Ideation and Innovation in Constitutional Rights

A Defining Constitutional Innovation

Studies of innovation in markets often draw on Joseph Schumpeter, who placed his conception of “innovation” at the center of his analysis of economic and social change.^[7] Schumpeter’s view of market innovations included new products, new means of production, and new ways to organize business. He emphasized the role of entrepreneurs in creating “new combinations” from existing resources. Understandably, the operative word in his well-known concept of “creative destruction” is “creative.” But crucially, for Schumpeter, it is not merely the conception of a new idea that matters (which he calls “invention”), but instead its application to the market (“innovation”). In the Schumpeterian view, the latter process of the application and marketing of the idea—a phase that would seem derivative and unimaginative—was especially responsible for economic impact.

Schumpeter’s study of markets identified sources of resistance to innovation, namely the inertia and habits of existing players. Innovation requires something of a conceptual breakthrough, which may be marginal or more radical. Sometimes a series of innovations co-occur in a particular time or space, and if significant enough, these can create technological revolutions. Indeed, a close inspection of most “inventions” reveals multiple contemporary “inventors.” For example, both the Wright Brothers and the Brazilian Alberto Santos Dumont were working on early aeronautics at the same time.^[8] The former are celebrated in North America while the latter is a national hero in Brazil. Personal computing had numerous inventors. And indeed, the emphasis on innovations as abrupt shifts may also lead an analyst to ignore the cumulative effect of smaller, incremental changes.^[9]

Drawing on this literature, we might start by thinking about the written constitution as a commercial product. The product analogy draws on similar work in the field of contracts that date back at least to Arthur Leff, who observed that consumer agreements and physical products have many similarities.^[10] There are competitive markets for both contracts and products, and both often come mass-produced rather than custom-designed.^[11] Further, producers of both types of products, argued Leff, fail to internalize the cost of the product’s harm to consumers.^[12] Due to these similarities, treating form contracts and boilerplate terms as attributes of physical consumer products is the basis for much legal commentary.

A constitution is, of course, a special type of product, if you think of one of its goals as organizing further production, namely that of public goods within the state. The analogy to contracts is admittedly imperfect: there is not a competitive market for terms, since people cannot easily move to the jurisdiction whose constitution they like best. Unlike a form contract, the constitution functions on an ideological level and is supposed to be custom-designed for a particular people at a particular time. Yet there is a good deal of copying among constitutions. Prior research has shown that even two centuries after these concepts were first introduced into constitutions in 1787, “We the People” is still the most popular phrase in constitutional preambles; 25 is the most common minimum age for a legislator, and a 2/3 majority is still the most common legislative threshold for constitutional amendment.^[13] Written constitutions display far less originality than might otherwise be expected.

Furthermore, like products, constitutions are produced: meaning that they are the outputs of a production process. Following Schumpeter, we can distinguish between innovation in the constitutional product, innovation in the production process, and new supporting institutions to make those institutions effective.

For each of these, we define a constitutional innovation as the first appearance in the corpus of written constitutions of an idea or institution, or a first use of a mechanism to produce constitutions. Following Schumpeter, we recognize that this does not mean that the concept was invented out of whole cloth; indeed, it is possible that very few ideas (if any) were first introduced in such an important venue. Instead, it is the combination of the instantiated constitution with the existing norm or institution, which might have existed in ordinary law or in political discourse, that marks innovation for our purposes. To illustrate, in 2008, the Constitution of Ecuador introduced the notion of the rights of nature into global constitutional discourse.^[14] In doing so, it deployed the indigenous term Pachamama, loosely translated as “Mother Nature,” which had spiritual significance for the peoples of the Andes, along with the western concept of rights. The combination marked the first appearance in national constitutions, and so marked an innovation, which in this case has garnered a good deal of attention. The creation of a new bearer of rights is an example of an innovation at the level of the constitutional product. Pachamama may have puzzled constitutional observers at the time (and even now), but since then philosophers and legal scholars have propounded a full theory of “personhood” for hippos, chimpanzees, and other non-humans.^[15] Rights for organisms throughout the plant and animal kingdoms amount to a real trend.

Constitution-making processes can also be the locus of innovation. Processes involve a series of stages in which ideas are consolidated into a text that is ratified in some generally accepted manner, and then enters into force. Process innovations can include new ways of seeking input, including from the public, expert groups, or outside advisors. The various modalities of constitution-making bodies—constituent assemblies, constituent legislatures, expert commissions, and others—are sites of innovation. New mechanisms of selecting drafters and adopting a final text also count as innovations. Below we describe the innovation of public participation in constitutional design by way of example.

