Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Studying constitution-making during post-communist transition can be regarded as particularly insightful for testing the veil of ignorance hypothesis proposed by Buchanan and Tullock (1962) because of the manifold transition that took place in these countries. These countries were simultaneously moving from one-party system to multi-party system, from command to market economy, and from government unconstrained by laws to the rule of law (see Sunstein 1991: 371). Hence, one would expect constitution-makers in these countries to have been facing extreme uncertainties with regard to their future power positions. This chapter examines constitution-making in one particular transition country—Estonia. After the declaration of independence (from the Soviet Union) in 1991, the constitutional assembly (CA) drafted a new constitution for the restored republic. What makes the Estonian case special compared with other instances of constitution-making in Central and Eastern Europe during the transition period is that the 1992 Constitution of Estonia was prepared by a separate CA (rather than by an existing legislature) and the resulting constitution was approved by means of a referendum, implying that it could be viewed almost an “ideal” case if viewed in the light of normative propositions for constitution-making brought out in the literature on constitutional political economy. Therefore, the Estonian constitution-making can be considered to be a particularly interesting case for constitutional political economy in general and for exploring the opaqueness of the veil and its effects on the content of the constitution in particular.

This paper will explore what kind of motivations of the framers can be inferred from the text of the Estonian constitution as discourse (about power relations) and how opaque was the veil of uncertainty behind which the members of the Estonian Constitutional Assembly (1991–1992) drafted the new constitution.Footnote 1 According to the theoretical framework proposed by Imbeau (2009) and by Imbeau and Jacob (2011, 2015), if the veil of uncertainty in constitution-making is opaque, the predominant power relations in the constitution should be based on political power, pertain to social relations and be concerned with denying capacities to act (rather than attributing capacities to act). Thus, this paper examines to what extent these theoretical propositions are corroborated in the Estonian case.

This chapter is structured as follows. The first section gives an overview of the historical setting in which the CA was convened, how the constitution was drafted and what the motives and objectives of the framers were. The second section describes the content of the Estonian 1992 Constitution in terms of the power relations framework and addresses the following sub-questions: (a) What proportion of power relations in the Estonian constitution is based on political power (as opposed to economic power and perceptoral power)? (b) Among the power relations, what is the proportion of social and instrumental power relations? (c) Is the Estonian constitution concerned more with negative or positive power relations? The third section discusses the findings of the Estonian case study in the light the theoretical propositions of the Veil of Ignorance Project. The final section concludes.

As will be shown, the Estonian constitution does indeed emphasize power relations based on authority, but at the same time, it is more concerned with positive power relations than with negative ones and entails a lower proportion of social power relations than could be expected behind the veil of uncertainty. This paper also shows that, while for the most part, the opacity of the veil does indeed facilitate the reaching of agreements, there can also be situations where the transparency of the veil can contribute to reaching agreements among the framers, and conversely, that in some situations or issue areas, the veil can be so thick that the uncertainties involved make it very difficult to reach consensus. The Estonian case also points to the importance of the prevailing normative discussions on constitutionalism at the time of constitution-writing and to the role of path-dependencies in shaping the content of the constitution.

2 Historical Context of the Drafting and Adoption of the Constitution

2.1 The Historical Setting: The Decision to Convene the Constitutional Assembly

The decision to convene the CA (together with the Declaration of Independence) on August 20, 1991 can be considered as the culmination point in Estonia’s struggle for independence from the Soviet Union, which had already started at the end of 1980s.

The Supreme Council of Estonia made use of the power vacuum in the Soviet Union—created by an attempted coup d’État in Moscow—and seized the opportunity to restore Estonia’s independence and hence end the Soviet occupation, which had lasted since 1940.Footnote 2

The task of drafting a new constitution for the country was given to the CA, which was to include members from the two competing parliamentary bodies in Estonia at the time: the Supreme Council and the Estonian Congress (30 from each body).Footnote 3

