Last December, the EU's leaders formally agreed to expand their union in 2004 from 15 to 25 members -- a historic broadening of one of the world's most exclusive clubs. Europe's politicians also set a schedule under which two more countries, Bulgaria and Romania, would be brought into the fold three years later.

During the months leading up to the December decision, Recep Tayyip Erdogan -- leader of Turkey's Justice and Development Party, which won control of Turkey's parliament in November -- energetically toured Europe's capitals, urging EU leaders to include his nation in their expansion plans and to set a definite date to begin accession talks. Ankara's lobbying got strong backing from Washington: President George W. Bush even made a personal telephone call to the EU's then president, Danish Prime Minister Anders Fogh Rasmussen, to make the case. After much debate, however, when the EU announced its decision at a summit in Copenhagen in December, Turkey was turned down. Instead of offering a concrete date, the EU's leaders, in somewhat Delphic language, merely promised that if Turkey fulfilled the so-called Copenhagen criteria on human rights and democracy by December 2004, accession talks could then begin "without further delay."

By waffling, Brussels in effect managed to push down the road what has become a fundamental debate on the continent: should Turkey ever be admitted to the EU? Brussels' ambivalence reflected what has become the position of many of Europe's individual leaders -- an attitude that can best be described as "yes-but." Many of Europe's politicians now seem willing to recognize Turkey as an official candidate -- but only once it becomes more like them. This means greater respect for human rights and a reduced role in government affairs for Turkey's military. And it also means that Ankara must demonstrate sustained economic growth, enough to minimize the flood of Turkish emigration that many fear will result from its admission to the EU.

Some European leaders have also expressed darker concerns about Turkey. Most declarative, perhaps, have been the views of France's former president, Valery Giscard d'Estaing, the man now in charge of overseeing the drafting of a "constitution for a United Europe." On November 8, speaking to an interviewer from Le Monde, Giscard d'Estaing flatly asserted that Turkey simply is not a European country. "[Turkey's] capital is not in Europe," he declared, "and 95 percent of its population is outside Europe. [It has] a different culture, a different approach, and a different way of life. It is not a European country." EU membership for Turkey, he further declared, would mean "the end of Europe."

Giscard d'Estaing is not alone in such sentiments. His comments have been echoed by West Germany's former chancellor, Helmut Schmidt, who publicly voiced fears that the admission of Turkey "would open the door for similarly plausible full membership of other Muslim nations in Africa and in the Middle East. That could result," he argued, "in the political union degenerating into nothing more than a free trade community."

As such comments suggest, the debate over Turkey's admission to the EU has come to involve far more than simple economics, and this complexity makes Brussels' current ambivalence more understandable. As historians are quick to point out, European perspectives on Turkey have been colored by centuries of often troubled relations and by venerable, sometimes contradictory images of "the Turk." European elites long defined themselves in contrast to what they saw as the decadent, effete, depraved, and weak societies of "the East," dominated by the Ottoman Empire. At the same time, "the Turk" was also often portrayed as a powerful Mediterranean seafarer or a Barbary pirate: a shrewd and cruel warrior and a potent enemy to be feared and respected. Both images, although forged in the eighteenth century, have persisted into the current one, and for many they continue to define Turkey as outside the scope of European culture.

And yet there are also now strong arguments in favor of admitting Turkey to the EU. Turkey has a long record as a loyal NATO ally in a violent and dangerous part of the world. Turkey's admission would also allow the EU to send an important signal to the world: that it is open to Muslim as well as Christian societies. And in practical terms, advocates insist that Turkey's admission might not pose serious difficulties; they point to the EU's past success in integrating poorer countries such as Greece, Portugal, and Spain as proof that admission, accompanied by economic assistance, can set in motion virtuous cycles that speed up economic growth and constrain migration.

Would Turkey's admission, then, truly represent the "end of Europe," as Giscard d'Estaing has warned? Or would it offer a much-needed embrace to the Islamic world? With credible arguments on both sides, the answer probably lies somewhere in between. Turkey has much to offer the EU, and vice versa. In fact, Turkey's admission is probably only a matter of time. Having said that, rushing the process before Turkey is ready will benefit no one, and the best approach would be a measured and precautionary one. Before forcing the question, Turkey should improve conditions at home, making itself a "European" country in spirit and form. Undertaking such reforms will not be easy. But it is possible, especially if Turkey and the EU work together to make it happen.


