Introduction

It has been difficult to determine what is the real socioeconomic impact of eminent domain in Brazil because the Brazilian juridical area is widely dogmatic and never gave much attention to the question. Much of the academic production on the theme focuses on the specific statutory norms, accentuating the "simplicity" and "naturality" of the concept as if it were universal and ahistorical. Celso António Bandeira de ^{Mello (2015)}, one the most influential Brazilian Public Law professors, summarizes only four relevant questions for eminent domain judicial review: "a) pricing; b) procedural irregularities; c) the accordance between the Public Power claim to land, and [d)] the legal hypothesis" (p. 915). The only doctrinal controversy on the theme of eminent domain, according to him, would be the right to retrieve when the use of land is different from the initially stated by the Public Power, even though it has some public utility. Another noted author, José Carlos de Moraes Salles (2006), specifically on eminent domain, makes a "historic record" beginning by assuring that Romans already "faced" eminent domain, because "people who built numerous and important public works obviously would face opposition from property owners" (p. 67).

In practice, Brazilian legislation forbids questioning eminent domain, which makes it a secondary matter for lawyers. Article 20 of Dec.-Lei 3.361/41 states that "pleadings [during the eminent domain mandatory lawsuit] should only discuss judicial procedure irregularities or pricing; any other questions demand an ordinary lawsuit" Nevertheless, eminent domain is present as one of the most discussed social science topics-agrarian reform. The theme is sensible, widely regarded in Brazil as something "communist" or "leftist" since the 1964 Coup d'état. The military dictatorship imposed a strong and illusory "war against subversion" (legally embraced by the surviving National Security Act) that led to bloody persecution against left parties and collectives, social movements, and even extremely impoverished people.

The article, following André-Jean ^{Arnaud (1988}; ¹⁹⁹¹; ¹⁹⁹²), is a sociolegal study, positioned as a legal research enriched by insights from diverse social science domains (^{Villas-Bóas Filho, 2018}, p. 260). Arnaud cautions against an overly simplistic interdisciplinary approach, critiquing what Bailloux and Ost term an "ingenuous model of interdisciplinarity" based exclusively on a mere superposition of studies sometimes very heterogeneous (apud ^{Villas-Bóas Filho, 2020, p. 537}).

Some researchers, according to ^{Arnaud (1992, p. 25)}, operate within their own disciplinary canons without considering the object's original provenance. Lawyers interpret law texts according to juridical hermeneutics, historians may misconstrue law to gauge historical mentalities, and social scientists treat it as a social fact, among others (^{Villas-Bóas Filho, 2018}, p. 252). Arnaud also advises legal researchers to abandon the illusion of conducting deep historical or sociological analyses and proposes interdisciplinary collaboration through fixing a common object for all disciplines: the juridical reality. The collaborative framework, creating a "carrefour interdisciplinaire" (^{Arnaud, 1988, pp. 7-8}), is embodied in a sociolegal study. In other words, this article aims to study the interconnection of eminent domain and agrarian reform as a legal reality present in Brazilian Era Vargas, in its socioeconomic and historical dimensions.

Additionally, the research mobilizes a history of concepts as theorized by Reinhard ^{Koselleck (2004)}. Koselleck's perspective allows for an accurate diachronic study of concepts, unifying semantic considerations with social contextualization (^{Villas-Bóas Filho, 2010}, p. 30), which opens a great opportunity for sociolegal studies that demand cautiously working with juridical concepts and some method to appropriate more detailed historical results made by real historians into law discourses.

The result of using the history of concepts in research is to discover and interpret social conflicts from the past in their conceptual context, linking the meaning of a concept with its contemporary reality (^{Jasmin & Feres Junior, 2006}, pp. 31-32). Koselleck's method, in short, consists of first translating the concept-not a word, as it has a kind of intentional persistent polysemy, aggregating the totality of sociopolitical circumstances in it (^{Koselleck, 2004, pp. 109}) from one time to the present, then isolating a concept from its "situational" context, and finally organizing its presence through time in chronological order. Afterward, the final step is the evaluation of the duration and social impact of the concept, as well as its structures (^{Koselleck, 2004, p. 105}). This was performed, restricting the analysis to the Era Vargas period (1930-1964), a critical moment where agrarian reform was at the same time an official agenda and subject to powerful opposition.

At last, to understand the whole direction of urban politics in this period, the concept of spatial Keynesianism, by Neil ^{Brenner (2004}; ²⁰¹⁹), is useful. Brenner works with the function of scale-fixing in urban uneven development, using concepts from the critical theory of urbanization, especially those from David Harvey and Neil Smith. In short, Brenner tends to classify historical "modes" of urbanization and urban policy-making by the ways each epoch deals with the hierarchization of geographical scales, constituting a scalar fix conducted by the needs of capital accumulation in and through space (^{Brenner, 2019, pp. 61-62}). Spatial Keynesianism, for example, gave huge priority to the national scale, adopting a bunch of (legal, social, political) instruments to equalize spaces inside national-state borders (^{Brenner, 2004, p. 116}).