The Supreme Council (105 members) had been elected in the first multiparty parliamentary elections in 1990 by the electorate that consisted of all adult inhabitants registered in Estonia.Footnote 4 The Estonian Congress (499 members) had been elected in 1990 by all who had been citizens of Estonia prior to the Soviet occupation (and their adult descendants). These parallel elections to two different representative bodies mirrored the competition of two different pro-independence groups in Estonia, which had begun already in 1988–1989, when the more active movement toward independence started (Raitviir 1996: 353; Pettai 2001). On the one hand, the Estonian Popular Front advocated a more compromise-oriented approach to re-establishing Estonian independence, implying the utilization of the existing political institutions, gradual takeover and reform of the existing organs of state power, formal secession from the Soviet Union, and the proclamation of a new Estonian Republic. On the other hand, the movement called the Citizens’ Committees (consisting of the Estonian Heritage Society, the Estonian National Independence Party (ENIP), and the Christian Union) vehemently disagreed with such a “conciliatory” strategy; instead, they called for setting up new institutions and restitution of the pre-war Estonian Republic (on the basis of legal continuity). As a result of the elections to these two bodies, the Popular Front became a dominant force in the Council, while the National Independence Party, together with the Estonian Heritage Society, became an important power in the Estonian Congress. Although the Estonian Congress remained a kind of “shadow parliament” to the Supreme Council (which had de facto powers and passed the laws), it was supported by the public and could not be ignored by the Council on important issues (Pettai 2001: 112–114; Taagepera 1993, Chap. 7; 1994: 213–214; Raitviir 1996: 353; Lauristin and Vihalemm 1997: 90).

The freely elected Supreme Council provided an opportunity for the Estonian political elites to experiment with democratic institutions. Altogether, it was a curious mix of elements of Western democracy and residues from the Soviet past. For example, although in principle the political system in place in 1990 was parliamentary democracy and the parliamentary majority appointed the cabinet, the members of parliament were not allowed to become cabinet ministers. There was no de jure head of state, although the Chairman of the Presidium of the Supreme Council (Arnold Rüütel) came to serve as a de facto head of state in 1990 and 1991. Legislative activity of the new parliament was bogged down by quorum and majority requirements, which were unrealistically high—another leftover from the Soviet-type system (Taagepera 1991: 478). Despite these hurdles, the Supreme Council managed to pass a significant number of laws, some of which (e.g., on the head of state, the parliament, the electoral rules) amounted to a temporary constitution (Taagepera 1991: 480). Parallel to the work of the Supreme Council, the Estonian Congress also debated constitutional issues and commented on the laws adopted by the Council.

The decision of the Supreme Council to convene the CA has been termed as a “miracle compromise” (Pettai 2001: 111), since it succeeded in striking a balance between the diverging plans of the Supreme Council and the Estonian Congress on the question of how to proceed with constitution-making. Until then, some members of the Supreme Council had suggested that the Supreme Council should be declared a constitution-making body, while the leaders of the Estonian Congress demanded new elections for a CA or, alternatively, re-enacting the 1938 Constitution, which had been the last constitution in force before Estonia was occupied (Pettai 2001: 114).Footnote 5

In sum, the constitution-making in Estonia was to take place in the context of high uncertainty as Estonia was in the early phase of a manifold transition. Internationally, it was not clear how the breakup of the Soviet Union would evolve and what implications these developments would have for Estonian foreign and internal politics. Domestically, while some of the political delineations had already been formed, the party landscape was still in the process of being formed, which, in turn, made strategic calculations more difficult for the actors involved. Further, with regard to socioeconomic issues, the uncertainty was even more pronounced. Given that during the Soviet regime, private property had been limited and privatization process had not been clearly outlined yet, it would have been rather difficult for the framers to evaluate their own socioeconomic position in the future.

2.2 The Constitutional Assembly at Work: The Drafting of the Constitution

The elections to the CA took place on September 7, 1991, in the Supreme Council and the Estonian Congress. The resulting composition of the Assembly mirrored relatively proportionally the strength of the different groupings in the two bodies (Taagepera 1994: 217). Among the total of 60 CA members, about 20 were from the Popular Front, another 20 could be labeled as national radicals, 13 were various Estonian moderates and reform Communists, and 7 were representatives of the Russians.Footnote 6

Altogether, the CA of Estonia met at 30 different sessions between September 1991 and April 1992. The Assembly’s work was divided between plenary sessions and committee work. The committees were to focus on more detailed discussions of the topics and the issues that could not be resolved within the committees were brought to the plenary session.

At the second session of the CA, two draft constitutions were submitted. The first one had been drafted by a working group of legal experts led by the Minister of Justice Jüri Raidla. The second draft was the 1938 Constitution of Estonia. By the end of September, three further draft proposals were submitted. The main competing drafts were the draft of Raidla and his team (putting forth a more presidential system) and the draft of Jüri Adams (putting forth a more parliamentary system by mending the 1920 Constitution of Estonia with some more balanced provisions and incorporating some provisions from the 1938 Constitution).Footnote 7 In October, the CA chose the draft of Adams (a member of the Estonian Congress and the chairman of the ENIP) as the basis for further deliberations (with 29 vs. 22 votes).

