Democratic Innovations

Insights about the first-ever EU-wide citizens’ assembly: Considered opinions on further EU enlargement

11 Dec 2024

In October 2007, the first EU-wide and in fact the first transnational, Deliberative Poll®, called Tomorrow’s Europe, was organized. Imagined and convened by Stephen Boucher, then at the Jacques Delors Institute, in partnership with Professor James Fishkin of Stanford University, it gathered a random sample of 362 citizens from all 27 EU member states to the European Parliament building in Brussels, where they spent a weekend deliberating about a variety of social, economic, and foreign policy issues affecting the European Union and its member states.

An earlier version of this paper was prepared for presentation at the annual meeting of the International Society of Political Psychology, Paris, France, July 9-12, 2008.

Co-authors: Dr. Robert C. Luskin (University of Texas at Austin), Prof. James S. Fishkin (Stanford University), Stephen Boucher (Dreamocracy), Henri Monceau

In October 2007, the first EU-wide, indeed the first transnational, Deliberative Poll®, called Tomorrow’s Europe, gathered a random sample of 362 citizens from all 27 EU member states to the European Parliament building in Brussels, where they spent a weekend deliberating about a variety of social, economic, and foreign policy issues affecting the European Union and its member states. The deliberation, in a total of 23 languages, with simultaneous translation, alternated between small group discussion led by trained moderators and plenary question-and-answer sessions with leading policy experts and prominent politicians. The participants’ were queried about their views on first contact, before being invited to the deliberative weekend, again on arrival, and again, finally, at the end. (For more on the method and the rationale, see Fishkin 1997; Luskin, Fishkin, and Jowell 2002.)

The object, as always in Deliberative Polling, was to estimate what the public would think about the issues if it thought, knew, and talked much more about them and how that would differ from what they currently think about them (what ordinary polls measure). This Deliberative Poll, however, was unprecedented in bringing together a random sample from all the EU’s member states. The public whose views were being measured was not that of Germany, France, the U.K., or any other single member state but of the whole EU.

The issues deliberated included what the EU should do to preserve its pension systems, what role it should play in the world, how it can remain competitive in an increasingly global economy, and what if anything it should do about admitting additional member states. The results shed light on deliberation’s effects on all these issues. They also shed light on the possibilities of creating a European public sphere and on deliberation’s effects on mutual respect across national boundaries.

Here, however, we focus on the results concerning the issue of enlargement, which were striking and ran counter to what many of the EU’s supporters would have anticipated. On three of the four enlargement questions we asked—about enlargement as a general proposition, about admitting Turkey, and about admitting Croatia—the participants cooled on enlargement. We examine some possible reasons below.

We should emphasize that this is still a very preliminary report. The models may be refined, the extensive excerpts from the small group discussions more thoroughly folded into the overall discussion. Indeed we expect to try incorporating variables based on the codings into these or parallel models.

Sampling and Representativeness

The rationale of Deliberative Polling requires beginning with a sample representative of the public as it is—knowing, having thought, and having talked about the issues only as much as usual in everyday life, which in most cases is not much. In this case, the first step was parallel random sampling of all 27 member states, conducted by TNS Sofrès (the firm responsible for the Eurobarometer). We then randomly invited a subset of the 3,550 interviewees to the Deliberative Poll. Of these 362 made their way to Brussels for the weekend.

The aim was to wind up with a sample in which each country’s representation would be roughly proportional to the size of its delegation in the European Parliament. This is also roughly proportional to the country’s population, with some intentional over-representation of small countries, just as in the European Parliament itself, to ensure that they too have some voice.

Table 1 shows the distribution by country of those who showed up for the weekend. All 27 member states were in fact represented, and the percentages of the sample and of the European Parliament from given countries are indeed a very close match. In no case is the difference statistically significant. Even the match to the percentage of the EU population is close. None of these differences is statistically significant either.

But of course country is hardly the only dimension on which we should want the participants to be representative. One way of addressing the question of how well the participants represented the population of Europe is to compare them to the “nonparticipants”: the respondents to the initial survey who either were not invited or declined to attend.