According to Brenner, the main scope of spatial Keynesianism was to promote balanced development and territorial redistribution at each scale. It implied several State organizations with redistribution aims, including:

Within this nationalized spatial matrix, each unit of state administration was to be equipped with relatively analogous, if not identical, institutional arrangements, and comparable infrastructural facilities, public service relays, and socioeconomic capacities were to be anchored within each local and regional economy. At the same time, the compensatory, redistributive approach to intra-national territorial inequality associated with spatial Keynesianism was seen to secure a stabilized, reproducible pattern of industrial development, to promote the efficient allocation of public services and to maintain national political and geographical cohesion. Spatial Keynesianism thus represented a historically unprecedented constellation of state programs to mold the geographies of capital investment, infrastructure provision, and public services into an equalized, balanced, and relatively uniform grid of national state space (^{Brenner, 2004}, p. 116).

The four main characteristics of spatial Keynesianism also acknowledged as a State mode of production are (1) mobilization of space itself as a productive force; (2) the State guarantees the conditions of social reproduction; (3) the State mediates uneven development with national urban systems; (4) spatial fix for capitalist growth is secured by the State (2004, pp. 125-126). In other words, spatial Keynesianism supposes centralizing decisions to distribute spaces among national territory.

The 'Era Vargas' worked with a developmentalist paradigm, designed by Celso ^{Furtado (2013}) and the ECLAC. It has sensible differences about economic strategies, scopes, and the role of the State. They resemble in many points, including land policies but could not be efficiently implemented in Brazil, which is the focus of the present article.

I - Vargas and development

In 1930, amid an economic and political crisis, Getúlio Vargas rose to power. Coming from the economic elite from Rio Grande do Sul, Getúlio was affiliated with his state's branch of the Republican Party (rp). The republicans' ideology was based on Comtean positivism. Originally president of the province, after he was nominated Minister of Finance at the Washington Luiz government, he became president of the Republic by vote, later perpetuating his position by coup d'état.

In the seven years before the Estado Novo's definitive formation (his dictatorship, from 1937 to 1945) and as an intervenor (from 190 to 1937), Getúlio's greatest challenge was price control, especially to face coffee's devaluation- Brazil's most produced commodity during that period and a declining product in the world's market. The 1929 crisis hit Brazil with some delay, late 1929 and the beginning of 1930. The crisis in the capitalist center, coupled with the irrationally continuous production of coffee farms and the lack of diversity in export products, led to a sharp drop in prices affecting the whole of the Brazilian economy.

In 1929, after reuniting Julio Prestes' (from the Paulista elite) presidential candidacy opposition, Getúlio presented a discourse planning, as a financial and economic platform, a nationalist defense of export products' value and a solution to steel industry problems (^{Fausto, 2006}, p. 37). As underlined by Gilberto ^{Bercovici (2020)} regarding the politics to protect the value of coffee:

Coffee stocks could not be sold, as this would lower prices even further. It was not possible to borrow money internationally, so it was necessary to raise internal funds. The government did the opposite of what the liberal theory would advise: it expanded credit (starting what was called the "socialization of losses") and devalued the currency. More coffee was being harvested than could be sold. For the producer not to harvest, prices had to be lowered further, which was not desired. After all, by guaranteeing minimum prices to the producer, employment levels would be maintained in the export sector and, indirectly, in the sectors linked to the domestic market itself. Since coffee was the center of the Brazilian economy, if the level of employment in the coffee economy fell, it would fall in the whole economy. Part of the production had to be withdrawn from the market. Thus, the destruction of the surplus was necessary. The State started buying all the non-exportable surpluses of coffee production, which after some time were simply burned. (p. 191)

It is estimated that the Brazilian government, between 1930-1947, burned three times the annual world coffee production (^{Fausto, 2006}, p. 44). However, the protection of the economy was not restricted to price regulations. It was part of a continuous concern with the country's industrialization. According to Wilson Suzigan, from 1920 to 1930, the industry's 8 % growth occurred almost without direct state intervention. In this period, the construction of urban infrastructure was the central asset of the State's collaboration with industry, without any planning (Suzigan, 1988, p. 6). Much of the surplus, until 1940, was also invested by rentiers -Nadia Somekh notes that some of the industrial capitalists also invested in the rentier sector, such as the Crespi and the Matarazzo (1997, pp. 221-223). Nabil ^{Bonduki (2017, p. 226)} also draws attention to the volume of investment made in this area up to the 1940s.