An example of an innovation in the domain of supporting institutions is the constituitonalization of a new body or agency. Constitutions written today are full of many new bodies that were not present when the era of modern-constitution-making took shape in Philadelphia, Paris, and elsewhere in the late 18th century: election commissions, media bodies, audit agencies, ombudsmen, and constitutional courts. Each of these institutions were the product of an innovation. The example of constitutional courts shows how institutional innovations can be repurposed. The original design was created by Hans Kelsen for the Constitution of Austria of 1920, primarily as an umpire to resolve disputes about federalism.^[16] But following World War II, the institution was borrowed by an executive-wary Germany and access significantly broadened, turning the court into a central locus of rights protection. Our Schumpeterian perspective would recognize not only Kelsen but also his postwar followers as innovators to the institution of the constitutional court. On a lower level is the 11-year term for constitutional court justices—making it the first and only constitution to choose this particular number of years for term length.^[17] This too was an innovation, albeit a minor one, and surely one less worthy of study and apparently emulation.

These examples show that innovations can operate at various levels and may depend on how broadly or narrowly one chooses to look. This range is actually a strength of the product analogy: innovations can include whole product categories such as cellphones; a discrete type such as the iPhone; or a component innovation, such as the development of the camera technology for the iPhone 14 model. It can involve streamlining the production process to make it more efficient, or the creation of a new supporting institution such as the Apple Store where one can fix and accessorize phones.

B Predicting Innovation: Constraints

For constitutions, the micro-foundations behind such new innovations are mysterious. Constitution-making processes are notoriously complex and contingent.^[18] They involve a sometimes-unpredictable collaboration among individuals with various motivations and constraints who are sometimes working under unusual conditions. As a result, any theory of constitutional innovation must start with the production process and the motivations and constraints facing drafters. That is the subject of this section. We recognize that market analogies for analyzing constitutions are imperfect. We begin by acknowledging that the incentives of constitution drafters differ from those of innovators in a marketplace.^[19] In a market for products, the innovator may be rewarded with increased profits, and so has some incentive to come up with the proverbial better mousetrap. Producers of law—whether legislators, constitution-makers, or judges—are not able to capture the benefits of a revenue stream. They have no pecuniary interest in the products they produce.^[20] Nor can they claim a property right if they innovate and are unable to control subsequent “use.”^[21] Furthermore, law has a particularly rigid structure in that new “producers” must fit into pre-existing frameworks. “(O)nly in law,” wrote Judge Posner, “is ‘innovative’ a pejorative.”^[22]

On the other hand, there may be reputational benefits that can “compensate” the creator of new legal concepts. As noted above, Hans Kelsen is forever known as the inventor of the model of the constitutional court. Similarly, a great judge may live on in history for his role in inventing a new doctrine. Judge Learned Hand, for example, bestowed his name on the so-called “Hand Formula” taught to first year law students. Judge Roger Traynor of California—whose obituary called him “one of the greatest judicial talents never to sit on the United States Supreme Court”–is known to lawyers for his role in forming modern product liability law and a host of other innovative decisions. Legislators may live on in the names of legislation: the Financial Reform and Consumer Protection Act in the U.S., for example, is usually referred to as the Dodd-Frank Act; the Sarbanes-Oxley Public Accounting Reform and Investor Protection Act is another eponymous piece of legislation well known in financial circles. For the most part, however, those who contribute constitutional ideas to the world are lost to history. Whoever was responsible for inserting consumer rights into the 1917 Mexican constitution, for the first time in constitutional history, surely set an important precedent, but their act–like others–will go largely unsung.

However, the very fragility of constitutions might alter the probability and benefits of innovation. Although, a new constitution can be expected to last 19 years on average, many new constitutions last less than a decade and around 10% perish within just one year of being put into force.^[23] Fragility may mean that drafters are discouraged from experimenting with new institutions. After all, why invest any political energy into a constitutional project that is here today gone tomorrow? If no one expects a constitution to last, it may produce a self-fulfilling equilibrium of serial, short-lived documents. Such contexts may provide little incentive to invent new concepts and institutions, or to disaggregate older rights into new ones. On the other hand, constitutional churn can open up windows of opportunity, so-called “plastic hours,” in which innovators can inject their ideas. The persistence of the U.S. Constitution has meant that world innovations have left U.S. higher law in a state of either perennial underdevelopment or bedrock stability, depending upon one’s perspective.

