Burak Cop printable version send your friend

I. Introduction

The debate over the relationship between political parties and election systems which implies an "egg and chicken" type problematic still maintains its importance in political science. The question addressed by various scholars is can the number of parties that compete in a given polity be explained by the existing electoral system or is it the former that determines the latter? Or to put it more simply, what determines the number of major parties in a given polity?

The divergence lies in choosing the independent variable. Scholars advancing on Duverger's path take the electoral system as an independent variable and analyse how it shapes the party system. However another group of academics intend to turn this relationship upside down by demonstrating that it is the parties that determine the election system. Needless to mention, those who emphasize the impact of pre-existing social cleavages over the number of parties can be considered to stand relatively close to the second group of scholars since they do not regard the electoral system as the factor that primarily determines the party system.[2]

This paper seeks to analyse the outcomes of the 1999, 2002 and 2007 elections in Turkey in light of the debate mentioned above. What is striking about the Turkish case is that some major political parties that suffered heavy electoral defeats have been virtually wiped out off the political arena, whereas some other heavily defeated big parties have managed to turn into major political actors in subsequent elections. It is hoped that the examination of the Turkish case may constitute a humble contribution to the debate being mentioned.

II.A concise look at the theoretical debate

Duverger’�s Law is well known in electoral studies: "The simple-majority single-ballot system [i.e., simple plurality rule in single-member districts] favors the two-party system".[3] The same argument inevitably implies that a proportional representation (PR) system favours multi-partyism. Duverger’�s Law constitutes the classical starting point from which the two main approaches depart -namely the one that considers the electoral system to determine the party system and the opposing one that proclaims otherwise.

The logic behind Duverger’�s thesis explains the presence of a two-party system in terms of the particular perception that the plurality system imposes on both voters and political elites: Voters would seek to ’avoid wasting their votes on hopeless third party candidacies’� in a plurality system, and similarly, elites would tend to avoid wasting their sources and assets ’in launching what the voters will perceive as hopeless candidacies’�. Voters’� perception that voting for small parties means wasting the vote is the psychological aspect of the matter, as put by Farrell.[4]

Advancing on Duverger’�s path, Rae stresses that PR formulas and high district magnitudes generate more proportional outcomes than the parliamentary configuration created by plurality systems. And more proportional results surely decrease the deviation between votes’� and seats’� percentages.[5]

The district magnitude is an element of electoral systems other than the formula whose role over the number of parties is examined by "Duvergerian" scholars. Rae points to the ’curvilinear’� relation between district magnitude and the proportionality by identifying that the ’disproportionality decreases at a decreasing rate’� when ’district magnitudes increase’�. ’As district magnitude approaches infinity, so the outcome produced by every PR formula approaches perfect proportionality’� maintains Gallagher, and small parties can hardly win any seat in districts where there are less than four seats, even though the formula is a PR one, suggests Lijphart.[6]

Whatever its features are, what makes the PR fundamentally different from the plurality formulas is the assertion that ’wherever there is a proportional electoral system there is a greater likelihood of finding more parties represented in the parliament, and wherever there is a non-proportional electoral system, we are more likely to find a two-party system’�.[7] It could be said that this proposition is broadly shared by the authors who view the electoral system as a phenomenon that determines the number of parties in a given polity.

As for the "second group", i.e. academics who embrace the pre-existing party system as the determinant of election system, Colomer should precede others in mentioning. He posits that electoral systems ’are chosen by already existing political actors in their own interest’�, in other words Duverger’�s Laws should be looked at ’upside down’�: ’[I]t is the number of parties that can explain the choice of electoral systems, rather than the other way around’�. Therefore the argument is that ’multiparty systems already existed before the introduction of PR, and were, thus, more a cause than a consequence of the adoption of such inclusive electoral rules’�.[8]

With regard to shifting the system from a majoritarian formula to a PR one, Colomer suggests that majoritarian systems are likely to be set up and retained when the effective number of parties (ENP) in a polity is below four. Otherwise, he argues, maintaining a majoritarian system would be risky for the governing party because when the ENP is above four, ’any party can risk becoming an absolute loser’�. And furthermore it could not be viewed as feasible.[9]

Another proposition based on the conception that the party system prevails over the electoral system is developed by Boix. Pertaining to circumstances where a new party (or parties) is (are) about to emerge and likely to threaten the dominance of pre-existing parties, Boix posits that a number of possible scenarios could materialise. ’When the new parties are strong, the old parties shift from plurality/majority to proportional representation if no old party enjoys a dominant position’� he suggests, and adds: ’[However] when the new entrants are weak, non-PR rule will remain in place, regardless of the structure of the old party system’�.[10]

