"Ignorance breeds fear. Knowledge drives those fears out."

India, Kazakhstan to jointly build satellites

2007-10-01T10:38:00.000-04:00

Submitted by Indian-Muslim on Mon, 10/01/2007 - 02:27.
India News
NEW DELHI, Sept 30 (KUNA) -- Oil rich Central Asian Republic of Kazakhstan, which houses the famous Baikonur satellite launching facility, is considering to partner the Indian Space Research Organisation (ISRO), for joint projects for construction of various types of satellites and training of engineers and experts. A five-member high-level delegation led by the Chairman of National Cosmic Agency of the Republic of Kazakhstan, Talghat Mussabayev, is currently in India for discussions with ISRO Chairman G Madhavan Nair on the bilateral partnership, Indias leading English daily "The Hindu" reported Sunday.
During the week-long visit, the delegation will visit the ISRO headquarters, the ISRO Satellite Centre, the ISRO Telemetry, Tracking and Commanding Networks Headquarter and the Regional Remote Sensing Service Centre in Southern Indian city of Bangalore, the Vikram Sarabhai Space Centre in Thiruvananthapuram (Southern Indian state of Kerala), and the Satish Dhawan Space Centre at Sriharikota and the National Remote Sensing Agency in Hyderabad (both in Southern Indian state of Andhra Pradesh). Kazakhstan was looking for long-term cooperation with India as the two countries had a lot of expertise to share in terms of construction of satellites as also their launching, The Hindu said.
Kazakhstan is developing a new generation rocket launcher that would be eco-friendly. Called Baiterek, the new launcher will be capable of placing 30-tonne charge into orbit. It will be launched from a specially designed launch pad at Baikonur. The Central Asian Republic hoped to construct a dozen telecommunication satellites and put in place a satellite navigation system by 2020, the daily said.

http://www.indianmuslims.info/news/2007/sep/30/india_kazakhstan_jointly_build_satellites.html

Әлемнің ең кір қалалары: Өскемен алғашқы 30дың ішінде.

2007-09-15T14:25:00.000-04:00

Блэксмит Институты жыл сайын әлемнің экология жағынан ең кір қалалары жайында зеріттеулер жүргізеді. Қытайдың жане Ресейдің қалалары көбіне алғашқы 10дыққа кіреді. Бұл жылғы алғашқы 30дықтың ішінде Қазақстанның Өскемен қаласы да кірді. ПДФ файлын мына жерден ашып қараңыздар. http://www.blacksmithinstitute.org/wwpp2007/finalReport2007.pdf

Kazakhstan: Nazarbaev's Regional Tour Shows Growing Economic Influence

2007-09-14T15:25:00.000-04:00

By Gulnoza Saidazimova

Kazakhstan's Nazarbaev (left) and Turkmenisatan's Berdymukhammedov signed an agreement on a gas pipeline to Russia

(official site)

September 14, 2007 (RFE/RL) -- Kazakhstan has for years wanted to be the flagship country in Central Asia. It now seems to be realizing that ambition.

With its vast hydrocarbon resources, Kazakhstan has achieved nearly 10 percent annual economic growth in recent years, mostly due to its oil industry. Rising oil revenues have also allowed Kazakh companies to invest in neighboring countries.

Kazakh President Nursultan Nazarbaev visited both Turkmenistan and Tajikistan this week, holding talks focused mostly on economic issues. The visits are seen as the latest moves in Kazakhstan's increasing economic expansion in the region.

Reaching Out To The Region

In Dushanbe, Nazarbaev promised on September 13 to provide Tajikistan with grain -- a timely offer for a country where prices for bread and flour have risen sharply in recent weeks.

Nazarbaev and his Tajik counterpart, Emomali Rahmon, agreed to set up a joint investment fund, with Kazakhstan providing 80 percent of the money. "We agreed to establish a special investment fund of $100 million," Nazarbaev said. "The Kazakh side will contribute its significant part. The fund will work for the benefit of the Tajik economy. I believe it will be a good support."

Nazarbaev also said that Kazakhstan is ready to finance the construction of the Nurobod hydroelectric power plant in northern Tajikistan.

Rahmon said Kazakh-Tajik trade has increased 57 percent in January-July 2007 compared to the same period last year.

Economic issues were also at the center of Kazakh-Turkmen talks in Ashgabat on September 12. Turkmen President Gurbanguly Berdymukhammedov said economic cooperation with Kazakhstan is a "priority aspect."

Nazarbaev and Berdymukhammedov also discussed the construction of a gas pipeline that would deliver Turkmen gas to Russia via Kazakh territory. Kazakh energy companies have expressed interest in investing in the development of Turkmenistan's oil fields, too.

The two sides also discussed the railroad project going through Uzen-Gyzylgaya-Bereket-Etrek-Gorgan -- stretching from Kazakhstan to Iran via Turkmen territory.

But Kazakhstan's regional economic aspirations are not only limited to Tajikistan and Turkmenistan. In recent years, Kazakh companies have made significant investments in Kyrgyzstan's economy, with the banking sector leading the trend.

Leaving Traditional Rival Behind

Sergei Luzynanin, a professor at the Moscow Institute for International Relations (MGIMO), tells RFE/RL that Nazarbaev's recent trips to neighboring countries are part of Astana's policy to establish itself as a regional leader.

"All this proves that Astana wants to be a center of influence -- actually it has already become one. Kyrgyzstan and Tajikistan are willing to cooperate [with Kazakhstan]. They have received Kazakh loans and investments. They have gained some advantage" from the cooperation, Luzynanin says.

"But others, like Uzbekistan, are cautious about Astana's initiatives. There's been some rivalry and jealousy [between Kazakhstan and Uzbekistan]," he adds. "Tashkent has often criticized Astana's actions. It demonstrates an old, traditional antagonism."

Uzbekistan is Kazakhstan's rival for regional leadership, but it lags behind Kazakhstan in economic development. Astana introduced free-market economic reforms shortly after gaining independence in 1991, whereas the Uzbek economy remains state-controlled and still bears resemblance to the Soviet command-administrative economic system.

Luzyanin says Uzbekistan could strengthen its position in the region in a decade or more. "But Kazakhstan undoubtedly has taken a leading position in Central Asia for the time being," he says.

Russia, China Also Look To Move In

But it is not only officials in Astana who aim to develop economic ties with its neighbors. Kazakh companies -- state-owned as well as private ones -- have shown the interest and financial capacity to invest in neighboring countries' energy sectors, construction businesses, and general services. Luzynanin says the political will of the authorities has coincided with the companies' pragmatic interests at present.

Kazakhstan is not the only country expanding its economic presence in Central Asia. Russia has had a strong economic position in its former "soft underbelly." The Russian state gas monopoly Gazprom has been the major foreign partner of Turkmenistan and Uzbekistan. Russia is also Tajikistan's biggest trade partner.

In recent years, China has also greatly strengthened its economic position in the region. Many Central Asians have even expressed fears of China's "creeping expansion."

Turkmenistan has reached an agreement with Beijing on the construction of a gas pipeline that is expected to start transporting Turkmen gas to China by 2009, while Russia's Gazprom also has an agreement to export Turkmen gas.

Will these developments produce a clash between the economic interests of Astana and those of Moscow and Beijing in Central Asia? Luzyanin is optimistic that they will not.

"Undoubtedly, there are contradicting [interests]. But Astana, Moscow, and Beijing are bound by dozens and even hundreds of bilateral contracts, agreements, and treaties as well as big institutional projects, like the Shanghai Cooperation Organization [SCO]," he says. "The SCO summit in Bishkek [in August] proved that -- at least in the energy sector -- China, Russia, and Kazakhstan are interested in settling those contradictions on Turkmen gas."

As the gap deepens between the rich and poor in all Central Asian countries, there are also growing discrepancies between Kazakhstan, on the one hand, and poor countries like Kyrgyzstan, Tajikistan, and Turkmenistan on the other. Luzyanin says that this makes regional stability more vulnerable.

He adds that Kazakhstan "should help its neighbors by investing not only in, let's say, the gas sector, but also build hospitals, schools, and roads."

Біздің Университет

2007-09-14T03:05:00.000-04:00

Бингхамтон Университеті (Нью-Йорк Мемлекеттік Университеті).

Бұл суреттер өткен жылы күзде түсірілген еді. Күз мезгілі өте адемі өтеді Нью-Йорк штатында, себебі барлық жерлер орманды болып келеді де, күзде түрлі түске боянады. Біздің кампусымыз да орманның ішінде орналасқан.

Жатаханалар

Бір Аллаға сыйынып, Кел балалар оқылық!

2007-08-29T15:13:00.001-04:00

Ыбырай Алтынсарин:

"Бір Аллаға сыйынып,
Кел балалар оқылық!
Оқығанды көңілге,
Ықыласпен тоқылық!"

Өте адемі, арі өте ақылды айтылған сөз. Бұрында Совет Одағы кезеңінде бірінші сөйлем жоқ еді, Кел балалар...дан басталатын еді бұл тақпақ. Мектептің сыныптарында, кітаптарда жазылып тұратын. Алғашқы сөйлемнің маңызы мынада: иман мен білім бірге жүруі керек. Бірі болмаса екіншісі көпке жарамайды.

Fund of Science JSC presented in Astana

2007-08-21T15:27:00.001-04:00

21.08.2007 / 19:01 Fund of Science JSC presented in Astana

ASTANA. August 21. KAZINFORM /Aigul Tulekbayeva/ – National scientific-technological holding Samgau held a presentation of Fund of Science JSC in Astana on August 21.

Kazakh Prime Minister Karim Massimov took part in the presentation.

“Science Fund will promote Kazakhstan’s applied science and creation of scientific-technical research”, Kazakh Education and Science Minister Zhanseit Tuimebayev noted.

The Fund signed memorandums of cooperation with the National Information Technologies Center of the Russian Federation, National Center of Biotechnologies and Cisco Systems Finance International Company.

The Science Fund is a new institute for financing Kazakhstani science. It was established by the decree of the President of Kazakhstan in the end of 2006. In May 2007 the Fund went in Samgau State Holding.

Source: inform.kz

Videolar

2007-08-17T02:05:00.000-04:00

Keyde Youtube'tan karagan kyzyk videolardy koyip turamyn

Evolution of dance degen video kazirge deyin en kop karalgan video eken, 55 mln'nan kop karalgan.

Burynda osy videony favorite'ima koygan edim, en kop karalgandardyn arasyna kiripti. Myna jigit tamasha oynaydy gitaramen.

Балалар Жайында

2007-08-12T14:05:00.000-04:00

Қалай үйретіп қойған екен. "мына папам қандай ақымақ, қандай қандай дауыстар шығарып жатыр?" деп күліп жатқан шығар :)

Германия

2007-08-12T13:40:00.000-04:00

Келн қаласындағы катедраль. 1200ші жылдары соғылған. Өте үлкен бір ғимарат екен.

Леверкузен қаласы. Кіші гірім бір қала.

Франсузша өлеңдер

2007-08-09T01:22:00.000-04:00

Франсузша өлеңдер ұсынамын. Франсуз тілі ұнайды.

Анн Арбор, Мичиган Штаты

2007-08-08T22:04:00.000-04:00

Анн Арбор қаласында уақытша тұрып жатырмын, Мичиган штатында. Бір ай осындағы университеттен ілгері статистика мен саясаттанудағы математикалық моделдер сабақтарын алып жатырмын.
Үйдін бассейіні бар еді, сол жақтан шомылып келе жатсам шөптердің арасынан қызық дауыстар шықты, ашып қарасам богомол бір шегірткені ұстап алыпты. Шегіртке қанаттарын тыпырлатып жатыр еді, мен фотоаппаратымды әкелгенше шаршап қалыпты, богомол асықпай шайнап жеп жатыр екен.

Амстердам - Каналдар

2007-08-08T21:46:00.000-04:00

Амстердамның каналдары адемі, таза. Қаланың орталығында да шетінде де көруге болады. Былай қарап отырсаң Голландиянын көптеген атағы бар екен: каналарды, тюльпандары, сырлары, марихуанасы, проституциясы, бір жыныстық үйлену. Голландия дегенде ең алғаш еске келетіндер осылар.