At the end of October 1991, a delegation of foreign experts from the Council of Europe attended the Assembly’s sessions and voiced their opinions on the draft. The first draft of the constitution was released to the public on December 21, 1991. One of the committees was given the task to sort through the proposals submitted by the public and accommodate these proposals, where they considered appropriate (for a more detailed discussion, see Raudla 2010b).

In January 1992, the Supreme Council decided that the CA should continue its work but cooperate more closely with Raidla (the Minister of Justice) and his expert group, who had been involved in formulating the more presidential draft (but which had been rejected by the CA) (Aaskivi and Pärnaste 2002: 60). As a result of the cooperation, several additions were made to the draft, in particular to the chapter on the fundamental rights and duties (Hänni 2002: 164).

In February, the CA approved the draftFootnote 8 it had formulated by then and sent it to the Supreme Council, who, in turn, sent it back to the CA, claiming that it did not correspond to the preferences of the Estonian people (especially on the question of how to elect the president) and that it did not have the support of the majority of the Assembly. According to the decision of the Supreme Council, the CA was to “coordinate” its draft with the expert group consisting of Council’s lawyers and prepare a revised draft by mid-April (Taagepera 1994: 226–227; Aaskivi and Pärnaste 2002: 68; Adams 2002: 151; Hänni 2002: 166). The CA obliged, although mostly in form rather than substance, as witnessed by the fact that as a result of this “coordination,” only minor wording changes were made to the draft constitution, but, in terms of content, no extensive changes were made. In particular, the CA was not willing to give in on the issue of how the President should be elected: it stayed with its initial choice of having the president elected by the parliament (and, if the parliament fails to achieve 2/3 majority, by an electoral body composed of members of the legislature and representatives of the local governments). The only more substantive change was the right given to the president to appeal to the Supreme Court after his veto is overridden by the parliament. The Assembly reaffirmed its support for the draft constitution at the 30th session (on April 10, 1992), and the Supreme Council decided to put the Constitution to referendum. On June 28, 1992, the Constitution was approved in a referendum where only Estonian citizens could vote. The list of citizens included 669 080 persons; 446 708 took part in the referendum, of whom 91 % voted in favor of the Constitution.

It is worth mentioning here that together with the core text of the Constitution itself, the CA also worked on what was called the implementation law (the so-called Chap. 17 of the constitution) which entailed provisions on how to phase in and implement the new constitution. The most controversial aspect in preparing the implementation law was the proposal to “block former Communist power-holders from running for office during the first ten years” (Taagepera 1994: 225). After heated debates on the issue, the CA concluded that it was “too political for a constitution-making body to decide” and proposed to the Supreme Council to include it as a separate question on the constitutional referendum ballot (Pettai 2001: 126). However, by the time the implementation act was submitted to popular referendum this “lustration” provision had been dropped. The implementation act also foresaw different rules for the first elections of the parliament and the President. The first parliament was to be elected only for 2.5 years (instead of 4). The President was to be elected for 4 years (instead of 5) and according to a different election procedure: in the first round, the public could directly vote for the candidates, and if no candidate received more the 50 % of the votes, the parliament would elect the president. The different method for the first election of the President was introduced in order to appease the public, but also several members of the Supreme Council and the Government, who favored direct elections of the president.

2.3 The Framers of the Estonian Constitution: Their Motivations and Objectives

The constitution-making in Estonia in 1991–1992 was characterized by a complex configuration of actors with various powers. The representatives from the existing legislature (the Supreme Council) and the Estonian Congress had agenda-setting, deliberative power, and voting power. The Supreme Council as a whole had process-guiding powers. The local constitutional experts who were involved in the work of the assembly through advisory functions (although they did not have voting powers as such) had some agenda-setting powers and powers emerging from their expertise. The foreign constitutional experts had the power of persuasion and were regarded with great respect by the Estonian framers because of their assumed neutrality and long-term expertise (see also Raudla 2010a). And, in the end, the Estonian citizens had the power to approve or reject the constitution in a referendum.Footnote 9

Altogether, the framers came from either the existing political elite (e.g., the representatives of the Supreme Council) or emerging elite (the representatives of the Estonian Congress), among whom several had been dissidents under the Soviet regime. In other words, the members of the CA were, to a large extent, holders of significant political and perceptoral powers. It can be conjectured that a significant proportion of them expected to become members in the future parliaments and governments and thus hold positions of power—especially political power and perceptoral power. With regard to the position of wealth, then because of the transition from one economic system to another, the framers had only limited possibilities to foresee their future positions: There were simply too many uncertainties surrounding the regime change.