A viewpoint situated in between the two debating approaches is developed by Neto and Cox. They posit that the ENP in a given political arena does not depend on the ’additive function’� of social heterogeneity and electoral permissiveness, but it is rather related to their ’product’�. They conclude that ’a polity can tend toward bipartism either because it has a strong electoral system or because it has few cleavages. Multipartism arises as the joint product of many exploitable cleavages and a permissive electoral system’�.[11]

III. The electoral system in Turkey

Before analysing the effects of the electoral outcomes over parties’� strength in the political arena in the past 10 years in Turkey, outlining the general features of the current Turkish electoral system could be a good base step in order to furnish the unfamiliar readers with a better command over the topic.

In Turkey different versions of the PR formula have been applied since the 1961 elections. Before this particular stage, the system was a plurality one applied in multi-member districts resulting in highly disproportional results in the competitive elections between 1946 and 1957. It can be said that the PR is now well-institutionalised in Turkey, but the election system in general is somewhat not well-institutionalised. This is the outcome of a series of system changes; four major changes and several minor ones have been realised since 1961.

A purely d’�Hondt system was in use between 1969 and 1980; however, the military coup which took place on 12 September 1980 brought about a d’�Hondt system combined with two thresholds. On the district level, parties and independent candidates were required to pass the Hare quota (total number of votes divided by the total number of seats) to be eligible to win seats in the parliament. On the national level, a 10 percent threshold has been prescribed. Parties whose votes stayed below 10 percent on the national level would gain no seat, whatever their performances in particular districts are.

In 1995 the Constitutional Court abolished the district threshold but did not touch the national one. Many parties with considerable popular support, especially the Kurdish ones, have suffered non-representation in different elections since 1987. The 10 percent national threshold is the highest example of its kind in the world. There are some other countries where national thresholds are also administrated. An example for that is the German election system which has a 5 percent national threshold. However, the German system is a mixed-member one, and half of the parliamentary seats are allocated by the plurality system in single-member constituencies. In addition to that, any party failing to reach the 5 percent threshold on national level could anyway gain representation if it manages to secure at least three seats from the single-member constituencies.[12] Thus the German system, whose threshold is already half of the Turkish one, gives the regionally strong parties the opportunity to pass over the threshold. In the Turkish system there is no such by-passing path.

Depending on the number of parties being excluded from the parliament by the maxim of threshold, elections in Turkey are likely to incur highly disproportional results. The deviational (that is the deviation between votes and seats) potential the Turkish election system embraces makes it a semi-majoritarian PR system. The reason I prefer to term the Turkish system semi-majoritarian is that the very majoritarian "loser loses all" notion is infused to it. In plurality systems winners take all and losers lose all. As for the Turkish system, it could be said that winners cannot take all due to the allocation of seats according to the d’�Hondt calculation method, but losers lose all given that even a party that wins 9.99 percent of votes is not entitled to be represented in parliament.

IV. The 1999 elections

Before the 1999 elections both Turkish society and polity were under the strong effect of polarisation around the Secularism-Islamism cleavage. The Prosperity Party (Refah Partisi ’� RP) came first at the 1995 elections with slightly more than 21 percent of votes and became the major partner of a governing coalition. The minor actor of the coalition was the centre-right True Path Party (Dogru Yol Partisi ’� DYP). The latter formed a short-lived coalition with the Motherland Party (Anavatan Partisi ’� ANAP) first, another centre-right party, before becoming the RP’�s partner. The DYP-ANAP coalition would not last for a long period of time primarily due to well publicised severe personal clashes between the leaders. The DYP leader then went on to make an agreement with the Islamist RP’�s leader to form a coalition in July 1996 which would later prove to be short-lived.

The RP’�s accession to power generated intense resentment and a high level of opposition amongst the public sensitive to secularism. Parts of the population sensible to the secular character of the regime have mainly consisted of the urban population and the middle to upper-middle class strata of the people, as well as the Alevi religious minority that has faced various forms of oppression and persecution over several centuries by the Sunni Ottoman State (and by some actors of the Sunni community during the Republican era). The scepticism towards the RP was that it had a secret agenda about "bringing back" the Islamic order, namely the Sharia, which is believed to remain from Ottoman times. Some incidents to showcase the close ties between the party and Islamic references have contributed to the deepening of the negative perceptions of the secularist masses.

