Kulturkampf in Turkey: The Constitutional Referendum of 12 September 2010

Abstract

The referendum of 12 September 2010 saw the resurfacing of the main cultural cleavages in Turkish society, resulting in a major kulturkampf between the more socio-cultural liberal and secular coastal provinces and the more religious conservative hinterland. The AKP and some Islamist and Turkish nationalist fringe parties supported the ‘yea’ vote, the rest of the political parties supported ‘nay’, and the Kurdish nationalists boycotted the referendum. Individual-level statistical analysis of the vote indicates that partisanship, ideological splits, economic satisfaction and religiosity played a major role in influencing vote choices. A highly divided society, producing a 60–40 divide across a host of political issues, seemed to have emerged, helping the governing AKP get its way through heightened conflict.

Keywords:

• Turkey
• Referendum
• Secular–religious Divide
• AKP
• Economic Satisfaction
• Identity
• Party Identification

Acknowledgements

The author wishes to thank TUBITAK and Sabanci University Istanbul Policy Center for their financial help for the collection of the data and TUIK for their help with the drawing of the sample of the field survey, and INFACTO for their help in the collection and coding of the data.

Notes

 [1] For a thorough examination of this case by a journalist, see Jenkins (2009).

 [2] An Islamist journalist had been charged with sexual harassment of a fourteen-year-old girl a few months before, and the overall attitude of the right-wing press and the AKP spokespersons on that issue hardly showed any sensitivity to the victim's rights. Many brought up the overall attitude of the government and the media to that court case during the referendum campaign.

 [3] For an analysis of the pre-referendum process, see Turan (2010).

 [4] Adnan Menderes was Prime Minister of the Democrat Party government from 1950 to 1960. He was arrested after the military coup of 27 May 1960, tried by the military government, found guilty and executed with two other former DP government members. Since 1961, Menderes has been considered a ‘martyr’, especially in his home province of Aydın.

 [5] For an application of this model to the 2007 elections see Kalaycıoğlu (2010a).

 [6] See the following articles for further analysis: Kalaycıoğlu (1994), Esmer (2002), and Çarkoğlu (2007a).

 [7] This proclivity was first identified as part of a multivariate statistical analysis by the mid-1990s and later on rediscovered in several studies. See Kalaycıoğlu (1994) passim, and Çarkoglu (2005) passim.

 [8] The survey adopted the NUTS-2 regions of the Turkish Statistical Institute (Türkiye İstatistik Kurumu-TUIK). The target sample was distributed according to each region's share of urban and rural population on the basis of recent registered voter records. We applied probability proportionate to the population size (PPS) principle in selecting household addresses from each TUIK-2 region of urban, township (belde) and village localities. All neighbourhoods were separated into urban, township and villages of NUTS-2 regions and, within these, household addresses were selected on the basis of the application of the proper PPS ratios per region. We used no replacements and requested TUIK select 3500 household addresses, visiting each of them up to three times to conduct face-to-face interviews. Unfortunately 172 of the addresses turned out not to be households, and in 408 households either the occupant was away or no one seemed to live there at the time of the visits. We could not enter a further 142 walled residences. So we had to operate with 2778 addresses in total. The response rate was 60 per cent. A nationally representative sample of 1665 voting-age adults (18 years old or older) were interviewed at their households, and the resulting sampling error was about+/ − 2.5%.

 [9] For more on the aggregate results of the referendum, see Kalaycıoğlu (2010b).

[10] Turkish public university presidents (officially defined as Rectors) are appointed by the President of the country. In 2002–07, the AKP government functioned with President Ahmet Necdet Sezer, who was criticised for appointing staunch laicists as rectors. Since August 2007, it seems that the new President, Abdullah Gül, has been acting with the same kind of partisanship, appointing candidates who are conservatives, even when the majority of the university faculty failed to support their candidacy. The continuation of this practice has created an overall impression that Presidential appointments in Turkey are more often than not based on non-meritocratic criteria, such as ideological or partisan affiliations.

[11] Indeed the election of the new members of the HSYK created a shock even among those judges who had pushed for these changes. Surprisingly, the High Election Council of Turkey, the body that organised the elections, banned any propaganda by the candidates. The government seems to have drawn up a list of candidates, which was leaked to the press and received a huge amount of support at the polls. It looked as if overwhelming number of judges and prosecutors voted for the Director of the Personnel Office of the Ministry, the undersecretary of the Justice Ministry and other top-level bureaucrats. The opposition seemed to have a point. The High Election Council seemed to have failed to take the necessary measures to render the ballots secret. The judges and prosecutors, especially in smaller provinces where only five or six of them served, assumed that their votes would be as good as public. They seemed to have voted in an environment of intimidation. See Milliyet daily at http://siyaset.milliyet.com.tr/hsyk-seciminde-olanlara-bakin/siyaset/siyasetdetay/23.05.2011/1393487/default.htm which reports a forthcoming book by Orhan Gazi Ertekin, entitled Yargı Meselesi Hallolundu (Judicial Problems Dissolved), which focuses on the efforts of the Ministry of Justice to rig the HSYK elections.