Drafters face other constraints as well. National constitutions are often negotiated in complex, contentious processes involving intense political struggle. Writing is costly in time and negotiations; creating something new is especially costly. The problems associated with scarcity of time are exacerbated by the fact that most constitutional drafting efforts have tight deadlines.

Another major challenge for constitutional drafting is that of knowledge.^[24] Most constitution-makers have never engaged in the project before and will never do it again. This is not to deny that there are real benefits to amateurism, in the sense of G.K. Chesterton.^[25] Indeed, our tool, Constitute, is designed with just such a premise of D.I.Y. constitutionalism.^[26] Still, we recognize that learning constitutional design from scratch has its own start-up costs and disharmony, as does, say, learning a musical instrument. Furthermore, in the constitutional context, it is enormously difficult to identify whether a particular institution is likely to be effective or not, as that may depend on a complex interaction between institution and context. Social science has not generated very many “best practices” for constitution-makers, even if policy makers continue to search for them. And best at what? Clearly, constitutional choices will have multiple, sometimes orthogonal effects on whatever outcome one wishes to maximize. There is thus incomplete information on the consequences of choices. Whether a particular institution works will depend in part on the institution, but in part on the context. If a drafter adopts an institution and observes failure, it is hard to know whether that failure is because of the institution and its inherent properties; because of the context, in which any plausible alternative would also have failed; or some interaction between the two. There is a lack of usable social science knowledge available; indeed, one motive for our broader Comparative Constitutions Project is to facilitate greater learning about the consequences of constitutional choices.

C Predicting Innovation: Political Considerations

All of these reasons suggest that innovation will be undersupplied in the constitutional marketplace, just as Schumpeter observed in private markets. When, then, might we observe innovation, despite inertia? An obvious starting point is that innovation will occur when political actors mobilize to demand it, and when a pivotal player inside a constitution-making body has a particular interest in accepting that demand. The Pachamama provision mentioned above was the result of enhanced power of indigenous groups in the constituent assembly of 2006–2007, as well as the lobbying efforts of an interest group, Fundacion Pachamama, that pressed for inclusion of the language. The first constitution to regulate the internal character of political parties was that of Mexico in 1917, which banned associations of a religious character.^[27] One of the central demands of the revolution was for disenfranchisement of the Catholic Church, and the drafters understood they had to ensure that church officials stayed far away from politics.

We also note that innovations may reflect local power politics, although with global ideas. In 1993, in the height of the neoliberal wave of reforms sweeping Latin America, Peru adopted a new constitution. The prime mover behind the document was President Alberto Fujimori, who had just committed his “auto-golpe” by suspending the legislature. The new Peruvian constitution protected freedom of entrepreneurship and extended the international law norms of national treatment and nondiscrimination to foreign investors.^[28] One of the constitutional innovations was a clause that essentially stabilized contracts, immunizing them from new regulation.^[29] Another clause of the constitution prevents the state from establishing new monopolies.^[30] These provisions reflected attempts to bracket the state out of the market, embodying salient ideas into constitutional text for the first time.

The presence of normative materials and models from outside the country, and the importance of local politics, do not tell us when and where constitutional innovation will occur. Which constitution-making situations are most prone to innovation? Our prediction is that we are more likely to observe innovation in situations in which there is a radical break from a prior order. Aversion to a recent past can help to overcome the general inertia and the use of boilerplate in constitutions, and accordingly create more demand for innovation. Kim Lane Scheppele may well be right that constitutional designers have a much better sense of what they do not want, as opposed to what they do want.^[31] In that sense, aversion may have a repulsive effect as well as opening a “window of opportunity” for the reception of some new ideas, whatever they are.^[32] The examples of the postwar German Constitution, with its sharp turn toward rights provides a well-known example. Similarly, the process innovations in Iceland occurred after a major financial crisis, which led to a set of protests called the “pots and pans revolution.”^[33] The adoption of constitutional courts in many countries in the 1990s may have reflected a new focus on rights in the aftermath of the Cold War.

All these examples concern the creation of institutions and processes that lean toward the direction of rights protection. But aversive constitution-making can also work in the direction of rights-reduction. The Constitution of Hungary adopted by Viktor Orbán in 2012 was notable for consolidating political authority and breaking with the liberal jurisprudence of the constitutional court that had existed for two decades.^[34] Orbán has mobilized the psychology of aversion to his own political ends. Interwar fascist movements in Germany and southern Europe mobilized aversion to liberalism along the same lines.^[35]

With these considerations in mind, the next sections turn to four examples of constitutional innovation in the area of human rights: the emergence of the Bill of Rights as a form, which is an example of a major product innovation; the constitutionalization of particular rights, which reflect lower levels of product innovation; the idea of a right to public participation, an example of process innovation; and the constitutionalization of human rights commissions, an example of an innovation of supporting institutions.^[36]

