Turkey's EU accession prospects

10

+

Turkey's EU

Accession Prospects

Serdar

$.

Gilner

This chapter develops a two-stage game of incomplete information to generate insights about future trajectories of Turkey's EU acces­ sion process. It is assumed that Turkey perceives the EU members as divided into two groups: Type 1 and the Type 2. The Type 2 members never endorse Turkish accession whereas the Type 1 members do. The types move in the first period of the game and Turkey reacts in the sec­ ond. The analysis shows that the Type I's time preferences do not count for its cooperative conduct as long as the new Turkish foreign policy (TFP) continues to improve Turkish power and its reputation, domestic reforms are conducted at a steady rate, and Turkish economy continues to thrive. The future potential issues for Type 1 members are, if the new TFP capsizes, domestic reforms stall, or Turkish economic stabil­ ity becomes deeply shaken. A successful Turkey can quit the process at the slightest belief that no EU member supports its EU bid. It becomes more prone to perceive the Type 1 as the.Type 2. A failing Turkey will in turn leave the accession process if both types defect even if it believes that it interacts with the Type 1.

202 Serdar $. Giiner

INTRODUCTION

Would other EU members align with Austria, France, and Germany that oppose Turkey's accession to the European Union? How. could Turkey respond to such an enlarged opposition? Would it continue or quit nego­ tiations if it perceived that no EU member favors its EU bid? The chapter aims to shed light on these questions by adapting a two-period game of incomplete information proposed by Tirole (1997: 439-440).

Our model constructs Turkey's EU accession as a strategic interaction. It allows the analysis of the effect of Turkish misperceptions and the EU members' discounting of future payoffs upon the accession process. The game equilibria help to understand how the new Turkish foreign­ policy orientation, Turkish economic successes, and Turkish failures in conducting domestic reforms affect EU members' future evaluations of defecting and cooperating with Turkey as well as Turkish attitude towards the European Union. Turkish successes have become more vis­ ible given financial failures of several EU members. The equilibria also clarify why and under what conditions Turkey quits or continues its full membership quest.

Turkey, an area where the Middle East and Europe intersect, consti­ tutes a unique case in the expansion of the European Union. Turkey is the largest Muslim state that has applied for EU membership. A full accession of Turkey could represent a genuine dialogue between the two cultures, and refute the well-publicized argument about a clash of civilizations (Huntington 1993). A better understanding of Turkish accession to the European Union will help us think about how cultural-religious differ­ ences fuel misperceptions and affect international interactions. Therefore, the fate of Turkish accession is immensely consequential for the European continent, if not the world.

The game combines three features. First, EU-Turkey relations are not a one-shot deal. The game represents the European Union as acting in the first period and Turkey as reacting in the second one. Second, Turkey is doubtful whether it will be an EU member; the European Union does not seem to be a solid group with respect to its accession.1 It perceives the Eu­ ropean Union as divided over its membership, observing staunch French and German opposition. Yet it also suspects that no EU member genuinely supports its accession.2 Accordingly, the game models how these Turkish beliefs affect EU-Turkey interactions. Finally, a player can estimate future payoffs as lesser than or nearly equal to present payoffs. The unusual length of open-ended Turkish accession implies a breeding ground for such estimations. Patient players estimate future payoffs as nearly equal to present ones. Future does not represent any value for those who are

Turkey's EU Accessio11 Prospects 203 impatient. The game allows for how the EU members' future estimations or patience, that is, time preferences, influence the process.

This paper is organized as follows. The first section motivates the game. The second section presents the game and the equilibria. The third section discusses and interprets the findings. The fourth section concludes by indicating possible extensions of the model.

STRATEGIC BACKGROUND

Turkey's Europeanness is a topic of discussion within the European Union. The European Unjon perceives Turkey as a non-European coun­ try. Turkey's large and predominantly Muslim population represents a tremendous challenge for the European Union's enlargement (Mtifti.iler­ Ba� 2000: 22). European discourse about Turkey stimulates fear. In this view, Christian Europeans should be protected against Muslim Turks (Rumelili 2004: 45; Yegenoglu 2006: 253; Wodak 2007). Identity differen­ tiation in European populations towards Turkey in terms of geography and culture is particularly strong (Burgin 2010). Duff (2011) notes:" Anti­ Turkish or anti-Islamic sentiments in many EU countries are scarcely inviting." There exists a strong Turco-phobia in the European public (Grigoriadis 2006). The work by De Vreese, Boomgaarden and Semetko (2008) further explores the feelings of identity and and how attitudes towards immigrants play a decisive role in the EU citizens' position in respect to Turkey's accession. Similarly, Gerhards and Silke (2011) find that economic benefits of Turkish accession, cultural differences, politi­ cal ideology, and general attitudes towards the European Union shape EU citizens' preference for keeping Turkey outside of the union. Hence, Europeans' attitude is primarily fueled by cultural and religious dif­ ferences. Turkish chances of membership are dim, as European public opinion constitutes the overriding determinant of Turkey's accession (Koprtilti 2009: 198).

Turkish leaders believe that their country is segregated on religious and cultural bases.3 Recent polls conducted in Turkey also reveal that Turks believe that the European Union applies double standards with respect to their country's membership. In short, Turks do not believe their country will eventually become an EU member (Karda� 2010: 125; Larrabee 2010: 173). Several factors exacerbate Turkish distrust and suspicion: the length of the accession process, the EU enlargement spanning the accessions of the Republic of Cyprus, and questionable membership experiences of other Eastern European countries (Oni� 2009: 26; Oni� 2007: 258-260; Kiri��i 2004: 89-94). The more the accession drags, the more Turks

204 Serdar $. Gii11er

believe that the European Union employs political hypocrisy. Accord­ ingly, Turkish enthusiasm for the European Union wanes Oung 2001; Gordon and Ta�pmar 2006; Kentmen 2007; Ytlmaz 2008; Oni� 2009, 2010).

Indeed, Turkey is the only candidate that was offered a privileged partnership as formulated by German Christian democrats in 2b02. The proposal was then supported not only by Austria, Germany, and France but also by the Czech Republic and Denmark (Karaka� 2006). The problem is that even the proponents of the proposal do not agree about its content (Pope, 2009). Turkey, as a privileged partner would have no rights to be a part of the EU's decision making processes; it would only have an observer status, which Turkey already occupies as an associate member of the union that is also an official candidate (Karaka� 2006 and Pope 2009).

