Desirability of Unamendable Provisions

Recall that the first potential reason to make provisions unamendable is to protect them from the personal interest of subsequent legislatures, as well as from pas­sionate decisions that would be regretted later.

However, the use of unamendable clauses for this purpose is problematic. First, if we consider the use of unamendable provisions to provide protection from self-interest, one would need impartial drafters to achieve impartial provisions in the first place. While many authors argue that drafters are less self-interested than politicians in normal times,^[557] this by no means implies that constitutional drafters do not pursue their personal aims while drafting a constitution. It can be assumed that drafters of constitutions are also, at least partially, motivated by their private interests, and thus cannot be expected to impose selfless clauses. While it is easy to see that this problem is more likely in nondemocratic settings with an unelected constitutional assembly, the case for selfish drafters does not disappear in a rep­resentative, elected assembly. Constitution-making is a rare event, which reduces the possibility for citizens to hold drafters accountable since there is no option to vote them out of office.^[558] More generally, the notion that politicians are rational and self-interested is not novel; it is the main foundation for the research field of public choice and political economy.

Historical evidence tends to support the view that drafters are selfish. Historically, personal motives have played a key role, for example, in the drafting of the US constitution.^[559] Furthermore, few countries have a political elite that is large enough to prevent drafters from subsequently entering the political arena. Selfish drafters might attempt to improve their future status by using unamendable provisions to raise the future cost of changing provisions. This is particularly problematic if an elite in power abusively inserts an unamendable provision in order to protect itself against the opposition in conflicts with high stakes.^[560] It has been argued that an unamendable provision is a tool that allows selfish constitution-makers to install a preferred power asymmetry for the majority.^[561] In this light, it is doubtful whether the availability of unamendability will make self-interested decisions of politicians more or less likely, taking the drafters’ own interests into account.

Regarding passions, times of constitution-making are rarely times of calm and rational reasoning. Moments of constitution-making, in fact, are at an increased risk of being times of heightened passions. In the past 40 years, more than 200 con­stitutions have been written in times of crisis.^[562] This implies that drafters are unlikely to be particularly cool-headed in their decision-making during the drafting process.^[563] Therefore, since the process itself is likely to be passionate, it seems incongruous that the moment of drafting is the optimal time to install a device against passions. In the words of Elster, “[i]t is mainly if the framers are impartial and know that impartiality may be lacking on future occasions that they will have an incentive to pre-commit themselves. Although this case cannot be excluded... there is no reason to think that it is typical or frequent.”^[564] Combining the two arguments presented above, one can conclude that drafters are unlikely to act without self-interest and, even if they do so, their passion would be another obstacle to create unamendable provisions that protect from these same passions and interests.

The second reason to motivate the use of unamendable provisions for pre-commitment is nonstrategic time inconsistency. Discounting payoffs over time means that individuals often have less utility or disutility from distant consequences than from present ones. One relevant reason for this is that waiting is costly, and another relevant reason is that, over time, payoffs present the risk of either disap­pearing or depreciating. When either of these two effects is strong, they can lead to nonstrategic time inconsistency, which produces a switch in choice among varying delays while maintaining stable preferences.

Regarding the first reason to discount, a large number of studies have shown that people often display self-control problems, which lead them to choices favoring immediate gratification over welfare-enhancing alternatives.^[565] This is the classic problem of the dieter. While a person trying to lose weight is perfectly aware that the long-term effects of a salad are welfare-enhancing, the direct gratification of eating the French fries is too tempting to resist. For these cases, pre-commitment is welfare-enhancing, since it eliminates temptation problems hence allowing the agent to choose the welfare-enhancing option, by disallowing him or her to choose the tempting but welfare-decreasing option.

Regarding the second reason to discount, it has been shown that agents who do not face behavioral biases will reverse their choice when the probability of the payoff disappearing is uncertain.^[566] For these agents, the welfare-enhancing mech­anism, as opposed to pre-commitment, is an increase in flexibility that allows them to update their choice upon the availability of new information.^[567]

From this perspective, a pre-commitment device would be useful to solve nonstrategic time inconsistencies in constitutional choice if and only if these inconsistencies are driven by behavioral biases and not by the uncertainty of the future.