Амстердам - Марихуана қаласы

2007-08-07T22:55:00.000-04:00

Амстердамның көшелерінде дүкендерде, кафелерде, барларда канапля қосылған азық-түліктің неше түрлісін табуға болады, конфеттерде, нандарда т.б.. Марихуана музейлері бар, бұрыннан бері келе жатқан бір адет ғұрыптары екен ғой :)

Амстердам - Көшелер

2007-08-07T22:24:00.001-04:00

Амстердам қызық бір қала. Теңізді жауып үстіне салынған қала. Көшелері өте тар, тек бір ғана машина жүре алады. Оған қарағанда велосипедпен қатынайды адамдар, олар үшін светофорлар, жолдар, жол белгілері... барлық жағдайлар жасалған. Машиналары да кіп-кішкентай, көбіне купэ. Джиптер жоқ десе болады. Суреттен көрген боларсыздар, үлкен машиналар ол көшелерге сыймас та.

Амстердам

2007-08-07T22:16:00.001-04:00

Амстердамда бес күн болдым, көп жерін аралап шықтым. Өте адемі қала: каналдары, үйлері, марихуана шегетін кафелері...

Винсент Ван Гог музейі. Ең ұнаған шыгармасы Алмонд Блоссом еді, яғні өрік ағашының көктемдегі гүлдерін салған. Соны сатып алдым мүзейден (оригиналын емес әрине).

Париж Суреттері - Модерн Ғимараттар

2007-08-07T21:57:00.000-04:00

Париждің тек тарихи ғимараттары ғана емес, жаңалары да адемі салынған. Яғни қаланың архитектурасына жақсы көніл бөлінген. Мына арктін аты Үлкен Арк. Шынымен де биіктігіне жане үлкендігіне таң қаласын. Ең үстіне де шықтым, сол жерден суреттер бар. Эйфель көрініп тұр. Триумф аркі де көрінеді. Осы екі арк бір біріне карама карсы салынған. Жалпы бұл жердің аты Ла Дефанс.

Париж Суреттері - Триумф Аркі

2007-08-07T09:13:00.000-04:00

Париж Суреттері - Сена өзені

2007-08-07T09:02:00.000-04:00

Сена өзенінің жағалауы өте адемі. Әсіресе кешке күн батқанда өзеннің үстінде Парижді аралап кешкі тамақ жегенге не жетсін?

Париж Суреттері - Лувр Музейі

2007-08-07T08:28:00.000-04:00

Мона Лиза да осы музейде, бірак оны суретке түсіруге рұхсат жоқ. Тамаша бір музей.

Париж Суреттері - Көшелер мен дүкендер

2007-08-07T00:34:00.000-04:00

Шанз-Элизе Авенюі. Білмедім дұрыс жаздым ба қазақшасын. Триумф аркінен басталып Лувр музейіне дейін жалғасады. Париждің ең атақты көшесі болар.

Шанз-Элизе

Мына ғимаратқа караңыздаршы.

Шанз-Элизе

Лафаетте Галлереялары: өте үлкен бір дүкен, Париждегі ен атақты маркаларды табуга болады. Бағалары қымбат болса да скидкалар болады. Бірақ соның өзінде де қымбат.

Париждің көшелері өте адемі. Әдеттегі үйлердін озі тарихі архитектурамен салынған. Көшелермен жаяу қыдырсаң ертегіде жүргендей боласың. Күлмеңдер, шын айтамын. Қазірге дейін көрген ең адемі калам менің. Кейін Парижде тұрғым келер еді (Гаухар келіссе арине).

Опера театры

Әсіресе үйлердің ең жоғарғы қабаттары тамаша салынған.

Мына отельде мен тұрған едім. Орталыққа жақын, метроға жақын. Арзан. Тағы не керек?

Парижден Суреттер - Эйфель

2007-08-07T00:14:00.001-04:00

Эйфель өте тамаша бір шығарма. Жан жағы да, үстінен көрініс те оте ұнады. Париждің ен атақты жері шығар.

Атлантик Сити суреттері

2007-08-05T02:28:00.001-04:00

Нью Джерси штатындағы Атлантик Сити деген қалашықта, Атлантикалық мұхиттың жағалауында Арнур деген досыммен демалған едік.

Political Repression and the Theory of Political Survival

2007-02-11T00:32:00.000-05:00

Galymzhan Kirbassov
PLSC 663X: Dictatorship
Political Repression and the Theory of Political Survival

Abstract: Previous literature on the level of political repression finds controversial results. Only few of them try to explain repression looking at domestic institutions of a state. Proponents of pacifying effect of democracy on the level of political repression identify the effect of domestic institutions but the results are controversial. In this study, I use the theory of political survival and observe institutions that are necessary for a leader of a state to come to power and keep his/her position. The size of the winning coalition, which is a group of regime supporters, is found to be significant in explaining the varying levels of political repression.

Introduction

Relationship between political repression and related variables has been studies extensively in political science literature. “What causes political repression”, “what is the relationship between repression and economic development, regime type and internal institutions of a state” are some of the questions scholars have sought to answer. In this sense, they find relatively consistent results in terms of explanatory factors which might cause political repression. For example, democracy has a negative influence on political repression according to many studies (Henderson 1991, Davenport 99, Davenport and Armstrong 2004, Regan and Henderson 2002, Poe and Tate 1994, and Poe, Tate and Keith 99). This notion, however, is challenged by other scholars that posit that the relationship between democracy and political repression is not a linear. Rather, it is non-linear since level of repression is positively related to threat opposition groups make to the ruling coalition (Davenport 95, Davenport and Armstrong 2004, Gartner and Regan 96, and Regan and Henderson 2002). Another explanatory variable scholars examine to understand political repression is the level of economic development. Some of them find that the higher the level of economic development the less political repression is used by the governments (Mitchell and McCormick 1988, Henderson 1991, Poe and Tate 94, and Poe, Tate and Keith 1999). British colonialism has pacifying effect on the likelihood the government use political repression. States which were colonies of Britain in the past had relatively less political repression (Mitchell and McCormick 1988, Poe, Tate and Keith 1999).

Literature on political repression and its causes is very rich in terms of different studies and empirical investigations as briefly presented above. However, most of these studies lack a convincing theory why the government or ruling elites would sometimes use political repression and sometimes not. In addition, they examine only a short time period for their empirical research. This is understandable since data on repression and other variables was very limited until recent developments in data collection.

In this study, I provide an alternative explanation why some states repress their citizens more than others using a theory of political survival (Bueno De Mesquita, Morrow, Siverson and Smith 1999, 2002) and model of dictatorships, particularly tinpots (Wintrobe 1990, 1998, and 2001). I argue that the size of group of people whose support is necessary for leader to stay in power can explain whether the leader represses citizens or not. When it is smaller, the governing elites have to repress people who are outside of this winning coalition in order to decrease the probability of emergence of opposition. Also, the level of economic development is very essential to understand the level of political repression. When a state is economically developed, people who are outside of the winning coalition do not organize among themselves in order to overthrow the current regime because the benefits of establishing a new coalition are relatively low. I test my hypotheses and find empirical support for my argument.

Literature review

Literature on repression is relatively well developed. There have been many scholars who identified different factors that influence the level of repression. Mostly, they focus on a regime type, particularly autocratic and democratic regimes, and level of economic development. I shall review these two determinants of repression in different sections.

Regime Type, Threat to the Regime and Political Repression

One of the important determinants of political repression is a regime type of state. Many scholars focus on democracies and find that probability of political repression in one state is low when the state has established democratic institutions (Henderson 1991, Davenport 99, Davenport and Armstrong 2004, Regan and Henderson 2002, Poe and Tate 1994, and Poe, Tate and Keith 99). Henderson (1991) for example argues that in democracies are more responsive to the demands of the public and use other means to react. But he provides very little theoretical argument in this study. Poe and Tate (1994, p.856) view the relationship between democracy and repression as a tautology if simply look at democracy as a broad concept without clear definition. As they argue, democracy, as a concept, is thought to have a respect to human rights and therefore, when one puts it as an explanatory variable to understand repression, it may become a tautology. They operationalize democracy differently using the Freedom House indicators and find significant and negative relationship.

Not only type of the regime matters when we try to understand the level of repression, but also threat to the regime and the nature of this threat are significant according to some scholars. Fore example, Davenport (1995) looks at multidimensional aspect of threat. He argues that level of repression can be understood by looking at the nature of threat which comes from oppositional groups or public in general to the ruling government. Democratic governments respond to this threat within certain rule established in the constitution. They tolerate demonstrations and other oppositional activities. Groups that oppose to the government also use non-violent tools which can be seen as a threat to the ruling coalition. Because elections and other institutions allow political opponents and people to punish the government in the next elections, they do not use violent tools against the ruling elites. Davenport (1995, p.686) identifies three aspects of the threat. First is the presence of violence. Using these violent strategies, oppositional groups may kill people and create a state of chaos and provoke citizens against the government which can not provide enough security. Second aspect is “the variety of strategies used by dissidents.” Availability of different strategies opposing groups may utilize is what the government pays attention. The last aspect of the threat is “cultural limits” of dissident behavior which means the level of government tolerance to the activities which is against it. Davenport argues that these aspects influence the level of repression in a state. Larger number of violent activities, availability of many strategies and deviation from the cultural limits of the government would increase the level of repression. He runs a multivariate regression and finds statistical support for this argument except for the presence of violence aspect of the threat.

Gartner and Regan (1996), building their argument on previous literature, particularly on Davenport’s 1995 article, argue that the level of repression as a reaction to the threat of opposition depends on costs imposed by domestic and international actors. To them, relationship between repression and threat is non-linear which is contrary to what Davenport argued. Davenport (1995) sees repression as a linear function of different dimensions of threat. As these dimensions or aspects of threat increase, the level of repression in a state increases as well. By contrast, Gartner and Regan (1996, p. 276) posit that the governments do not necessarily use repression in cases of threat or sometimes they escalate the level of repression and other times do not. It depends on cost benefit calculations of the government. The model works as following: opposition group makes demand and want to challenge the status quo. The government reacts to this demand and on the extreme edges either makes a total concession and gives up or initiates a civil war to eliminate the opposition. But this is not enough to understand when and to what level the ruling coalition uses political repression. Domestic economic and political elites and international actors can influence the level of repression. If the cost they impose is very large and the benefits of repression are relatively low, then the government does not use very extreme tools to repress. International and domestic costs are also a function of demands and level of repression (p.279). International actors’ tolerance to the repression is high when the government represses to the less point than the opposition demands. But if repression exceeds the point of demand, then international actors may not tolerate the repression and thus impose a large cost to the ruling coalition. Domestic economic elites, on the other hand, tolerate the repression when the level of repression is much higher than the level of demand. Since they want more economic stability they impose higher cost when the government does not respond excessively.

If the level of threat determines the level of political repression, then the nature of threat is a significant explanatory independent from the regime type. Gartner and Regan (1996) argue that whether it is democracy or autocracy, if the level of threat is low, the government does not respond repressively.

In 2002, Regan and Henderson retested their previous argument with a larger number of observations. They looked at the regime types of states since Fein’s (1995) theory predicted that semi-democracies are more likely to be repressive. In addition, Huntington (1968) suggests that underdeveloped institutions result in political instability and repression. Why would this be a case? The reason is that the threat of opposition in autocracies is very low because opposition is relatively weak. In democracies, demands of opposition can be satisfied with different non-violent tools. Also, opposition does not threat the government violently since they have a chance to be in power in the next election. This situation can be understood by using Przeworski’s (1991) argument. He posits that institutions in democracies give relevant groups chance to win in elections time to time (Przeworski 1991, 33). This gives the opposition incentive not to use violent and costly tools against the government. After empirical analysis, Regan and Henderson (2002) found that the relationship between regime type and repression is inverted U type which means that the level of repression is lowest in two full democratic and totalitarian regimes and highest in intermediate regimes with semi-developed institutions.
A challenge to these arguments came from Davenport and Armstrong (2004). They disagree with the notion of inverted U shaped relationship between democracy and repression. In their model, they decompose democracy into several categories with 1 meaning underdeveloped democracy and 10 the most developed democracy. They find that when democracy variable has a score of 7 and lower, these categories of democracy do not have statistically significant coefficients. This implies that only highly developed democracies do not repress. But those states where institutions are not developed may or may not use repression as a tool to react to the demands of opposition. The relationship between democracy and repression, as they conclude, is linear.