Turkey remains enthusiastic despite the European objections to its accession. Turkish Prime Minister Recep Tayyip Erdogan confirms that Turkey will continµe the negotiations until it becomes a full member. He rejects the privileged partnership status.4 Thus, Turkish preference to finalize its European status has strong bases. Turkey will not be entrusted in the Western world without an EU membership (Mtiftuler-Bac; 2000; Erdogdu 2002; Gordon and Ta�pmar 2004; Hadjit and Moxon-Browne 2005; Oni� 2007). The Justice and Development Party (known as AKP, its Turkish acronym) in power since 2002 has a moderately Islamic politi­ cal tone. One might expect that such a party would oppose Turkey's EU membership. However, the AKP did not give up the project, perhaps because an EU membership would strengthen religious actors socially through higher democratic norms that expand religious freedom (Bech and Snyder 2011: 201).5

The length of the process is staggering. EU-Turkey relations cover both Cold War and post-Cold War periods. Turkey started to interact with the European Union-the EEC at that time-in 1959 with an aim to join the organization. It became an associate member of the EEC through the Ankara Treaty of 1963. Turkey is the only candidate country that signed customs union agreement with the European Union, and, after more than half a century, it still strives to become a full member.

Moreover, the Turkish accession process is open ended. Turkey still might not achieve its full membership goal, even if it successfully con­ ducts all reforms needed to fulfill the Copenhagen criteria. Hence, there is a high level of uncertainty about whether Turkey will join the European Union in the foreseeable future. In contrast to Turkey, some former Com­ munist Block countries did not wait as long: they became members less than two decades after the dissolution of the Soviet Union. Hungary and Poland became EU members in 2004; Bulgaria and Romania waited just three more years for full accession.

Turkey's EU Accession Prospects 205 The recent EU enlargement supports the view that there are no scales of measuring a candidate's performance along the fulfillment of Copenha­ gen criteria. The ripeness of economic and political conditions prevailing in the candidate country cannot be accurately quantified for accession. The acceptance of a new member is merely rooted in the EU members' power, that is, their interests and ability to influence the accession pro­ cess. In short, the fulfillment of the Copenhagen criteria is not a sufficient but a necessary condition for accession (Mi.ifti..iler-Ba<; 2008; Mi..ifti.iler-Ba<; and McLaren 2003). To illustrate, Romania had a long way to go about the fulfillment of the Copenhagen criteria. But it was given candidate status in the summit of Luxembourg (Mi..ifti.iler-Ba<; and McLaren 2003: 28). Greece became a member in 1981, though it was negatively evalu­ ated in 1976. Erdogdu (2002: 43) argues that the customs union agreement between Turkey and the European Union was made after the European Union promised Greec� that it would start accession negotiations with Cyprus; Greece would have vetoed the agreement otherwise.6 Secondly, unlike Turkish Cypriots, Greek Cypriots voted against the Annan Plan aiming to unify the island. They became EU citizens just a week later while the island was still divided. As a result, Turkish misperceptions about the EU double standards and prejudices became deeper (Karda� 2010: 116; Oni? 2011).1·

Currently, the largest obstacle to Turkish accession remains Turkey's non-recognition of the Republic of Cyprus, sole-recognition of Turkish Cyprus, and denial of its ports to trade with the Greek south of the island.11 Cyprus is responsible for many frozen chapters in accession negotiations; only one out of thirty-five chapters is negotiated and closed (Christensen 2009; Kettle 2010; Tocci 2010). Greece and Cyprus both declare their sup­ port for Turkey in its EU membership quest. Yet they also have their own conditions for a successful process: Turkey's diplomatic recognition of Cyprus Republic, Turkish acceptance of Greek Cypriot trade at its ports together, and the withdrawal of Turkish troops from Cyprus.9 _{Unless sig­}

nificant Greek and Cypriot concessions are given, no Turkish government seems willing or able to fulfil) these conditions.

The 2010 progress report of the European Commission urges Turkey to remove all remaining restrictions on the free movement of goods, includ­ ing restrictions on means of transports regarding Cyprus, to implement the customs union fully. The report indicates that: "Despite repeated calls by the Council and the Commission, Turkey still has not complied with its obligations as outlined in the declaration of the European Community and its member states of 21 September 2005 and in the council conclusions including those in Decembers of 2006 and 2009. It does not meet its obli­ gation of full, non-discriminatory implementation of the Additional Pro­ tocol to the Association Agreement, and it has not removed all obstacles

206 Serrlnr $. Gii11er

to the free movement of goods, including restrictions on direct transport links with Cyprus."10 Thus, the report affirms that the Cyprus issue will

continue to block Turkey's adhesion.

Any collapse of Turkish accession due to Turkish misperceptions or Eu­ ropean intransigence would have regional and even global repercussions. If rejected, Turkey could discontinue its progress toward adjustment with European norms and democratization; a move that could socially and politically destabilize the periphery of the European Union. The rejection of Turkey would support clash of civilizations thesis and demonstrate that political-economic blocks can emerge in terms of religious and cul­ tural differences. Two recent developments must be taken into account. Firstly, in terms of Turkey's traditional-Western orientation and connec­ tions, Turkish foreign policy recently became more volatile and rampant. Secondly, Turkey is currently the sixteenth largest economy in the world. Current financial difficulties of Greece, and, future prospects of Spain, Portugal, Ireland, and even Italy facing similar problems, threaten Eu­ rope's common currency zone. Turkey remains almost unaffected by the 2007 global crisis in sharp contrast with the majority of the European Union. These facts rule out the possibility of a poor country aiming at an expected economic growth through EU membership.11

To summarize, Turkish misperceptions of a European tendency to leave Turkey out, and the division within the European Union about whether to eventually accept Turkey are the primary determinants of the relationship between the two sides. It is unknown, however, how Turkish successes and failures in political economic reforms and foreign policy would reflect on this relationship. The European Union might appreciate future cooperation and favor a successful Turkey for membership, and Turkey could reciprocate. On the contrary, a failing Turkey can choose to leave the negotiation table facing European opposition. Our game model sheds light on these interactive dynamics.

THEGAME

Turkey is assumed to perceive the European Union as being divided into two groups of states denoted as the Type 1 and the Type 2. Turkey perceives that some EU members can issue declarations of strong support for Turkey even as they recognize multiple obstacles against full Turkish membership. These countries make up the Type 1. In contrast to the Type 1, the Type 2 consists of those countries that openly declare that Turkey has no place in Europe whatsoever, and Turkey should instead become a privileged partner.

Both types indicate that negotiations will take a long time and that Turkish membership is not a sure prospect. According to Turkey,

impos-Turkey's EU Accessio11 Prospects 207

sibility of full accession can also be realized, and negotiations could be terminated. Turkey has this perception because the vast majority of the EU citizens do not want Turkey in the union, and member states have to factor in their constituents' opinions. This is especially true in the after­ math of the French and Dutch rejections of the EU constitution (Gordon and Ta$pmar 2006: 58).

The European Union is assumed to be informed ofTurkish perceptions. The European Union is aware of Turkish leaders' suspicions that the Eu­ ropean Union negotiates ultimately to create a neighbor that shares Eu­ ropean norms but not to integrate it. A Turkey that improved conditions in human rights, civil-military relations, and minority rights is preferable to one that did not.

