Like sunset clauses, experimental legislation is far from being an invention of our times; instead it has a long tradition in Europe. The first signs of experimental legislation in France date back to the kingdom of Louis XVI and appear to have been motivated by ‘the wish to improve the effectiveness of the administration [of the country]’.
‘Experimental legislation is the very method of social growth’, affirmed W.S. Jevons in 1880. This scholar argued that the employment of the experimental method, already successfully used in natural sciences, was applicable to law: legislators should undertake small steps in the enactment of any new law, submit it to the test of reality, adapt it to local specificities and assess its effects both empirically and statisti- cally. More than one century later, experimental legislation is still far from being a commonly used instrument in most jurisdictions. Nonetheless, the idea of experimenting with laws has drawn more interest in the twentieth and twenty-first centuries.
John Dewey was one of the twentieth century scholars who advocated the use of an experimentalist approach to policy and lawmaking. Dewey supported a pragmatic and ‘experimentalist’ problem-solving method to remedy various equal protection problems. Dewey considered that legal precepts, like every social policy should be subject to ‘a more experimental and flexible logic’ in which rules and principles should be treated as ‘working hypotheses’ that should be constantly tested. Dewey was ‘a man ahead of his time’ who did not have in mind the concrete case of experimental legislation or the characteristics of experimental designs. Instead, as Bradley Karkkainen explains, Dewey simply advocated ‘a dynamic, experimental, iterative, and adaptive problem-solving approach’. Dewey’s experimentalism has been described as a ‘day-today process of self-government, which he equated with public problemsolving uncovering social needs and troubles and best judging how the trouble is to be remedied’.
Dewey’s experimentalism cannot be confused with a plea for laboratory experiments. Instead, the philosopher insisted on the need to ‘keep premises and conclusions open and subject to constant testing and revision’. Dewey conceived state and regulation as dynamic realities that must evolve as the underlying social conditions do; and such evolution can be executed by constant experimentation. Dewey had in mind a continuous process of learning by doing, accompanied by a cyclical system of planning, implementation, evaluation, and planning of alternative policies. One can almost see in Dewey’s pragmatism a translation of the theory of evolution to the lawmaking process. Policies and laws are assumed to be fallible and should therefore be systematically tested and corrected.
The re-emergence of an experimental approach to lawmaking in the twentieth century has equally much to owe to Justice Brandeis and his ‘states-as-laboratories’ metaphor. In Chapter 1, Brandeis’ dissenting opinion in New State Ice Co. v. Liebmann has already been mentioned, but its multiple interpretations were not explored. Brandeis stated:
It [was] one of the happy accidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory, and try novel social and economic experiments without risk to the rest of the country.
Justice Brandeis firmly believed that legislation should be supported by social facts, experience and statistics. In this states-as-laboratories metaphor, Brandeis praises state experimentation as a method of enhancing the ‘triumphs of invention’ in social and economic fields while safeguarding the legal order against the harms of failure. Should the experiment fail, then the damage would be limited to the ‘courageous state’ that decided to experiment. Brandeis’ dissenting opinion can first be interpreted as a plea for the use of experiments as instruments of innovation: states should have the right to enact experimental legislation and innovate beyond federal policies. This idea of states as laboratories of democracy was earlier present in the work of James Bryce who observed that federalism enabled ‘a people to try experiments in legislation and administration which could not safely be tried in a large centralized country. A comparatively small commonwealth like an American state easily makes and unmakes its laws; mistakes are not serious, for they are soon corrected’.
On the one hand, Bryce’ and Brandeis’ perceptions of ‘states-as- laboratories’ and experimentation can be interpreted as methods of converting the lawmaking process into a learning process. New laws are first tried on a small-scale basis, shaped by the facts, allowing ‘other States [to] profit by the experience of a law or a method which has worked well or ill in the State that has tried it’. On the other hand, a second interpretation of this metaphor could be associated with a plea for federalism. The ‘states-as-laboratories’ assertion can be read in the context of a Madisonian defense of diffusion of power to states and communities. This is a secondary interpretation, which is not developed in this study.
In the 1950s, F.K. Beutel conceived the idea of a ‘Science-State’ guided by strict rationality where policies would be enacted according to the scientific method, tested and improved in conformity with the results observed. Legal experiments would be particularly appropriate for fields characterized by frequent technical and social changes, such as road traffic and educational policies. Experiments in law were ‘the way’ to implement innovative social reforms. Beutel did not conceive experimental legislation in its modern form as a temporary derogation of rules. Rather, his theory of the ‘Science-State’ referred to a more global approach to both experimental policy and lawmaking as a concretization of a learning approach.