Those two sectors, industry and renting, despite apparently converging in people, ended up being placed at opposite poles in the federal policy of the 1940s. The organization between the capital sectors, by the Estado Novo, was indispensable for the economic growth of the following years. Vargas started a policy of import substitution driven by the currency devaluation. The government, for example, prohibited the import of machinery, except to replace obsolete goods, between 1931 and 1937 (^{Fausto, 2006}, p. 45).

It also limited banking activity with the Usury Law, in 1933. Thus, far from restricting banking activity itself, it ensured that the capital generated by the industry was reinvested in its productive activity. The expansionist credit policy took place through interest rate control by Banco do Brasil and specific credit programs for industry and food production. In fact, the diversification of the economy also presupposed, among other policies, the guarantee of food for the population. Its importance is clear: one of the consequences of import substitution was uncontrolled inflation, centered especially on the rise in prices of basic products. Vargas' economic victories were Pyrrhic for the lower part of the population at certain moments precisely due to the difficulty in acquiring the basic consumer goods for their reproduction.

The promise to nationalize the steel industry was also fulfilled. Despite controversies over its location, the development of the base industry was guaranteed with the inauguration of the Companhia Siderúrgica Nacional (csN) in Volta Redonda (rj). It emerged from a previous agreement between Brazil, the USA and European countries during the war, in which the us government received a cession of Natal military base and compromised to buy Brazilian steel, allowing the nationalization of North American and English steel companies. The agreement was both the base for the creation of Companhia do Vale do Rio Doce (now Vale s/A) and the CSN (^{Aguiar, 2022}, pp. 51-52).

After the end of the Estado Novo, industrial policy continued to expand. Even the Dutra government was elected with the promise of maintaining and expanding industrial policy, mainly supported by the "O Petróleo é Nosso" campaign (^{Fausto, 2006}, pp. 171-175). He also inaugurated, in 1952, the Banco Nacional de Desenvolvimento Económico (BNDE), responsible for industrial financing. Its absolute priorities for investment were in basic industry activity and infrastructure construction, especially energy and transportation. The use of expansionary monetary and fiscal policy continued to underpin the economy throughout the period (Suzigan, 1988, p. 7).

The 1950s were directly inspired by the developmentalist theory. Celso ^{Furtado's (2013}) approach to the federal government was responsible for the definitive introduction of the economic thought of the United Nations Economic Commission for Latin America and the Caribbean (ECLAC) in Brazil's economy and legislation.

After Vargas, the industrial policy, which took its first steps with him, was maintained. Juscelino Kubitschek approved the Plano de Metas (an economic plan), which provided a comprehensive basis for industrial development for the next two decades. Later, one of Vargas' most faithful followers, João Goulart, would continue the two political platforms defended in the letter-testament written by Vargas before his suicide in a chamber of the government palace: the Base Reforms (electoral, tax, agrarian, and urban) and the internalization of the decision centers (^{Fausto, 2006}, p. 197). Those two points were decisive for the Military Coup of April 1st, 1964.

The balance of the period, in sum, was economically positive: "The participation of the transformation industry in the GDP, which was 12.5 % by the end of the 1920s, reached 20 °% in 1955, jumping to 26 °% in 1960! The participation of non-durable consumer goods, which in 1920 was 78 °%, fell to 45 °% in 1960" (^{Cano, 2017}, p. 270).

On the other hand, the developmental policy was also responsible for an inflationary process, especially in basic necessity items such as food and clothing. Paradoxically, Brazil remained very dependent on foreign production in these sectors, and industrialization would not be able to rapidly catch up with international production. So, protectionist measures tended to have an impact on some points, one of them being price increases. Other points would only become clear after the 1970s, during the dictatorship period -with the excess of protection leading to a lack of innovation and technology enhancement. The impoverished had more difficulties, which resulted in the weakening of Vargas' power before his suicide (^{Fausto, 2006}, pp. 114-115). Some of the great development dilemmas, still present today, were in the area of urbanism, with the spread of housing in distant and inappropriate localizations.

Inserted in the international context peripherally, Brazil followed, in the Estado Novo, some of the economic and urbanistic ideas of the central countries, adapting the solutions to the context of an economy still poorly industrialized. In short, we can say that the initial project of Getúlio Vargas, passed on to his successors, was quite coincident with the policy of spatial Keynesianism: an agrarian reform to ensure food provision, nationalization of the production of capital goods and raw materials, a state development bank, etc. Culminating, in the area that interests us, with proposals for nationwide urban plans, contemplating mobility and housing, with attempts to renew the legislation applicable to the theme. We saw that a large proportion of the economic projects were effective, but agrarian and urban reforms were never put into practice.