It is worth noting here that despite the fact that about one-third of the population in Estonia was made up of Russians, they were, to a large extent, excluded from constitution-making. Although seven Russian-speakers were formally members of the CA, they found it difficult to participate in its work because they did not speak Estonian, which was the working language of the CA. According to Taagepera (1994: 221), the Russian members “did not seem to have much interest in details, apart from the rights of citizens and noncitizen residents.” In addition, as Pettai (2001) notes, the Russians were also unmotivated to participate in constitution-writing because the citizenship laws adopted in the autumn of 1991 foresaw that Estonian citizenship would be automatically granted only to the citizens of the inter-war republic and their descendants (and thus declaring the 500,000 Russians living in Estonia non-citizens). Thus, the Russian representatives were not overly enthusiastic to work on a constitution for a country where they would be non-citizens.

The “subjective motivations” of the Estonian framers varied to a significant degree. Some of the framers (from ENIP) argued for the restoration of the inter-war republic, together with the 1938 constitution: for them, the 1938 Constitution carried a symbolic importance as it would have implied the continuity of the Estonian Republic and signaled to the world that the Soviet occupation had been illegal. Most of the framers, however, viewed the 1938 Constitution as problematic (since it had been written within the context of an authoritarian regime) and called for finding a balance between the ultra-democratic 1920 Constitution and the super-presidential 1938 Constitution. In general, when reading the verbatim notes of the CA, one gets the feeling that the framers, for most part, did indeed set out to find the most “fitting” constitution for the Estonian context, while drawing on Estonia’s own historical experiences with the different inter-war constitutions, the experience of other countries and the propositions of constitutional law and political science. In fact, at the first session of the CA, the members had to give an oath to the effect that they would leave aside day-to-day politics when deliberating the draft constitution. And indeed, the verbatim records show that to a large extent the deliberations took place at the level of “constitutional theories” rather than being expressed in the form of naked “interests” (Raudla 2010a).Footnote 10 As Pettai (2001: 138) puts it, CA’s “attention was focused on how institutions could be structured through incentives and deterrents to produce reasonable and balanced effects.”

The retrospective evaluations of the members of the Assembly (in their memoirs) indicate diverging views about the extent to which the discussions were dominated by current politics and interests. On the one hand, Kask (2002: 178) states:

The important component of the success of the Assembly was the atmosphere that was not dominated by the politics of the day, there were attempts to interrupt it but these attempts were relatively unsuccessful. The discussions both in the Assembly and the committees were often very heated but were carried less from the political fights of the day than we had been afraid of.

On the other hand, Adams (2002: 139) argues that:

Of course, it was natural that a majority of the members of the Assembly looked at their work not only on a theoretical level but also from the point of view of current politics and practical future implications for themselves.

Indeed, there were a number of issues where the day-to-day politics (and “objective interests”) entered the constitution-making, namely in the choice between parliamentarism and presidentialism, the powers of the president and the mode of electing the president (directly by the public or by the parliament). As can be seen from Table 1, however, the preferences of the framers with regard to the powers of the president were far from being uniform. The only grouping which appeared to have strongly uniformed preferences was Free Estonia, consisting of top leaders of the Communist Party and managers of state factories. They all favored a strong president because of their connection to Arnold Rüütel, the most likely first President.Footnote 11

Table 1 Group preferences for presidentialism versus parliamentarism in the constitutional assembly
Full size table

Within ENIP, the dividing lines between those preferring presidentialism over parliamentarism were driven by very different considerations: The ENIP members who preferred presidentialism did it mostly because they were in favor of restoring the 1938 Constitution. More moderate members of the ENIP and also the members of the Estonian Heritage Society were more in favor of parliamentarism.

When looking at the debates in the CA, the main argument of the proponents of parliamentarism in the CA was that parliamentarism is more conducive to the survival of a young democracy; references were often made to the studies by Matthew Shugart and his colleagues (including Rein Taagepera, who was himself a member of the CA). It was feared that strong presidentialism would give rise to conflicts between the president and the parliament, leading to instability. Based on the verbatim records of the debates in the Assembly alone, one could conclude that the empirically proven constitutional theories and the desired goal of a stable democracy played a major role in the constitutional choice between presidentialism and parliamentarism. The memoirs of the members of the CA, however, point to a different conclusion. As a prominent member of the CA, Kask (2002: 180) admits in his recollections of the work of the Assembly that the constitutional preference of the majority of the Assembly for a strong parliamentary system was driven by the current interests of the members rather than by the weight of the arguments. Lang (2002: 219), who was an expert to the Assembly, states that in preparing the Constitution, there was a competition between the different forces who wanted to secure their power for the future: “Therefore, every legal construction that concerned the mechanisms of power were viewed through a political prism.” Hänni (2002: 164), another prominent member of the Assembly, notes that although the members of the Assembly had given an oath not to consider the issues of current politics, these considerations rose high in discussions over the institution of president. A number of CA members (both from the Congress and the Council) have admitted that they did not want to design a strong institution of the president, since they did not like Arnold Rüütel, the most likely candidate for that position and the party he was affiliated with (see, e.g., Kask 2002: 180; Lang 2002: 219; Hänni 2002: 164). Arnold Rüütel himself was formally a member of the CA but could not take part in its discussions very actively, given his obligations to act as de facto head of state. Still, through his position as the head of the Supreme Council, Rüütel could influence the reactions of the Council to the draft constitutions (Raudla 2010a: 259). Supported by several groupings in the Council who wanted to secure that Rüütel would become the first president, the Council repeatedly sent the draft constitution back to the Assembly and demanded that the CA revised the articles concerning the institution of president (Laar 2002; Ruutsoo 2002a, b).