However, the most influential de facto political actor behind the secularist reaction to the RP has been the military. The military in Turkey can be described as a self proclaimed defender and the protector of the secularist regime. It has a considerable political weight in Turkey ’� a weight that became fairly institutionalised after the 1960 coup d’�’�tat. ’One analyst suggests that there is "a de facto dual-track government" in contemporary Turkey in which the elected political leaders are constrained to operate within parameters maintained by the military’�.[13]

Hence a broad anti-RP campaign with a "civilian" tone and participation to some degree was initiated and coordinated by the military. Although there was a ’strong support of secular civilians in the media, the universities, labor unions, and among businessmen and the intelligentsia, the fact that none of these civilian elements apparently had the political resources to promulgate such measures without the active involvement of the military’�. Eventually Necmettin Erbakan, the Prime Minister and leader of RP, had to resign on 18 June 1997 because his ’position became untenable’�. A three-party coalition led by the ANAP would be formed instead.[14]

In the lead-up to the 1999 elections, alongside the secularist-Islamist cleavage, another factor that was becoming effective in Turkish politics was the rise of nationalism. The Kurdish armed organisation the Kurdistan Workers’� Party (Partiya Karkeran Kurdistan ’� PKK) which was founded on secessionist grounds in 1978 was pursuing guerrilla warfare (rarely combined with terrorist attacks in big cities) against the Turkish state. Its leader Abdullah ’�calan was expelled from Syria in the summer of 1998 following Turkey’�s war threats against this country. After being shuttled between a number of countries; the saga being apparently facilitated by intelligent officers and political figures, he went on to the Greek embassy in Kenya where he would soon be captured by the CIA and handed over to the Turkish government.

’The capture of the PKK leader Abdullah ’�calan and his trial’� helped increase the nationalist fervour in the country’�. ’�calan’�s capture took place two months before the elections, and the rise of nationalism served most the interests of the Democratic Left Party (Demokratik Sol Parti ’� DSP) and the Nationalist Movement Party (Milliyetci Hareket Partisi ’� MHP). The former was led by the charismatic leader B�lent Ecevit, and was in government as the minority government party with the duty to bring the country to early elections. This was a party which ’had virtually abandoned social democracy [and] reinvented [it]self as an ardent nationalist’�. On the other hand, the nationalist MHP was a former neo-fascist party whose members committed many atrocities during their armed struggle against the militants of the left in the 1970s. However, since the death of its charismatic leader, Alpaslan T�rke, in 1995, the MHP has gone through a massive restructuring and reshaped itself ’with a more centrist electoral appeal’�.[15]

Both the ANAP and DYP, the two centre-right parties were eroding in terms of political strength and influence. Mesut Ylmaz, the leader of the ANAP and Prime Minister of the three-party coalition that was founded after the collapse of the RP-DYP government, ’was brought down by an opposition censure motion that charged him with corruption and links with the ’mafia’� ’�. Thus, before the elections the ’public agenda was being shaped around discussions of increasing corruption scandals’� along with other themes mentioned above.[16]

The elections were held on 18 April 1999 under these circumstances and have resulted in the following outlook:

Table 4.1: The 18 April 1999 elections’� results

The party	Vote share (%)	No. of MPs	Seat share (%)	Under- or over- representation
DSP	22.2	136	24.7	+ 2.5
MHP	18.0	129	23.4	+ 5.4
FP	15.4	111	20.2	+ 4.8
ANAP	13.2	86	15.6	+ 2.4
DYP	12.0	85	15.5	+ 3.5
CHP	8.7	-	-	- 8.7
HADEP	4.7	-	-	- 4.7
BBP	1.5	-	-	- 1.5
’�DP	0.8	-	-	- 0.8
Others	2.6	-	-	- 2.6
Independents	0.9	3	0.6	- 0.3

Source: Adapted from Erol Tuncer and Necati Danac, �ok Partili Dˆnemde Se’�imler ve Se’�im Sistemleri (Ankara: TESAV Publications, 2003), p. 44.

Information about the parties mentioned for the first time: FP (the continuation of the RP which was closed down in January 1998, Islamist with a slightly more moderate image than the RP), CHP (centre-left, highly secularist), HADEP (Kurdish party, with some links to PKK), BBP (nationalist and highly conservative) and ’�DP (socialist, with a Marxist predominance).