The Initial Legal Framework

As far as the urban planning law is concerned, this translated into a discourse on planning, where the discussion on eminent domain ( desapropriação) is rooted. Two documents from this period could be cited as paradigms of the position that the eminent domain law would have in the agrarian and urban reform projects.

In 1930, Vargas' first year in power, at the First Pan-American Congress of Architects, the issue of transforming the urban environment was put in very strong terms. The conference's importance was strong not only because it brought together some avant-garde international professionals of architecture, but also because it was set up as a representative event of the us "good neighbor policy" under the Monroe Doctrine. From the conclusions of Congress, one can see the direct influence of the New Deal. The North American urbanistic solutions were highlighted as applicable also to the Latin American reality in general. They included the basis of the "new" Keynesian urbanism of the us matrix almost as an imposition: zoning, national and regional planning, "modern" legislation, etc.

Congress did not fail to observe the need for new eminent domain laws in the Latin American context of civil law to make urban reforms possible:

Theme V - Urbanism and landscape architecture [...]

11° - The Congress draws attention to the need for American countries to create more effective expropriation laws necessary for modern city governments. (I Congresso... , 1940, p. 29)

Despite the generic character of the published conclusions, this observation illustrates the question posed almost a decade before the Brazilian current statute on eminent domain was edited, in a context with particular contours. The 1905 Act was conceived, in its entirety, based on the hygienist paradigm. The 1941 Eminent Domain Act, though not revolutionary, was based on urbanism itself, with less of a sanitary focus and more of an urban spatial planning approach. Congress cited the need for this paradigm shift for Latin American cities, and the alteration of the Brazilian eminent domain law seemed a nodal point.

Another event, historically closer to the drafting of the Eminent Domain Act, was the First Urbanism Congress, in 1941. Although it thematized the subject of eminent domain, this time it was even more succinct. Published by the Departamento de Urbanismo do Centro Carioca in the journal Acrópole as "eminent domain and police power" (using a literal translation of the English word "domínio eminente" despite the Brazilian Portuguese word "desapropriação") the event's panel only concluded on the need for "the formation of a technical body of inspectors of urban observances, determined in the plans and legislations" (Conclusões..., 1941, p. 395). The title says more than the text itself in suggesting the introduction of the distinction, built by us courts' precedents, between takings and police power to implement zoning.

Congress enshrined the concern with zoning still in other panels. This importance would be reflected in the subsequent development of the concept of eminent domain, which was later downplayed as something nearly solved from a theoretical and practical point of view, or even restricted for political reasons, and subordinated to zoning acts efforts.¹

In the first moment, nothing was done in terms of changing the eminent domain act basic liberal paradigms, and Brazil has never had any alternative for regulating property. The status of private property and ownership have never been effectively questioned. Even Orlando Gomes, of Marxist beliefs, defines property according to its essential elements of use, enjoyment, fruition, free disposition, and right to claim it from third parties.

The Second Empire's doctrine was quite influential concerning property. The discussion regarding the issue of the Land Law was concomitant to the confrontation between conflicting schools of liberalism. The most influential Land Law theorizer, Constitutionalist Lafayette Rodrigues Pereira, Prime Minister between 1883 and 1884, defended that two of the "fundamental characteristics of dominion" were its unlimited and exclusive nature, which he defined as the right to "exclude from the thing the action of strange persons." He thus defined dominion as "the right that binds and legally subjects to the absolute power of our will the corporeal thing, in substance, accidents and accessories" (Pereira, 1942, p. 98). He also maintains that the right to property, "from a generic point of view, covers all the rights that form our patrimony, that is, all the rights that can be reduced to a pecuniary value" (Pereira, 1942, p. 99).

The work was considered basic for the study of the law of things for decades, having directly influenced the discussions of the 1919 Civil Code. Besides, one can see how Constitutionalist Lafayette was aware that the essence of private property was its transformation into merchandise, at a time when private property was only formally and comprehensively in force in Brazil. Clóvis Beviláqua, creator of the 1919 Civil Code, himself explains, on the subject, that he preferred a less absolute legal definition for property, understood considering its "limitations", but that the conjuncture did not allow the adaptation of the wording:

On the other hand, the Brazilian Project, with the intention of undoing wrong assumptions of excessive individualism attributed to the jus abutendi of the Romans, which, of course, corresponds to our disposal, preferred to say "use their property". The thought, thus expressed, was intended to emphasize that the right to property did not have the absolute character that some attributed to it; that it suffered restrictions imposed by life in common, restrictions resulting from the neighborhood, public hygiene, the collective interest, when in conflict with the private, as in the case of expropriation, moreover disciplined by law, the shape of public streets in cities and towns, the safety of buildings. (^{Beviláqua, 2003}, p. 135)

Nevertheless, the indicated orientation did not prevail, and the Civil Code presented the property right as unrestricted. This author also criticized the original text of art. 527 of the 1916 Civil Code, inspired by the work of Lafayette Pereira, observing that "the current conception of dominion no longer has it as absolute and unlimited" (^{Beviláqua, 2003}, p. 139), preferring the expression "full." In this sense, he underlines that the owner preserves the property of the thing even when a third party's right conflicts with it, reducing the enforcement of the dominion. The difference is clear. While Lafayette Pereira submits any third-party right to the will of the owner of the domain, in Beviláqua's conception, third parties and the community could overlap and even exclude the powers inherent to the domain in some cases.