After the debates on the choice between parliamentarism and presidentialism and the powers of the president, the most heated discussions arose on the issue of how the president should be elected. For most part, it seemed that those who favored Rüütel as president supported the election of the president by the public (given how popular Rüütel was among the public at large) and those who disliked the idea of him becoming a president advocated indirect elections of the president. The CA’s insistence on indirect elections (i.e., by the parliament) was also one of the reasons why the Supreme Council kept sending the draft constitution back to the CA. As pointed out in Sect. 2.2, eventually, the issue was “resolved” by prescribing a different mode of electing the president for the first election taking place after the adoption of the constitution. Ruutsoo (2002a: 204) argues that this provision was “engineered” by the representatives of the Congress with a goal to guarantee that Rüütel would not become president. The supporters of Rüütel, however, still expected him to become the president, even with a system like that. This shows that there were some opaque “spots” even in the relatively “thin” veil that covered the institution of the president: It was still difficult for the framers to foresee how many presidential candidates there would be, how exactly the distribution of public support for the candidates would look like, and what the composition of the first parliament would be.

While the debates over the election-mode of the president were permeated by current politics and “objective” interests of the framers, then on many other issues, their decisions were driven by constitutional theories and the desire to put together a good constitution for Estonia. For example, as Raudla (2010a) argues, despite the prevalence of delegates from the existing legislature in the CA, the judiciary was given extensive constitutional powers and guarantees of autonomy.Footnote 12 The general attitude among framers was that guaranteeing the independence and autonomy of the judiciary was extremely important for establishing a genuinely democratic order and avoiding the politicization of courts, which had characterized the Soviet period. With regard to constitutional review, Pettai (2005: 93) notes that during the discussions of the CA, “the need for some kind of constitutional review was never disputed”, and the debates were confined to the specific “mechanics” of how the procedures and institutions would look like. As Raudla (2010a: 259) notes, the discussions on constitutional review in the CA appear to provide evidence for the arguments by Shapiro (1999: 218), who suggested that constitutional review has been adopted by framers in many countries partly because it has become to be perceived as an obligatory component of constitutionalism.

3 Description of the Content of the Estonian 1992 Constitution in Terms of a Power Relation Framework

The 1992 Constitution of Estonia is made up of 168 articles (11,181 words in English) and it is divided into 15 chapters. Three chapters pertain to the powers of the parliament, president, and cabinet (encompassing 52 articles) and the longest chapter (48 articles) entails the bill of rights. The remaining chapters are shorter and are concerned with foreign relations, local government, courts, defense, and finances.

The content analysis of the Estonian constitution shows that it provides for altogether 530 units of power relations, which is quite close to the average of the sample of 16 countries analyzed in the Veil of Ignorance Project (564.6). At the same time, the density of power relations in the Estonian constitution is 47.4, which is very close to the sample average (46.7). Table 2 provides an overview of the power relations found in the Estonian constitution, broken down by the direction of power, source of power, and type of power.

Table 2 Characteristics of power relations in the Estonian 1992 constitution
Full size table

Altogether, with regard to the source of power, 80.6 % of the power relations are based on authority (which is higher than the sample average of 75 %), 0.9 % are based on wealth and 7.2 % on knowledge. Based on these ratios, one would argue that the Estonian constitution was drafted behind a rather opaque veil (at least to a large extent, this was indeed the case in Estonia as shown above in Sect. 2 although there were also some transparent spots in the veil).

With regard to the type of power, the division of social and instrumental power is almost equal, with social power recorded in 48.3 % of the units and instrumental power in 51.6 % of the units (this coincides with the sample average). Given the opaqueness of the veil in the Estonian case (as demonstrated by the high ratio of power relations based on authority), one would have expected the ratio of social power relations to be higher.