The European Union and Turkey are also assumed to interact through two periods. In period one, the European Union moves between defection and cooperation that respectiv.ely correspond to objections against Turk­ ish accession including priviledged partnership proposals and efforts for accession negotiations. This assumption simplifies and categorizes the European Union's options towards Turkish accession. Turkey is assumed to move after the European Union by deciding to continue or to halt ne­ gotiations. Therefore, the action sequence models Turkish reaction to the European Union moves displaying an action-reaction process.

Let A, denote the value of Turkish unilateral termination of negotia­ tions, A2 the value of cooperation with Turkey, and A.1 the value of de­ fection, that is, the opposifion against full Turkish membership, for both types. Hence, Type 1 and Type 2 preference orderings are assumed to be A , > A1 > A.1 > 0 and A , > A3 > A2 > 0, respectively. Both types prefer Turk­ ish unilateral end of negotiations while their preferences are in conflict over defection and cooperation. Turkish preference ordering is simpler. Turkey gains B, if the European Union cooperates but suffers 8₁ facing EU defection. Accordingly, Turkish preference ordering is assumed to be as B, > 0 > B_i

-The rules of the game and preferences imply European Union and Turkish actions that are best replies against each other, that is, equilib­ rium. The equilibrium consists of European Union and Turkish actions, updated beliefs of Turkey about the preferences of the European Union after observing EU actions.

Separating Equilibrium

The types are separated in terms of actions they take in this equilibrium. The Type 1 cooperates and the Type 2 defects in the first period. Turkey reacts to these actions under its updated beliefs.

Let Turkey believe that the EU is the Type 1 with a probability of p and the Type 2 with a probability of 1 - p before the EU acts. These

208 Serdar $. Giiner

probabilities represent Turkey's prior beliefs about the EU preferences over cooperation and defection. They are updated given observed EU actions. More formally, we have:

µ ( 1 1 C ) = [ p ( C J 1 ) . p ) / [ p ( C I 1 ) . p + p ( C I 2 ) . ( 1 - p ) ) = 1 . p / [ 1 . p + 0 . ( 1 - p ) ] = 1 µ ( 2 1 D ) = [ p ( D 1 2 ) . ( 1 -p ) J / [ p ( D 1 2 ) . ( 1 - p J + p ( D 1 1 ) . p ] = 1 . ( 1 - p ) / [ 1 . ( 1 - p ) + O . p ) = 1

where µ ( 1 I C ) and µ ( 2 I D ) are Turkish updated posterior belief

that the EU is the Type 1 given cooperation is observed and that the European Union is the Type 2 given defection is observed, respectively. The updated beliefs are conditional probabilities computed by applying Bayes's formula. As to the actions of the two types, p _{( C I} 1 ) denotes the likelihood of the Type 1 choosing cooperation and p ( D I 2 ) denotes the likelihood that the Type 2 defects. Similarly, p ( D I 1 ) denotes the likeli­ hood of Type l's defection and p ( C I 2 ) denotes the likelihood of Type

2's defection. As the Type 1 never but the Type 2 always defects, we have

p ( C I 1 ) = p ( D I 2 ) = 1, and p ( D I 1 ) = p ( C I 2 ) = 0.

Therefore, if the European Union cooperates, Turkey updates its prior belief about the European Union and presumes it is certainly facing the Type 1. Otherwise, if the European Union defects, Turkey updates its belief and understands that it is dealing with the Type 2 by certainty. The calculations demonstrate how Turkish beliefs of p and 2 -p change observing EU behavior.

The Type 1 cooperates in the first period and obtains A₂• Believing that it deals with the Type 1, Turkey does not exit the process in the second period. Consequently, the Type 1 obtains another A₂ in the second period. However, the values of cooperation with Turkey in the first and the second periods are different for the European Union. The values of present and future cooperation are unequal, as a player can estimate future payoffs as representing lesser values compared to present ones. An uncertain future weighs less than the present (Gardner 2003: 186; Rasmusen 1989: 89-91). Therefore, the second-period payoff is scaled down by a discount param­ eter b where O < b < 1. lf the parameter takes values close to zero, the future does not matter. The player does not care about future interactions. If, on the other hand, the parameter takes values close to one, player can be inter­ preted as being patient to such an extent that the future is almost equally important as the present, that is, future payoffs are not highly discounted.

Turkey's EU Accession Prospects 209 The Type l's cooperation payoff becomes A2 + bA2 as it obtains A2 in both periods but the second-period payoff is discounted. If the Type 1 defects, it obtains A₃ in the first period. Turkey then suffers 82 in the first period, updates its beliefs, and presumes it is dealing with the Type 2. As a result, Turkey stops negotiations in the second period. Turkish exit move generates a payoff of A, for the Type 1. Thus, the Type 1 receives a payoff of A3 + b A, from defection. The Type 1 cooperates only if coopera­ tion yields a higher payo,ff than defection, that is, if A3 + b A, < A2 + b Ar Rearranging, we obtain b A, -b A1 < A1 -Ar Hence, we have b (A_{1 -}A) < A₁ -Ar Dividing both sides of the inequality by A, -A_2, the discount factor condition for the'Type 1 to cooperate with Turkey becomes:

The condition (1) is necessary for the separating equilibrium to exist but not sufficient as we need to specify Turkish equilibrium actions also. Sup­ pose now that the condition (1) is fulfilled. Hence, the Type 1 cooperates. The Type 2 always defects and obtains its next-most preferred outcome in period 1 and the most preferred outcome in the second period. Future discounts of payoffs do not matter for the Type 2.

Turkey reacts to defection by interrupting negotiations given that B, > 0 > 8,. These actions and Turkish beliefs constitute a separating equilib­ rium.-Type 1 cooperates and Type 2 defects. Turkey continues to negoti­ ate if the European Union does not defect; interrupts otherwise. Turkey believes that it interacts with the Type 1 observing EU cooperation; oth­ erwise it believes that it interacts with the Type 2.

Semi-Separating Equilibrium

It is also possible that the Type 1 becomes indifferent between defection and cooperation so that it can use either action while the Type 2 always defects. The condition for the Type l's indifference is:

Turkey would then stop observing defection believing that it is interac t ­ ing with the Type 2, or continues to negotiate believing that it interacts with the Type 1 given EU cooperation· under the condition (2).

Pooling Equilibrium

Both types take the same action in this equilibrium. Thus, both types d e ­ fect, as the Type 1 imitates the Type 2 who never cooperates with Turkey.