As for the legislation on eminent domain, the 1905 Act was still in force, approved in the context of the hygienist paradigm (to which Bevilácqua expressly adhered), directly inspired by baron Haussmann's remodeling of Paris during the II French Empire (some Parisian colleagues of the designers of the III French Empire brought to Brazil practices and technologies). The takings resulting from urban revitalization projects in Brazilian major cities, all quite ad hoc and focusing on the 'beautification' of already rich areas through the opening and widening of streets and equalization of facades, helped accelerate the process of confusing sprawl toward the periphery.

Until 1941, at least, the privilege for "ruinous" buildings referred to in the 1905 Act remained, for the detriment of low-income occupants. One of the articles of the Act referred to tenements and boarding houses in a pejorative way, equating them to unoccupied buildings and authorizing a reduction in the applicable compensation. Thus, the opening of avenues "drained" the lower classes to the periphery. Later, other factors contributed to the expulsion of the lower- and middle-income population from city centers. In fact, throughout the entire period, through Law or Exception, the dispossessed were being pushed away, without compensation.

Another element of expulsion and reconfiguration of the city of São Paulo was the Tenancy Act of 1942. Using as justification the economic and, consequently, housing crisis generated by the Second World War, the Estado Novo passed a bill freezing the price of rents throughout Brazil. In practice, since they were not corrected for inflation, prices would fall to derisory values during the years the Act was in force. This Act would replace, at least initially, a more vigorous proposal for housing provision through the urban reform Vargas dreamed, limiting the rentier force and reconducting its surplus to industry, including the building material industry and the private housing industry in the unregulated peripheries.

According to ^{Bonduki (2017)}, the freeze, despite being formally favorable to tenants, ended up causing a high number of evictions during the 1940s. The tenements were demolished, and less popular properties were built in the urban centers to replace them. Downtown real estate, ambiguously, appreciated in value as the utility for rent decreased and the utility for sale increased. The capital previously invested in rental properties went heavily into industry, including construction. But in addition, many of those who sold their land and real estate began to invest directly in the industry, buying investment papers, advertised in newspapers of the time (pp. 249-252). It was a government's project to implement an individual housing ideology, which, at once, encouraged productive activity and warded off communist pretensions of provate property abolition. The anti-communism of the Estado Novo presupposed the generalization of private property in a more universal way (^{Bonduki, 2017, pp. 90-97}).

The public provision, however, did not happen until 1964, when the Sistema Financeiro de Habitação (sfh) was created. Previous investment in production through the Institutos de Aposentadoria e Pensão (iaps) in the 1930s-with money diverted from social security contributions, excluding most sectors of the formal economy, informal workers and the unemployed-was restricted, and the Fundação da Casa Popular, constituted in 1946 by the Dutra government, was a failure, having, until its replacement, produced mere 18,082 units (^{Bonduki, 2017}, p. 139).

Because of the Tenancy Act and the punctual attempts to provide housing without urban reform, the freezing of rents "joined" the expulsions required to expel the inhabitants to the periphery, where they would buy plots and build their own houses, often in a "mutirão" (community action) scheme. In the absence of formal production, the governments encouraged informal self-construction by loosening building inspection systems and mediating policies of financing construction materials. In a process that lasted throughout the industrialization period, the influx of migrants corresponded to the creation of a cheap labor force for construction and services, just as it happened in the countryside. To keep the low cost of reproduction, self-construction by the workers was an essential element (^{Oliveira, 2003}, positions 461-462). During the dictatorship, the picture would acquire a tragic contour, with the construction of units tied to the SFH in places without any infrastructure connecting them to the urban network in general or, worse, in places with serious environmental risks (^{Rolnik, 2017}, pp. 45-47). Eminent domain without planning or zoning binds also contributed to the emergence of favelas (^{Bonduki, 2017}, p. 272).