Example 1: Emergence of the Bill of Rights as a Form in the United States

One of Schumpeter’s modalities of innovation is that of finding new ways to organize and present a product. So we start with the basic idea of a bill of rights as a discrete listing of rights. Such a document is a kind of meta-innovation at a higher level (specifically, a new genre), which invites innovation in terms of the kinds of rights and ideas contained therein. In much the same way literary historians describe the rise and fall of the novel, a literary form so ubiquitous that one cannot imagine literature without it, so too with the bill of rights.^[37]

The genre defined by a discrete listing of rights dates from at least the Agreement of the People of 1647–1648 in England, which set out complaints against the monarch before enumerating a set of rights. The first document called a “Bill of Rights” is thought to be that adopted by the English Parliament in 1689, as part of the Glorious Revolution.^[38] While it was not entrenched because it was an ordinary Act of Parliament, it is considered to be part of the “English Constitution” and an important milestone in the evolution of constitutional rights. It represents an important blow to absolute monarchy in the United Kingdom, as later codified in the Act of Settlement of 1701. Using the occasion of the arrival of a new foreign monarch, William of Orange, the House of Commons articulated a set of complaints against the prior King James.^[39] Framed as a set of “ancient rights and liberties,” the Bill of Rights includes such familiar provisions as the right to petition, the right to be free of cruel and unusual punishments, and a prohibition on excessive bail, along with collective claims to parliamentary power. The document was an obvious inspiration to the American colonists with their own litany of complaints a century later against George III.

The period between the English Revolution and the drafting of the American Constitution was an extraordinarily important period for thinking about rights.^[40] It saw the gradual evolution from the language of privilege to that of rights, which became implicated in the colonialists’ struggle for independence. Inspired by Locke in particular, the colonialists clothed their claims in a universalist language of men’s inherent equality and rights to life, liberty, and property. But they also appealed to the rights of “Englishmen.”^[41] As Hutson has written, “the 18th century was a period (not, perhaps, unlike our own) in which the public’s penchant for asserting its rights outran its ability to analyze them and to reach consensus about their scope and meaning.”^[42] De Bolla (2013) traces the 18th century shift from naturalistic claims about “rights of man” to “the rights of men” (with the definite article), which are enumerated products of particular societies.

This dual structure of natural and civil rights helps explain why the American revolutionaries became committed to the articulation of rights in written constitutions.^[43] As Hunt has argued, if all rights are self-evident, then there was little reason to write them down, much less to enumerate them in some sort of list.^[44] But writing implied more specificity than did prior notions of natural law, and its concreteness allowed colonists to debate the particular rights to be adopted. Moving rights to positive law also allowed some diversity among the different colonies, reinforcing the notion that they were distinct political communities. Finally, the revolution incentivized writing because it allowed for the voicing of specific claims against the English crown, just as had Magna Carta and the English Bill of Rights before it. Again, aversion to antecedent abuses is a powerful motivator for rights innovation.

The initial efforts in this regard were the Bills of Rights of various American colonies, particularly the Virginia Declaration of Rights of 1776, which spoke in terms of natural rights and served as the inspiration for similar efforts in other states and at the national level. There was great diversity among the states: Vermont’s Constitution, for example, outlawed slavery in 1777, which was hardly the approach taken in Virginia.^[45] But there was also a tremendous amount of innovation in the kinds of claims articulated as rights, many of which had no precedent in earlier English thought. Freedom to be free from the quartering of troops, for example, represented a specific response to a felt need. And freedom of assembly seemed to be of particular concern to the colonists. The rights of Englishmen, in turn, were responses to specific abuses by the Crown or sheriffs, which generated demands for protection.

As the American founders gathered in Philadelphia in the summer of 1787, rights were not foremost on all of their minds. While the final constitution included only a limited set of rights interspersed in its other obligations (such as the proscriptions against ex post facto laws and bills of attainder), the omission of a larger list immediately provided fuel to the anti-Federalists, who were mobilizing to block the Constitution’s ratification.^[46] A promised Bill of Rights became the price of adoption, and its ten articles contain a wide array of rights that have inspired citizens ever since. It marks a significant step in the evolution of constitutional rights. The form of the Bill of Rights drew directly on the colonial constitutions of 1776–1777, which in turn reflected the idea that articulating rights was what a people should do when they constituted a civil government.