210 Serdnr $. Giiner

An European Union move to suspend negotiations with Turkey because of its dispute with Cyprus exemplifies such a defection. Turkey cannot now update its beliefs. Turkish posteriors remain equal to p and 1 - p

that are prior Turkish beliefs of interacting with the Type 1 and the Type 2, respectively. Again, more formally, we have:

µ ( 1 I D ) == [ p ( D 1 1 ) . p ] / [ p ( D 1 1 ) . p + p ( D I 2 ) . ( 1 - p ) ) = 1 . p / [ l . p + l . ( 1 - p ) ) = p µ ( 2 I D ) = [ p ( o· 1 2 ) . ( 1 - p ) ) / [ p ( D I 2 ) . ( 1 - p) + p ( D I 1 ) . p ] == 1 . ( 1 -p ) / [ 1 . ( 1 - p ) + l . p ] = 1 - p

where µ ( 1 I D _{) denotes Turkish updated posterior belief that the Eu­} ropean Union is the Type 1 and µ ( 2 I D ) is Turkish updated posterior belief that the EU is the Type 2 given defection is observed. Remark that

p ( D I 1 ) = p ( D _{I 2 ) = 1 in this proposed equilibrium. The calculation} demonstrates that Turkish updated beliefs observing EU defection re­ main equivalent to Turkey's prior beliefs about the type of the European Union it faces. Turkey's observation of EU defection does not produce any additional information about the preferences of the European Union with respect to its accession. Hence, Turkish prior beliefs about the type of the European Union it faces do not change after the Eur,opean Union opposes Turkish membership.

The Type 1 obtains A₂ from cooperation but A₃ from defection. We have

A₂ > A₃ > 0 by assumption, so the Type 1 suffers a net loss of A₂ -A₃ from defection. The Type 1 recovers from the cost and makes some benefit if Turkey stops in the second period so that it obtains A_{1• If Turkey does}

not stop, the equilibrium collapses as defection becomes costly for the Type 1. The Type 1 would then have no incentive to behave like the Type 2. Thus, the Type l's payoff from defection must exceed its payoff from cooperation, that is, we must have: A₃ + b A₁ > A₂ (1 + b). The solution of the inequality for the discount factor b implies the following defection condition for the Type 1:

Turkey walks away only if ending negotiations generates a payoff greater than or equal to the uncertain prospect negotiating with the Type 1 or the Type 2. Turkey must decide whether to continue or exit the n e ­ gotiation process without knowing exactly the type of European Union it

Turkey's ELI Accession Prospects 211 interacts. Turkish expectation of continuing negotiations is p B₁ + ( 1 -p ) Br If the expectation constitutes a loss, Turkey stops to avoid a�y more losses under the following condition: 0 > p B₁ + ( 1 -p ) B_r The inequal­ ity implies that:

The threshold 8₂

I

Consequently, both types defect and Turkey stops provided that con­ ditions (3) and (4) are satisfied. There are infinitely many pooling equi­ libria where both types choose to defect and Turkey randomly chooses either to continue or -to stop. Turkey becomes indifferent between in­ terrupting negotiations and continuing under its misperceptions with respect to the European Union. Turkish indifference implies that Turkish belief of facing the Type 1 must fulfill the condition of p = 82 / ( B2 -B1 ).

Table 10.1 summarizes all equilibria.

Semi-separating and infinitely-pooling equilibria support indifference between actions. They can explain all actions of the European Union and Turkey. Yet w� need explanations of clear policy directions. Con­ sequently, we focus on the separating and the pooling equilibria in the following section.

DISCUSSION AND IMPLICATIONS

Turkish successes and failures in meeting the Copenhagen criteria affect Type l's evaluations of A₁₁ A₂, and A_J' that is, Turkish decision to aban­

don the accession process, cooperation, and defection, respectively.12 _The criteria emphasize economic and political successes such as achieving a functioning market economy, capacity to compete within the European Union, higher standards of democracy and human rights. A wealthier and a more prosperous Turkey is a more valuable candidate. Turkey's economic performance is indeed impressive given financial crises some EU members face such as Ireland and Greece. Turkey is the "rising power on the Bosphorus" (Steinvorth and Zand �011) and a "trading state" (Kiri�c;i 2009: 33). Nowadays, Turkish economy is the 6th largest economy among the members and candidates of the European Union and its infla­ tion rate has been in single digits for several years.13 It is an influential economic and political powerhouse in the Middle East, Caucasus, and

212 Serdm· $. Gii11er

Table 10.1. Game Equilibria _{· ·-··-}

·-Equilibrium Separating Semi-separating Pooling Infinitely-many pooling Type 1 Type 1 Both types Both types

cooperates, indifferent defect, Turkey defect, Turkey is

Type 2 defects. between stops. indifferent

cooperation and between quitting

defection; ir can and continuing

EU Actions rake either _{actions. Type 2} negotiations.

defects. Type 1 0 A? - Ai _{0 = A}2 -A., _/j > _�L:-31. o > A?-Ai discount < · -factor Ai -A2 A1-A2 A1-A1 A, -A1 conditions

Turkey is cerrain Turkey is certain Turkey believes Turkey believes to interact with to interact with that it interacts that it interacts

Type 1 Type 1 with Type 1 with Type 1 with

observing observing with p and p and Type 2 cooperation and cooperation and Type 2 with 1 with 1 - p certain to certain to - p probabilities. interact with interact with probabilities.

Turkish Type 2 Type 2

beliefs observing _defection. observing _defection.

Turkey stops Turkey stops Turkey stops Turkey is

facing defection; facing defection; provided that indifferent continues continues _Bi between quitting

otherwise. otherwise. _{p < - --·} and continuing

Turkish 82 -8, negotiations

actions _{provided that}

B, p =- - ·-₈

2-81

the Balkans. As a result, Turkish elites expect that the long-term Turkish assets will form "a hedge against EU rejection" and will make Turkey "more attractive to Europe" (Danforth 2011: 21 ). Adverse developments can harm Turkish economy. Wall Street Journal indicates that the Turk­ ish economic boom is not without structural weaknesses, as the value of Turkish lira continues to drop against the US dollar.14 Hence, Turkish economic successes imply higher values of A2 but lower values of A1 and A1 in contrast to Turkish economic failures.

A counter argument is that the accession of Muslim Turkey still re­ mains controversial even if economic barriers are worked out (Nelsen, Guth, and Cleveland 2001: 22). Economic successes do not have any im­ pact on A,, A2, and A1• Cultural differences, which are not included in the Copenhagen criteria, offset any increase in the value of cooperation with an economically successful Turkey. The problem is that the critical

thresh-Turkey's EU Accessio11 Prospects 213

old ( A2 -A1) / ( A1 -A) remains constant under such a cultural-religious

argument. As a result, it is impossible to analyze changes in the Type l's behavior towards Turkey. The Type 2, for example, is assumed to oppose Turkish accession regardless Turkish economic failures and successes.

Hence, the counter argument rather supports Type 2's behavior.