Table 2 presents these results for sociodemographic variables. As can be seen, there were a fair number of statistically significant differences between the participants and the nonparticipants. Most, however, were relatively small. Somewhat more of the participants were men. Somewhat more of them were single, somewhat fewer of them widowed. Distinctly more of them were working full-time, distinctly fewer retired, slightly fewer unemployed and looking for work, somewhat more completing their education full-time, and somewhat fewer looking after the home. Perhaps the largest and most important differences, predictably, were with respect to education. Decidedly fewer of the participants had only a secondary education or less, decidedly more a university education or more. To the extent that the better educated start off knowing and having thought more about the issues, this probably makes the observed knowledge gains and attitude changes conservative.

We can also compare the participants and nonparticipants with respect to their pre- deliberation attitudes, as we do in Table 3 for four of the most relevant policy attitude indices. Here too there are some statistically significant differences, but here too they tend to be modest. As can be seen, the participants were significantly more in favor of admitting Ukraine than the nonparticipants, but neither more nor less in favor of admitting Turkey or the idea of enlargement in general. The participants were also less euro-skeptic than the non-participants on a series of ten questions asking about where decisions in given policy areas should be made, from 0 (by the individual member states) to 10 (by the EU)—rescaled in the table to 0 (EU) to 1 (individual member states). But these differences were modest. Across all 59 individual policy items we asked, not just those in these four indices, the average difference between participants and nonparticipants on these questions was only 4.0 % of the maximum it could possibly have been (given the ranges of the scales).

Initially published by the Center for Deliberative Democracy (Stanford University)

Attitudes toward EU Enlargement

As always in Deliberative Polling, there are two basic questions to be asked about deliberation’s attitudinal effects: (1) What do the participants think after deliberating? What, i.e., is the distribution of post-deliberation attitudes? (2) How did deliberation change what the participants thought? How, i.e. did the post deliberation distribution of attitudes differ from the pre-deliberation distribution?

The Tomorrow’s Europe survey contained four questions about what the EU should do about enlargement: a general question asking whether “additional countries that meet all the political and economic conditions for membership should be admitted to the EU” and three country-specific questions asking whether Turkey, Ukraine, and Croatia “should be admitted to the EU,” in each case, “if it meets all the political and economic conditions for membership. Originally on 0-10 scales, these, like, all our policy attitude measures, have been linearly translated into 0-1 scales, where 1 represents the highest level of whatever the variable’s name suggests (here, the most pro-enlargement attitude), 0 the lowest level (here, the most anti- enlargement attitude), and .5 neutrality.

We asked these four questions at three points: an initial interview (T1), by phone in most countries, in-person in four; a self-completion questionnaire on arrival (T2); and another self- completion questionnaire at the end (T3). The one exception was the Croatia question, asked only at T2 and T3. There may well be some learning and some attitude change from T1 to T2, a period during which the prospective participants tend to start paying heightened attention to media stories concerning the topics they knew they would be talking about in Brussels, to talk more about those topics with family, friends, and coworkers, and even, in some cases, to research them, in the library or on the web. We certainly expect there to have been learning and some attitude change from T2 to T3, the period of the deliberative weekend.

On all three questions asked at T1, the participants started with attitudes that, on average, tilted toward enlargement (as indicated by mean scores exceeding .5). The same was true of the Croatia question as of T2. There was less support for admitting Turkey than for admitting Ukraine (or for admitting Croatia at T2) or for the general idea of enlargement. Attitudes on all but one remained pro-enlargement following deliberation. The exception was the Turkey question, where the pre-deliberation mean favored enlargement, but the post-deliberation mean was right around (insignificantly to the anti side of) neutrality. On all three items we can track from T1 to T3 (the general, Turkey, and Ukraine questions), there was a statistically significant and sizable decrease in support for enlargement, largest for Ukraine, but appreciable for Turkey and the general question as well.

On the general and Ukraine questions almost all this change occurred during the anticipatory period from T1 to T2. On the Turkey question, by contrast, the bulk of it occurred during the deliberative weekend, from T2 to T3. On the Croatia item, which we can track only from T2 to T3, the was no sizable or statistically significant change, but the timewise pattern of change for the general and Ukraine items, where almost all the change occurred before the weekend, suggests that there could have been a decrease in support for admitting Croatia as well, just one that occurred before the weekend (in this case, before we began measuring).

