost all of the polling stations, has not offered its services.

Given all this, it is very difficult to predict the results of the referendum. It is, however, possible to say that for the first time in 15 years, members of the ruling party are less than certain of victory. Some are even privately admitting that the ‘Noes’ might actually win. Their majority would be small, but this is the first time that the ruling party has even contemplated defeat. It is telling that on 29 November, PM (and leader of the ‘Yes’ campaign) Hovik Abrahamyan, during a visit to Gyumri, Armenia’s second largest city, unexpectedly announced that the ‘Yesses’ expected to win only 30% of the vote in the local Shirak region. The polls suggest it could be a mere 14%.

Cancelling the vote?

In this situation, it is not out of the question that the referendum might be cancelled. It’s unlikely, but it is possible.

The referendum might be cancelled. It’s unlikely, but it is possible.

On 25 November, Armenia’s Special Services unexpectedly discovered the presence in Yerevan of an armed crime gang, more than 20 strong, led by a certain Artur Bardanyan (who lived in Spain for many years).

Bardanyan had supposedly gone to Syria as a volunteer to fight in defence of the local Armenian community. According to press reports (which have remained unchallenged), the gang was planning armed attacks on Abrahamyan, the Armenian Parliament building and other government structures. As a result, the law enforcement agencies and Special Services had strengthened the guard on government buildings and introduced additional security measures.

If we add to this the steep increase in tension around Syria and the Russo-Turkish stand-off over the Turkish downing of a Russian warplane, not to mention skirmishing in the Nagorno-Karabakh conflict zone and the Armenian-Azeri border, the gang’s arrest could provide an excuse to cancel the referendum. Other pretexts that might be used would be a terrorist threat and increased danger from outside Armenia.

Armenians, in other words, need to be ready for any eventuality.