Concerning the direction of power, in 95.1 % of the units the direction of power is positive and only in 4.9 % of the units it is negative (which is a little higher than the sample average of 3.8 %). Based on the framework of the VOI Project, one would have expected the ratio of negative power relations resulting from constitution-making behind the veil of uncertainty to be higher.

The main agents identified in the constitution are the parliament (the Riigikogu) and its membership, the President, the Government (i.e., the cabinet of ministers), the courts, and the Chancellor of Justice. In addition, in the bill of rights part of the Constitution, the citizens and residents of Estonia are the main agents, in their capacity to use their rights (and hence “everyone” and “Estonian citizens” or other specified groups of individuals have been identified as agents). Table 3 provides an overview of the main agents and the power relations associated with these agents in the Estonian constitution.

Table 3 The powers of the main agents in the Estonian constitution (number of occurrences)
Full size table

The main capacities of the Riigikogu are to adopt legislation, to authorize the candidate for Prime Minister to form the Government, to elect the President, and to appoint to office the Chief Justice of the Supreme Court, the Auditor General, and the Chancellor of Justice (and other high offices). As can be seen from Table 3, with regard to the power relations that the constitution attributes to the parliament, just below half of the power relations are based on social power (53), just above half are based on instrumental power (57). The main capacities of the President are to represent Estonia in international relations, to proclaim laws, to review the constitutionality of laws, and to make proposals to the parliament for appointments to high offices such as the Chief Justice, the Chancellor of Justice, and the Auditor General. Again, an overwhelming proportion of the power relations is positive.Footnote 13 The main capacities of the Government are to execute the policies and to administer the implementation of laws and issue regulations. In the case of government, the constitution foresees 47 social power relations and 31 instrumental power relations; 81 positive power relations and 5 negative power relations.Footnote 14 The courts are attributed predominantly social power relations and only one provision refers to negative power.Footnote 15 It is interesting to note that while for the parliament, instrumental power relations dominate, then for the government, the president and the courts, social power relations are more frequent. In the bill of rights, in 40 cases “everyone” is identified as an agent and they have been accorded capacities in terms of positive power relations. In 15 units, the Constitution, instead of stating “everyone,” accords rights to the Estonian citizens (those rights mostly pertain to political rights like voting, membership in parties, etc.).

4 Discussion

The veil of uncertainty influenced constitution-making in Estonia at different levels. First, it can be argued that the decision of the Supreme Council to convene a CA—a separate entity that would also involve the representatives of the Estonian Congress—for drafting the constitution (rather than having the constitution written by a committee of the Supreme Council) was made possible by the extreme uncertainties that characterized the situation of Estonia and its government during the putsch in Moscow in August 1991. Before the coup, the divisions between the different representative bodies and the groupings they entailed ran deep and Estonia was “seemingly nowhere near ready to undertake the solemn and consensus-obliging task of constitution-making” (Pettai 2001: 112). However, the demonstration of unity between the Supreme Council and the Estonian Congress was needed to secure both domestic and international credibility of the declaration of independence. In order to reach the compromise, both sides were willing to make concessions that would have been difficult (or even impossible) during the times of “normality,” which is why it has been termed a “miracle compromise” (Pettai 2001: 115). By agreeing to share the constitution-making with the Congress, the Supreme Council ceded a significant amount of its formal power and gave up the proposal to declare the Council as a constitution-making body. In return for being included in the making of the new Constitution, the representatives of the Congress gave up their ideas of re-enacting the 1938 Constitution or calling for general elections of a new constitution-making body. Taagepera (1994: 216) remarks that after the threat of the coup passed, some representatives on both sides began to feel they had conceded too much, but the agreement had already been signed. Given that the resolution to convene the Assembly was in the same decision with the declaration of independence, contesting one would have jeopardized the other.

With regard to the work of the CA itself, the making of the 1992 Constitution by the CA in Estonia was driven by a complex configuration of subjective motivations, objective interests, constitutional theories, symbolic considerations, and historical experiences. As Raudla (2010a, b) emphasizes with regard to the role of institutional and partisan interests in the constitution-making in Estonia, there is certainly some evidence for their influence, these factors cannot account for the “entire story.” A mix of additional considerations (theories of political science, the substance of inter-war constitutions and the types of polities these previous constitutions gave rise to, symbolic considerations, and the prevailing international discourse on constitutionalism) have to be taken into account in order to understand the causal mechanisms behind the proposals, choices, and decisions of the Estonian framers. In other words, as Raudla (2010b: 74) notes, the Estonian case provides evidence for “the interaction between strategic calculations and symbolic meanings that the actors attach to particular choices in the making of the constitution.” Thus, while strategic calculations certainly played a role in delineating the powers of the parliament and the president (and the preference for a strong parliament was certainly influenced by the fact that most of the framers were members of legislature-type body at the time of constitution-making), then various symbolic considerations shaped constitutional choices as well. For example, the symbolic meanings attached to the inter-war constitutions played an important role in the institutional choices. Because the 1938 Constitution was seen as overly authoritarian, many of the framers were reluctant to copy the presidential system it had entailed. At the same time, because the inter-war constitutions had another symbolic connotation—that of re-asserting the continuity of the Estonian statehood and sovereignty—the members of the CA did “borrow” a significant number of provisions from these constitutions. Also, the desire to establish a clear break with the Soviet heritage shaped a number of institutional choices, like granting strong guarantees of independence for the judiciary.