The two centre-right parties, the ANAP and the DYP were eclipsed by the DSP and MHP ’because voters were tired of the corruption and bickering of [their leaders]. The Islamist vote had also declined’� but the party was still a force to be reckoned with’� HADEP, the pro-Kurdish party, had failed at the national level’�.[17]

The CHP’�s failure to gain representation in parliament was noteworthy. It was the first time that the CHP found itself out of the parliament in a competitive election it participated in since 1946. It was one of the two main political currents in Turkish history for a long time. However its leader was not a popular figure compared to the other centre-left (or supposed to be centre-left) party’�s leader Ecevit. Its electoral base had obviously moved to the DSP throughout the 1990s. Thus the traditionally firm supporter of secularism, the CHP, ’seemed to offer nothing to the voter and failed to enter parliament’� as put by Ahmad.[18]

V. The 2002 elections

Following the 1999 elections a coalition composed of three parties came into existence. The major partner was the "nationalist left" DSP, the middle one was the nationalist MHP, and the junior one was the liberal ANAP. Although these three parties did differ from one another regarding their programs, ideologies and especially historical backgrounds; they pursued a considerably coherent performance in terms of EU membership reforms. However the Turkish economy was structurally fragile, and a devastating economic crisis which broke out in February 2001 signified the beginning of the collapse for the coalition partners.

The inter- and intra- harmony of the three parties deteriorated in the lead-up to the 2002 elections. In fact there was no election scheduled for 2002, but since the relations between the MHP and ANAP were particularly bad, fearing that a conspiracy was in effect against his party, the MHP leader announced he wanted the elections to be held earlier than what the normal schedule foresaw.

Meanwhile, an internal crisis broke out in the DSP. Half of the DSP deputies guided by three prominent politicians including Kemal Dervi, the non-partisan minister of economy and former Vice-President of the World Bank, formed a new party under the name New Turkey Party (Yeni T�rkiye Partisi ’� YTP). However, Dervi soon became a CHP member, and since he was quite popular amongst the public opinion, that manoeuvre became a "life kiss" for the CHP. The DSP and YTP would therefore contest in the elections as two desperate parties.

On the Islamist part, the Virtue Party (Fazilet Partisi ’� FP) had already shared the RP’�s fate and been closed down. Two different parties arose from the ashes of the FP. One was the traditional-Islamist Felicity Party (Saadet Partisi ’� SP), and the other being the Justice and Development Party (Adalet ve Kalknma Partisi ’� AKP). The AKP did seem to be different from the other Islamist parties. It was moderate in discourse, and economically far more liberal than the RP-FP-SP current which traditionally favours protectionist and state-interventionist economic policies. The AKP also seemed to abandon the anti-West rhetoric of previous Islamist parties. It was formed in August 2001 and soon started to attract a high level of public support.

A surprisingly emerging new party was the Young Party (Gen’� Parti ’� GP). That was a highly populist party led by Cem Uzan, an ambitious ’business tycoon’�. ’Professional advisers had run his campaign and given the voters musical concerts and free food, as well as much publicity in the Uzan-owned media’�.[19]

The elections took place on 3 November 2002 with the following outcomes:

Table 4.2: The 3 November 2002 elections’� results

The party	Vote share (%)	No. of MPs	Seat share (%)	Under- or over- representation
AKP	34.3	363	66.0	+ 31.7
CHP	19.4	178	32.4	+ 13.0
DYP	9.5	-	-	- 9.5
MHP	8.4	-	-	- 8.4
GP	7.3	-	-	- 7.3
DEHAP	6.2	-	-	- 6.2
ANAP	5.1	-	-	- 5.1
SP	2.5	-	-	- 2.5
DSP	1.2	-	-	- 1.2
YTP	1.2	-	-	- 1.2
BBP	1	-	-	- 1.0
Others	2.9	-	-	- 2.9
Independents	1.0	9	1.6	+ 0.6

Source: Adapted from Tuncer and Danac, p. 46

Information for DEHAP: The DEHAP was the "umbrella party" of a coalition that the cadres of the banned HADEP formed with a number of politically marginal Marxist parties.