New versus Old Member States

These results pertain to the sample as a whole, but they hide some important variation. On some of these questions, the participants from new versus old member states reacted somewhat differently. On the general enlargement question, essentially all the change came from the new member state participants. The old member state participants changed scarcely at all. The new member state participants decreased their support sharply—by a bit more than 15% of the scale. About half of that decrease came before arrival, and the other half over the weekend.

By contrast, the decreases in support for the admission of Turkey and Ukraine occurred both among the new and old member state participants. Regarding Turkey, the decrease, among both sorts of participants, came almost entirely during the weekend. Regarding the Ukraine, it came almost entirely before the weekend among the old member state participants but both before and during (though still more before than during) the weekend among the new member state participants.

Explaining Attitudes toward Enlargement and How They Changed

The results so far have been descriptive. The next questions are explanatory: (1) Why did some participants emerge with more pro- or anti-enlargement attitudes than others? (2) Why did some participants’ attitudes change in more a pro- or anti-enlargement direction than others? In the first case we look simply at T3 attitudes; in the second, ideally we should look at the differences between T3 and T1 attitudes. Unfortunately, many of the relevant explanatory variables were not present in the T1 questionnaire, so we look instead at the differences between T3 and T2 attitudes. That is, we examine the changes over the deliberative weekend, from arrival to departure, rather than the changes from the moment of first contact. We examine the T3 attitudes and T3 – T2 changes for the general enlargement, Turkey, and Ukraine items. We set aside the Croatia item, since attitudes toward admitting Croatia hardly budged.

What may account for the individual-level variation in T3 attitudes and T3 – T2 attitude change? We consider a mix of relevant empirical premises (arguments) about enlargement’s effects, values, attitudes toward particular countries, and sociodemographic characteristics:

Old member state. This is a dummy variable scored 1 for participants from old member states and 0 for participants from new ones. The differences of means above suggest that the new member state participants may have particularly reduced their support for enlargement.

Education. Another dummy variable, defined as 1 for those having a university education or more, and 0 for everyone else.

Adding a Muslim country would improve the EU’s relations with the Muslim world. This is the extent to which the participant agreed with this statement, the first of several empirical

premises about the consequences of adding (certain kinds of) countries. These are arguments for (in this case) or against (in other cases) enlargement.

Adding a Muslim country would make the EU too diverse. A second empirical premise/argument.

Adding more countries would help our economy. Another.

Adding more countries would help our security. Yet another.

Adding more countries would make it more difficult for the EU to make decisions. And another.

Keeping prices down. This is the first of several items and indices capturing relevant values—the importance attached by the participant to things many people think good for themselves or society. These were originally gauged on 0-10 scales, translated to 0-1.

Helping people in other parts of the world. Another value.

Having Europe play a larger role in the world. And another.

Traditionalism. Another.

Autonomy. Another.

Economic growth. Another.

Personal economic security. Another.

Protecting the less well off. And another.

Liking/Disliking Turkey. This was one of series of questions asking how much the participant liked or disliked various nationalities. This one was obviously relevant for attitudes toward admitting Turkey.

Liking/Disliking Russia. Another, relevant to attitudes toward admitting Ukraine.

Attitude towards migration policy. The thought here was that enlargement could affect immigration and internal migration and that attitudes toward migration might therefore affect attitudes toward enlargement.

Your country’s ability to provide for its own security. The thought here was that people from countries they see as less able to protect themselves might favor enlargement to the extent they think it serves their country’s security (or oppose it to the extent they think it undermines their country’s security).

Europe’s dependency on Russian energy supplies. Seeing this as a problem might have affected attitudes toward admitting Ukraine.

Russian interference in Eastern European and Central Asian. Seeing this as a problem might have also affected attitudes toward admitting Ukraine.

Most of these variables are part of the explanation for all three of our dependent variables (general enlargement, Turkey, Ukraine). Some are part of the explanation of only some of them (for example liking/disliking Turkey for admitting Turkey and liking/disliking Russia for admitting Ukraine). We estimate two models for each dependent variable, one for its value at T3 (post-deliberation), the other for the change from T2 to T3 (over the course of the weekend). In the first case, we use the explanatory variables at T3 as well. In the second case, we use the T3 – T2 changes in the explanatory variables, explaining change with change.