Perhaps the one thing that both supporters and opponents of Turkey's admission can agree on is the unique geographic and historical position the country occupies. Although part of its largest city does sit in Europe, nearly all of the modern Turkish Republic's landmass is in Asia, and throughout history, Turkey has served as both a barrier and a bridge to Europe. Whatever happens with Turkey's EU candidacy, the country is likely remain distinctive, even unique, for some time.

Already large by the standards of the EU's 15 (soon to be 25) members, Turkey's population is growing rapidly. At present there are about 67 million Turks; of these, 3.5 million live abroad, two-thirds of them in Germany. Germany, by contrast, is currently the largest EU state with about 83 million residents, but it is shrinking by about 82,000 a year.

Under the EU's founding Treaty of Rome, citizens of member states are guaranteed the right to move freely to any other EU state and to seek jobs there on an equal basis with the locals. Indeed, the freedoms of movement and employment are among the cornerstones of the union's constitutional principles. When the EU has expanded in the past, however, freedom of movement for new members has often been delayed for a period of seven years in order to prevent mass emigration. Hence, if Turkey were to be admitted along with Bulgaria and Romania in 2007, its citizens could similarly be denied the right to freely emigrate for seven years -- until 2014, which, coincidentally, is about the same year Turkey is projected to become more populous than any EU member state.

Turkey is not remarkable just for its size; its history is similarly unique. The region was already of vital importance when Emperor Constantine established Constantinople in ad 330, and for the next millennium the city remained the capital of the Eastern Roman (Byzantine) Empire. In 1453, the Ottoman Turks captured Constantinople and renamed it Istanbul, and for the 450 years that followed, the cosmopolitan city presided over the Ottoman Empire as it expanded its control into the Mediterranean and central Europe.

The empire collapsed after World War I, however, and the modern secular state of Turkey emerged from its ruins -- founded in 1923 by a hero of that war, General Mustafa Kemal. Known as AtatÜrk, Turkey's patriarch was deeply suspicious of Europe's postwar attempts to dismember his country, and for most of the next 40 years, Turkey and Europe remained wary of and ambivalent toward one another. Despite his suspicions, AtatÜrk was a relentless modernizer, and his top-down transformations of Turkey often involved the adoption of European norms -- as, for example, when he outlawed the fez and the veil or changed Turkey's alphabet from Arabic to Latin.

Ataturk died in 1938, and in 1945 his successor allowed other political parties to compete for power. In modern Turkey's first free election in 1950, the Democratic Party led by Adnan Menderes came to power. This election, unhappily, also inaugurated an era of political instability, and the country has suffered four coups d'etat in the ensuing half-century. First Menderes was overthrown by the military and executed in 1960; then, although civilian rule had been quickly restored, his successor, SÜleyman Demirel, was similarly deposed in 1971. Civilians once again returned to power three years later, but the coalition governments that followed proved unstable, and a third coup ensued in 1980. Turgut ...zal governed as civilian prime minister from 1983 until his death in 1993. But not long after the Islamist Welfare Party came to power in 1996, it was forced out by the military (in a bloodless putsch that has come to be known as Turkey's "soft" coup of 1997).

There are now signs, fortunately, that Turkey's nationalistic and fiercely secular military is reducing its influence over politics. In last November's national elections, Erdogan's Justice and Development Party (known as the AKP in Turkish) -- formed from previously banned Islamic parties -- won 34 percent of the popular vote with promises to end corruption and revive the stagnating economy. Under Turkey's electoral system, this plurality translated into a clear majority of 363 (64 percent) of the seats in Turkey's 550-seat parliament. This huge share has allowed the AKP to form Turkey's first single-party government since 1987, and thus far, the military has shown no signs of meddling. For example, although Erdogan, the AKP's leader and a former mayor of Istanbul, was initially banned from formal office because of a 1999 conviction for "inciting religious hatred" (the result of having read at a political rally a poem declaring, "The mosques are our bayonets, the domes our helmets and the believers our soldiers"), when his party changed the constitution to allow him to run for parliament in a March 2003 by-election, the generals did not openly object. As this article went to press, Erdogan had just won the by-election and looked likely to become prime minister.