In this context, the Eminent Domain Act's draft was inserted, which would be replicated in almost identical form when approved as ^{Decreto-Lei n. 3.365 of 1941}. Its author, Carlos Medeiros Silva, was a lawyer historically linked to institutionalism. A supporter of state intervention in the economy, Carlos Medeiros was, following the preliminary draft, a legal consultant to the Commission for Defense of the National Economy, and the Departamento de Administração do Setor Público (DASP).² The new statute was part of a more general reform of public administration, centralized by the federal government around DASP, provided for in the 1937 Constitution and created in 1938. This department was supported by a Weberian idea of setting up and regulating a state bureaucracy to enable the state to act in its role of rationalizing society and the economy (^{Fausto, 2006}, pp. 93-94; ^{Nohara, 2012}, pp. 35-39). Virtually all legislation on Administrative Law until the 1990s was conceived and edited in the wake of the restructuring of the time.

The purpose of the Eminent Domain Act was not disguised by Medeiros. In an article from 1945, he states about the context in which the decree was granted:

In the imminence of undertaking the construction of large-scale works, the public authorities have, on several occasions, found themselves forced to establish rigid criteria that allow a global estimate of the expenses to be incurred. Thus, the law of 1855, in which the limitation appears for the first time, is linked to the undertaking of the construction of the first railroads in the country; the law of 1903, to the remodeling of the capital of the Republic, and that of 41, to the monumental works designed by the City Halls of the Federal District [then Rio de Janeiro] and São Paulo. (^{Medeiros Silva, 1945}, p. 103)

It should also be noted that the infrastructure built during the period included takings of gigantic dimensions, such as that for the installation of the CSN, in Volta Redonda, or the construction of Brasília, as well as the us air base in Natal (RN).

The text of the Act, however, had few novelties in comparison to the previous law. Besides a less hygienic approach, only some essential points were significantly changed. For example, the Decreto-Lei indemnity valuation based on rents, almost a year before they were frozen. In this way, the amount of the indemnities would decrease year by year, making it possible to accelerate public investment in urban infrastructure works. However, the policy of cost reduction ended up being barred by the Judiciary, which started to disregard the legal criterion invoking the formal "exceptionality" of the Tenancy Act. Carlos Medeiros, in an article published in 1952, criticized the Judiciary for preventing the eminent domain prevision to produce its normal effects.

Conservative judges and jurists in general sustained that the Constitution would guarantee the exact correspondence of indemnity and market price of the property and that the freeze would cause the obsolescence. The Third Chamber of the Rio de Janeiro Court of Appeals collected dictionary definitions of the word "indemnity" -preferred by Brazilian legislators since the slavery abolition discussions to the less conservative word "compensation" (initially to protect slave owners, afterward overprotecting landowners)- to assert its view:

But indemnity, which is the condition imposed by the constitutional precept for eminent domain to be constitutionally admissible, is a word that has a perfectly defined grammatical and legal meaning, of restitution or composition of damage suffered, or the loss of a lucrative right already acquired: it is to render indene, that is, free of loss or damage, "without prejudice". It follows, therefore, that indemnity, in its legal and grammatical meaning, is a constitutionally imposed condition for eminent domain to be lawful, which must be based on public necessity or utility, and all expropriation laws must inevitably be subordinated to this condition. (^{Medeiros Silva, 1945}, pp. 93-94)

As much as the argumentation had some basis, one can see how the clash among jurists about property rights in Brazil was still impregnated by the absolutizing notion of this right. While in Europe and the us alternative property concepts were adopted according to the socioeconomic context, here the attachment to the absolute powers of the owner was almost unanimous. By reforming the eminent domain law without amending the Civil Code and the Constitution, the ambitions of the new Act were decisively limited, beyond the inherent flaws in the conception of the text. This limitation reflected that even changes disguised by the legislator in the eminent domain regime failed. This explains the two main complaints of the mayor of Rio de Janeiro, Henrique Dodsworth, about Decreto-Lei n. 3,365 of 1941, which came into effect during the execution of Rio de Janeiro's improvement plan for the opening of the Avenida Presidente Vargas.

As previously mentioned, Carlos Medeiros and the federal government were aware that urban reforms in Rio de Janeiro and São Paulo were underway, led by the City Halls of the interventors Prestes Maia and Henrique Dodsworth. In the case of Rio, there was a discussion about obtaining the funds for the construction to occupy the esplanade resulting from the dismantling of Morro do Castelo. The plan was drawn up in 1927, by French architect Alfred Agache, but would only be put into practice in the late 1930s-1940s (^{Furtado & Rezende, 2016}, p. 177).

Two legal options, directly linked to eminent domain law, already existed to finance public works. The first instrument was the special assessment tax, article 124 of the 1934 Constitution. As in other countries, this tax is calculated by a percentage of the estimated evaluation of the property, made after the conclusion of the public works. The second was the so-called "taking by zone," which consisted of the expropriation of an area larger than that intended for the public works with the prior intention of selling the properties on the real estate market to finance the works. The inspiration to consider its application, at first glance, might seem to come from the previous experience of the "boot down." However, it came from the American excess condemnation, an experience in which the discussions at the urbanism congresses were mirrored. Unlike the improvement contribution, this method was expressly provided for in the General Achievement Program prepared by the Municipality, with implicit reference to the us experience (^{Furtado & Rezende, 2016}, p. 180).