Although these arguments discussed above are significant in understanding repression there is at least one weakness which is overlooked or not emphasized. The weakness is the notion of seeing relationship as a demand and reaction without taking credibility or commitment problem into account. This argument resembles to study of Acemoglu and Robinson (2006) who also argue that as threat of revolution in one state increases, the ruling elites make concessions and democratize. The weakness of the argument comes from the notion of what kind of demands and commitments are credible and do impose a threat to the government. Can all opposition groups threat credibly so that the government would respond by repression? The ruling elites may not necessarily react to the demand of opposing groups if the threat of these groups is not credible. Commitment problem may come from different sources. For example, if there is a strong incentive for opposing groups to act collectively and low probability of free riding in this group, then the threat they impose on the ruling coalition is credible. The government, confronted with this credible threat, may use repressive tools to protect its loyal supporters and its ruling position. This implies that the ruling coalition may also threaten the opposition by different signals. For instance, even if the demand of opposition is low and may not necessarily challenge the status quo, the government may repress extensively and violently. This would signal how the government is ready to react to any oppositional activities and consequently make government’s commitment to repress credible. Recent repressive action in Andijan, Uzbekistan can be an example of the argument. Government’s reaction to the demonstrations was very severe relative to the demand. Thousands of demonstrators wanted some prisoners to be released, but the government shot dead hundreds of them and signaled to the public that it will not tolerate these kinds of activities in the future.

There have been also studies related to regime change. Autocratization and democratization, which are changes, were investigated instead of just a regime type. For example, Davenport (1999) argues that autocratization increases the likelihood of repression. By contrast, democratization pacifies repressive behavior of state. After empirical analysis, he finds support for his arguments. But what is not clear is that whether change of a regime or just a type of a regime at time t influences repressive behavior. In their recent study, Davenport and Armstrong (2004) find that democracy, under certain level of development, does not have any effect on repression. Regan and Henderson (2002) find that states in the middle point of democratic development are more likely to repress their citizens. As a result, it is hard to distinguish between effects of democratization and democracy.

Economic Development, Economic Growth and Repression

Scholars examine the level of economic development and economic growth separately. In most studies, their effect on the level of repression is in different way. Henderson (1991) argues that the more developed a state, the less likely the government to use repression. However, economic growth increases repressive behavior of the ruling group. The reason of different effects is not clear. He argues that nations, when experience economic growth, have a high inflation and people tend to think that the wealth disproportionately goes to the hands of rich. Repression is used by the government in order to prevent unionization, political parties or other institutions that would allow poor to demand some portion of the wealth (Henderson 1991, p.126). Level of economic development, however, has a negative effect, implying that the more developed countries use less political repression. Underlying theory of his argument is that citizens of economically developed countries are more satisfied and therefore less repression is needed (Henderson 1991, p.126). The results of their quantitative analysis show that economic growth increases the level of repression, while economic development has no statistically significant effect.

Poe and Tate (1994) have also examined this relationship along with other explanatory variables. They argue that economic growth makes a state unstable and this creates a condition where poorer people are dissatisfied with unequal distribution of wealth. The level of economic development, on the other hand, has a pacifying effect on the repression because of the factors Henderson (1991) argues. These authors find contrary results. Particularly, economic growth has no statistically significant effect on repression. However, economically developed countries have lower level of repression.

Theoretical argument of the relationship between repression and economic factors is unconvincing. Does economic growth occur only in underdeveloped states? Authors of studies mentioned above argued that economic development decreases repression while economic growth does the contrary. If a rich and developed state is growing in terms of its economy, should that effect repression positively or negatively? Results of regression models of these studies show that one of these two variables does not have statistically significant impact on repression. This implies that controlling for one of these variables, another has no effect. For example, Poe and Tate (1994) found that economic development decreased the level of repression. But economic growth, which was in the same model, had no influence, implying that it is the level of economic development, not economic growth, what matters to understand repression.

Wintrobe (1990, 1998), on the other hand, provides more convincing theory how economy of a state, particularly in tinpot autocracies, effect the level of repression. According to his model, economic growth decreases repression since people get more economic assets in exchange to loyalty. The more loyal are citizens, the less repression is necessary to stay in power. I will discuss Wintrobe’s model in detail in the next section of this study.

Theoretical Argument: Political Survival of Ruling Elites and Economic Performance
First of all, concept of political repression has to be defined in order not to confuse it with other repressions such as economic or financial one. I use Henderson’s (1991, 1993) definition of political repression which is common in the literature.

“Political repression is the use or threat of coercion in varying degrees applied by government against opponents or potential opponents to weaken their resistance to the will of the authorities. Repression can take multiple forms, but certainly this policy can include arbitrary arrest, disappearance, detention, torture, and political killing.”

This definition explains political repression in terms of weakening opponents or potential opponents which is relevant to my argument where I use the theory of political survival (Bueno de Mesquita, Morrow, Siverson and Smith 1999, 2002) and Wintrobe’s (1990, 1998) model of dictators. First, I shall review theory of political survival and model of dictators, particularly tinpots, and then build my argument on these theories.
Bueno de Mesquita et al. (BdM2S2 hereafter) (1999, 2002) argue that in all nations, regardless of regime type, there are a number people whose support is essential for the government or ruling coalition to stay in power. To survive in political arena, both in dictatorships and democracies, leaders of nations have to get loyalty and support from certain number of people. Their argument was controversial to the existing literature since dictators, be them a monarch, totalitarian dictator, or tinpot, in many studies were not accountable to anyone in the nation. BdM2S2 argue that even in dictatorships, leaders are not totally free of influence from different groups or people; but rather, they are accountable for their supporters even if it is a small number of elites.

Theory of political survival is based on two major concepts: winning coalition and selectorate. Selectorate is a subset of the whole population who has constitutional rights to choose nation’s leadership. It is up to nations whom to grant that privilege of being in the selectorate. England in 1199 had only 197 barons who had right to select the king. In democracies, almost all adult population has right to choose their government (BdM2S2 1999, p.148). In the Soviet Union, selectorate was very broad. All adult citizens had right to be a member of the Communist Party. Thus, selectorate can be very small or very large, or in the middle of these two, depending on how institutions are established. Winning coalition, however, is smaller than selectorate. In democracies, winning coalition can be understood as people who voted for the current executive. Their vote or support is necessary for the leader to come to office and stay there. In autocracies, winning coalition is much smaller and usually is economic elites and leader in other spheres. If they do not support the incumbent leader, then it is almost impossible for the leader to stay in power.

What winning coalition gets for their support or loyalty? The leader who is dependent on the loyalty of winning coalition distributes goods which are valuable to the members of the winning coalition. These goods are private goods and only given to the winning coalition to maintain their support. By contrast, public goods such as defense, political freedom and economic growth go to all citizens of the nation. Leaders of the nations have to give sufficient private goods to the members of the winning coalition since there are also oppositional groups and elites who want political power. If incumbent leaders do not provide sufficient amount of benefits, then rival can relatively easily get support of those members of the current winning coalition. But the problem is credibility of rival’s commitment to take potential defectors in the current winning coalition in future winning coalition. Rivals are always in disadvantage against incumbents as long as the incumbent leader provides enough private goods to the members of the winning coalition. Therefore, members of the current winning coalition expect to receive more benefits under current leadership than under potential leader because of the notion of credible commitment (Bdm2S2, p.151). As a consequence, leaders also prefer small size of the winning coalition because the risk of defect is much lower. What does the size of the selectorate mean in this context? As the size of the selectorate increase, holding the size of the winning coalition constant, probability of defector’s being in rival’s winning coalition decreases. This is because there are many rivals who compete for political power and if the number of eligible supporters (selectorate) is very large, then the defector of incumbent will have as many alternatives as the number of rivals. These two effects imply that the safest system for the leader is to have a small size of winning coalition and large size of selectorate. Leninist-system, in this sense, is an efficient way to survive in a political arena (BdM2S2 1999, 153).

As the size of winning coalition increases, amount of private goods to be provided to the member of this coalition decreases. As a result, there is no incentive for a leader under a large size winning coalition to provide private goods. It is better to direct resources to the public goods which are public policies. Thus, in democracies where the size of the winning coalition is very large in proportion to the adult population, members of the winning coalition can relatively easier defect the incumbent and support the rival. This is because they do not receive a significant amount of private goods, and the risk of defecting is relatively small. They can defect the incumbent if the public policy is not sufficient and get the same or more amount of public goods from the challenger once he/she is in power. In contrast, members of the winning coalition in autocratic states do not care about the public policy and do not defect as long as they get sufficient private benefits.

What does it imply? It implies that in democracies, leaders have to focus on good public policy in order to maintain support of the winning coalition. If they can not provide it, the likelihood of defection of member is very high. Thus, poor public performance is an invitation to be overthrown in democratic states (BdM2S2 1999, p.152). As a consequence, the authors of this argument posit that the larger the size of winning coalition, the better public policy is produced. They found empirical support for this claim (p.157). This means that autocratic leaders, with small size of the winning coalition, do not spend many resources on public goods since they care only about members of the winning coalition, not about the selectorate. Even if economic performance is very poor in autocratic states, leaders are not at the risk of being overthrown as long as they provide sufficient private goods to his/her supporters. This argument is controversial to the literature on political economy of dictatorships. For example, Olson (2000) argues that dictators with encompassing interest would provide a good economic policy in order to continue to get economic benefits. Budget of a dictator is very important since they have to buy loyalty and have resources to repress which require a significant amount of financial assets (Wintrobe 1998). BdM2S2 (1999) argue that in regimes where the size of the winning coalition is small, leaders have to provide private goods to his/her supporters to stay in power. But controversially, they also posit that these leaders do not care about production of good public and economic policies. Where do the resources to provide private goods come from? If leaders are concerned to survive in political arena, how long can he/she provide private goods? Wintrobe (1998) and Olson (2000) suggest that leaders with small winning coalition have to have good economy to generate wealth to either maximize power, maximize personal consumption or to buy loyalty and tools for repression. Maintaining some amount of loyalty and repression is necessary for a leader to stay in power and weaken oppositional groups.
These arguments predict different outcomes in terms of public policy and particularly economic policy. While BdM2S2 (1999) argue that autocrats with small winging coalition do not care about economic polices because they are just concerned about staying in power, Olson (2000) and Wintrobe (1998) suggest that autocrats have to produce a good economic policy since they have an encompassing interest and want to continue to generate economic wealth to buy loyalty and repression instruments. In this study, both of the models are important to build my argument and I do not see them very distinct. Rather, they complement each other. Therefore, I suggest that the argument whether autocratic leaders produce poor economic policies or not largely depends on availability of other resources apart from taxes generated from the population. If an autocratic state has, for example, natural resources which have a high demand in international markets, then the leader may not necessarily care about producing a good economic policy. By contrast, a leader of a nation where there is not much natural resources has to produce good economic policies because they do not possess resources to buy loyalty and repression tools except for taxes. The better the economy of a state, the more a leader can extract taxes and spend on his personal consumption or other expenditures. A state does not necessarily have to possess natural resources to generate wealth to buy loyalties. A strategic geographical location of the state can also produce a good deal of wealth. For example, if a state is located on major transportation ways, it can extract resources from those transporters. South Africa was in a very strategic location in this sense, especially before the construction of Suez Canal in 1869. Imperialist powers such as England had to transport their goods to and from India and Middle Eastern states through oceans, using ports in several locations. Port in South Africa was one of the important stops in their way. In terms of natural resources, Saudi Arabia has one of the world’s largest oil reserves. This gives the governing elites incentives not to produce good economic policies since they can extract sufficient resources to distribute as private goods to the members of the winning coalition. As we observe many states such as Turkmenistan, Iraq, Iran and some other Middle Eastern states that own rich natural resources reserves, general population of these countries are relatively poor which is the consequence of poor economic policies of ruling elites. Leaders in these states do not depend on taxes, rather they depend on natural or other strategic resources that can generate sufficient wealth to maintain support of the winning coalition.