Madison (who had been forced to reverse his position on rights to win election to a House seat) led the drafting process.^[47] Eight states proposed a series of more than 200 amendments to the Constitution, and these contributed to Madison’s drafting.^[48] Drawing on foundational work by Levy, we can identify 90 discrete ideas in these 200 proposals. 22 of which made it into the Bill of Rights, and 68 that did not; the latter category includes some of the more obscure rights, as well as some of the proposals that in modern terms seem to focus more on institutions than rights.^[49] Because we have the bills of rights of the American colonies, the rights included in iconic prior documents, as well as the specific proposals of the ratifying state legislatures, we can glean some sense of the factors that led some rights to gain inclusion, and thus be part of the innovation.

Figure 1 provides an illustration. Using the various documents identified by Levy as containing proposals for the U.S. Bill of Rights, we can look at the pairwise correlations among documents and the final bill of rights as ratified by the various states. Whether a document contains a right is binary in this analysis, and we score the inclusion of a right in a proposal with a 1, and the absence with a 0.^[50] We also calculate the total number of colonial proposals that include a particular right, which allows us to assess whether those rights that were prevalent were more likely to be included in the U.S. document. The correlations show that some documents, such as the famous Virginia Bill of Rights, were indeed highly influential on what came afterwards. The correlations between that document and the ratified Bill of Rights exceeds 0.50. On the other hand, some proposals seem to have been very idiosyncratic and not correlated with those adopted a the federal level. The amendments proposed at the ratifying conventions in New York and North Carolina are negatively correlated with the U.S. Bill of Rights, meaning that the pairs of documents included or omitted very few of the same rights. Both conventions had expansive understandings of rights, including limitations on standing armies and requirement of publication of a government journal, that did not make it into the national bill.^[51]

Figure 1:

Influences on the U.S. Bill of Rights.

The exercise also shows that consensus among proposers is impactful. The more states that proposed a particular right, the more likely it was to be adopted by the House in the final Bill. Those rights that were included were endorsed by an average of four different proposals, while those that were excluded were endorsed by only 2.2 on average. A visible model, such as that of the Virginia Bill of Rights, exerts a powerful focal influence, but so too does a consensual idea espoused by many proposers. In this sense, the Bill of Rights was reflective of a consensus among discrete political entities, the peoples, and legislatures of the various American states, rather than a document that constructed an original menu of rights. This role, incidentally, is very similar to that of the Universal Declaration of Human Rights,^[52] an influential mid-century rights document that should be seen as influential, but still very much a product of its time.^[53]

In an effort to understand the sources, evolving conceptualization, and what we call the “edit flow” that formed the U.S. Bill of rights, we have analyzed and depicted its various draft versions in a series of interactive data visualizations. Our chain of events starts with the source documents—the colonial constitutions and the ratification conventions—through Madison’s set of proposals, through the editions made by the House and Senate chambers in their deliberation of Madison’s proposal, and finally to the set of proposals approved by Congress and ratified by the states. For each of the ten amendments, as well as five of Madison’s proposals that were not adopted, we traced the ostensible path of the core ideas, and their meandering language.

Our intellectual product, in this case, was an interactive museum exhibit that we (with Robert Shaffer) mounted in 2017 at the National Constitutions Center, together with Jeff Rosen and the NCC research staff (and still available online.)^[54] The true experience, of course, is an interactive one, but Figure 2 provides a snapshot of the edit flow in the case of the Seventh Amendment. Some notation: the putative path of the core ideas (as suggested by Cogan) is noted by line segments and the shading of bars indicates the level of similarity (darker is more similar).^[55] For each proposed phrasing, we calculate its similarity to the ultimate amendment based on their shared inclusion of politically meaningful words. Online, one can also click on a document a view its text in comparison with the final formulation.

Figure 2:

The edit flow of the seventh amendment. N.B. An interactive exhibit installed at the National Constitutions Center in 2017 and available online.

The exercise reminds us of the diversity of ideas within the founders’ deliberations. In the case of the Second Amendment, each antecedent document contains some version of the right to bear and/or keep arms, along with related ideas about standing armies and common defense. However, the construction of these rights is often quite different. Consider some highlights in the edit flow of the Second Amendment. The final formulation of the 2nd amendment is, of course well known:

A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.

But if we start with the source documents, of which there are some 20 documented by Cogan with elements of the right, we see that the Second Amendment could have been quite different. Take the formulation from the Pennsylvania Minority in 1787:

That the people have a right to bear arms for the defense of themselves and their own state, or the United States, or for the purpose of killing game; and no law shall be passed for disarming the people or any of them, unless for crimes committed, or real danger of public injury from individuals.