Despite fears that the AKP would Islamicize Turkey, since taking power its top priority has been to win EU membership, and the new government has taken immediate steps to make the necessary changes. In November, for example, Turkey's state broadcasting authority responded to EU concerns about the treatment of Turkish Kurds by announcing that it would for the first time start allowing limited broadcasts in Kurdish.


Politics aside, perhaps the major obstacle to Turkey's EU entry today is the country's unproductive and unstable economy, and the related threat that with accession to the EU, millions of Turks in search of jobs and higher wages would emigrate to Germany and elsewhere in Europe. In order to determine whether such fears are merited, it helps to review Europe's experience with the last great influx of Turkish workers: namely, under the "guest worker" programs, which often led to unexpected results for the countries involved.

In 1960, Turkey adopted a new constitution that, for the first time, guaranteed citizens the right to a passport and to travel abroad. The timing was auspicious; Turks gained this freedom just when European, especially German, employers, were rapidly expanding under Europe's "economic miracle" and were starting to look abroad to recruit foreign guest workers (Gastarbeiter) to staff their assembly lines. The supply of East German migrants had been curtailed by the building of the Berlin Wall, and thus in October 1961, Turkey and Germany signed a bilateral labor recruitment agreement that allowed German employers to hire Turks under temporary (one-year) work permits. If employers certified that they still needed their Turkish workers after the one-year period expired, the work permits could be renewed for up to two further years, and the workers' families were allowed to join them in Germany. After five years in Germany, guest workers became entitled to change employers and to remain in the country even if they lost their jobs.

The program proved very popular in Turkey, and the number of Turkish guest workers recruited by Germany rose sharply: from 9,000 in 1961 to 66,000 in 1964, and then to 130,000 in 1970. The flow peaked in 1973 at 136,000, the same year the German government decided to stop recruiting abroad. By then, however, some 805,000 Turks had already officially moved to Europe under the Turkish Employment Service, and 500,000-700,000 more had gone abroad without work permits (most European countries had not yet required Turks to obtain tourist visas), and then found jobs that allowed them to stay.

In the beginning, at least, both Europeans and Turks described the guest-worker programs as a win-win policy. The system, they argued, allowed northern and western Europe to achieve sustained noninflationary economic growth by importing un- and underemployed workers from southern Europe. Meanwhile, southern Europe benefited by reducing its persistent under- and unemployment while profiting from the hard currency sent home by workers abroad as well as from the return of former peasants who had become experienced factory laborers.

The deal, however, depended on the assumption that the migration would be temporary. Guest workers were expected to rotate in and out of jobs on assembly lines, construction sites, or mines. When the economic boom finally waned and unemployment rose, guest workers who lost their jobs were expected to act as shock absorbers for European labor markets by naturally choosing to return home to take advantage of lower living costs, thereby keeping unemployment rates in northern Europe low.

This "worker rotation" principle was first tested in Germany during the recession of 1966-67, and at first it seemed to work: the number of foreigners employed in Germany did decline, the German unemployment rate stayed under two percent, and economic growth and guest worker recruitment resumed in the late 1960s. Soon, however, the skeptics who argued that worker rotation would not work over the long term turned out to have been right after all. Neither guest workers nor their employers desired strict enforcement of the rotation policy. The workers became accustomed to wages that were eight to ten times higher than those at home. Employers, for their part, had little incentive to send trained workers home and then pay to recruit and train replacements. Guest workers who stayed often reunified their families in Europe, and thus the number of nonworking dependents climbed steadily. As a result, whereas in the early 1970s two-thirds of foreigners in Germany were employed, 20 years later the figure had sunk to one-third.

Today, 30 years after the formal guest worker programs ended, some 50,000 to 80,000 Turks continue to migrate to Germany each year. Most come today under family unification programs, when Turks in Germany choose to import their spouses from home. A second way in is through political asylum; during the 1990s, an average of 25,000 Turks a year, especially Kurds, applied for the protection of asylum status in Germany. (Only a minority of such claims were approved, but many whose claims were rejected stayed on anyway.) Finally, there is a large level of illegal Turkish immigration adding to the influx.