Predictions of the Theory about the Sizes of Winning Coalition and Selectorate and Their influence on Political Repression

Having established and explained my general argument which is build on the theory of political survival of BdM2S2 and Olson/Wintrobe’s theories of dictator’s behavior, I shall discus what this argument implies in terms of political repression. I argue that the smaller the size of the winning coalition and the larger the selectorate, the more repressive is the regime. The logic is intuitive: leaders who have small number of supporters and large selectorate repress the selectorate in order to weaken their ability to organize collectively and oppose the government. Wintrobe (1998) suggests that the more an autocrat represses the population, the more loyalty is generated. But I argue that an autocrat does not repress all the population. Rather, he/she represses the selectorate which has eligible right to support or not and do not repress members of the winning coalition as long as they provide loyalty. By repressing the selectorate, leader can substantially decrease the likelihood of emergence of opposition. As the size of winning coalition increases, the leader can no more provide private goods and better off to focus on public goods. Repression in these states with large winning coalition should decrease since the leader relies on the effectiveness of the public policy rather than repression. If leader is able to provide a good public policy, he/she will protect his/her position in the office. Also, leader can not repress the selectorate and buy loyalties of the winning coalition since the size of the winning coalition is too large. A rational leader would rather decrease repression and increase public goods. Thus, the cost of repression decreases which is a better for leader’s budget. In autocratic states, where the size of winning coalition is small and that of the selectorate large, leader increases repression to prevent collective action among the selectorate.

Olson (2000) suggests that collective action is a function of the number of people in a group and amount of benefit which would come from the collective action. In states where leader does not repress the selectorate, the probability of collective action and emergence of opposition is higher. To reduce this probability, leader, as argued before, distribute private goods to the members of the winning coalition sufficiently enough in order to decrease the likelihood of defection, and also repress the selectorate in order to decrease the probability of collective action. Thus, in small size group or winning coalition, members may act collectively to overthrow the leader, but do not do that because the leader already gives them enough private goods and privileges, which is not guaranteed in different winning coalition. Why would leader repress the selectorate if Olson’s theory predicts that large number of people can not act collectively? To answer this question, we need to examine another aspect of Olson’s theory and combine it with BdM2S2’s theory. Large number of people in the selectorate can act collectively only when the benefit of the action is sufficiently large. Citizens in the selectorate under the dictatorship have incentive to act collectively since establishing another winning coalition and overthrowing the current provides very huge amount of benefit. When the dictator does not repress the selectorate, collective action is very promising in terms of benefits. Benefits of being in the winning coalition under the dictatorship are much larger than that under the democracy. This is why dictators use repression tools in order to weaken the selectorate and decrease the probability of emergence of opposition.

How does an economic performance influence the level of repression? I prefer to use the level of economic development rather than economic growth because of the concerns mentioned before. Economic growth is just one or more points in the level of economic development. In states where the size of the winning coalition is small and that of the selectorate is large, the higher the economic development the less a leader repress the population. The reason is that the leader can spend more on private goods and buy more loyalties from the members of the winning coalition. Doing so, leader ensures that the members will not defect. The selectorate is relatively more satisfied with the current leader because of the economic well-being. This is because relative benefit for the selectorate to constitute a new winning coalition and overthrow the incumbent decreases when the members of selectorate become wealthier due to the economic development. Knowing this, rational leader do not spend much money on repression tools and decreases the level of political repression.

In states where the size of winning coalition is very large, on the other hand, economic development decreases the level of political repression because leader relies on good public policy to protect his/her position in the office. Leaders of these countries do not repress the selectorate. The size of the winning coalition is very large and the likelihood of emergence of opposition is very high. Rational leader would not spend resources on private goods and repression. Instead, he/she would focus on public policy in order not to lose the support of the winning coalition who wants better public goods.

Model and the Hypotheses

From the argument I have made above several testable hypotheses can be drawn. First I state the hypotheses and then I provide a model to test these hypotheses.
My first hypothesis is drawn from the argument about the size of the winning coalition and the level of repression. I argue that the smaller the size the more repressive is the government. The expected relationship is negative.

Hypothesis 1: The smaller the size of the winning coalition, the higher the level of political repression.

Second hypothesis is about the size of selectorate and its impact on repressiveness of a state. Holding the size of the winning coalition constant, leaders in states where the size of selectorate is large are more repressive since they want to decrease the probability of collective action and emergence of opposition. But when the size of coalition is not constant and positively varies with the size of selectorate, the level of repression should decrease. But this is not an argument I make in this study.

Hypothesis 2: The larger the size of the selectorate, the higher the level of political repression.

Another hypothesis is that the level of economic development decreases the level of repression since the ruling coalition does not want to spend resources where the likelihood of emergence of opposition is relatively low, and incentive for collective action is not that powerful. Being in the new winning coalition is not as beneficiary as being in that coalition when the level of economic development is low, because people themselves are relatively wealthy.

Hypothesis 3: The higher the level of economic development, the lower the level of political repression.

Political Repressiont = α + β1Political Repressiont-1 + β2Winning Coalitiont + β3Selectoratet + Economic Developmentt + Regime Typet + et

Table 1 about here.

Findings

In my analysis I have used OLS model to estimate effects of the winning coalition, selectorate and the level of economic development on the level of repression. Table 2 shows the results which partially support my argument. I have expected the relationship between the size of the winning coalition and the level of political repression to be negative since to stay in power a leader have to use two tools, loyalty and repression. In states with a small size winning coalition leader represses more in order to prevent the emergence of opposition. Result of the regression shows that the expected relationship holds. The smaller the size of the coalition the more likely leader use political repression. In both models coefficient of the winning coalition variable is statistically significant and negatively related to the level of repression.

Table 2 about here

Selectorate, however, does not have any effect on the level of political repression, disappointing my expectations. One of the possible explanations of this result is that the size of selectorate does not much change when the size of the winning coalition changes. And the leader does not only repress the selectorate, but also those people who are outside of the selectorate. This is why selectorate does not have any effect on the political repression in my model.
Third independent variable which is the level of economic development is statistically significant at .01 level. This result is consistent with previous researches. As the level of economic development arises, citizens get wealthier do not consider to rebel against the government. Benefits of establishing a new coalition decreases when people become richer then before. This provides citizens disincentive to collectively oppose the regime.
The effect of the control variable, level of repression in the last year is statistically significant and self explanatory. The level of previous repressions predicts the level of future repressions.

Conclusion

In any regime, leader needs some level of support from people to stay in power. In democracies the number of these supporters is very large while in autocracies leaders have just few supporters. If a leader has and needs just a few supporters, then the number of those who does not support him/her is very large. This gives incentive to repress people who are outside of the winning coalition. However, high level of economic development decreases the level of repression. Since the benefits of being in a winning coalition are very high, it is attractive for people to establish their own coalition and overthrow the incumbent. Collective action problem is relatively low in these cases because of the large portion of benefits. But when people get richer the benefits of being in the new winning coalition decreases and people are relatively less motivated to oppose the regime.

References

Acemoglu, Daron, and James A. Robinson. (2005). Economic Origins of Dictatorship and Democracy : Economic and Political Origins. Cambridge University Press
Bueno de Mesquita, Bruce, James D. Morrow, Randolph Siverson, and Alastair Smith. 1999. “An Institutional Explanation of the Democratic Peace.” American Political Science Review 93(4): 791-807

_____. 2002. “Political Institutions, Policy Choices, and the Survival of Leaders.” British Political Science Review 32(4): 559-590
Davenport, Christian. 1995. “Multi-Dimensional Threat Perception and State Repression: An Inquiry into Why States Apply Negative Sanctions.” American Journal of Political Science 39(3): 683-713
Davenport, Christian. 1999. “Human Rights and the Democratic Proposition.” The Journal of Conflict Resolution 43(1): 92-116
Davenport, Christian A. and David A. Armstrong, II. 2004. “Democracy and the Violation of Human Rights: A Statistical Analysis from 1976 to 1996.” American Journal of Political Science 48(3): 538-554
Fein, Helen. 1995. “More Murder in the Middle: Life-integrity Violations and Democracy in the World.” Human Rights Quarterly 17:170-191
Gartner, Scott Sigmund and Patrick M.Regan. 1996. “Threat and Repression: The Non-Linear Relationship between Government and Opposition Violence.” Journal of Peace Research 33(3): 273-287
Henderson, Conway W. 1991. “Conditions Affecting the Use of Political Repression.” The Journal of Conflict Resolution 35(1): 120-142
Huntington, Samuel. 1968. Political Order in Changing Societies. New Haven, CT: Yale University Press
Olson, Mancur. 2000. Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships. Basic Books.
Poe, Steven C. and C.Neal Tate. 1994. “Repression of Human Rights to Personal Integrity in the 1980s: A Global Analysis.” The American Political Science Review 88(4): 853-872
Poe, Steven C. and C.Neal Tate and Linda Camp Keith. 1999. “Repression of the Human Right to Personal Integrity Revisited: A Global Cross-National Study Covering the years 1976-1993.” International Studies Quarterly 43(2): 291-313
Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge University Press.
Regan, Patrick M. and Errol A Henderson. 2002. “Democracy, Threats and Political Repression in Developing Countries: Are Democracies internally less violent?” Third World Quarterly 23(1): 119-136
Wintrobe, Ronald. 1990. “The Tinpot and the Totalitarian: An Economic Theory of Dictatorship.” The American Political Science Review 84(3): 849-872
Wintrobe, Ronald. 1998. The Political Economy of Dictatorship. Cambridge University Press.
Wintrobe, Ronald. 1999. “How to Understand, and Deal with Dictatorship: An Economist’s View.” Economics of Governance 2:35-58

Democratization, Competitiveness of Political Participation and War

2006-11-15T12:57:00.000-05:00

Galymzhan Kirbassov
Extensions of Linear Model

Democratization, Competitiveness of Political Participation and War

Abstract: This study aims to test the relationship between democratization, competitiveness of political participation and the likelihood of war. Literature on this topic is not relatively rich. Many of studies focus on the direct effect of democratization on the likelihood of initiation a war. However, proponents of this argument posit that democratization leads to the emergence of different groups which compete with each other in political arena, and this competition results in dangerous wars. In this study, I test both effects of democratization and the number of political groups on the likelihood of initiation a war. Results suggest that democratization does not have any effect on war. However, the number of competing groups does matter and negatively affects the likelihood of war. The more competing groups in a state, the less likely that state to initiate a war. This result is contrary to the argument made before which were not empirically analyzed.

Introduction

As democratic peace theory predicts, democratic states do not fight each other. When they have disputes they negotiate and come to agreement in a peaceful way. But is it true for newly democratizing states? What is the behavior of a state in international arena which recently experienced a regime change from autocracy to democracy? Regime change implies that either policy makers or policy making procedure have changed dramatically which means that the state is expected to behave differently than in the past. This phenomenon has become an issue of interest for many scholars. Some of them, for example Mansfield and Snyder (1995, 2002), argue that this regime change toward democracy makes states more aggressive in international realm. To them, foreign policy of the United States which considers democratization of autocratic states one of its priorities has to be implemented carefully since newly democratizing states are more war prone than established democracies. Adamson (2001) and Lomagin (1999) apply the argument to the cases of Turkish and Russian foreign policies and the former finds support. However, this argument was challenged by many other scholars such as Ward and Gleditsch (1998), Crescenzi and Enterline (1999), Enterline (1998), and Thompson and Tucker (1997a, 1197b). They test the proposition empirically and conclude that democratization do not necessarily lead to aggressive behavior of states.

What lacks from these studies is either they focus only on one case or do pure quantitative analysis on the effect of democratization on war overlooking the causal mechanism. Mansfield and Snyder (1995, 2002) argue that there will be many competing political groups in newly democratizing states. Some of them are elites of the old regime and some are newly emerged groups that oppose old elites. In this competition, elites in the government pursue nationalistic policies in order to mobilize mass support. They get involved in interstate disputes or wars to attract mass support and to strengthen their authority over constituencies.