Note that the Pennsylvania Minority here would have something for both sides of today’s gun debates: on the one hand, an individual right of self-defense and recreational use, with abolition expressly denied the state, and on the other, grounds for regulation on the basis on public health and misuse. Seemingly a workable compromise, as least as some of us see it.

Many of the source documents—and amendment proposals in the House and Senate—included language that further elaborated, clarified, or specified the meaning of these ideas. For example, a Senate amendment considered on September 4 of 1789 suggested that conscientious objectors would not be required to bear arms, something also reflected in two proposals from the House, as well as from some of the colonial and state proposals.

All told, these documents provide a sense of what could have been, had the collective drafting efforts gone differently. They also make clear that the Amendments adopted in 1790 were products with a long and pervasive history in North America. This may well be what “innovation” looks like, once we take a closer look. Innovation, as the rolling out of an invention, takes the form of experimentation across a larger set of authors and jurisdictions.

In short, the idea of a constitutionalized Bill of Rights was an innovation born of politics and struggle. It illustrates Schumpeter’s point that innovations involve the combining of elements already present into new forms; drawing on existing models is probably necessary for innovations to occur. And it illustrates aversion as a motivator for innovation. Furthermore, the production process matters. What is constitutionalized is (often) negotiated and results from the specific configuration of conflicting interests sitting around the proverbial negotiating table. Rights will emerge only when some actor is willing (and able) to push them.

Example 2: Worldwide Constitutionalization of Rights

Once the Bill of Rights canonized the notion of rights in constitutions, the idea took off (even if some early constitutions such as Poland 1791 did not use it.) And as constitutions spread around the world, they invited further innovation. By constitutionalizing certain rights—that is raising them to the level of higher law—drafters marked the right as particularly important and worthy of protection outside the realm of ordinary politics.

Drawing on the data of the Comparative Constitutions Project, we can examine when particular innovations first entered the genre of modern constitutions. We identify a set of 117 constitutional rights, which we had included in the project’s survey instrument. Note that these 117 do not represent the entire universe of rights, however conceptualized, but we intend the set to be as comprehensive as possible. Invention and innovation continue unabated and our taxonomy of rights undergoes regular enrichment, in response. For example, the Chilean constitutional draft of 2022—which ultimately failed in a referendum—enumerated a set of rights not in our taxonomy, such as the right to sports. We recognize that our taxonomy, with which we index the world’s constitutions on the site Constitute, often serves as an ideational resource for constitutional drafters. As such, we take conceptual integration, discovery, and enrichment seriously.

Our original data identifies the provision (or not) of the 117 rights for a sample of the more than 800 national constitutions written for independent states since 1789.^[56] The sample also includes several constitutional documents from the United Kingdom, including Magna Carta and the Bill of Rights. (Our selection for the UK, which always presents a constitutional identity challenge, relies on an interesting codification project of University College London).^[57] Of course, some of these innovations diffused widely: over 90% of constitutions include some provision for free speech. Others did not diffuse at all.

Table 1 ranks the 117 rights by when they first appeared in the corpus of national constitutions.

The exercise is interesting for what it says about the emergence of ideas. It confirms that some of the canonical constitutional texts, namely Magna Carta and the 18th century constitutions of the United States and France, were indeed loci of great innovation. But we also observe a tremendous amount of innovation in more peripheral environments, in keeping with Colley’s historical account.^[58] One might not have expected the right to found political parties to be constitutionalized first in Cuba, for example.

Haiti in the early 19th century stands out as the site of innovation for 10 constitutional rights, and then in 1867 it became the first country to constitutionalize a right to shelter. This is an extraordinary record of generativity for a small country occupying half of an island, and it is worth describing some of its history in more depth. It is, we believe, a strong illustration of the importance of aversion in incentivizing innovation.

Rights articulation in Haiti was shaped by its colonial legacy; independence was triggered by the French revolution and the glaring contrast between the ideals of equality and colonial life in Haiti. The French revolutionaries had spoken in the name of the citizen, while the middle-class American revolutionaries to the north spoke in the name of the individual. Haiti’s dominant class was a group of former slaves whose first demand was simply recognition of their humanity, and this was the initial motive for making rights claims. It is worth emphasizing that this emphasis on humanity as the source of rights was an innovation.^[59] At the same time, human rights were articulated through a demand for nationality, first French but later Haitian. Article 3 of Toussaint L’Overture’s pre-independence document said that “Slaves are forbidden in this territory, servitude is abolished forever. All men born here live and die free and French.” Nationality was a mode of ensuring equality.