From a practical perspective, it has been argued that the risk of binding future generations is only of limited importance given the fact that a constitution’s average lifespan is 19 years, which is roughly one generation.^[570] Following this line of reasoning, unamendable clauses may be a more fitting name than “eternity” clause, since the average lifespan of a constitution is nowhere near eternal. The problem with this argumentation stems from the effect unamendable clauses have on a constitution’s lifespan. If we turn the argument around, the question becomes whether unamendable provisions have an effect on a constitution’s lifespan. Looking at the empirical evidence at hand, the closest proxy for a direct effect of unamendability is the effect of the ease of amendment. Empirically, ease of amendment has an inverse U-shaped effect on the constitutional lifespan.^[571] Since unamendable provisions tend to take the ease of amendment to one extreme end of this variable, the evidence at least hints at a negative effect of unamendable pro­visions on constitutional longevity.

The third reason to use unamendability is to overcome the problem of strategic time inconsistencies, which could potentially be mitigated through the existence of pre-commitment devices.

In other words, unamendable provisions can be seen as a separation of powers done over time. When a certain clause is made unamendable, parts of the political power, namely the power to amend, remains with the original drafters and prevents future governments from using it. This separation of powers over time has the advantage that the commitment is credible, assuming the constitution is enforced and no new constitution comes into action.^[573] Using the case of the Weimar Republic in Germany as an example (which had very low requirements for amendment),^[574] it becomes clear how this commitment can add to constitutional stability. However, this separation over time has a problem: its effect in times of crisis or in case of substantial changes. While an “orthodox” separation of powers still allows for all powers to act in unison if circumstances require it, a separation of powers over time, such as the one generated by unamendable clauses, is unable to allow for unified action in times of crisis. It has been argued that unamendability acts like a lock on the door, which is effective in normal times but cannot withstand extraordinary force.^[575] While vertical and horizontal separation of powers resembles locking a door with more than one key and distributing them among people with different interests, making certain provisions unamendable can be seen as locking the door and throwing the key away.

The second issue is using unamendable provisions to make constitutional agreements intended to end conflicts more credibly. Typical examples of this issue are amnesty clauses protecting the former government from prosecution after a regime change and gag rules to prevent existing disputes from worsening. Amnesties are a case where a strategic time inconsistency arises because the opposition to the old government has incentives to agree to an amnesty clause when the threat of conflict still exists, and then renegotiate on their promise once they are in power.^[576] Being aware of this risk, the incumbent government might prefer a violent conflict as opposed to stepping down and facing prosecution despite the promise of an amnesty. In this situation, making the promise as an unamendable constitutional provision can increase the promise’s credibility and thereby tip the scales to prevent violent conflict. However, this argument only deals with the case of amnesties that are in the interest of all parties. The flipside of making amnesties unamendable is the risk that an autocratic government will use amnesties to absolve themselves from crimes committed during their reign.^[577] In this way, they protect themselves in case of a transition by having raised the costs of removing the amnesty, since rewriting a whole constitution is always costlier than amending a single provision.

Even in the aforementioned case where the opposition agrees to provide an amnesty, the victims of the crimes may not be represented by either of the two groups. One can imagine a scenario where two elite groups struggle for power and use an unamendable amnesty even though victims in the general population are strongly opposed. Recognition of these types of risks is one of the reasons why international law tends to disallow amnesties.^[578]

The rationale for gag rules is to increase the credibility of compromises between conflicting groups by protecting them with unamendability and thereby also silencing future motions to renegotiate them. An interesting example of a gag rule in a different context can be found in the status quo rule from 1852 of the Church of the Holy Sepulchre in Jerusalem. The different Christian groups sharing the church would resort to violence in an attempt to obtain more of the exclusive rights for the use of specific parts of the church. The Ottoman ruler at the time fixed the status quo and decided that the rule would never be changed in the future, thereby making it effectively unamendable.^[579]

The problem with the use of gag rules is twofold. First, both necessary com­promises as well as issues where one can find clear winners and losers can be gagged. Using the gag rule in this way might fix a temporary power difference between groups in society and prevent renegotiation, but it might also lead to dangerous differences between de jure and de facto power in the future. Second, silencing an issue does not necessarily help to solve its underlying problems and may even exacerbate it. In the example of the Holy Sepulchre, violent outbreaks due to the underlying conflict have not completely stopped, despite more than 150 years of having the gag rule in place. Altogether, both examples of conflict-solving mechanisms show that they are problematic as commitment devices to overcome strategic time inconsistency.

Finally, unamendable provisions could be used to ensure efficiency gains. This would be done by expanding the time horizon and by preventing the cyclic amending of constitutional provisions following changes in political power.