As pointed out by �arkolu and Kalaycolu, ’the rapid rise of the AKP support in November 2002 election marks another step in the electoral collapse of centrist politics in the country’�.[20] The two centre-right parties, ANAP and DYP were eventually pushed out of the parliament after having been gradually discredited in the eyes of the electorate due to their decade-long wearisome struggle against each other and the poor performance of the ANAP in government as well as its image associated with corruption.[21] The MHP lost more than half of its vote share since the 1999 elections also because of its poor performance in government. The DSP, for the same reason plus the internal crisis it experienced and the weakening it suffered due to his leader Ecevit’�s serious health problems, was wiped out of the political arena. The voters who chose the DSP in 1999 obviously preferred voting for the CHP this time. The CHP became a concentration point for voters who were worried about the AKP’�s probable Islamist policies.

The Prime Minister and DSP leader Ecevit’�s comments on the election night were dramatic: "We (i.e. the coalition partners) have virtually committed suicide" he said, regarding the decision to go to early elections. He accused his coalition partners of asking to go to early elections before the coalition could benefit from the harvest of the works it had undertaken.[22]

Thus the following question comes to mind at this point: Why did the DSP-MHP-ANAP coalition decide to go to elections? It has already been referred to Colomer, who argues that a majoritarian system is unlikely to be retained when the ENP is above 4. Although the Turkish electoral system is technically a PR system, the effects of the 10 percent threshold make it semi-majoritarian, as has been previously argued. The national threshold is a majoritarian element that dramatically diminishes the proportionality of the Turkish system. Hence maintaining the current threshold can be said to correspond to "maintaining the majoritarian system" in Colomer’�s terminology. The majoritarian aspect of the Turkish electoral system became so apparent in the 2002 elections, in which more than 45 percent of votes could not gain representation in parliament.

The ENP in the 2002 elections was 5.4.[23] This figure is considerably higher than 4. But the coalition partners which decided to hold early elections did not change the electoral law, although their legislative strength was sufficient for that. Here it can be argued that they were poorly informed about voters’� preferences or behaved non-rationally. Or, alternatively, one can suggest that the Turkish case constitutes an exception to Colomer’�s proposition.

As to Boix’�s arguments, the Turkish electoral case of 2002 seems to deviate from the theory once again. Boix suggests that when new and powerful actors come into the political arena, a shift from majoritarianism to PR occurs if none of the pre-existing parties is in a dominant position. However, no change takes place if new entrants are relatively weak.

The birth of YTP, rise of CHP, and unexpected emergence of GP can perhaps be disregarded given that all these phenomena took place when the country had already been on the path to elections. However, the rise of the AKP following its birth was something already apparent. And more importantly, all the three members of coalition were weakening gradually since the economic crisis of February 2001. Hence the Boix arguments also are far from explaining what happened before the 2002 elections.

At this point perhaps there is no need to be critical towards Colomer’�s and Boix’�s arguments. The Turkish system is in fact a sui generis example of electoral systems since it is technically proportional, but depending on parties’� vote shares, it can become highly majoritarian in some particular elections. The problem lies perhaps in the insufficient interest of electoral studies literature towards Turkey. As an example, Lijphart and Farrell have not taken Turkey’�s election system into consideration. The reason behind that seems to be purely methodological since Turkey was termed as a partly free country by the Freedom House throughout 1990s, and Lijphart for example, looks at the election systems of countries which are considered free by the Freedom House.[24]

VI. The 2007 elections

Although the Effective Number of Parties was as high as 5.4, the 2002 elections brought about a two-party system in practice. Only two parties gained representation in parliament; the AKP which obtained enough seats to form the government alone, and the CHP as the main and sole opposition party.

The tension severely rose in Turkish domestic politics throughout 2007, the year of the new elections. The governing AKP, a conservative party executing highly neo-liberal economic policies but whose roots are linked to former Islamist parties, was challenged by both the nationalism and secular sensitivity which were on the rise. The rise of both sentiments seemed to favour the CHP, which had completely abandoned social democracy. Nationalistic fervour was also to the benefit of the MHP.

The elections on 22 July 2007 produced the following outcomes:

Table 5.1: The 22 July 2007 elections’� results

Party	Votes	Percentage	Seats
AKP	16 million	46.6	341
CHP	7 million	20.9	112
MHP	5 million	14.3	70
Independents	1.900.000	5.3	26
DP	1.900.000	5.4	0
GP	1.000.000	3	0
SP	820.000	2.3	0
Others	780.000	2.2	0

Source: Adapted from ’Election Resources on the Internet: Elections to the Turkish Grand National Assembly, at http://www.electionresources.org/tr/assembly.php?election=2007. (Figures are approximated by the author)

Information for DP: The DYP was renamed Democrat Party (Demokrat Parti ’� DP) in June 2007 after a failed attempt of merger with the ANAP, the other main centre-right party.