However, this causal mechanism is seen as a given in these studies. None of the studies examine if a large number of competing political groups in democratizing states leads to interstate disputes or even wars. I argue that the larger the number of political elites that compete for political power or the government, the less these states are likely to engage in interstate disputes or wars. This is intuitive since power is not concentrated in only one hand in the societies where many groups compete for political power. Rather, power is relatively decentralized and distributed. Military leaders and bureaucrats of the old regime and economic and religious elites of new regime are dependent on each other in many affairs. Especially, when the government decides to initiate a war it has to get support from the military elites and other groups that have control over financial and crucial natural resources. This interdependency constrains the government to initiate wars or disputes. I examine this relationship in my model and find that the number of competing groups in democratizing states has a negative relationship with probability of initiating an interstate dispute.

In addition, I argue that not all democratizing states are dispute prone. Rather, major powers behave differently than other smaller states and are more aggressive in their foreign relations. When major powers experience regime change toward democracy, they are likely to initiate interstate disputes. And when the power declines, aggressive behavior of states decreases as well. This implies that looking at only major powers misleads us in concluding that democratization affects war proneness of states. The smaller a democratizing state, the less it becomes war prone.
In this study, I first review the literature on this issue and criticize arguments and methodology of previous scholars. Second, I challenge theoretical argument of Mansfield and Snyder and examine causal mechanism between democratization and state behavior. I suggest an alternative explanation why democratization does not always increase the likelihood of states to engage in international conflicts. Third, I derive hypotheses from my theoretical argument and test them empirically. Findings of the research support my argument.

Literature Review and Criticism

Mansfield and Snyder’s (1995, 2002) argument about democratization is a challenge for policy makers since they base their foreign policies on the democratic peace theory which assumes that domestic institutions are mature and stable. Voters elect leaders through fair elections and may punish them in the next elections by not giving votes. But are there only two categories of states: democracies and autocracies? What about states which are in transition from autocratic regimes to democracies? They have democratic institutions that allow voters to elect their leaders. But do we see or expect leaders of democratizing states to behave in the way as mature democratic leader do? Do institutions constrain leaders of these countries? Mansfield and Snyder (1995, 2002) argue that democratic transitions from autocratic regimes to democracies increase the likelihood of interstate wars. States that are democratizing do not have mature and stable institutions as consolidated democracies have and therefore leaders do not have constraints. Furthermore, the most important, leaders of the old regime and some of the new regime pursue nationalistic policies to overcome collective action problem in the absence of perfect institutions (Mansfield and Snyder 2002, p.299, 303) and to mobilize mass allies and support (Mansfield and Snyder 1995, p.19). Using nationalism, leaders are more inclined to get involved in interstate wars because they want to strengthen their authority over constituencies by being successful in foreign affairs (p.20). They also exploit imperfect institutions in order to control political agenda and influence information in the media and to mobilize mass public opinion (p.7, 28). In a new regime, several different elite interest groups emerge and their demands are satisfied by policies which are results of logrolling (p.31). In addition to their empirical findings, the authors support their argument with four case-studies. They investigate how democratization affected war in mid-Victorian Britain, the France of Napoleon III, Bismarckian and Wilhelmina Germany, and Taisho Japan (p.20).

There were many counter-arguments against studies and finding of Mansfield and Snyder (1995, 2002). Wolf (Wolf et. al. 1996, p. 179) argues that post-communist states are different from feudal nations and will not be war prone as they start democratizing because of the difference between ruling old elites of these two types of regimes. It seems to be plausible since the elites of old communist regimes did not possess a huge amount of property compared to the leaders of feudal states. In the period of regime change, communist elites privatize state properties in favor of themselves and therefore are not very strict in making concessions and redistributing goods and benefits gradually. In contrast, elites of feudal states used to possess a significantly larger amount of property including land before the regime change. And in case of threat from citizens after the democratization, they were reluctant to make compromise and tried to use different policies in order to escape public attention.

Most of criticisms, however, were focused on the research design rather than theoretical plausibility of the argument. Enterline (1996) claims that Mansfield and Snyder’s (1995) research design is unable to capture some periods of transition. He redesigns the model that finds that democratization does not have a significant impact on war proneness of sates. In 1998, Enterline tests the argument again by analyzing a time-series-cross-sectional dataset and examining state-year as a unit of analysis. His findings support the previous research and he concludes that “democratization is not a catalyst for subsequent short-term interstate aggression; either in the form of interstate disputes or wars” (Enterline 1998, p.403). However, Enterline does not give any theoretical explanation why democratization or autocratization would have different impact on war proneness of nations. In other words, question “what is the linkage between regime change and war proneness of states?” is not answered. He just replicates previous studies changing methods of the research. But models of his studies are left without theoretical explanation.

Thompson and Tucker (1997a) argue that war proneness is weakest immediately after regime change and greatest over the ensuing decade when states becomes stronger economically and militarily (p.144). They criticize firstly because Mansfield and Snyder’s (1995) research results show that not only democratization is dangerous in terms of war proneness, but also autocratization has influence on higher probability of war. Secondly, Mansfield and Snyder (1995, 2002) use major imperialistic powers of 19th century in their illustrative case-studies. But same theoretical argument may not be applicable to 20th century relatively small states (Thompson and Tucker 1997a, p.442). The most important drawback, Thompson and Trucker argue, is the index of democracy they have used, which was borrowed from Maoz and Russett !993) Using PCON x (DEMOC – AUTOC) index, two identical states which have equal score on democracy and autocracy, might have been considered as two different regimes because of different power concentration score (Thompson and Tucker 1997, p.447). In turn, Mansfield and Snyder (1997) tried to justify their research design but could not satisfy Thompson and Tucker (1997b, p.468) who brought additional criticism such as using nation-states as a unit of analysis instead of dyads. Mansfield and Snyder (2002, p.300) do not see any conceptual mismatch between monadic argument and democratic peace literature.

Ward and Gleditsch (1998) examine monadic aspects of democratization and war (disputes with 1000 and more deaths) and find that democratization decreases “the probability of state being involved in a war, either as a target or an initiator” (p.59).

Gleditsch and Herge (1997) have examined the relationship between democratization and interstate conflict from systemic approach. As Waltz (1979) and other realists argue, behavior of states are influenced by some characteristics of international system. Uncertainty and security issues force countries to behave in certain ways. As a result of regime change, the number of states in the system might be increased or decreased. Consequently, newly emerged states are mystery for others in terms of which foreign policies they will pursue. Therefore, pre-established balance of in the system may change.

Gleditsch and Herge (1997) ask three questions about democracy and war in three different level of analysis, namely dyadic, national, and system levels. Contrary to Mansfield and Snyder’s argument, their research result shows that democratization, at the system level, occurs simultaneously with decreasing violence, especially where there are more democracies in the system (Gleditsch and Herge 1997, p. 308). Does this imply that democratization decreases the probability of states to involve in interstate disputes?

Could there be some factors that led the authors to this result and conclusion? In cross-temporal analysis, there are some threats to internal validity of research. One of them is the threat of historical facts that happened independently of causes scholars examine. World War I and II may very much affect the likelihood of conflict involvement of those states that were democratizing, independently of the causal mechanism. In addition, there was much less numbers of democracies in the periods of “the first and second wave” of democratization. In 1942, for example, there were only 12 democratic states in the world, while in 1960s the number was about 36 (Huntington 1991-1992, p.579). Both of these situations, considered together, may have significant effect on Gleditsch and Herge’s findings, as they claim that the more the number of democracies in the system the less the likelihood of conflict involvement of democratizing states. There have been more democracies after 1946 and the number of wars has decreased. Consequently, causal factors, whatever they might be, may not be measured appropriately when the threat of history is not considered in the model.

Other scholars, Crescenzi and Enterline (1999), examine the systemic and regional relationships among democracy, democratization and interstate conflict using vector auto-regression approach. They find that the proportion of democracy in a system and the number of war have a parabolic relationship. In terms of the relationship between democracy and democratization, in African and Middle Eastern systems, democratization positively affects the number of democracies, and the proportion of democracy positively affects democratization in European, Asian and Western Hemisphere systems (Crescenzi and Enterline 1999, p.92).
Maoz (1989) conducted one of the earliest and extensive studies in this. He sought to find out whether post-independence state formations and regime changes had effect on conflict proneness of these states, and how newly founded states influenced conflicts at the system level. According to the author, states were formed generally in two ways: in the long period of time in the process of evolution; and sharply and violently revolution. This categorization is important in terms of first, their engagement in conflicts, and second, their impact on system level conflicts. Evolutionary states did not increase conflicts since their expectations were stable and adaptation to the existing system was gradual and systematic. By contrast, revolutionary states that were formed as a result of violent struggles were more conflict prone and were seen as potential enemies by the existing states. Uncertainty that new leaders of these revolutionary states cause in the system forces other states to be cautious. Moreover, redistribution of benefits might be an issue by the formation new states. Maoz found that new and old revolutionary states were effective in increasing interstate conflicts, while new and old evolutionary states were not. And the number of states in the system did not itself explain international conflicts, but rather the number of new revolutionary states entering the system was a significant cause of these conflicts (Maoz 1989).

Not only did scholars study Mansfield and Snyder’s argument quantitatively, but also some of them have examined some cases. Lomagin (1999) applies the theory to Russian foreign policy and argues that nationalism has been rising in Russia since its independence days. But what is interesting is that public is less nationalist than elites according to some public opinion surveys. There are many elites or nationalist groups competing in political arena of Russia and nationalism serves as a substitute for communism for binding new, old and military elites. Lomagin describes nationalistic tendency in Russian politics but he does not study whether this nationalism among political elites resulted in interstate wars or important disputes. On the contrary, he states that Russia avoided “bloody ethnic conflicts” which has affected other post-Soviet states. This implies that even though democratization leads to nationalism among political elites it does not necessarily result in interstate wars as Mansfield and Snyder argue.

Adamson (2001) applies this argument to the case of Turkish foreign policy in 1974 when there was a dispute among Greece and Turkey because of the situation in the independent island of Cyprus. In July 1974, there was a military coup in this island which was sponsored by military regime in Greece. Turkey intervened Cyprus in mid-July and August to defend Turkish population in the island. Adamson concludes that democratization process led to interstate war in this case. However, as Adamson herself states (p. 299), not mid-July, but mid-August intervention can be explained by Mansfield and Snyder’s argument since the government of Turkey did not seek to intervene immediately in mid-July. Rather, they wanted to resolve the problem peacefully by negotiations. After negotiations failed, Ecevit, the prime minister of Turkey at that time, decided to use military force. Why first intervention can not be explained by the argument stated above and the second can be? Why democratization did not cause aggressive foreign policy in the first case and did in the second? It is arguable that the second intervention is a result of democratization and nationalistic policies because Turkey and Greece had many other disputes and wars in the past. History of Turkey and Greece is full of small and large disputes since the times of the Ottoman Empire. This made it possible to accumulate nationalistic feelings in these two states against each other. And therefore there was a huge public support for these two interventions in Turkey. Moreover, public was even afraid that Ecevit would not initiate an intervention. The government would have been punished if it did not intervene.

On the whole, literature lacks to test causal mechanism between democratization and conflict proneness of states. Does democratization cause interstate dispute? Are there many groups competing for the political power in democratizing states? If there are, then do these groups pursue nationalistic policies to get mass support, resulting in interstate disputes and wars? Mansfield and Snyder (1995, 2002, Snyder 2000) try to explain the causal mechanism with nationalistic policies of leaders of new states. They partly rely on diversionary theory of war (Mansfield and Snyder, 2002, p.304) which states that leaders, troubled by low performance in domestic politics, want to involve in interstate disputes and pursue adventurous foreign policies in order to get support of domestic actors such as voters and interest groups (Smith 1996). Using nationalistic foreign policies, leaders seek to get mass support. However, one of the assumptions of diversionary theory is that democratic institutions are consolidated, elections are held and fair, voters, media and interest groups may punish their positions. Do institutions in newly formed democratizing states or in states where the regime has recently been changed functions perfectly? The answer is obvious: no. In their studies, Mansfield and Snyder (2002, p.299, 303) state that media and other institutions in democratizing countries are imperfect and leaders usually exploit them to get mass support. But controversially, they use diversionary theory of war to justify their argument. Why would leaders of newly democratizing countries involve in costly interstate wars to improve their image when there are many other “cheaper” alternative ways, such as misusing the media, organizing unfair elections, consequently when not being threatened to lose their position? Gelpi (1997) examined the relationship between regime type and diversionary foreign policy, and found that democratic leaders are more prone to use diversionary policies than autocratic leaders. Diversionary theory may be applied to countries with a significant level o democratic consolidation. Miller (1995) showed that in democratizing states or in more autocratic nations, leaders do not necessarily involve in wars. In short, nationalistic foreign policies adopted by leaders of democratizing states may not be accounted as diversionary foreign policies, implying that the relationship between democratization and nationalistic foreign policies is unclear and unexplained in the studies of Mansfield and Snyder.