Cyclic amendment would only be a problem if the shifts in majorities were large enough to fulfil the amendment requirement in the first place, which is typically a two-thirds majority requirement. Even if a society has big swings in its majorities, making provisions unamendable in the constitutional drafting process would either require selfless drafters, which seems to be an unlikely assumption, or simply establish a “first-mover advantage” to the faction which is enjoying the majority during the constitution’s drafting, which is not necessarily a more desirable alternative.

It has been shown that the effect of the flexibility of the amendment process on a constitution’s longevity follows an inverted U-shape.^[580] In other words, constitu­tions that are very easy or very difficult to amend are more likely to have a shorter lifespan. This empirical observation fits with the theoretical prediction that super­majorities, providing a middle ground for the amendment process between simple majority rules and unamendability, give great stability to a constitution.^[581] Therefore, choosing the most extreme version of a difficult amendment procedure, namely unamendability, seems to be a suboptimal measure to increase the time horizon, all other things being equal.^[582]

Beyond these potential reasons for commitment outlined by Elster, protection of the constitution’s core is yet another motive discussed in the literature about una­mendable constitutions. In other words, unamendable provisions might be con­sidered a credible commitment to a particular set of values. While amendment powers are argued to be implicitly unable to dismantle a democracy,^[583] it is unclear how this implicit ban would be enforced and how to distinguish which amendments would constitute a destruction of democracy and which would not. Unamendable provisions might be used to protect the spirit of a constitution, and thus democracy in an explicit way.

The core or spirit of the document is not necessarily at risk from any of the four reasons discussed previously, but might nevertheless be a value that drafters want to commit to. It has even been argued that the main reason to use unamendability is to protect the founding myth of the constitution from change. In other words, una­mendability is the guardian of the constitution’s identity.^[584] Once this protection fails and the relevant provisions are changed, the society might veer into chaos and possibly civil war.^[585]

The upside of protecting a set of core values is that the risk of abusing of the amendment process that destroys constitutionalism or democracy is mitigated, assuming that the constitution is set in a way that these values are included and protected by unamendable provisions. From the perspective of normative individ­ualism, this approach leads to problems. The constitution in itself has no inherent value besides the effect it has on the individuals who are living under its’ rules. It is important that a society is not seen as just a different individual, but as an aggre­gation of the individuals living in it.^[586] Therefore, protecting the constitution’s spirit or integrity is only valuable if it has a positive overall impact on aggregated welfare in society and is not necessarily positive or negative per se.

There are, however, two potential downsides when considering the overall impact on the aggregate welfare of a society. The first downside becomes imme­diately clear, namely that the tool of unamendability can also be used to protect autocratic values. As an example, a dictatorship oppressing a minority can enshrine constitutional provisions that will perpetually limit the minority’s power. This not only justifies the dictatorship oppressing the minority group but also makes it harder for the minority group to find legal channels to challenge the oppression. Constitutions are a part of many different forms of government. It is unclear how to determine which constitutions incorporate a “good” founding spirit and which ones are more on the side of “evil”. Autocracies might create constitutions that protect their core values and the sources of their power simply to increase the cost of transitions in case of an uprising against them. It is difficult to see how a govern­ment formed by selfish actors can be stopped from using unamendability simply for their own benefit.

Another dangerous downside of using unamendability to protect the core or spirit of a constitution relates to the fact that constitutional bargaining is costly. It is argued that protecting the core set of values is not a problem as long as citizens still have the constitutive power to replace the old constitution with a new one.^[587] Now, assuming that most provisions of the old constitution are acceptable to the society, but a change in preferences has made one or more of the “core” provisions unpopular. All citizens would like to change this provision, but to do so they must draft a new constitution. Even if everyone is satisfied with the other parts of the constitution as they are, placing them back on the negotiation table by redrafting the whole constitution opens the door to strategic bargaining.^[588] This cost arises because choosing among constitutions can be likened to a scenario called the “Battle of the Sexes” in game theory.^[589] Even when all members of society are better off with any possible constitution, each member still ranks the different options in different ways. For example, assume that the old constitution did not include a positive right to work. While left-wing as well as right-wing groups would prefer to have a constitution, the left-wing group would like to include such provisions and the right-wing would prefer to keep the old setting. In this way, costly negotiations due to strategic options would arise. If these negotiating costs are high enough, a society will refrain from drafting a new constitution even if all of its members dislike certain core unamendable values of the current constitution. This argument presents additional support for the classical critique of “dead-hand” constitutionalism and casts some doubts on whether the protection of the constitution’s spirit is something beneficial to society.^[590]

5