As the constitutional political economy literature in general and the VOIP framework would have predicted, in the Estonian case those issue areas where the veil appeared to be the least opaque gave rise to the most extensive discussions. These were the areas where it took the longest to reach a compromise among the framers: the mode of electing the president and the powers of the president. Given the popularity of the head of state (Arnold Rüütel) during the work of the CA, many framers thought that he would become the first President. That, in turn, motivated his proponents to argue for strong presidency and his opponents for a presidential institution with relatively limited powers.

On the other hand, one can also argue that in some areas of debate, it was the reduction of uncertainty that facilitated agreement in the CA. This was particularly the case with the adoption of the citizenship law by the Supreme Council (already during the beginning of the work of the CA), which foresaw the granting of automatic citizenship only to the citizens of the inter-war Estonian Republic and their descendants. That effectively meant that a significant proportion of the Russian-speaking inhabitants would be excluded from participating in national elections and the Assembly was essentially crafting a political system that “would be dominated and run by Estonians” (Pettai 2001: 120). That in turn meant, as Taagepera (1994) and Pettai (2001) note, that the members of the CA did not have to think about arrangements that would have accommodated the Russian-speaking population (e.g., in the form of consociational governance structures), which, one can argue, made the reaching of “agreements” on the constitutional structures easier. As Pettai (2001: 120) puts it, “the subsequent process was in fact more akin to writing a constitution for a homogeneous nation-state than for the multiethnic republic that the country actually was.” Thus, it appears that in some situations reduced uncertainty can facilitate the reaching of constitutional agreements. This, at least at first sight, appears to go against the predictions of the constitutional political economy literature. However, one should not forget that this higher “certainty” was achieved in Estonia by effectively excluding a sizable minority from political decision-making and hence could not be viewed as a consensus (which is a recommended decision rule in the constitutional political economy framework).

What is also interesting in the Estonian case is that although there was extremely high uncertainty with regard to the future positions of wealth and economic power among the framers (both on individual and group level), the number of power relations that pertain to wealth is very limited in the constitution. One possible explanation is that the framers in Estonia had a pretty clear idea of what kind of elements should be included in the constitution and viewed the constitution primarily as a document that deals with the delineation of political authority rather than economic aspects of social life. In addition, one could argue that extreme economic uncertainty—when the framers do not even know how the entire economic system is going to look like in the near future—can render it difficult to reach consensus on those power relations that have to do with wealth. Because of the prevailing uncertainty (or one could even say ignorance) with regard to how the market economy would operate and to what kind of outcomes it would lead to and how the state-market nexus would look like, it was probably very difficult for the framers to theorize—even in general terms—what the effects of provisions pertaining to wealth relations could be, how the causal mechanisms could play out, etc. In other words, because of the lack of “constitutional theories” in that regard, it was difficult for the framers to even start a negotiation or bargaining process over constitutional provisions concerning power relations based on wealth.

It is also worth noting that even provisions concerning social rights in the Estonian constitution are rather constrained. Despite the opaqueness of the veil (and hence uncertainty with regard to their own position in the future), the framers were reluctant to include provisions pertaining to extensive welfare entitlements in the constitution. As Pettai (2001: 123) explains, this was in part due to “the legacy of Soviet socialism, which many liberal-minded Estonians did not remember fondly.”