In all, then, we estimate six models, which we write as linear regression models and estimate by ordinary least squares. To maximize the effective sample size, we place “no opinion” responses to the attitudinal questions in these analyses at the midpoint and impute responses when the whole questionnaire is missing, as a number at T2 were, owing to an administrative glitch. The imputation is described in an appendix.

The results, in Tables 6-12, are generally satisfying and enlightening. The F’s and 2’

adjustedR ssuggestthatthesemodelsperformquiteattheexplainingpost-deliberationattitudes and moderately well at explaining post-deliberation attitudes. (It is generally harder to explain change scores.) Many of these explanatory variables have some significant effect, and almost always in the direction one would have expected. The tables also show the results when the equations are estimated separately for participants from new versus old member states.

Enlargement in General

We begin with the post-deliberation attitudes toward enlargement in general. Participants who thought that adding a Muslim country would improve the EU’s relations with the Muslim world or that adding more countries would help its economy or its security smiled distinctly more on the idea of enlargement. Those who thought that adding a Muslim country would make the EU too diverse frowned distinctly more on it.

The one significant but apparently anomalous coefficient estimate in these results belongs to the empirical premise that adding more countries would make it more difficult for the EU to make decisions. The more the participants endorsed this proposition, the more they wanted to see the EU admit new member states. The anomaly disappears for the equation explaining the pre- to post-deliberation change in attitudes toward enlargement in general, does not appear in either of the equations explaining attitudes toward admitting Turkey, but then reappears in the equation explaining post-deliberation attitudes toward admitting Ukraine. We are unsure what to make of this, but one possibility is that some segment of the sample would prefer that the EU have a hard time making decisions—that decision making rest as much as possible with the individual member states. From that point of view, admitting more countries, if it impaired EU-

level decision–making, might be a plus. It may be worth noting that this effect appears to be confined to old-member-state participants.

But what of the change from pre-deliberation attitudes? Here too a belief that adding a Muslim country would improve the EU’s relations with the Muslim world was important. Those who came to believe this more came to approve more of enlargement. So did those who came to place a higher value on traditionalism or their personal economic security. In addition, the more the participants believed their country could take care of its own security, the more favorably they viewed the prospect of enlargement. It is worth noting that the effect of personal was confined to the participants from the new member states, suggesting that part of the slide in support for enlargement was a matter of these participants realizing that their countries’ contributions from the EU might be reduced if they had to be shared with additional new member states.

Turkey

Many of the influences on the post-deliberation attitudes toward admitting Turkey are similar. Again most of the empirical premises matter. The more participants believed that adding a Muslim country would improve relations with Muslim world, that adding more countries would help our economy, or that adding more countries would help our security, they more they tended to favor admitting Turkey. The more they believed that adding a Muslim country would make the EU too diverse, however, the less the tended to favor it. The more participants valued helping people in other parts of the world, Europe’s playing a larger role in the world, or traditionalism, the more they also tended to favor admitting Turkey. Finally, unsurprisingly, the more the participants liked the Turks, the more they favored admitting Turkey.

Again the influences on the changes in attitude toward admitting Turkey look similar. Participants who came to believe more that adding a Muslim country would improve the EU’s relations with the Muslim world or that adding more countries would help our security came to look more kindly on admitting Turkey, while those who came to believe more that adding a Muslim country would make the EU too diverse came to look more askance at admitting Turkey. People who came to place higher value on helping people in other parts of the world came to favor admitting Turkey more, as f course did people who came to like Turks more.

Ukraine

Attitudes toward admitting Ukraine were affected by education. In addition, those who believed that adding more countries would help the EU’s economy, help its security, or (again perhaps anomalously) make it more difficult for the EU to make decisions were more inclined to support admitting Ukraine, as were those who cared more about keeping prices low.

As for changes in attitudes toward admitting Ukraine, the more one came to believe that adding more countries would help our economy and, similarly, the more one came to value keeping prices down, the more one came to favor admitting Ukraine.

Implications

The statistically significant (and near-significant, since there is nothing God-given about the conventional .05 threshold) coefficient estimates suggest potential levers for moving public opinion under the sort of good conditions—balanced, informative, mutually respectful— Deliberative Polling tries to create. For example, attitudes toward enlargement in general could be made more positive to the extent that more people could be persuaded that adding a Muslim country would improve the EU’s relations with the Muslim world, that adding more countries would help its economy, or that adding more countries would help its security—or made more negative to the extent that they could be persuaded of the opposite.