Theoretical Framework

Mansfield and Snyder (1995, 2002), in developing their argument about how democratization increases the probability of conflict, do not give satisfactory explanation about why democratizing states pursue nationalistic policies and as a result of it, engage in interstate conflicts. They lack to explain why some of “third wave” (Huntington 1991-1992, 1991) democratizing states do not necessarily pursue nationalistic politics and as a consequence, initiate interstate disputes. They argue that Kazakhstan, Kyrgyzstan and some other post-Soviet states were late developers because of relatively late achievement of literacy and high employment in agricultural sector, and that’s why these states did not need nationalistic politics (Snyder 2000, p.196). However, at the time of the regime change, from communist system to democracy, these nations had the same level of literacy as other post-Soviet states.

Mansfield and Snyder give Azerbaijan, Armenia and Serbia as examples of recently democratizing states which pursued nationalistic politics and got involved in interstate disputes. If historical facts are taken into account, these countries had a lot of conflicts and large scale disputes with each other. Azerbaijan and Armenia, during the collapse of the Ottoman Empire, had many hostile conflicts which were calmed down in the Soviet era. But as these states became independent, conflicts over territory started again. Serbia, one of the former states of Yugoslavia, also had many violent conflicts, particularly with Muslim nations, in the last periods of the Ottoman Empire. Balkan region was especially unstable in 18th and 19th centuries and had many conflicts among nations living there. So, Serbian invasion to Bosnia in 1995 was not just a pure nationalistic policy of Serbian leaders to get mass support, but rather, a continuation of historical disputes.

Other states Mansfield and Snyder analyzed were mid-Victorian Britain, the France of Napoleon III, Bismarckian and Wilhelmina Germany, and Taisho Japan (Mansfield and Snyder 1995, p.20). It is obvious that these states were involved in many conflicts in different periods since they were imperialistic and major powers of that period. It is hard to differentiate between influences of imperialistic and nationalistic politics on motivations of leaders’ motivations to get engaged in international wars.

Theory and the Model

First of all, I argue that Mansfield and Snyder’s (1995, 2002) model is not a proper one to quantitatively evaluate their argument. In their studies independent variable is regime change. However, they explain that war proneness of a new regime is caused nationalistic policies elites tend to follow in the first phase of democratization. Nationalism itself is caused by democratization. Then, the relationship between democratization and war proneness of a state is not direct. Rather, after democratization, many competing groups emerge in a new regime and the government pursues nationalistic policies to get their support and to unite them. In the end, this behavior causes an aggressive international behavior. This implies that there are several stages in their argument. Democratization causes the larger number of competing groups, this results in choosing nationalistic policies, and as a result, a state initiates an interstate dispute or war. However, they estimate direct model and do not include the number of competing elites and whether a new state pursues nationalistic policies. Rather, regime change is the only independent variable. This is why I argue that their model does not capture elements of their argument.
I support the idea that many or at least two competing groups emerge as a result of any democratization. This is intuitive since concept of democracy implies that there are one government and some other opposition groups in political arena. These groups or elites which are outside of the government may or may not necessarily establish political parties because in the early phase of democratization institutions may not work properly to make it possible for groups to establish parties. Groups may include different elites depending on what kind of regime was the old one. For example, there could be military leaders who had strong influence on the previous government. Even in autocracies there are many groups that affect policy choices of the regime. A leader of an autocratic regime has to get support of these groups in order to survive. These groups or individual elites are similar to voters or supporters of the government in democracies. The regime has to take them into account when choose one or another policy. Wintrobe (1998) and Acemoglu and Robinson (2006) also elaborate this phenomenon and conclude that the dictator in autocratic regimes has to get loyalties of these groups in order to stay in power. In the Soviet Union, there were many bureaucratic elites who became very strong and influential as a result of horizontal collusion in the system (Wintrobe 1998, Olson 2000). In any system where bureaucracy exists, vertical or hierarchical relations begin to weaken as elites in different levels or horizontal lines of this system start to organize among themselves and collude. It is possible to overcome collective action problem when the number of members in a group is small. Even in the strong dictatorial Soviet regime were groups able to organize and collude (Olson 2000). This implies that when a state experience a regime change from autocracy to democracy, these elites or groups do not disappear, but rather establish themselves as new elites of the new regime. They still play an important role in policy choices of the new government. In addition to these old elites, new groups and elites emerge as a result of democratization including business elites who control substantial amount financial and natural resources.

What this argument implies in respect to this study? I argue that the larger the number of groups or elites in new regime, the more it becomes difficult for the government to implement its ideal foreign policy choice. In a spatial model of policy choices, we can easily see that a specific policy is chosen through concessions. As the number of veto players increases, the more concessions have to be made from the ideal point. This implies that foreign policy choice, one of the most important policies within a state, is not very easily implemented. Even in non-perfect democratic institutions can veto players affect the government and make it difficult to use foreign policy on its behalf. This is especially the case when different groups compete for political power. When the government misuses the resources or pursues policies that do not necessarily favor voters, opposition groups use this as a tool to get public support, making it more difficult to pursue policies at ideal point. As a consequence, the government does not use aggressive foreign policies and get involved in very costly wars.

This notion can be also derived from Olson’s (2000) theory about collective action in small and large groups. He argues that small group can overcome collective action problem relatively easily since group members have direct benefits from the action and free riding is difficult. On the contrary, groups with large number of members can not act collectively because of small amount of benefits and incentive for free riding. To apply this theory to the argument of this study, when there are many opposing elites in the political arena, it is difficult to overcome collective action problem. The government has to get support from some of these elites including military and business elites to initiate costly interstate wars. But these elites also compete for the political power and have incentive to oppose the government. This notion makes it difficult for a state to get involved in interstate wars. Only with less number of competing elites can a government overcome collective action problem and initiate disputes. Therefore, the relationship between the number of competing groups and the probability to initiate a dispute is, as I argue, negative.

Hypotheses

From the theoretical argument I have made above I derive my hypotheses. Many scholars did not find statistical significant relationship between democratization and war initiation, while others found a negative relationship. In this context, I do not argue that a regime change toward democracy has some kind of relationship. My intuition is that the regime change does not affect aggressiveness of state’s behavior. What is important is the causal mechanism which is argued above. I state the same hypothesis as many other scholars did.

Hypothesis 1: democratization has no relationship with the likelihood of initiating wars.

My main argument is that if there are large number of groups competing for political power, initiating a costly war is difficult for the ruling leader. This is a causal mechanism which is stated and argued in length, but never empirically observed by Mansfield and Snyder (1995, 2002). As democratization occurs, there will be many groups competing and the ruling elites will use nationalistic policies to get mass support and this, in the end, results in aggressive foreign policy such as war. I test the relationship between the number of competing groups and the likelihood of initiating a war.

Hypothesis 2: the larger the number of competing groups for political power, the less the likelihood of initiating wars.

Model and Measurement

To find out if the results of empirical analysis support my argument, I construct several models.

Model 1:
PWart = α + β1DEMOCt + β2CAPt + β3WW2t + β4WW1t +ε t

Model 2:
PWart = α + β1COMPARt + β2CAPt + β3WW2t + β4WW1t +ε t

Model 3:
PWart = α + β1DEMOCt + β2COMPARt + β3CAPt + β4WW2t + β5WW1t +ε t

where,
PWart: Probability of initiating a war in year t.
DEMOCt : Democratization in year t.
COMPARt: Competitiveness of Participation which is the number of influential groups competing for political power in year t.
Control Variables:
CAPt: Capability of a state in year t.
WW2: Dummy variable for the World War II in year t.
WW1: Dummy variable for the World War I in year t.
ε t: Stochastic error term in year t.

My unit of analysis is a nation-year, which I have generated with 3.100 version of Eugene software (Bennett and Stam 2000). I use Militarized Interstate Dispute (MID) dataset of the Correlates of War project (Small and Singer 1982; Singer and Small 1994) for my dependent variable, which is the probability of initiating a war. Since I argue that democratizing states are not necessarily war prone, war initiation would be an appropriate variable to use. MID dataset has all interstate militarized disputes occurred in 1816-2001 period. In spite of the limitations of the MID dataset, all studies which are discussed above use it. To have a comparable argument, I had to use the same dataset. MID dataset has several hostility levels of interstate disputes varying from conflicts without any military intervention to war. . Since I and other scholars argue about the initiation of war, I measure my dependent variable as war which is the highest level of hostility.

To operationalize my independent and control variables, I used Polity IVe and MID datasets. Democratization in year t is obtained by lagging polity scores of states in year t. Polity score of one state is the difference between democracy and autocracy scores which varies from -10 (full autocracy) to +10 (full democracy). (Polity Scorei = Democracy Scorei –Autocracy Score I ). Democratization, in this study, means 3 point change of a polity score toward democracy. This is consistent with studies of democratization. Also, Marshall and Jaggers (2005, p.28) measure transition as a 3 point change on the polity score. If the change of polity score is at least 3 and is positive, then it implies that state is moving towards democracy and I put a dummy 1 and 0 to capture it. In a different model, which I did not present in this study, democratization variable was not 0 1. Instead, it was difference of autocracy and democracy scores. But the results were the same and I preferred to use 0 and 1.

To observe impact of initial phase of democratization on war initiation, after coding democratization I have added two more dummies. Thus, probability to initiate a war is observed on the 1st, 2nd and 3rd year of democratization. As Mansfield and Snydar (1995, 2002) and other scholars suggest, there is a relationship between the initial stage of democratization and war.

Competitiveness of participation variables measures the number of competing groups which are influential in political arena. This is central to my argument and needs a special attention. To operationalize this variable, I have used Polity IV dataset. The variable measures the level of competitiveness varying from 0 to 5. For example, 1 means that “no significant oppositional activity is permitted outside the ranks of the regime and ruling party” (Marshall and Jaggers, 2005, p.25). The highest level of this variable is 5 which is defined as “there are relatively stable and enduring, secular political groups which regularly compete for political influence at the national level” (p.26).

Other control variables are taken from MID dataset. Capability variable indicates military capability of a state. It is a composite index which is calculated using several indicators such as the number of military personnel, military expenditure, energy consumption and iron and steel production. World War 1 and 2 variables are dummies for years these wars occurred.

There have been 163 wars in the world during the period I am observing. This is not exact number since a war involves at least two states. My unit of analysis is in a monadic level and therefore includes just states and years. As a result, approximately half of 163 which is somewhat 80 is realistic number of wars happened from 1816 to 2001.

334 states in this period experienced democratization. Democratization, as discussed above, is a 3 point change on regime score toward democracy.

Table 1 about here

Since I want to examine the probability of initiation of interstate dispute, my dependent variable is binary, 0 and 1. Ordinary Least Square model would not be an appropriate method to estimate slopes of the independent variables. Instead, I use logit regression model, which is relevant and suggested by Gujarati (2003).

Findings and Discussion

Democratization does not lead to wars. I reject my null hypothesis since my first hypothesis states that there is no relationship between democratization and war initiation. In table 2, we can see that the coefficient of democratization variable is not statistically significantly different from zero. This finding is consistent with variety of studies of democratization and war which are discussed in the literature review section of this study. One of the important limitations of using regime changes as an explanatory variable is that in coding procedure, even if a score of a regime changes 3 points toward democracy, it still may be an autocratic state. And also, it is probably a regime type, not a regime change, what matters in terms of the probability of initiation a war. If a regime changes from -9 to -6, it still represents a regime type of -6. Then it is hard to argue that this change made certain influence. Democratic peace theory predicts that regime type is significant in state’s behaviors.