After independence in 1804, Toussaint’s successor Jean-Jacques Dessalines produced the constitution of 1805, a revolutionary document that called for freedom of religion, racial equality, and non-discrimination on the basis of color.^[60] Property was sacred, but whites could not possess it, with the exception of white women and naturalized Germans and Poles. This racialism was sticky: it was a feature of many of subsequent constitutions of Haiti up until 1867.^[61] The 1805 Constitution also provided that Haitian citizenship was lost by emigration.

After the assassination of Dessalines by his own advisors, a new constitution was adopted in 1806, one that emphasized legislative power and institutional structures. The split of the country that followed led to an authoritarian constitution centered on a powerful executive in the North, led by Christophe. The southern part of Haiti was governed by Petion, Dessalines’ successor. Although Colley (2021) emphasizes Christophe as the innovator of authoritarian constitutions, it was Petion’s Constitution of 1816, promulgated in the southern, republican part of the island that marked the bigger shift. The primary motive for this document was ongoing conflict with the North, and Petion’s reversal of his earlier position to make himself president for life. This draft became one of the prototypes for the many subsequent constitutions in Haiti as the country toggled back and forth between institutional variants for much of the next century.

In its rights provisions, the 1816 Constitution followed two main sources: the earlier constitutions of 1801 and 1805, and the French Declaration of 1793. Serial continuity was combined with external borrowing. It abolished slavery in the very first article, repeating a feature of the 1801 constitution and the French Declaration. (It also retained, in Article 38, the idea that “No white, whatever his nation, will set foot on this territory, as master or owner.”) Article 6 repeated the classic French formula that “The rights of man in society are: liberty, equality, security, and property.” And several other provisions were borrowed more or less verbatim.^[62]

On the other hand, the 1816 document made a number of important innovations. The equality formula, “Equality admits no distinction for birth or hereditary powers,” though inspired by France, was novel in its mention of heredity. Article 53 is directed to the rights of children born out of wedlock, apparently in response to a serious social problem that had already been identified, but had been left to ordinary law in the 1805 Constitution. Citizenship and property receive special treatment, including provision for naturalization of Africans and Indians. Article 218 provided that “Foreigners as well as their place of business are placed under the protection of the nation.” And it provided that even foreign descendants of Africans or Indians would be recognized as Haitian and after one year could acquire citizenship.^[63] This new constitutional provision was litigated within three months in a prominent case in which Petion refused to return seven Jamaican escaped slaves to their owner.^[64] He justified his refusal on constitutional grounds, as well as general principles of asylum.^[65] This effectively projected the principles of the Haitian revolution outside its own borders, an early example of externalities imposed by rights innovation.

One may have expected the inventory of rights to have been exhausted by the 21st century. Actually, innovation has continued – with concerted leaps and bounds – since the early constitutions of the early 1800s. Indeed, we find ourselves in an era of tremendous rights innovation. Besides the rights of nature mentioned above, we have seen new emphasis on the rights of sexual minorities, since they were first recognized in 1988 in Brazil. The Brazilian document was clearly an aversive constitution, marking a sharp break with the previous era of dictatorship. Nepal’s 2015 constitution allows special laws to be passed for the protection of sexual minorities and guarantees them a right to participate in state bodies. Thus, as society has recognized new interest groups, their identity gets reflected in practice.

The constitutional innovation of habeas data occurred in Brazil in 1988. Data privacy had a prehistory in the case law of the German Constitutional Court, but the Brazilian constituent legislature turned this into a full constitutional right for the first time.^[66] Following this example, Colombia incorporated the habeas data right in its 1991 constitution, from whence it spread to Paraguay (1992), Peru (1993), Argentina (1994), and other countries in Latin American thereafter. Again, the aversive politics of reaction to military dictatorship, along with new ideas available from outside, combined to generate an important constitutional innovation, one that subsequently diffused.

Example 3: A Right to Participation in Constitution-Making

In recent years, there has been the consolidation of a view that constitution-making processes should include widespread public participation. One important scholar in this regard was Vivien Hart, who argued that the production of new constitutions should always include extensive public participation, and that the public enjoys a proper right to do so.^[67] She inferred this right from the “right to take part in public affairs” in the International Covenant on Civil and Political Rights,^[68] Thomas Franck’s idea of a right to democracy, and an examination of state practice including constitutions produced under United Nations auspices.^[69] She placed great weight on the South African constitution-making process of 1992–1996, which included extensive efforts to include and inform the public. This normative work of constructing a new right applicable to constitution-making is an example of a process innovation, that presumably would affect the quality of the product.