Knowledge Gains

One way of assuring ourselves that these policy attitude changes were not adventitious, that the participants were indeed seriously deliberating is to examine their scores on a series of factual and quasi-factual knowledge questions. The questionnaire included factual knowledge questions about the European Union and the policy areas under discussion, as well as two quasi- factual questions asking respondents to place the views of Nicholas Sarkozy and Gordon Brown on a left/right scale (where placements of Brown on the right and of Sarkozy on the left may be regarded as wrong). All these questions are described in an appendix. Table 3 shows that the participants learned a lot. The participants from increased their knowledge scores by about 15%, a sizable increase. (Think of it as a 15% increase in the average mark on an exam.)

Note the results for the Brown and Sarkozy placements are, on average, very similar to those from the purely factual items. The percentages knowing the correct side of the left-right scale rose by roughly 9% for Sarkozy and by roughly 15% for Brown. This even though no leader’s ideological position (or party affiliation) was mentioned in the briefing document or likely to surface in the discussion and even though only a decided minority of the sample came from Britain or France. This result suggests the effect that the prospect of attending the Deliberative Poll had on the participants’ attention to political news.

Since, as we have seen, there were some differences between the participants from new versus old member states when it came to policy attitude change, Table 6 also examines the knowledge gains separately for the two groups. Although the new member state participants started from a somewhat (roughly 5%) lower level, the two groups learned about equally. Averaging across all the knowledge items, the new member state participants gained 15.1%, the old member state participants 15.5%. The only large differences were with respect to the Role of EU in employment benefits, where the new member state participants gained 20.9%, and the old member state participants only 10.5%; with respect to 1/3 of EU Budget, where the new member state participants gained 26.4%, and the old member state participants only 13.3%; and with respect to EU vs. US Foreign Aid, where the new member state participants gained only 10.5%, and the old member state participants only 30.5%.

What Was Said: Insights from the Transcripts

We are in the process of examining the transcripts of the discussion of enlargement to identify the opinions expressed and arguments used by participants. We adopted a coding scheme distinguishing 16 themes, including: whether the European Union is adding too many countries too fast, whether some countries are too different, whether enlargement would make the EU’s decision-making capacity easier, and whether enlargement would improve the EU’s relations with the Muslim world. Below we present the frequencies of statements exhibiting these themes and provide examples.

All 19 statements made about the timing of EU expansion expressed the view that the European Union is adding countries too quickly. For example, one participant from Group 14 said, “…the EU should slow down the enlargement process, and it should provide us as members to improve…” On similar lines, a participant from Group 18 said, “First three, then one, then two, and then three again. And then, all of a sudden, ten countries joined…This was a very large step…And I think if one keeps pushing the borders further and further out, and keeps pushing faster and faster, it won’t be possible in this area to sufficiently attend to the countries to get them adjusted…”

A participant from group 16 offered a similar view, that it takes time to digest, indeed to unify or deepen the union, before further enlargement proceeds: “… a Community when it gets larger needs also to “digest” its new members and produce more unity by defeating reciprocal resistances. Once united it can then try to enlarge itself once again. I think that lately we have done more than it was necessary. Besides economic assessments, or evaluations based upon human rights standards, all things that are very important indeed, we should try to find a way to divide enlargement as according to a certain amount of time, that is for instance, that we cannot proceed every year to enlarge Europe. Otherwise, we risk to run into problems…”

Some of the concerns about enlargement were economic, 3 of 3 statements made about EU aid to EU’s current countries said that enlargement would decrease EU Aid to its current member states. On this topic, a participant from Group 18 said, “…5% is support for the joining candidates. Joining candidates at the time are Macedonia, Croatia, and Turkey…one has to subsequently think about the fact that, since these countries have been accepted and even more countries are going to join, the new federal states in Germany all of a sudden are above the average and actually will not receive aid anymore…”

Regarding the effect of enlargement on EU’s decision making capacity, the statements were mixed, with 5 statements indicating enlargement would make decision making easier, but 9 statements indicate decision making would be harder. For one participant from Group 7, the decision making capacity would be easier, because “…we might become less dependent vis à vis the United States. It might be nice to have – and possible, in that case – to define our own EU policy.” While on the other side, another participant from Group 7 said, “…there is not enough cohesion in Europe between the countries and that can be felt in our external policies because countries can have differences, but they should be more cohesive to make it stronger…”