Similarly, ·1:urkey's successes and failures in domestic reforms to improve minority group rights and to eliminate military interferences into domestic politics influence Type l's payoffs. There is lesser value in raising objections against a country sharing the EU's polit�cal values, if incompletely. Presidents of the European Council and European Com­ mission; Van Rompuy and Barroso noted in 2011 that Turkish progress in political reforms "should also give new impetus to the accession negotia­ tions with the EU."1' In fact, constitutional amendments started since 2001

and those following 2010 Turkish referendum hint at Turkey's commit­ ment for success along with the political criteria.

Yet Turkey's democratization is difficult to achieve as long as the Kurd­ ish question is not resolved and civilian control over the military is not fully established (Ulusoy 2008; Goney and Karatekelioglu 2005). There are several court cases against alleged military-coup leaders including civilians, bureaucrats, and journalists. The cases are still open while the intensity of armed clashes between Turkish Army and the PKK (Kurdish separatists) is on the· rise. Elected parliamentarians of the pro-Kurdish Peace and Democracy Party protested against the Turkish Grand Assem­ bly for a long time, and refused to take their parliamentary oaths. Some of them are detained and charged for having connection to the PKK. A brand new constitution the newly elected parliament would prepare may be a solution to the Kurdish problem. Nevertheless, it is unclear how the collective spirit in the parliament required for it would be created. Hence, political reforms to reach higher norms of democracy and human rights constitute the most difficult area for Turkey.

A third factor that influences Type 1's payoffs is Turkey's conduct of a new and active foreign policy aiming at "zero-conflict" with the neigh­ bors and establishing trade relations with them. A powerful and prosper­ ous Turkey would manage energy routes from Central Asia to Europe through Russo-Turkish energy cooperation prospects, nurture lucrative commercial interactions at different scales, which also repair old animosi­ ties and conflicts in its neighborhood. The new Turkish foreign policy (TFP) constitutes a "trading state" foreign policy evolving through com­ mercial links and in line with AKP political elite's cultural background

(Zalewski 2010: 99-100).

According to Ahmet Davutoglu, Turkish Minister of Foreign Affairs and the architect of the new TFP, Turkey constitutes a source of stability and wealth for many countries including former Ottoman dominions and

214 Serdnr $. Gii11er

others. This gives Turkey a tremendous "strategic depth" in the global scene (Davutoglu 2001, 2008). Davutoglu believes that Turkey is distin­ guished from other Muslim countries with its unique history and Turk­ ish speaking minorities around the world, especially in Europe. Turkey's imperial history, geopolitical position, cultural affinities, and ties can help policymakers in the Middle East and the Balkans to realize how con­ nected their countries are with Turkey. Some analysts call the new TFP "soft Euro-Asianism" due to the belief that the EU accession is no longer a priority for Turkish decision makers who are more interested in building commercial and political connections with China and Russia (Oni� and Yilmaz 2009: 20). Another camp argues that Turkish leadership's commit­ ment to the European Union remains to be strong. Turkey's alternative options to increase its wealth and political influence complement, not substitute, its EU ideal (Zalewski 2010: 102).16

The new TFP became more volatile and diversified than the country's traditional Western orientation. Some analysts believe that the drift in TFP constitutes a reaction to the European Union and American policies (Cook 2009: 116). Consequently, there exists a tension between Turkish efforts for accession and the new TFP orientation (�agaptay 2009; Oni�, 2010; Pope 2010). Buzan and Diez (1999: 47) argue that the accession of Turkey would lead to the EU entanglement in centuries-old rivalries that ravage the Middle East and the Caucasus. Turkey's accession would not necessarily entail the EU entrapment in such conflict epicenters consider­ ing cooperative successes of the new TFP. Nevertheless, the current dete­ rioration of Turkey-Israel and Turkey-Syria relations would revive such worries. These relations demonstrate that the new TFP can spectacularly fail. Therefore, it is likely that the new TFP successes and failures can strengthen or undermine anti-Turkish sentiments within the European Union. Now, we can examine how Turkish failures and successes in eco­ nomics, domestic reforms and foreign policy influence the threshold and therefore cooperative behavior of the Type 1 towards Turkey.

The separating equilibrium reveals that the Type 1 cooperates if its discount rate remains below the critical threshold ( A2 - A3 ) / ( A, - AJ

Turkish successes and failures would cause changes in A _1, A_2, and A_3• The changes would consequently affect the threshold while the prefer­ ence orderings of both types remain fixed. If Turkey is successful, A2

increases and approaches A,, A, decreases and approaches A₂, or A₃ de­ creases. As a result, the threshold increases and Type l's range of time preference expands.

A numerical example illustrating the effects of higher A₂ values is use­ ful here. If A2 > ( A, + A3)

I

I

Turkey's EU Accessio11 Prospects 215

1. Assume that A,= 2 and A₃= 0 but A₂ varies. The threshold becomes 3/5,

1, and 3, for A, values of 3/4, 1, and 3/2, respectively. All discount rates would satisfy the conditions b < 1 and b < 3, as O < b < 1 by assumption. Therefore, the Type l's time preferences would not matter for its coopera­

tive conduct as long as the new TFP continues to augment Turkish power and reputation, domestic reforms are conducted at a steady rate, and Turkish economy continues to thrive. The Type 1 would always cooper­ ate with a successful Turkey regardless its evaluations of future payoffs.

The threshold decreases, however, if A₂ decreases and approaches Ay

A, increases, or A3 increases and approaches Ar To outline these relations

more clearly, assume again that A,= 2 and A3 = 0 while A₂ gets smaller. The threshold gets lower, a's A, values decrease and approach 0.17

Table 10.2. Time Preferences and Discount Factor Intervals A1 = 2 Ai = 1 A,= 0 A1 = 2 A,=

f

f

-

_-·=

---

The table shows that the Type l's range of time preferences shrinks, as the discount rate b must be lower than smaller thresholds for the Type 1 to cooperate. The Type 1 cooperates with Turkey by becoming progres­ sively impatient as cooperation payoff loses value. We can therefore as­ sert that the Type 1 cooperates only if it discounts future payoffs to an extent that varies with Turkish failures. If the newTFP capsizes, domestic reforms stall, or Turkish economic stability becomes deeply shaken, time preferences of the Type 1 start to matter for its behavior towards Turkey. The Type 1 avoids an uncertain future so that the future casts a shadow

216 Serdar $. Gii11er

upon present relations. One would intuitively expect that if future mat­ ters, some EU countries would have a higher tendency to cooperate with Turkey. Unfortunately, this might not be the case. Turkish failures induce the Type 1 to cooperate only if it evaluates future gains of the accession process as not valuable as the present ones.

What about Turkish reactions to the European Union? According to the pooling equilibrium, Turkey reacts by walking out the accession process if the European Union defects, for example, if the European Union de­ cides to suspend accession negotiations as Turkey does not open its ports

to Greek Cypriot vessels.18 _{The values of} B

1 and 82, Turkish evaluations

of cooperative and defective moves emanating from the European Union respectively, determine the critical threshold for Turkey to stop negotia­ tions with the European Union.