In their 40 years in Europe, Turkish guest workers and their families have had considerable difficulty integrating successfully into European societies. These problems have been widely recognized across Europe's political spectrum and are cause for further concern about Turkey's admission to the EU, as many European societies fear that additional Turkish migration will produce hard-to-integrate minorities that will threaten social peace and stability.

Numerically, Turks have never represented more than a third of all the foreigners in Germany, but in many respects they have been the most visible and least integrated. Turks were the last guest workers to arrive in large numbers, the poorest, the least educated, and the most different in cultural and historical terms. Their large numbers and low levels of income and education meant that Turkish migrants were more likely to reside in enclaves beset by high rates of poverty and joblessness. Their integration was also impeded by sharp differences between Turkish and European cultural views on the roles of men and women, by the deep significance of Islam in the daily lives of many Turks, and by the persistent and sometimes violent political divisions within Turkish society (for example, those between Kurdish and other Turks). The failure of assimilation was also compounded by the policies of both governments: Bonn made naturalization extremely difficult and stressed Turkish-language education even for German-born Turks, while Ankara encouraged foreign-born Turks to think and act as Turkish citizens.

Both countries, moreover, had unrealistic expectations about the benefits of guest worker migration. The Germans overlooked the social costs of the program, and Turkey overestimated the positive impact it would have on its own development. Ankara expected that migration's three Rs -- recruitment, remittances, and returns -- would catapult Turkey into the ranks of industrial countries. That leap never materialized, however, and given how high the expectations were, disappointment was all but inevitable. European employers naturally wanted to recruit only the best and brightest workers. The Turkish government tried to counter this trend with various schemes designed to promote candidates from its poorer regions in the eastern part of the country, and fostered the creation of Turkish Workers Companies (TWCS) designed to pool hard-currency remittances to establish factories and other operations in the migrants' area of origin. Most such plans, however, failed to achieve their development objectives. Rather than work in a factory for far lower wages than they had received abroad, most returning migrants preferred to open a small store or operate a truck or taxi. Hence their return never enhanced the productivity of local industry, as had been hoped.


What do such experiences augur for the future, if Turkey is admitted to the EU? Turkey's own demographics also bear on this question. During the 1950s, the population of Turkey grew at the exceptionally high rate of nearly three percent a year, as death rates declined and fertility stayed high. Fertility rates eventually declined to moderate levels, yet between 1950 and 1990, the country's population almost tripled, from 21 to 56 million, and it continues to increase by about 800,000 a year today. As mentioned above, this growth means that by 2014, if it is admitted, Turkey would become the most populous country in the EU.

The country's work force, meanwhile, remains heavily agricultural: of the 20 million employed Turks, 35 percent work in agriculture, compared to 25 percent in industry and 40 percent in services. Within the EU, by contrast, only 4 percent of the population works in agriculture; even in Poland, the most agricultural of the EU's 10 new members, the figure is only 19 percent.

Past experience with EU entry indicates that admission would accelerate labor displacement from agriculture in Turkey. Ankara provides a higher level of economic aid to its agriculture sector than does any EU country: Turkey's agricultural support programs, in fact, account for 4.3 percent of GDP, versus 1.7 percent in the EU. Nonetheless, incomes in Turkish agriculture remain low, and Turkish entry into the EU would therefore likely provoke large-scale movement off farms. It remains uncertain whether those leaving agriculture, which is concentrated in the east and southeast of Turkey, would move only to Istanbul -- or all the way to Germany.

Further raising the likelihood of migration, not to mention economic chaos in general, is the fact that, if admitted, Turkey would be the poorest EU state by far. The Organization for Economic Cooperation and Development reports that Turkish per capita GDP was only $2,100 in 2001 -- a tenth of the $21,000 average in the EU. Turkey's economy has also proved alarmingly unstable in the last decade. It contracted by six percent in 1994, expanded by six to seven percent a year between 1995 and 1997, and then shrank again in 1999-2000. In 2001, Turkey suffered its worst postwar economic crisis, as the economy contracted ten percent and the official unemployment rate exceeded ten percent; the economy was eventually stabilized in 2002 only thanks to a $16 billion loan from the International Monetary Fund.