When discussing the issue of enlargement and relations with the Muslim world, 7 of 7 statements said enlargement would improve relations with the Muslim World. In Group 14, a participant said, “…if we accept Turkey in Europe, then all Muslims would understand and therefore we would gain respect. This is the reason for having Turkey in the EU. I know how they live, let’s try to mentally accept them…there are at least 70 percent of Turkish that would join Europe. If we isolate Turkey than this would be a damage for us.” However, 35 of 50 statements made on the topic of whether a country is too different said that a particular country is too different from the existing member countries. In the same discussion group, Group 14, within minutes of the previous participant’s statement, another participant said, “They are extremist, they dialogue in a different way. I see them as very far away from us. I don’t see you for instance as very far from us, there might be other kinds of problems ok, different speeds, but all these problems can be solved. In their case, instead, they are very much far away, their religion is a very heavy impediment, they are Muslims and therefore very much different.” Meanwhile, 15 of the 50 statements said a particular country is not so different and an example of this view is from Group 17, where a participant said, “…I think that Turkey could add some color…it could allow Europe to initiate a dialogue with other countries…start a dialogue with Asian countries…So Turkey would add some differences. So the present member states of Europe have more in common…” Even though some participants feel a country like Turkey is quite different from the countries within the European Union, other participants believe that admitting a Muslim country would not be a problem. Of the 37 statements on admitting a majority Muslim country, 21 of the statements felt it would not be a problem. A statement from a Group 18 participant pointed out that “…Europe is already multi-cultural, so the Muslim culture won’t make such a difference…” The remaining statements, 16 statements, said that admitting a majority Muslim country would be a problem. Here is a statement from Group 17, “…I’m very skeptical about Turkish membership. Obviously, we must keep the door open and perhaps Turkey will one day be ready to join the European Union, but the fact that Turkey is an Islamic country, this is something that we really need to be cautious about and take a moment before we make any decisions…”

Regarding human rights, 19 of 19 statements made said countries that violate human rights should not be permitted to join the European Union. From Group 18, a participant said, “…I would definitely have a problem…the violations of human rights in Turkey, the set of problems concerning the treatment of Kurds in Turkey, and the reaction of Turkey to the US calling the crime against the Armenians genocide…the way Turkey reacted…”In addition, participants also expressed that being a part of the EU would reduce human rights violations in member countries. 5 of 5 statements made said enlargement would reduce human rights violation. Here is an example from a participant in Group 12, “Concerning Turkey, I think you have to think about the alternative if Turkey doesn’t become a member of the EU. Are they going to be pushed further eastwards? Is this going to be a disadvantage for Europe…”

Four topics covered the effects of enlargement on the economy. On the economy in general, the participants expressed mixed views. Half of the statements indicated enlargement would stimulate economic growth and the other half of the statements indicated enlargement would be an economic burden. A participant from Group 14 with the latter view said, “It is out of question that we should enlarge to Turkey and this is for economic reasons. Also with Muslims we have to deal since they cannot stand us, they feel uncomfortable with us. It is an economic reason…” While a participant from Group 1 felt that enlargement would stimulate economic growth, this participant said, “Well, when the EU went into effect back then, when the union started, I thought: Thank god, this has been due for a long time. That people find each other and go with each other more. Economically, it had incredibly positive effects.”

The second code for the economy was the personal financial impact of enlargement. Only two statements were made on this topic; one statement indicated its positive impact and the second statement indicated its negative impact. A participant from Group 1 had positive experiences with enlargement and said, that “…Since I am working in export.. I am happy about it, about every expansion, because they mean less work for me personally and less complication…you can simply order a truck and have things picked up, from Estonia, Latvia, everywhere we were doing trade…” While a participant from Group 7 felt that, “…If we allow a country like Turkey in right now, it’s going to generate enormous expenses …I think it’s going to make a nice difference in our personal budget.”