Another explanatory variable, competitiveness of participation, which represents the number of political groups competing with each other for political power, is statistically significant at .05 level of confidence. Its sign indicates that I reject my null hypothesis. The more political groups compete in political arena, the less that state is likely to initiate war. This causal mechanism was never been examined, as far as I know. Mansfield and Snyder (1995, 2002) argue that democratization leads to competition among groups who want to rule the nation. As competition gets stronger, the ruling government uses nationalistic ideology to attract mass support, ant this leads to war initiation. There are multiple steps from democratization to war initiation. However, Mansfield and Snyder and other scholars examined just the relationship between democratization and the likelihood of war initiation.

How much a competitiveness of political participation affects the likelihood of initiating a war? This can be calculated using the coefficient of Competitiveness of Participation variable. On average, one unit increase in the level of competitiveness decreases the likelihood of war initiation by .0013 units or %0.13. In substantive term, this magnitude is not insignificant since 0.13% is not a small change in the probability.

Control variables are statistically significant as expected. I include capability of a nation as a control variable because along with other factors, military capability is one of the most important factors that affect the probability of war initiation. On average, on unit change in the capability index results in 12.8% increase in probability of initiating war. This is a significant impact and should be included in estimations. Unfortunately, Mansfield and Snyder did not have control variables in their models when they were estimating the impact of democratization on the likelihood of initiating war.

World War I and II are included in the model since being in the period of these two major wars increases the likelihood of entering war. They are statistically significant at .01 level.

Table 2 about here

Conclusion

This study aims to test the causal relationship between democratization and the likelihood of initiating a war. Although many studies have been published on this topic, they mostly focused on the direct relationship. Mansfield and Snyder, however, argue that democratization leads to the competition among different new and old elites. It is intuitive that the regime change toward democracy will allow many groups to compete in political arena for becoming the government or staying there once they reach. What is arguable and not convincing in Mansfield and Snyder’s studies is that this competition will result in war. The ruling coalition or elites will use nationalism as a tool to get mass support and will initiate a war. I tested this relationship, namely, the effect of the number of competing groups in a state and the likelihood of initiating a war. The results of estimation suggest that democratization itself does not affect this likelihood in negative or positive way. Furthermore, the number of competing elites seems to decrease the likelihood of war initiation, which is contrary to what the authors argue.

Lastly, I propose some ideas that can be studies in the future. These ideas or propositions are related to this study and may be derived from the results I found. First, in the preliminary stage, democratization may increase the likelihood that democratizing state to join international organizations and be more cooperative. The reason is that the previous regime is gone and all international cooperative agreements are no more valid. Second, occurrence of war between two states before democratization may increase the likelihood of these two states’ involvement in conflicts after democratization. In case studies, Mansfield and Snyder (1995, 2002) provide many examples of wars initiated by democratizing states. But mostly, those states previously had conflicts and this could affect behaviors of leaders to initiate a war again after the regime change. Third, democratization may increase the likelihood of initiating wars if democratizing state is an imperialistic state.

References

Acemoglu, Daron, and James A. Robinson (2005). Economic Origins of Dictatorship and Democracy: Economic and Political Origins. Cambridge University Press
Adamson, Fiona B. (2001). “Democratization and Domestic Sources of Foreign Policy: Turkey in the 1974 Cyprus Crisis.” Political Science Quarterly, 116(2): 277-303
Bennett, D. Scott, and Allan Stam (2000). “EUGene: A Conceptual Manual.” International Interactions, 26:179-204
Crescenzi, Mark J.C. and Andrew J. Enterline (1999). “Ripples from the Waves? A Systemic, Time-Series Analysis of Democracy, Democratization, and Interstate War.” Journal of Peace Research, 36 (1):75-94
Enterline, Andrew J. (1998). “Regime Changes and Interstate Conflict, 1816-1992.” Political Science Quarterly, 51(2):385-409
Fordham, Benjamin O. (2006). “What Makes a Major Power?” Manuscript prepared for presentation to the annual meeting of the International Studies Association, San Diego, California.
Gelpi, Christopher. (1997). “Democratic Diversions: Governmental Structure and the Externalization of Domestic Conflict.” The Journal of Conflict Resolution, 41(2): 255-282
Gujarati, Damodar. (2003). Basic Econometrics, 4th Edition. New York: McGraw-Hill
Huntington, Samuel P. (1991). The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press
______. (1991-1992). “How Countries Democratize.” Political Science Quarterly, 106(4): 579-616
Lomagin, Nikita. (1999). “Democratization and Foreign Policy of Russia.” Abstract of the Final Report. Available at: http://www.nato.int/acad/fellow/97-99/lomagin.pdf
MacNair, Brian. (1991). Glasnost, Perestroika and the Soviet Media. UK: Routledge
Makarov, Valery L. (1988). “On the Strategy for Implementing Economic Reform in the USSR.” American Economic Review, 78(2): 457-460
Mansfield, Edward D., and Jack Snyder. (1995). “Democratization and the Danger of War.” International Security, 20(1): 5-38
______. (1997). “A Reply to Thompson and Tucker.” The Journal of Conflict Resolution, 41(3): 457-461
______. (2002). “Democratic Transitions, Institutional Strength, and War.” International Organizations, 56(2): 297-337
Maoz, Zeev. (1989). “Joining the Club of Nations: Political Development and International Conflict, 1816-1979.” International Studies Quarterly, 33(2): 199-231
______, Zeev, and Bruce Russett. (1993). “Normative and Structural Causes of Democratic Peace, 1946-1986.” The American Political Science Review, 87(3): 624-638
Marshall, Monty G. and Keith Jaggers. 2005. Polity IV Project: Dataset Users’ Manual. George Mason University
Miller, Ross A. (1995). “Domestic Structures and the Diversionary Use of Force.” American Journal of Political Science, 39(3): 760-785
Morgan, Clifton T., and Sally Howard Campbell. (1991). “Domestic Structure, Decisional Constraints, and War: So Why Kant Democracies Fight?” The Journal of Conflict Resolution, 35(2): 187-211
Olson, Mancur. (2000). ………………………………………
Russett, Bruce. (1993). Grasping the Democratic Peace: Principals for a Post-Cold War World. Princeton University Press
Singer, J. David, and Melvin Small. (1994). Correlates of War Project: International and Civil War Data, 1816-1992. Computer File. Ann Arbor, MI: Inter-University Consortium for Political and Social Research. ICPSR 9905
________, J. David, Stuart Bremer, and John Stuckey. (1972). “Capability Distribution, Uncertainty, and Major Power War, 1820-1965.” in Bruce Russett, ed., Peace, War, and Numbers. Beverly Hills, CA: Sage
Smith, Alastair. (1996). “Diversionary Foreign Policy in Democratic Systems.” International Studies Quarterly, 40(1): 133-54
Snyder, Jack L. (2000). From Voting to Violence: Democratization and Nationalistic Conflict. New York: W.W Norton & Company
Thompson, William R. and Richard Tucker. (1997a). “A Tale of Two Democratic Peace Critiques.” The Journal of Conflict Resolution, 41(3): 428-454
_______. (1997b). “Bewitched, Bothered, and Bewildered: A Reply to Farber and Gowa and to Mansfield and Snyder.” The Journal of Conflict Resolution, 41(3): 462-477
Waltz, Kenneth N. (1979). A Theory of International Politics. New York: Columbia University Press
Ward, Michael D., and Kristian S. Gleditch. (1998). “Democratizing for Peace.” The American Political Science Review, 92(1): 51-61
Wolf, Reinhard, Erich Weede, Andrew J. Enterline, Edward D. Mansfield and Jack Snyder. (1996). “Correspondence.” International Security, 20(4): 176-207
Wintrobe, Ronald D. (1998). The Political Economy of Dictatorship. Cambridge University Press

Did “the Clash of Civilizations” Find Empirical Support?: Impact of Civilizations on Militarized Interstate Disputes of the Post-Cold War Period.

2005-12-20T14:32:00.000-05:00

Galymzhan Kirbassov
PhD student of Political Science, SUNY at Binghamton.

Abstract. Huntington’s “clash of civilizations” theory has been one of the significant debates among political scientists, bureaucrats of foreign affairs and other social scientists. Many of his critics have responded in different publications, but few of them have conducted quantitative analysis. This article aims to evaluate Huntington’s argument by analyzing the impact of civilizations on militarized interstate disputes of the post-Cold War period. Finding of the study did not support clash of civilizations theory.

Introduction.
Samuel Huntington first published his article “The Clash of Civilizations?” in 1993. Later in 1996 he published a book The Clash of Civilizations and the Remaking of World Order, giving more detailed and broader explanation about his theory and providing more cases to support his argument. Debates among not only political scientists, but also scholars of different social studies and bureaucrats followed his both studies.

Huntington, in his studies, (1993, 1996) suggests alternative explanations to understand conflicts and disputes in the post-Cold War period. To him, “culture and cultural identities, which at the broadest level are civilizational identities, are shaping patterns of cohesion, disintegration, and conflict in the post-Cold War”. (Huntington 1996, 20).

Purpose of this paper is to test Huntington’s (1996) argument by addressing following question: “Does a civilizational difference cause militarized interstate dispute?” Significance to answer this question is obvious: if Huntington’s claim to be empirically supported, then the dominant theory of international politics, which do not base conflicts on civilizational differences, is in trouble. And Huntington’s theory will be useful to understand emerging world order and explain militarized disputes and conflicts occurred in the post-Cold War era.

Debate: Civilizations Matter v. Does not Matter.
There were many responses and objections to Huntington’s argument that resulted in extensive debate. Many of them claimed that civilization theory is unable to explain causes of conflicts and international order. In spite of large amount of responses and studies, few of them are supported by quantitative research.

One of the important studies that evaluated Huntington’s argument and deserve a close attention was conducted by Russett, Oneal, and Cox in 2000 (in Triangulating Peace 2001, 239-305). The main question to be answered was “Does a difference in civilization increase the likelihood that a pair of states will become involved in a militarized interstate dispute?” (Russett et al 2001, 248). To perform their study they used University of Michigan’s Correlates Of War (COW) database that provided Militarized Interstate Disputes (MID). To them, limiting their analysis to MID that occurred during 1950-92 period was appropriate, because “Huntington addressed events in these years when he first presented his argument in 1993.” (249). Variables of liberal and realist theories were also added to their analysis in order to observe which theory had had higher explanatory and predictive power. Results of logistic regression indicated that variables of liberal and realist theories had much more influence on probability of conflict than that of civilization theory. Belonging to different civilizations increased risk of dispute between two states only by 12 percent with .30 level of statistical significance. (255).

Huntington (2000, 609) replied to Russett et al suggesting to retest his argument since his argument was dealing with the post-Cold War period. Russett et al (2000, 611), in turn, argued that Huntington (1993) had used many disputes and conflicts that had not occurred in the post-Cold War era to justify and support his argument. That was the reason why they used MID in 1950-92 period.

Another study to test Huntington’s argument was conducted be Jonathan Fox in 2001. He tried to examine whether conflicts between civilizations, particularly the Western and Islamic, had increased with the end of the Cold War. (Fox 2001, 459). He used Minorities At Risk (MAR) dataset for his quantitative research and came up with interesting results. From global perspective, the ratio of non-civilizational conflicts to civilizational conflicts had not change significantly after the end of the Cold War. From Islamic civilization perspective, there had been little change in the amount of conflicts among Islamic and other civilizations. However, from perspective of the Western civilization, proportion of conflicts between the Western and Islamic civilizations had been significantly increased since the end of the Cold War. Findings illustrate, as Fox concludes, it is important from which perspective one looks at conflicts. If one looks from the Western perspective, he/she will find increased amount of conflicts among the Western and Islamic civilizations which supports Huntington’s argument. But in global and Islamic civilization perspective, little evidence can be found for “clash of civilizations” in the post-Cold War period. (459).