Parallels are frequently drawn between the relationship of Turkey and the EU to that between Mexico and the United States. And indeed, there are some important similarities. Turkey and Mexico both are developing countries with substantial populations: nearly 67 million in Turkey and 100 million in Mexico. Both countries have experienced large emigrations of their people toward richer countries to the north. In both cases, these migrations preceded economic integration with their northern neighbors, and one argument for closer economic integration (via the North American Free Trade Agreement or EU membership) has been to reduce migration pressure.

The Mexico-U.S. case, however -- not to mention other similar scenarios -- makes it overwhelmingly clear that the economic integration of lower-income countries with richer partners requires painful restructuring in the lower-income country. One typical result is that large numbers of workers are displaced from agriculture and from protected industries, leading to increased pressures favoring emigration over the short term. This so-called migration hump can easily last a decade or more.

Still, there are signs for hope. When Italy, for example, restructured itself and integrated with northern Europe in the early 1960s, at first large numbers of Italians left for France, Germany, and Switzerland. The surge of Italian migration was short-lived, however, because Italy managed to create new jobs for ex-farmers in the north and to launch major development projects in the south.

Could Turkey manage a similar feat, creating jobs for its own ex-farmers and internal migrants following EU integration? Unhappily, Turkey's recent track record on job creation has been poor. Over the last decade, as the working-age population has increased by a million a year, total employment has remained stagnant at 20 million-21 million. Only half of these jobs were in the formal sector, moreover, and many of these were in overstaffed and loss-producing State Economic Enterprises (SEES) -- public-sector companies involved with telecommunications, energy, iron and steel production, and transportation. In fact, it was Turkey's failure to privatize or restructure the SEES that contributed to its deep financial crisis in 2001 and 2002.

Not only has Turkey's domestic record been poor; it also has a history of failing to live up to commitments it has made to Europe in the past. Turkish and European diplomats have been anticipating Turkey's EU entry since the 1960s and negotiating agreements on that basis, trying to facilitate free movement for Turks in Europe. In 1963, before the social costs of Turkish integration had become clear, the then European Community signed an association agreement with Turkey that envisioned the mutual lowering of trade and migration barriers. In 1973, an additional protocol to the association agreement established a joint commission charged with removing migration barriers between the EC and Turkey by 1986.

The barriers were never removed, however. Turkey balked at lowering its trade barriers as required, and in 1982 the EC suspended relations with Turkey in the aftermath of the 1980 military coup. Nonetheless, on April 14, 1987, against the advice of Germany and other EC countries, Turkey formally applied for membership. This application was rebuffed on December 18, 1989, on grounds that Turkey had not fulfilled basic human rights criteria and because it was feared that Turkey's low wages and underemployment could lead to mass migration.

Today, Turkish leaders, like their EU counterparts, are actually divided about accession. Those in favor argue that the EU should accept Turkey for three reasons: to bolster those Turks who want to create a prosperous and secular state, to send a signal that other Islamic states could eventually join the EU, and to strengthen Turkey's economy and thus reduce unwanted emigration. Yet many Turkish leaders are less enthusiastic and emphasize that EU entry is desirable only if the EU understands Turkey's need to take a firm hand with its Kurdish minority and does not press too hard on Cyprus. Turks also worry about becoming the poorest member of Europe, when they could focus on leading the Islamic world instead.

Some Turkish leaders also resent what they perceive as Europe's condescending attitudes and complain that Islam is the reason why the EU is reluctant to admit them. Erdogan, for example, recently lamented that "Turkey has been waiting at the gates of the EU for 40 years, but countries that applied only 10 years ago are almost becoming members. We think we have to go beyond that and not look at the EU as a Christian club."

Erdogan does have a point. Although often avoided in public discussions of Turkey's admission, the place of Islam in Europe has become a highly sensitive subject. There are already 11 million to 12 million Muslims in Europe, including 5 million in France, 3 million in Germany, 2 million in the United Kingdom, and 1 million in Italy. The numbers, combined with problems surrounding integration of Islamic residents in some countries, the rise of Islamist militancy in Europe and elsewhere, and the uncovering of organized Islamist terror cells throughout the continent, have rendered the subject even more fraught.