But neither of these instruments was used. According to the research led by ^{Furtado and Rezende (2016, p. 179}), both were rejected for not providing a prior contribution, being useful only afterward. Furthermore, the tax would require prior discussion and approval in the House of Representatives, in addition to the costs of recollection, without any certainty about the effective evaluation of neighboring properties. The problems persist even today and are barriers that make the special assessment tax almost obsolete.

Other solutions were found, such as the issuing of securities called "Urbanistic Bonds" and, after the Decreto-Lei, the financing by Banco do Brasil.

In fact, the Decreto-Lei that would encourage the public works ended up hindering them. In an article, the former mayor of the Federal District [Rio de Janeiro] himself exposed the problem caused by raising the limits for calculating compensation in the course of the public works, affecting their cost (^{Dodsworth, 1955}, p. 3). This harm would have been "lesser" if the original intent of the law had not been stalled in the following years. The draft also reduced the validity of the public utility decree from three years to two years. As the decrees expired, new ones would have to be issued. These two points were included in an expedient addressed to Getúlio Vargas (^{Dodsworth, 1955, pp. 4-5}), but had only partial effect.

The City Plan Commission also pointed out another, more serious deficiency. The method of judicial expertise for the evaluation of the property would result in a lack of certainty about the budget needed to execute the work, as well as causing greater delay in the taking (^{Dodsworth, 1955, pp. 5-6}). Answering partially this claim, the Decreto-lei prohibited, according to its article 20, the revolving of merit matters regarding public utility in the eminent domain -now made obligatory- judicial suit. However, the evaluation method prevailed in Article 14 of the law: "When dispatching the complaint, the judge will designate an expert of his free choice, whenever possible, a technical expert, to proceed with the evaluation of the property".

From the point of view of the criticism of the original wording of Decreto-Lei n. 3,365/41, the experience of Rio de Janeiro and São Paulo illuminate crucial points. As a consequence of the flaws in the text and a conservative judicial context, complementary public works in Rio were suspended for a while. The execution of the Plan of Avenues, the main São Paulo urban restructuring project, on the other hand, would last for another three decades.

The problems were partially alleviated by two later legislative reforms. Decreto-Lei n. 9,282 of 1946 increased to five years the validity of state-taking decrees. The Lei n. 2,768/1956 enabled provisional immission in the possession of the property, upon deposit of the property's value according to a previous appraisal, authorizing, however, the expropriated party to immediately collect 80 % of the value. In addition, Article 26 was included, which excluded from the right to compensation any third-party rights over the property, a problematic aspect, as the working class has a home but is not property owner.

João Goulart, Agrarian and Urban Reforms and the Aliança para o progresso

The last change in the expropriation regime during the Vargas Era was enforced by Lei n. 4,132/1962, which regulated the social function of property. It was included in the 1946 Constitution through the concept of "social interest". Takings for social interest -alternative to public interest- regular takings, or public need -urgency takings- were set aside not only by Decreto-Lei n. 3,365/41 but subsequently postponed until 1964. Just at this point, it reappeared as a central issue of the short-lived João Goulart government.

Included in the 1946 Constitution, but discussed at least since the beginning of the Estado Novo, eminent domain based on social interest would be directly linked to the base reforms proposed by Vargas, which included: "Banking reform, tax reform, reform of the statutes on foreign capital, administrative reform, electoral reform, university reform, urban reform, and agrarian reform" (^{Bercovici, 2014, p. 98}). Of those, the most prominent and difficult, from the point of view of elite opposition, were exactly agrarian reform and urban reform, which would mean a complete revision of the eminent domain and property regimes:

By combating the monopoly of real estate ownership, urban reform aimed to expand access to urban real estate. Thus, it sought to combat real estate speculation, the implementation of an effective policy of planned popular housing with access to credit, the expansion of access to urban public services and a public transportation policy -which today we would call "urban mobility"-. (^{Bercovici, 2014}, p. 99)

Few people owned real estate, most of the population being leaseholders. A new modality of eminent domain and a group of institutions to promote it at the national level were not only necessary for the installation of infrastructure, but also to promote land distribution in urban and rural environments. As seen, since Vargas, a central organ that planned the housing production and made the poorest population owners was already being designed, not instituted. This logic of universalizing small private property was strategic in the fight against anarchists and communists and, as already mentioned, it was also an agenda of the agrarian reform. The conjuncture was always unfavorable for the creation of a centralized planning of the land, which only led to production using the Retirement and Pension Institutes. It may seem contradictory, but the 1946 Constitution, which for the first time consecrated takings for social interest, was also the one to make its realization impossible by imposing full indemnity in cash (^{Almino, 1980}, pp. 102-108, 226-228).