Although Fox’s study deserves attention, it does not examine conflicts among other seven civilizations, but only focuses on the Western versus Islamic civilizations. Furthermore, his units of analysis are conflicts within states, not interstate. Interstate disputes are important in analyzing Huntington’s argument because nation as a whole has a specific civilizational characteristic. According to Huntington’s map (1996, 26-27) of civilizations, each nation belongs to one of the nine civilizations and conflicts are mostly among them. Conflicts that occur within states would be only one minor part of Huntington’s argument.

Many other scholars have qualitatively criticized Huntington’s clash of civilizations theory. Ali Mazrui (1997) finds tree major problems in Huntington’s argument. First, there is a “factual fallacy”. “It may not be factually true that the main lines of conflict of the future following the Cold War will be lines of clash of civilizations. It could be states or economic blocs.” (Mazrui in Rashid 97, 27). Second, there could also be a “conceptual fallacy” in Huntington’s argument. Mazrui suggests “clash of races” as an alternative to the clash of civilizations. (27). Thirdly, there is a “temporal fallacy” that means that inter-civilizational conflicts are not phenomena of the emerging world but it has been the most important issue of the world for at least four to five hundred years, including the Crusades, “trans-Atlantic slave-trade and European colonization of much of the world”. (29).

In response to his critics (Foreign Affairs 1996, 67), Huntington believes that none of these scholars has come up with better alternative to explain conflicts of the new era. To him, clash of civilizations remains the most powerful theory to understand the post-Cold War period disputes.
In this context, this paper is of a crucial importance because 1) it will take militarized disputes of the post-Cold War period as units of analysis; 2) disputes are among states and civilizations they belong to; 3) it will conduct a quantitative method to analyze clash of civilizations. Thus, this paper considers weaknesses of previous studies and establishes a strong critique to Huntington’s argument.

Research Design.
This analysis uses Correlates of War (COW) project’s Militarized Interstate Disputes (MID). COW provides MID starting from 1816 to 2001. As Huntington (2000, 609) claims, clash of civilization theory is helpful to understand conflicts that happened after the end of the Cold War, specifically after 1989 (Huntington 1996, 21; Huntington 2000, 609). COW project had MID up to 2001 year, so I picked those MID which occurred from 1989 to 2001. Although the number disputes in this period seems to be small to test any theory, it should be quite enough to evaluate Huntington’s argument since post-Cold War disputes were basis of his theory. I had 620 militarized interstate disputes in total. Each dispute occurred at least between two states, one state in side A, and another on side B. Sides are coded as 0 and 1. So I had information on which pair of states had disputes between them. Some disputes had more than two sides. I rearranged these coalitions as if they had disputes separately as dyads. For instance if there were two states on side A and two states on side B in a dispute, than there would be four dyads having same dispute.

States in these disputes belong to one of the nine civilizations, which are Western, coded as 1, Latin American (2), African (3), Islamic (4), Sinic (5), Hindu (6), Orthodox (7), Buddhist (8), and Japanese (9). Huntington (1993, 1996) does not explicitly specify the number of civilizations. In his article (1993) he mentions about “seven or eight major civilizations”. To him, Buddhism “has not been the basis of a major civilization”. (Huntington 1996, 48). Japan has not also founded any civilization since there are not any nations that share same religion, language, cultural values with Japanese culture. However, a map 1.3 (The World of Civilizations: Post-1990) in his book (1996, 26-27) shows that there are nine civilizations in the world including Buddhist and Japanese. I used nine civilizations as shown in the map. Russett et al (2001, 251) used eight, excluding Japanese. I wonder how they dealt with disputes that had Japan in one of the sides.
Having nine coded civilizations and many more states in the dataset, I have coded all states as either of nine civilizations they belong to according to Huntington’s map 1.3. (Huntington 1996, 26-27). I had difficulties, as Russett et al did (2001, 251), in classifying some states because the map was not clear and Huntington did not explicitly mentioned about them in the text. For example Nigeria, according to map is separated into two civilizations, Islamic and African. Since religion is the most important factor that distinguishes each civilization, I have used CIA’s World Factbook to find out the proportion of religion in Nigeria, and classified it as Islamic. India is shown as a state of both Islamic and Hindu civilizations, but I classified it as Hindu. Sri Lanka was classified as Buddhist in relevance with Russett et al (2001). Philippines, in my classification go into Sinic civilization, but Russett et al (2001) classified it as the Western. The most interesting question in classification was which civilization Israel belongs to. The question is significant since Israel has had many disputes with its neighboring states. Huntington thinks Israel is in Islamic civilization according to the map. It is ironic that Israel is considered as an Islamic civilization first because it has nothing to do with religion Islam and second because it will decrease empirical support for Huntington’s argument. I classified Israel as Islamic, strictly following the map.

To answer my research question I have to have civilization dyads and observe if these dyads are prone to clash or not and how frequent they have disputes. Thus I arranged these nine civilizations in dyadic form and came up with 45 civilization dyads, extracting each one of the same dyads as 1,2 v. 2,1. Dyad is labeled 0 if civilizations in it are same and 1 if civilizations are different. After arranging civilization dyads I plugged each dispute in one of the dyads and summed in regard to 45 civilization dyads. Then I arranged number of disputes in each dyad as proportion of them to the total number of disputes.

Since I seek to find out if civilizations matter in interstate disputes my independent variable is civilization dyads (0 or 1) and dependent variable is dispute percentage. Huntington’s main hypothesis is that civilizations matter and have positive impact on disputes in the post-Cold War period. I doubt his argument and hypothesize that civilizations does not matter in interstate disputes and even further, predict that civilizations have a negative impact on conflicts.

Data Analysis.
After conducting regression analysis of variables with STATA, I tested my hypothesis using t test of the slope (ts) at .05 level of statistical significance. Since I predicted the negative impact of independent variable on dependent variable, I conducted one-tail test. In the end, my computed t was in the critical region and I have rejected my null hypothesis. What does it imply? It means that civilizations have negative impact in interstate disputes. Finding of this paper contradicts to what Huntington (1993, 1996) suggests to explain causes of interstate disputes in the post-Cold War era.

In addition, R2 is found .08 and standard error of estimate (se) 4.55 implying that only %8 of variations in dispute percentage can be explained by variation in civilization. %8 is insufficient to explain decrease or increase in interstate dispute percentage in regard to civilizations. There are some other factors that affect variations in disputes percentage.

Table 1. Militarized Interstate Disputes (MID) by Civilizations in 1989-2001 period.

Civilization Dyads --- Disputes --- Dispute Prcentage (%)
Western v. Western ---11--- 1.77
Latin American v. Latin American--- 30 --- 4.84
African v. African --- 37 --- 5.97
Islamic v. Islamic --- 126 --- 20.32
Sinic v. Sinic --- 26 --- 4.19
Hindu v. Hindu --- 4 --- 0.65
Orthodox v. Orthodox --- 31 --- 5.00
Buddhist v. Buddhist --- 7 --- 1.13
Japanese v. Japanese --- 0 --- 0.00
Western v. Latin American --- 11 --- 1.77
Western v. African --- 6 --- 0.97
Western v. Islamic --- 62 --- 10.00
Western v. Sinic --- 14--- 2.26
Western v. Hindu --- 1 --- 0.16
Western v. Orthodox --- 126--- 20.32
Western v. Buddhist --- 0 --- 0.00
Western v. Japanese --- 0--- 0.00
Latin American v. African --- 0 --- 0.00
Latin American v. Islamic --- 0 --- 0.00
Latin American v. Sinic --- 0 --- 0.00
Latin American v. Hindu --- 1 --- 0.16
Latin American v. Orthodox --- 2 --- 0.32
Latin American v. Buddhist --- 0 --- 0.00
Latin American v. Japanese --- 0 --- 0.00
African v. Islamic --- 32 --- 5.16
African v. Sinic --- 0 --- 0.00
African v. Hindu --- 0 --- 0.00
African v. Orthodox --- 0 --- 0.00
African v. Buddhist --- 0 --- 0.00
African v. Japanese --- 0 --- 0.00
Islamic v. Sinic --- 0 --- 0.00
Islamic v. Hindu --- 6 --- 0.97
Islamic v. Orthodox --- 68 --- 10.97
Islamic v. Buddhist --- 1 --- 0.16
Islamic v. Japanese --- 1 --- 0.16
Sinic v. Hindu --- 0 --- 0.00
Sinic v. Orthodox --- 2 --- 0.32
Sinic v. Buddhist--- 2 --- 0.32
Sinic v. Japanese --- 9 --- 1.45
Hindu v. Orthodox --- 0 --- 0.00
Hindu v. Buddhist --- 0 --- 0.00
Hindu v. Japanese--- 0 --- 0.00
Orthodox v. Buddhist --- 0 --- 0.00
Orthodox v. Japanese --- 4 --- 0.65
Buddhist v. Japanese --- 0 --- 0.00
Total MID between same civilizations --- 272 --- 43.87
Total MID between different civilizations --- 348 --- 56.13
TOTAL DISPUTES --- 620 --- 100.00

There have been total 620 militarized interstate disputes during 1989-2001 period according to COW dataset. (see Table 1). Some of civilization dyads did not have any disputes, whereas other dyads conflicted several times. Huntington’s argument that different civilizations clash does not find support according to the results of this study. We can see from the table 1 that 19 dyads that consist of different civilizations did not have any militarized dispute since the end the Cold War. How can Huntington explain this phenomenon on the basis of his clash of civilizations argument? Can interstate disputes be explained by civilizational differences? Answers to these questions are obviously negative. Different variables of realist and liberal theories may explain the finding, but it is out of scope of this study.

Limitations.
First of all, some parts of this study were completed manually because of the absence of appropriate software. Particularly, picking each dispute and plugging it in civilization dyads was the most tiring part of the study, and some minor errors might have occurred. However, these probable errors have very little influence to the result since there were hundreds of disputes.
Another limitation is absence of clear list of civilizations and states in these civilizations. If Huntington had provided these data there would not be dissimilarities among scholar who tried to test his argument. Some studies used eight civilizations while others used nine. Some states like Israel and Philippines were in different civilizations in different studies. This may influence results of these studies since these states have had many militarized disputes.

Additionally, I should have included relative weights of each dispute, such as fatality and cost, to see if difference of civilizations cause more important disputes.

Conclusion.
This study showed that Huntington’s argument is insufficient in explaining conflicts in the new world order. As he argues, “culture and cultural identities, which at the broadest level are civilizational identities, are shaping patterns of cohesion, disintegration, and conflict in the post-Cold War”. (Huntington 1996, 20). However, he did not do any quantitative analysis of his argument. He provided only case studies and qualitative analysis as support. Many studies, as well as this one, in response to his clash of civilizations argument were conducted and conflicts of the post-Cold War period were analyzed quantitatively. Neither of them found evidence to support Huntington’s argument. Finding of this study illustrates that difference in civilization is not sufficient alone to understand what causes conflicts in these days and does not have positive impact on militarized disputes.

References
· Foreign Affairs. 1996. Samuel P. Huntington’s the Clash of Civilizations? The Debate. New York: W.W. Norton & Company,
· Fox, Jonathan. 2001. “Two Civilizations and Ethnic Conflict: Islam and the West.” Journal of Peace Research 38(4): 459-472.
· Huntington, Samuel P. 1993. “The Clash of Civilizations?” Foreign Affairs 72(3): 22-49
· ______. 1993. “If Not Civilizations, What?” Foreign Affairs 72(4): 186-94
· ______. 1996. The Clash of Civilizations and the Remaking of World Order. New York: Simon and Schuster.
· ______. 2000. “Try Again: A Reply to Russett, Oneal & Cox.” Journal of Peace Research 37(5): 609-610
· Rashid, Salim. 1997. “The Clash of Civilizations?” Asian Responses. New York: Oxford University Press.
· Russett, Bruce M. and John R. Oneal. 2000. “A Response to Huntington.” Journal of Peace Research 37(5): 611-612.
· ______. 2001. Triangulating Peace: Democracy, Interdependence, and International Organizations. New York: W.W. Norton & Company,

Appendix A: Regression Analysis