Although proponents claim otherwise, in fact no one has any way of knowing whether admitting Turkey to the EU would reduce migration, as promised, or stimulate it. What we do know, however, is that the last time they had the chance, between 1961 and 1973, some 1.5 million Turks went abroad for employment. That figure was equivalent to 10 percent of Turkey's 1970 work force and 40 percent of its male workers aged 20-39. We also know that interviews with workers in the late 1980s, when Turkey first applied for EU entry, revealed that at least 20 percent of young men were interested in overseas jobs and that many young women would have liked to join them abroad.

Again, even within Turkey there is little agreement on the impact that EU admission would have on migration. Some Turkish experts anticipate it would lead to an emigration boom; these pundits note that there is already significant rural-to-urban and east-west migration within Turkey that could easily become international migration. Moreover, migrants who have settled abroad are seen in Turkey as economic successes. Finally, there is little prospect that enough formal-sector jobs will be created anytime soon for new entrants to the labor force -- especially for young women in urban areas.

Other Turks, however, deny these claims. They argue instead that Turkey would replicate the experiences of other, earlier, relatively poor EU entrants such as Italy, Greece, Portugal, and Spain. In those historic cases, the foreign investment and aid that accompanied admission created enough jobs to prevent large-scale migration. Similarly, most analysts predict the ten states joining the EU in 2004 to produce only 335,000 migrants -- despite the fact that these countries have a combined population of 67 million.1

The truth is that no one can offer credible predictions in the case of Turkey. Turkey's population in 2004 will exceed the combined populations of all ten of these new EU member states. Most analysts guess that if Turkey entered the EU, a large initial wave of Turks -- 20 to 30 percent of the country's young men and women -- would travel abroad to test the EU's labor markets as soon as they were allowed to do so. After this initial wave, however, migration would likely depend on the evolution of labor markets in Turkey and other EU nations. If Turkey managed to quickly create jobs and raise wages, departures would shrink, just as would be the case if EU labor markets offered few jobs for workers with limited skills and little education.


Turkey has a long history of struggle between modernizers and traditionalists in its government. The former now argue that the latter will become stronger and more assertive if the EU rejects or delays Turkey's admission. Although such arguments have a certain logic, however, the overriding fact is that no one knows what the consequences of rapid admission would be. Given such crucial uncertainty, prudence and good sense support a more cautious process. Certainly, Turkey has changed dramatically over the past 80 years, successfully transforming itself from an autocratic and decaying empire to a modernizing (if unstable) democracy. If these changes can be sustained, Turkey will gradually establish itself as ready for full EU membership.

Hasty EU admission, on the other hand, would be a gamble, one that could go badly wrong -- especially since reversal would not be an option. Particularly worrisome is the possibility that Turkey's military might once again take power, especially if it feels the country is threatened by resurgent Islamists. Should an Islamist takeover and coup (either "hard" or "soft") occur, EU leaders would find themselves facing an impossible choice: between endorsing a military takeover or accepting an Islamist regime in their largest member state.

Nonetheless, Turkey's admission to the EU remains an attractive idea in the long term. Those in Turkey and elsewhere who urge a rapid pace should not be disheartened when their friends tell them that Turkey must first undertake a process of adjustment -- economic, political, educational, and cultural. What Turkey really needs is more honesty about the need for such adjustments and about the considerable time they are likely to require. Such frankness, unpalatable though it may sound, would be far more helpful than pressure for a rapid timetable that will likely fail. Such a failure could do great damage both to Turkey and to the EU -- far more than would a more cautious, if less dramatic, process.


1. Even this low level of expected migration prompted Austria and Germany to insist that the EU prevent citizens of new eastern European members from migrating for at least two years. After this two-year wait, old EU members will still be allowed to individually prevent freedom of movement for up to another five years.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Michael S. Teitelbaum is Program Director at the Alfred P. Sloan Foundation.
  • Philip L. Martin is Professor in the Department of Agricultural and Resource Economics at the University of California, Davis, and the author of The Unfinished Story: Turkish Labor Migration to Western Europe.
  • More By Michael S. Teitelbaum
  • More By Philip L. Martin