We can offer two hypotheses assuming Turkish successes and failures shape B, and B₂ as well. First, if Turkey becomes successful, it could evaluate EU defection as representing lesser costs. According to Turkish leaders, Turkey having a better functioning democratic system, a stronger economy, and an increased influence and power through its new foreign policy might even not want an EU membership. Political reforms, for ex­ ample, are not necessarily conducted to meet EU criteria; they are tools to improve Turkish standards of living. Or, perhaps, Turks simply will not desire their country to become an EU member in the future. Some ana­ lysts think the EU membership is no more the principal Turkish national interest (Zalewski 2010: 97). Recent speeches emanating from the highest­ rank Turkish leaders indicate that a reversal in Turkish preference might actually occur in the long run.19 _{National priorities can change.}

Hence, if both types defect and Turkey is successful, 8₂ (assumed as a negative payoff) can approach zero. Turkish cost of facing EU defective moves becomes lower producing smaller values of the threshold B_{2 /} ( B₁ -B, ). Suppose that B, == 2 and B₂ takes the values of ( - 2), ( - 1) and (-�),

that is, the Turkish cost decreases. The threshold becomes equal to �' 1/J and 1/s, respectively. Turkey chooses to exit observing both types' defec­ tion provided that O � p � 1h, 0 � p � lh and O � p � 1/s. As a result, facing a full EU defection, the intervals of Turkish beliefs of interacting with the Type 1 and thus to quit the accession process shrink. At the extreme, Tur­ key is assumed to exit if it believes that it interacts only with the Type 2. Yet if it believes that there is a likelihood that it interacts with the Type 1, then shrinking intervals reveal that the maximum value of the likelihood for Turkey to quit diminishes. Thus, observing an EU suspension of the accession process, Turkey quits the accession process under progressively smaller beliefs of facing the Type 1. In other words, a successful Turkey becomes more inclined to perceive the Type 1 as the Type 2 observing defective EU moves.

Turkey's EU Accession Prospects 217

Second, the EU defection becomes costlier if Turkey fails. The European Union becomes the sole anchor for Turkey. The threshold then increases. Turkey quits the accession process through a wider range of beliefs that it interacts with the Type 1. Again, suppose that 8₁ = 2 and 8₂ takes the values of ( -1), (-2) and (-3). The threshold becomes equal to 1_{/:i, 'l2 and} 3_/s,"

respectively. Turkey then chooses to stop provided that O s p < 1h, 0 s p < 'l2, and O s p < 3/s, respectively. Therefore, if both types defect, a failing

Turkey can quit the accession process believing that the Type 1 has a higher tendency of not supporting Turkish membership.

Turkish successes and failures can affect payoffs at different rates as well. How the Type 1 ranks Turkish foreign-policy successes, domestic reforms, and Turkish booming economy would also shape its evaluations of A,, A_2, and A_3• Turkey can also fail in some domain but succeeds in others. For example, one could argue that failures in domestic reforms are inconsiderable next to Turkish economic successes for the Type 1. Or, one could argue that an economic setback would hurt domestic reform efforts or reversals in the new TFP would reflect upon economic performance of Turkey. We can propose th�t if Turkey succeeds in all three domains, the thresholds swiftly increases making time preferences obsolete for the Type 1 to cooperate. However, the critical threshold would grow at a lower magnitude if there is Turkish success in one domain only; not in all three. The future would relatively matter for the Type 1 to cooperate, for example, if Turkey's domestic reforms continue to stall, the new TFP fails more often, but Turkish economy continues to thrive.

The most prominent Turkish success is in the realm of economics fol­ lowed by foreign policy. There are almost no Turkish successes in do­ mestic reforms compared to economics and foreign policy. It seems quite difficult that Turkey becomes successful in all three fields. Thus, the Type 1 cares about the future values of cooperation with Turkey principally depending on how Turkey fails in domestic reforms. Foreign-policy and economic successes prevent the future's impact upon Type l's coopera­ tive behavior. However, the successes can produce Turkish incentives to perceive that even its traditional supporters can defect. Turkey can then give up its quest for full membership, if it slightly perceives that the Type 1 does no longer support its accession. If, for example, Erdogan's current visits in Arab countries and increased popularity of Turkey in the Middle East and the North Africa are assessed as Turkish foreign policy suc­ cesses, these developments can be argued to make the Type 1 as caring less about the future of the accession process but to make Turkey more sensitive to an EU defection. In gist, the equilibria reveal that the future becomes almost as valuable as the present for the Type 1 given Turkish successes while a successful Turkey tolerates EU defections less. Simi­ larly, the future loses its value for the Type 1 provided that Turkey fails.

218 Serrlnr $ . Gii11er

A failing Turkey quits the process only if it becomes more certain that the Type 1 opposes it accession similar to the Type 2.

Overall, Turkish successes and failures can be mixed. Turkish economic and foreign policy successes can induce the AKP leaders to become bolder during the accession process even if they fail to achieve higher standards of democracy and human rights. The leaders can assess EU defective moves as less costly compared to periods with no Turkish ac­ complishments in economics and foreign policy. The accession process would then become more conflictive than before. Hence, a Turkey with mixed successes and failures can demonstrate how cultural .and religious differences can divide states. The division can grow deeper destabilizing the EU p�riphery and show the world that a Muslim country can be a member of NATO, but not the European Union.

CONCLUSIONS

This chapter provided some insights about the EU-Turkey membership negotiations by modeling them as a simple decision game. The equilibria change when both types' preferences are fixed. The Type 1 must attri­ bute much lower values to the future benefits to cooperate with a failing Turkey; otherwise, if the future sufficiently matters for the Type 1, it will choose to behave like the Type 2. lfTurkish economic growth is sustained, Turkey achieves some success with domestic reforms towards satisfy­ ing political conditions of the Copenhagen criteria, and Turkish foreign policy continues to signal Turkey's growing power. That is, if Turkey is successful, then the division within the European Union with respect to the Turkish accession would persist as the separating equilibrium dem­ onstrates. Turkish beliefs of interacting with the Type 1 matter only when both types defect. If Turkey is successful, it becomes more sensitive to its beliefs of interacting with the Type 1. Surprisingly, the Type 1 should care more about the future to defect given Turkish failures.

Out of few game theoretic works on the EU-Turkey relations, the model resembles Ti.ire! (2005) in terms of strategies, Bayar and (ah�kan (2007) and K1bris and Mi.ifti.iler-Ba� (2008) in terms of misperceptions and informational asymmetries. None of these works presents a game that discusses two distinct periods of interaction, however. In addition, the game developed does not aim at empirical application, which is an overly ambitious goal. Game equilibria do not explain what actually happens and how the future will just look like. They solely provide conceptual clarification and insights. The pooling and the separating equilibria let us think about different scenarios; a more difficult, perhaps even impossible, endeavor under a unique equilibrium.