Evaluating the Estonian constitution-making in the light of the propositions of the contractarian approach suggesting that constitutional choice should take place behind the “veil of uncertainty,” which in turn is facilitated by the generality and durability of rules, an interesting feature of the Estonian case is that the “thin veil” behind which the discussions on the implementation law of the constitution took place actually facilitated reaching agreement on some provisions of the constitution negotiated behind a “thicker veil.” As mentioned above, the “implementation act” of the constitution (the so-called Chap. 17) foresaw different rules for the first elections of the Riigikogu and for the President. The different method for the first election of the President (especially having the first round as direct election by the public) was introduced in order to appease the public, but also several members of the Supreme Council and the Government, who favored the direct election of the president. The more “official” explanation for this option was that since the party system in Estonia was still in the process of being formed it would have been hard to expect that the political groupings in the parliament could immediately agree on a common President. Given that the constitution requires a 2/3 majority in the parliament for electing the President, it was suggested that the political system was too young to achieve such compromises only after the very first election. The implementation chapter was meant to “facilitate a gradual habituation to the new system” (Pettai 2001: 125). Altogether, the Estonian case shows, when there are some upstream and/or downstream constraints on the framers (who, as a result of these constraints, have to consider whether the produced document will pass through the institution acting as the constraint), a distinction between rules for the long term and rules for the short term can help to facilitate agreement. The Supreme Council (acting as an upstream authority) repeatedly sent the draft constitution back to the CA and sought to persuade the Assembly to strengthen the powers of the president (alongside with the requirement to replace indirect elections of the president with direct elections). The implementation law, with its different rules for the first election, facilitated the acceptance of the constitution by the upstream authority. Thus, curiously, the Estonian case shows that adopting two sets of “constitutional” provisions—“thick veil” provisions that take effect in a longer term and “thin-veil” transitional provisions that are enacted immediately—can facilitate the achievement of agreement between the conflicting parties (like between the CA and the institutions acting as upstream or downstream constraints on the constitution-making body).

What also seems to go against the propositions of the contractarian approach in the Estonian case are the provisions in the implementation chapter according to which, for 3 years following its adoption, constitutional amendments could be made with relaxed majorities in the parliament or initiated by popular initiative (requiring 10,000 signatures).Footnote 16 This was seen as necessary to allow flexible adaptation to changing circumstances, but from the point of view of the contractarian approach, it would have reduced the “durability” and “stability” of expectations, which are supposed to facilitate the thickness of the veil and the achieving of an agreement.

5 Conclusion

Altogether, the Estonian case corroborates the prediction of the veil of ignorance framework outlined by Imbeau (2009), and Imbeau and Jacob (2011, 2015) with regard to the proposition that predicts that in the context of uncertainty, the framers focus on those power relations that are based on authority (rather than on knowledge or wealth), but the case appears to go against the proposition that behind an opaque veil the framers write more negative power relations into the constitution. Also, the theoretical prediction stating that behind the veil of uncertainty, framers would include more power relations based on social power than instrumental power in the constitution was not borne out in the Estonian case. Whether this can be counted as evidence that there were some transparent holes in the predominantly opaque veil or whether the theoretical proposition itself should be revised would warrant further discussion in theorizing over the effects of the veil of ignorance.

As a whole, the case study indicates that constitution-making in Estonia took place in the midst of high uncertainty and that influenced the content of the constitution. The analysis above indicates that, while for most part, the opacity of the veil can indeed facilitate the reaching of compromise, there are situations where the transparency of the veil contribute to reaching agreements among the framers and, conversely, that in some situations or issue areas, the veil can be so thick that the uncertainties involved make it very difficult to reach consensus.

As the constitutional debates over the mode of electing the president indicate, the relative transparency of the veil (i.e., the predictability of the effects of particular constitutional rules on constellations of power in the immediate post-constitutional phase) can make it difficult to reach consensus on what the best rules would be for the long term. In the Estonian case, this conundrum essentially led to the adoption of the implementation act of the constitution (which foresaw different rules for the first post-constitutional election). Thus, the Estonian case demonstrates that in some cases, adopting two sets of “constitutional” provisions—“thick veil” provisions that take effect in a longer term and “thin-veil” transitional provisions that are enacted immediately—can facilitate the achievement of agreement between the conflicting parties involved in constitution-making.

The Estonian case also shows that even in the case of extreme uncertainty with regard to future positions of wealth—as prevailed at the wake of the transition period, with transition from planned economy to market economy—the framers may be reluctant to include a significant number of power relations based on wealth. Even if, in theory, the uncertainty with regard to future wealth position of the framers could make it easier to reach agreement on power relations based on wealth, the constitution-makers at a particular space and time may be more influenced by what they think should be included in a constitution and what the constitution should signal to the population at large and even to the international community.

In the Estonian case, the framers considered it important to focus on those power relations that are based on authority, because, in many ways, this is what the prevailing constitutional theories (and the international constitutional discourse) at the time “told” them to focus on. Also, given that the main reference constitutions the Estonian CA drew on—the inter-war constitutions of Estonia and the constitution of Germany—also focused on power relations that were based on authority, there were also some important lesson-drawing and path-dependence effects at play here.