The number of parties represented was actually more than twice what the above table indicates. Candidates from the Kurdish Democratic Society Party (Democratik Toplum Partisi ’� DTP) ran as independents in order to pass over the 10 percent threshold. Twenty-one candidates from the DTP got elected and formed the formal group of their party in the parliament afterwards. The leaders of the BBP (Muhsin Yazcolu) and ’�DP (Ufuk Uras), the latter being supported by the DTP, were also elected as independents. They soon rejoined their parties. And finally, 13 candidates from the DSP who ran on the CHP lists went back to their original party soon after being elected. The number of parties represented in the parliament therefore rose to seven.

In comparison with the results of 2002, the 2007 elections demonstrated that the party system in Turkey is not merely shaped by the election system. The 2002 elections generated a de facto two-party system; however, parties with firm popular support showed that they were still alive. The MHP, which stayed below the threshold in 2002, obtained 4.3 percent more votes than the required level in 2007. Social cleavages played a decisive role in the MHP’�s rise, since the rising nationalistic fervour allowed the MHP to drastically increase its votes.

The DP, although staying below the threshold once again, gained more than 5 percent of votes and demonstrated it still had some popular support. On the other hand, the ANAP could not even dare to contest the elections due to the lack of public support. Unlike the ANAP, the DP is a party whose traditional roots can be traced back to the Democrat Party of the 1950s. Such a traditional link seems to help that party maintain some of its electoral support.

VI. Conclusion

As for the Turkish political arena, it can be said that neither the Duvergerian nor the opposite paradigms explain broadly enough what has been going on. This phenomenon is mainly due to the sui generis character of the Turkish electoral system in which strong majoritarian elements are infused into an "at least technically PR system". However, the stated phenomenon is not absolute either. It rather depends on the vote share of the major parties in some particular elections.

The secondary character of the Turkish political arena that makes it difficult for both main paradigms to thoroughly explain the developments in the polity is the sui generis social cleavages. The Turkish political culture is keen to be polarised around issues related to secularism and nationalism. That potential of polarisation contributes to the volatility of parties’� popular support.

It can perhaps be said that the Turkish case seems to be relatively aligned with Cox and Neto’�s approach which takes the middle course between the "Duvergerian" and "Colomerian" paradigms.

References

- Ahmad, Feroz (2008) ’Politics and political parties in Republican Turkey’� in Resat Kasaba (ed.) The Cambridge History of Turkey Volume 4: Turkey in the Modern World. New York: Cambridge University Press.

- Boix, Carles (September 1999) ’Setting the Rules of the Game: The Choice of Electoral Systems in Advanced Democracies’�, American Political Science Review, 93.

- ’B�lent Ecevit: ntihar ettik’�’�, Radikal, 4 November 2002, at http://www.radikal.com.tr/haber.php?haberno=55403.

- Colomer, Josep M. (2005) ’It’�s Parties That Choose Electoral Systems (or, Duverger’�s Laws Upside Down)’�, Political Studies, 53.

- �arkolu, Ali and Kalaycolu, Ersin (2007) Turkish democracy today. New York: I. B. Tauris.

- ’Election Resources on the Internet: Elections to the Turkish Grand National Assembly’�, at http://www.electionresources.org/tr/assembly.php?election=2007 .

- Farrell, David M. (2001) Electoral Systems: A Comparative Introduction. New York: Palgrave.

- Gallagher, Michael (1991) ’Proportionality, Disproportionality and Electoral Systems’�, Electoral Studies.

- Lijphart, Arend (1994) Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945-1990. New York: Oxford University Press.

- Lijphart, Arend (2001) Patterns of Democracy. New Haven: Yale University Press.

- Neto, Octavio Amorim and Cox, Gary W. (January 1997) ’Electoral Institutions, Clevage Structures, and the Number of Parties’�, American Journal of Political Science.

- Rae, Douglas W. (1967) The Political Consequences of Electoral Laws. New Haven and London: Yale University Press.

- Tachau, Frank (2002) ’An Overview of Electoral Behavior: Toward Protest or Consolidation of Democracy?’� in Sabri Sayar and Ylmaz Esmer (eds.) Politics, Parties and Elections in Turkey. Boulder and London: Lynne Rienner Publishers.

- Tuncer, Erol and Danac, Necati (2003) �ok Partili Dˆnemde Se’�imler ve Se’�im Sistemleri. Ankara: TESAV Publications.