Turkey's EU Accessio11 Prospects 219

Our simple model reveals that Turkish beliefs and the EU time prefer­ ences are affected by 'l'urkey's future successes and failures in foreign policy, economic performance, and domestic reforms. As long as there is a division within the European Union about Turkey's place in the union, Turkey would reciprocate cooperative moves. The equilibria clarify fu­

ture concerns of traditional supporters of Turkey's accession and outline possible future paths depending on how Turkey fares.

The present model sbeds some light upon the present and the future EU-Turkey negotiations. Any game that models the EU-Turkey relations has to be a repeated game since actors must look forward about succes­ sive periods in order to make informed decisions as time passes. One­ shot games cannot answer questions like how defection would dissuade Turkey. If Turkey sustains its progress towards the European Union, then the accession process

will

NOTES

1. See www .curactiv .corn/ en/ eu-elcctions/ sarkozy-kicks-frcnch-cu-clcctions -c.unpaign/ article-182104 for Sarkozy's view and www.curactiv.com/ en/ eu-clec tions/ turkey-shockcd-franco-gcrman-clection-rhetoric/ article-182187 for Merkel's.

2. USAK, 4. USAK D1� Politika Alg1lama Anketi, Ankara, August 2(l09, avail­ able at: www.usak. org.tr/dosyalar/TDPAnket4_:rFP.pdf and German Marshall Fund, Transatlantic Trends, September 2010, available at: www.gmfus.org/ trcnds/2010/ . Sec also Transatlantic Trends: Key Findings 2009 (Washington DC:

The German Marshall Fund of the United States, September 2(X)9), 25. Available at: www .gmfus.org/ trcnds/2009 / docs/2009 _English_Key.pd f. Set? also Kent men (2008: 487- 510).

3. Available at: www.dunyabultcni.net/indcx.php?aType=habcrArchive&Art iclc10=34153.

4. Available at: www.abgs.gov.tr/files/strateji/yabs_enl .pdf. Sec also AKP Siyaset Akadcmisi Dcrs Notlan Kitapc;1g1, Arahk 2009, pp. 105-108. Available at: www.siyasctakademisi.org/habcrlcr/ Akadcmi_Kitap.pdf; "imtiyazh ortakltk kitab1m1zda yok," 10.02.2008. Available ·at: www.dunyabultenLnct/index.php

?a'J"ypc=haber Archive&ArticlelD=34153.

5. The decision of the European Court of Human Rights justifying headscarf ban in Turkey was therefore a big blow against the Turkish government.

6. The admission of the Republic of Cyprus as an EU member ,exemplifies the Greek veto as an extortionate threat (Schelling 1980: 5 0 -52).

7. Mr. Davutoglu, Turkish Minister of Foreign Affairs, complains exactly about such double standards and prejudices: www.kibrisgazetesi.com/index.php/ cat/2/ news/1 15626/ PageNamc/lc_Haberlcr.

220 Serdar $. Gii11er

8. Merkel re-iterated her position with respect to Turkish candidacy during her recent visit of Cyprus. Available at: www .neurope.eu/ articles/ 104242.php.

9. Available at: www .mfa.gov .cy / mfa/ mfa2006.nsf / cyprusl 7 _en/ cypms 17 _e n ?OpenDocument and www.mfa.gr/ www.mfa.gr/ en-USPolicy / Geo graphic+ Regions/South-Eastern+Europe / Turkey / Approach /Turkish +Accession+process/.

10. Available at: ec.europa.eu/cnlargcment/pdf/key_documents/2010/ package/ tr _rapport_2010_en.pdf, 32.

11. Erdogan RT (2011) "The robust man of Europe: Turkey has the vigor that the EU needs badly. Newsweek. Available at: www.newsweek.com/2011 /01 /17 / the-robust-man-of-europe.html.

12. European Council 2010 Progress Report on Turkey, pp. 9 - 12.

13. Avail�ble at: www.globservcr.com/en/turkey-economy.

14. Available at: online.wsj.com/ article/SB1000142405270230458400457641947 0604208418.html.

15. Available at: europa.eu/ rapid/ pressReleasesAction.do?reforence=MEMO /11 / 402&fonnat=_{HTML&aged=O&language=EN&guiLanguage=en.}

16. Turkey now has a ministry of EU affairs established after the June 2011 elections.

17. It is also possible to draw a graph with the ordinate and the abscissa indicat­ ing the threshold values and the values of cooperation for the Type 1, respectively. We think that the table conveys the relationship between payoffs and the thresh­ old better than a graph.

18. Davutoglu declared that EU-Turkey relations will be frozen if the Republic of Cyprus takes over the Presidency of the EU Council in 2012 without a solution to the Cyprus question. Available at: www.todayszaman.com/news-250301-eu -relations-to-be-frozen-if-greek-cyprus-takes-presidency-before- solution.html.

19. Turkish President Abdullah Gi.il's declarations in BBC's "Hard Talk" pro­ gram in 9 November 2010. Available at: www.intemethaber.com/abdullah-gul -ab-icin-u mu tsuz- konustu-308724h.htm # i xzz 1 J yHtQs40.

BIBLIOGRAPHY

Bayar, G. and C::ah�kan, 0. (2007) "Ti.irkiye-AB ili�kileri: Bir Oyun Kuram1 Yakla�1m1." ODTU Geli$me Dergisi, Vol. 34, No. 2, 19-62.

Bech C. E. and Snyder, J. (2011) "Religion's contribution to international relations theory." In Snyder, J. (ed.) Religion and l11ternational Relations Theory. New York:

Columbia University Press, 200-209.

Burgin, A. (2010) "Cosmopolitan Entrapment: The Failed Strategies to Reverse Turkey's EU Membership Eligibility." Perspectives: Central E11rapem1 Revil'W of International Affairs, Vol. 18, No. 2, 33-56.

Suzan, B. and Diez, T. (1999) "The European Union and Turkey." S11rvival, Vol. 41, No. 1, 41- 57.

(agaptay, S. (2009) "Is Turkey leaving the West?" Foreign Affairs (26 October).

Available at: www.foreignaffairs.com/ articles/ 65661 / soner-cagaptay / is-turkey -leaving-the-west?page=₂

Turkey's EU Accession Prospects 221 Christensen, B. M. (2(Xl9) "Mobilizing local networks for a better informed dia­ logue on Turkey's accession to the EU." Policy Brief, Danish Institute of Inter­ national Studies (May).

Cook, S. (2009) "Turkey's war at home." Survival, Vol. 51, No 5, 105-120. Danforth, N. (2011) 'Turkey: between paranoia and prosperity." Dissent, Vol. 58,

No. ·1, 18-22.