Although it is unclear whether Brandeis’ words produced any type of direct influence in Europe, experimental legislation emerged some decades later in the Netherlands and in Germany.
Experimental legislation emerged in Germany at a time of ‘legislative crisis’. In the 1960s, trust in the virtues of law and legislation was severely eroded. The progressive widening of legal subjects, lack of clarity and intelligibility of rules, and excessive workload of the ministerial departments in charge of legislating and overseeing the legislative process contributed to a true crisis in legislation. This awareness that the legislative process was ‘getting out of hand’ explains the enhanced attention for the field of Legislative Studies (Gesetzgebungslehre) in the 1960s.
Peter Noll became a prominent character in the German debate regarding the development of a theory of legislation, arguing that the latter should be supported by scientific work. This distrust in legislation and its implementation were at the root of the claims regarding the need to rationalize the lawmaking process, namely, by incorporating multidisciplinary and empirically-driven methodologies. Experimental lawmaking was then conceived as a process that would ensure that the regulation of society was based on facts and figures. This would be an alternative to the assumption that new laws purely based on theoretical knowledge would be effective. The main idea behind experimental legislation in Germany at the time was to submit laws to a reality check, gather information as to their effects, renew and optimize legislation through the incorporation of new elements.
Experimental laws fit Peter Noll’s discourse that ‘all new laws are experiments’. Every time the legislator implements a new law, it will inevitably draw valuable lessons from it. However, in practice, this was not always possible in the first examples of experimental laws in Germany. The experimental provisions on the qualifications of lawyers in the former Federal Republic of Germany in the 1970s illustrate this initial difficulty to ‘learn’ from experimental lawmaking (see example 2.1).
EXAMPLE 2.1 EXPERIMENTAL LEGISLATION ON LEGAL EDUCATION IN GERMANY
The first examples of experimental legislation in Germany date back to the 1970s, namely, to the experiment with the ‘one-step’ qualification of lawyers implemented in 1971. This experiment emerged in a period of student contestation and greater demand for a more dynamic regime for lawyers. A political compromise was then achieved allowing the Lander to experiment for five years with a different curriculum and approach to the qualification and training of lawyers. The ‘experimental character’ of these laws was not as explicit as it has been in more recent examples of experimental laws, but it clearly resulted from its context. Although this experiment was not a success and most of its participants abandoned it before the end of its duration period and evaluation, this was the first of many experiments at the state level.
The field of education was one the first fields to welcome experiments, not only in Germany but also in the Netherlands. In this country, the first experimental rules emerged in the 1960s as a means to enhance the quality of education by experimenting with new methods (see example 2.2).
EXAMPLE 2.2 EXPERIMENTAL LAWS ON EDUCATION IN THE NETHERLANDS
Among the first Dutch examples of legislative acts containing experimental dispositions is the Statute of 14 February 1963. This statute granted, on an experimental basis, a greater leeway to secondary schools to decide upon educational programmes. The Experimental Statute on Education (Experimentenwet Onderwijs) of 9 July 1970 opened the door to a number of experiments. This statute conferred competence on the responsible Minister, after having consulted the Education Council, to promote experiments with other forms of secondary school programmes diverging from ordinary forms of education. The statute did not specify exactly what type of alternative educational programmes could be experimented with. The initiative to promote an experiment could, however, be taken by numerous institutions as long as it was submitted for the approval of the Ministry of Education and Science and, as stated in article 2(2), the duration of the experiment did not exceed the period of ten years (which could be prolonged for a five-year period).
In the last decades, the interest in experimental legislation has increased in the Netherlands and Germany and examples of experiments can be found in multiple fields, including social and unemployment policies, deregulation of telecommunications and road traffic. These more recent examples, as well as the respective functions performed by sunset clauses and experimental legislation, are further examined in 2.4.
This brief overview of the historical background of sunset clauses and experimental legislation aimed to provide a glimpse of the context in which these instruments emerged. Although these legislative instruments still partially reflect this historical background, they have evolved and developed specific characteristics which allow us to identify them in legal texts and distinguish them from apparently similar instruments.