This innovation has been consolidated to some extent by the United Nations, which has had increased involvement in constitution-making processes in recent decades. To consolidate its own work, the organization produced a Guidance Note from the Secretary General on United Nations Assistance to Constitution-Making Processes in 2009; this was revised in 2020.^[70] The document recognizes that constitution-making is a sovereign process, and states as one of its overarching goals that the process “ensure national ownership and reflect local context.” Constitution making, the note states,

is a sovereign process that, to be successful, should be nationally owned and led. National ownership requires the engagement of national authorities, a broad range of political actors, ethnic, religious, and minority groups, civil society, including women’s groups, and the general public (emphasis added).^[71]

Thus, the requirement of public participation received significant external reinforcement.

There is a significant empirical debate on whether or not public participation in constitution-making is effective in any way. Abrak Saati argues against what she calls “The Participation Myth”, showing that participation has little subsequent effect on the consolidation of democracy.^[72] Other studies show some effect of the process on participants’ awareness of the text.^[73] It is safe to say, however, that the impact of participation is far from clear.^[74]

Beyond the normative consensus that participation is desirable, the precise mode of getting public input can be the subject of innovation. One can use surveys, focus groups, mini-publics, as well as mass media. Two recent examples of innovation come from Chile and Mongolia. In Chile, President Michele Bachelet launched a constitution-making process in 2015, designed to replace the Pinochet-era constitution with something more reflective of the times. Her team created a novel form of public participation, namely a set of private neighborhood consultations called cabildos, in which local citizens would gather to discuss ideas. Although the process failed to produce a constitution that could be adopted, it did generate ideas that informed the later process of constitution-making in Chile. Another example is in Mongolia, where Stanford political scientist James Fishkin introduced his idea of deliberative polling during discussions on constitutional amendments.^[75] The country subsequently adopted a requirement that any amendment proposals be subjected to deliberative polling before the public before adoption, perhaps the first time Fishkin’s method had been enshrined into law.^[76]

A related example of a process innovation is that associated with the Icelandic experiment of 2009–2011, which brought ordinary citizens deep into the production process.^[77] In that case, citizens were randomly selected to form an assembly to generate constitutional ideas. An election of citizens to do the drafting followed, in which political party leaders were not eligible. The 25 members of the constitutional council then proceeded to draft a text, with an online mechanism of soliciting comments on Facebook. This draft was ultimately approved in a referendum but failed to be adopted by Parliament. Still the process introduced several technical and conceptual innovations.

Example 4: Constitutionalization of Human Rights Institutions

One of the main institutional developments for the protection of human rights in our era has been the spread of National Human Rights Institutions (NHRIs), such as human rights commissions, specifically designed to protect and promote human rights. In 1991, the United Nations General Assembly endorsed these institutions and adopted the so-called Paris Principles. Subsequently they have spread rapidly. Linos and Pegram documented over 120 such bodies as of 2017.^[78] Our data shows least 39 of these have been constitutionalized, all since 1985.

NHRIs draw on a prior tradition of the ombudsman institutions, which have an ancient history in Scandinavia. The Swedish Instrument of Government of 1809 introduced the institution of the parliamentary ombudsman, designed to supervise the operations of government officials. But the modern form of a collective national human rights commission was that of Guatemala during the 1980s, established in the midst of the country’s civil war. That war involved wholesale massacres of Guatemalan peasants, and killed an estimated 100,000 and created many times that many refugees. The immediate impulse for constitutionalization of a human rights body occurred after a 1983 coup ousted General Efrain Rios Montt from power. His successor, Gen. Oscar Humberto Mejía Víctores, initiated a return to civilian rule for the first time since 1970. In 1984, a constituent assembly was elected in free elections, and it produced a constitution, even though the military remained a major presence and the war was ongoing. Aversion to unconstrained human rights abuses was a powerful motivator.

The 1985 Constitution created a Human Rights Commission, made up of one deputy from each party represented in the Congress, which in turn appoints a Human Rights Ombudsman (Procurador de Los Derechos Humanos).^[79] The Ombudsman serves a five-year term, and can receive complaints from any citizen, and make recommendations accordingly. These institutional innovations brought together three discrete elements: a specific focus on human rights, boosted significantly in the 1970s;^[80] a political body accountable to and located within the legislature; and a repurposing of the Scandinavian institution of the Ombudsman. To be sure, the innovations had a mixed record, caused by continued military interference in politics, and the military’s granting to itself of an amnesty to prevent prosecution for prior abuses. But again, as an example of a supporting institution to protect human rights, the NHRIs have successfully diffused.