The third code for the economy was the financial impact of enlargement on the participant’s own country. No participant indicated there has be positive financial impact for their own country, but two statements were made indicating enlargement had negative impacts on their own country. An example of this view comes from a participant from Group 14, the participant said, “Slovenia is part of the EU since 3 years, notwithstanding this the common opinion is that the EU has enlarged too much and has included too much diversity within itself. Due to this diversity, it has lost its strength as far as its economy is concerned as well as its unique position…”

The last code for the economy was the financial impact of enlargement on the European Union. For this topic, 6 statements said enlargement would have a good financial impact and 7 statements said enlargement would have a bad financial impact. A participant from Group 1 with a good outlook for the EU said, “…And I think, especially Turkey is indispensable for us…Turkey could offer us a lot of support or at least communicate experience in the direction of the Arabic world…Furthermore, one has to say, it has never been harmful for Europe to grow, when the market area grew at the same time..”. And, on the other end, a participant from Group 11 said, “…I just wanted to talk about the economic problem… As long as we haven’t successfully completed the standardization work …as long as we haven’t managed to raise the standard of living of those who just joined…bringing another country…would create more problems…”

In regards to military and security, 9 out of the 10 statements made said that enlargement would help EU’s military and security. An example of this view comes from a participant from Group 6, this participant said, “I wouldn’t think the approaches dictated with difficulty by the various nations could lead to the creation of a European army, an army whose objective would be to keep the peace. That would be its objective, an army to defend the European territory. Well, I see things this way: instead, we should try to encourage coordination of the police forces rather than the army. Police forces in charge of the security of the movement of people and goods within the European Union …” The one statement that felt enlargement would hurt EU’s military and security said, that “I think if certain countries try to achieve certain supremacies, whatever country that may be, that this is regarded as very suspicious by the others. I think that, in this regard, the citizens of Europe still have a long way ahead of themselves in order to really take a back seat, so that when some country Y makes a suggestion, they can all accept it.”

On the topic of EU’s influence in the world, 6 of 7 statements made on this topic said enlargement would increase EU’s influence in the world. A participant from Group 1 said, “Globally, it would certainly be desirable for the EU to have a considerably stronger position, since globally, nowadays everything happens with the US and China involved, and maybe India in the future…” The one participant that felt enlargement would decrease EU’s influence in the world said, “Starting with the enlargement, for me, there is not enough cohesion in Europe between the countries and that can be felt in our external policies because countries can have differences, but they should be more cohesive to make it stronger.”

Finally, on the topic of immigration and migration, two statements were made. One statement said immigration and migration would be a plus and this participant said, “…Turkey did not have to enter the EU but I have said that we have taken a commitment, it could be even a good thing, we have already Turks in Europe…” While the second statement argued that immigration and migration would be a minus, this participant said, “…But if they come and accept our law, then they are welcome….but at the same time in Germany they make immigrants sign an agreement where they endorse the responsibility to respect the German law. Whereas in Italy, they arrive, they want to work but there are no jobs even for Italians…so what should we do? They come from Slovenia, Romania from all countries of the world, I don’t know if they would accept this same situation…”

Conclusion

Among other things, Deliberative Polls are “polls with a human face.” The small group discussions allow us to see public opinion reshaping itself as people think, learn, and talk about the issues under good conditions. These good conditions include vetted and balanced briefing materials, the small group discussions with trained moderators, the question-and-answer sessions with competing experts and policy makers, and in this case, simultaneous interpretation, permitted transnational discussion. All this is deliberately counterfactual—an effort to see what the views of a more ideal citizenry would be like.

Tomorrow’s Europe was even more counterfactual than other Deliberative Polls, aiming, for its sample, for a weekend, to create an EU-wide public sphere that does not in fact exist in the real world. We hope that this may be merely the first of a series of pan-EU Deliberative Polls, as Europe develops its capacities for collective self understanding and political expression. In any case, the results of this maiden voyage are wide-ranging and rich. This report has focused simply on the issue of enlargement, but deliberation changed opinions on a variety of issues, not just enlargement, and tended to increase cross-national understanding and respect.

It is worth reiterating that the deliberations in Deliberative Polling provide a testing ground for the competing arguments. The regression coefficients suggest some of the empirical premises capable of moving opinion under these good conditions. So may the recurring themes that marked the discussions. The results thus help identify some of the levers proponents or opponents of enlargement might effectively use—in the sort of fair debate the conditions of Deliberative Polling are intended to create.