Davutoglu, A. (2(Xl1) Strntejik Deri11lik. istanbul: Ki.ire.

Davutoglu, A. (2(Xl8) "Turkey's new foreign policy vision." lnsigflt T11rkey, Vol. 10, No. 1, 77-96.

De Vreesc C. H., Boomgaarden H. G., and Semetko H. A. (2008) "Hard and soft: public support for Turkish membership in the EU." E11ropea11 Union Politics, Vol. 9, No. 4, 511- 53�).

Duff, A, (2011) EU and Turkey avoid last ditch Cypriot talks. Financial Times,

available at: www.ft.com/ ems/ s/0/ c7bf6898-2480-11e0-8c0e-00144fcab49a. html#ixzzl BflOPE6Z

Eralp-Ansan, N. (2010) "Turkiye A�1smdan Beyhude Bir AB Zirvesi Daha . . . . " TEP AV Degerlendirme Notu (Arahk). Available at: www.tepav.org.tr /upload/

files/12887076644.Turkiye_AB_lliskilerinde_ Yeni_Arayislar.pdf

Erdogan, R. T. (20'11 ) "The robust man of Europe: Turkey has the vigor that the

EU needs badly." Newsweek. Available at: www.newsweek.com/201 l /01 / 17 /

the-robust-man-of-europc.html

Erdogdu, E. (2002) "Turkey and Europe: undivided but not united." Middle East Review of l11temational Affairs, Vol. 6, No. 2, 40-51.

Gardner, R. (2003) Games for Business and Economics. New Jersey: Wiley.

Gerhards, J. and Silke, H. (2011) "Why not Turkey? Attitudes towards Turkish membership in the EU among citizens in 27 European Countries." /011mnl of Common Mnrket Studies, Vol. 49, No. 4, 741-766.

Gordon, P. and Ta�pmar, b. (2004) "T urkey's European quest: The EU;s decision on Turkish accession." Center on the United States and Europe, The Brookings Institution.

Gordon, P. and Ta�pmar, b. (2CXl6) "Turkey on the brink." Washington Qunrterly,

Vol. 29, No. 3, 5 7 -70.

Grigoriadis, N. (2(Xl6) 'Turkey's accession to the European Union: debating the most difficult enlargement ever." SAIS Review, Vol. 26, No. 1, 1 4 7 -160.

Gi.iney, A. and Karatckelioglu, P. (2005) "Turkey's EU Candidacy and Civil­

Military Relations: Challenges and Prospects." Armed Forces nnd Society, Vol. 31,

No. 3, 439-462.

Hadjit, A. and Moxon-Browne, E. (2005) "Foreign direct investment in Turkey: the

implications of EU accession." Turkish Studies, Vol. 6, No. 3, 321-40.

Huntington, S. (1993) 'The Clash of Civilizations?" Foreign Affnirs, Vol. 72, No. 3, 22-49.

International Crisis Group. Available at: www.crisisgroup.org/ en/ publication -type/ biogs/ eu-turkey-cyprus-triangle /privileged-partnership -offers-turkey

-neithcr-privilege-nor-partnership.aspx.

Jovanovich, M. N. (20Cl4) "Eastern enlargement of the EU: A topsy-turvy endgame.

or permanent disillusionment." Jo11rual of Eco110111ic lntegrntio11, Vol. 19, No. 4,

222 Serdar $. Gii11er

Jung, D. (2001) "Turkey and Europe: Ongoing Hypocrisy?" Working paper, Co­ penhagen Peace Research Institute.

Karaka�, C. (2006) "Gradual integration-a sustainable alternative for both sides."

Center for European Integration Studies (ZEI) ZEI Turkey Monitor 2 (4): 5-7. Available at: www.zei.de/ download/ zei_tur / ZEI_EU-Turkey-Monitor _ vol2nol .pdf

Karda�, $. (2010) "Turkey: redrawing the Middle East map or building sand­ castles?" Middle East Policy, Vol. 17, No. 1, 115-136.

Kentmen, <;. (2008) "Determinants of support for EU membership in Turkey:

Islamic attachments, utilitarian considerations and national identity." European U11io11 Politics, Vol. '9, No. 4, 487-510.

Kettle, M. (2010) "Disgracefully, Turkey's EU accession bid is going nowhere soon." Guardian, 29 October, 43.

K1br1s, A. and Mi.ifti.iler-Ba<;, M. (2008) "The accession games: a comparison of three limited-information negotiation designs." Unpublished manuscript. Sabanci University.

Kiri�<;i, K. (2004) "The December 2004 European Council decision on Turkey: ls it an historic turning point?" Tile Middle East Review of lntematio11al Affairs, Vol. 8, No. 4, pp. 87-94.

Kiri�<;i, K. (2009) "Transformation of Turkish foreign policy." New Perspectives 011

Turkey, Vol. 40, 29-57.

Kopri.ili.i, K. (2009) "Paradigm shift in Turkey's foreign policy." Brown Journal of lntematio11a/ Affairs, Vol. 16, No. 1, 185-201.

Kreps, D. and Wilson, R. (1982) "Reputation and imperfect information." /01m1al of Economic Theory, Vol. 27, No. 2, 253-279.

Larrabee, S. (2010) "Turkey's new geopolitics." Survival, Vol. 52, No. 2, 157-180. Morelli, V. (2010) "European Union enlargement: a status report on Turkey's ac­

cession negotiations." CRS Report for the Congress. Available at: www.fas.org/ sgp/ crs/row /RS22517.pdf

Mi.ifti.iler-Ba<;, M. (2000) "Through the looking glass: Turkey in Europe." Turkish Struf.ies, Vol. 1, No. 1, 21- 35.

Mi.ifti.iler-Ba<;, M. (2008) "Turkey's accession to the European Union: the impact of the EU's internal dynamics." lllternatio11a/ Studies Perspectives, Vol. 9, 201-219. Mi.ifti.iler-Ba<;, M. and McLaren, L. M. (2003) "Enlargement preferences and pol­

icy-making in the European Union: impacts on Turkey." European /11tegrntio11,

Vol. 25, pp. 17-30.

Nelsen, B.F., Guth, J. L., and Cleveland R. F. (2001) "Does religion matter?: Chris­

tianity and public support for the European Union." European Union Politics,

Vol. 2, No. 2, 191-217.

Oni�, Z. (2007) "Conservative globalists versus defensive nationalists: political parties and paradoxes of europeanization in Turkey." Jormrn/ of Southern Europe and tire Balka11s, Vol. 9. No. 3, 247-261 .

- - . (2009) "The new wave of foreign policy activism in Turkey: drifting away from Europeani:t..ation?" Danish Institute for International Studies Report, no. 5 (January).

--. (2010) "Contesting for Turkey's political 'centre': domestic politics, identity

conflicts and the controversy over EU Membership." /ormra/ of Co11tempornry