In this sense, it is important to compare the social function of property, the basis of eminent domain for social interest, with the experiences that allegedly inspired the 1946 Constitution. The Brazilian Constitutions since 1930 had the Weimar Republic as a horizon and, concomitantly, it also enshrined some of the most significant German innovations in their texts. Bismarck's Prussia was still trying to overcome its underdeveloped condition. In the middle of the nineteenth century, the property system was essentially feudal. Common law rights remained. The change was a long process, directed by the State. Between 1807 -the date of the decree abolishing the relationship of subjection between people and property- and 1857 -when the last feudal obligations fell, the Prussian state gradually limited previously existing rights over land, such as the right to graze, freed the population from serfdom and generalized the free disposal of property- (^{Rittstieg, 2000}, pp. 163-165).

After World War I, private property was still a recent reality in Germany, even if already rooted in the Courts, which gave rise to its instrumentalization as a way of solving the economic problems experienced in the interwar period, on the one hand, and resistance from conservative sectors, on the other (^{Rittstieg, 2000}, pp. 201-205). With the 20s and 30s crises, however, Economic Law in general emerged as a necessity. Thinking Law through its macroeconomic consequences was indispensable at a time of economic and social decadence and clash between empires not yet recovered from the depression (^{Comparato, 1978, pp. 455-462}).

The concept of the social function of property was originally issued by Léon Duguit already in France. Duguit, criticizing the liberal concept of private property, argued in his Les Transformations du Droit Privé that property would not be an inviolable and sacred right, but a social function. Inspired by Émile Durkheim, he argued that the property owner is never the individual, but the society in which he is inserted as an organ, and that the individual is merely a means to achieve society's objectives. According to the Frenchman, property and dominium are confused with sovereignty in the absolute monarchy. Thus, the social function would be directly connected to the concept of public service, which would be the new justification for the exercise of power by the Public Administration, replacing national sovereignty and its successor private property of land (Boccon-Gibod, 2014, § 7-8).

In the Weimar context, private property, still in a consolidation phase and heir of Bismarckian Prussia, already appears in its development connected to Economic Law. Thus, the constitutional expression "property obliges" (Eigentum verpflichtet) acquires a different scope from that of Duguit's thought. In fact, the Weimar expression will only be linked to the concept of social function in the 1950s. However, a new profile of property was established, connected to ideas of participation and social responsibility that would take a profile of what Karl Renner would later theorize as an organizational and economic function that would regulate the production and distribution of surplus in society. All subsequent experiences would identify the content of Weimar property with Duguit's concept of social function, adapting them according to each context (^{Polido, 2006}, pp. 16-22).

Two things are important here for the clarification of the concept of social function. The social function is a characteristic, an element of property - not a limit to property as police power. If the social function is not fulfilled, the subjective right to property does not belong to its apparent owner. Eminent domain and social function are distinguished by this fact: in eminent domain, there is property right and its equivalence is ensured through compensation. In the case of social function, there is the abandonment of the absolute character of private property for the benefit of the community, and it can be apprehended in the most diverse ways: prolonged occupation, destination by the government, cultural re-signification, etc. Therefore,

Regarding the social function of property, article 153 of the Weimar Constitution first establishes the guarantee and the binding effects (Bindungseffekte) of private property, especially arising from the expression "property obliges" (das Eigentumverpflicht). The model adopted there provides that property may be subject to taking by law, without possibly including a right to compensation. In the Weimar conception, property does not admit an individualistic, inviolable or sacralized approach, since it submits the exercise by the holder to the interest of the collectivity. (^{Polido, 2006}, p. 12)

The other aspect alluded to is largely ignored by the Brazilian civilist doctrine. It is not enough "that the land be productive to be constitutionally guaranteed" (^{Bercovici, 2014}, p. 104). The social function is directly linked to the Economic and Social Order, provided for in the Weimar Constitution and the Brazilian Constitutions since 1946 (^{Polido, 2006}, pp. 10-12). Both constitutions foresaw a socioeconomic project for the country, allowing much less room for subjectivity than what occurred in practice following their promulgation. This is incompatible with the absolute freedom to use private property theorized by liberal thinking in the 18^th century and adapted and included in Code Napoléon and the Brazilian Civil Code. A good example of disrespect to social function is the case of parking lots common in Brazilian large cities' downtowns. The strategy is quite common: occupying land with a parking lot or car wash avoids squatting and, today, the notification for compulsory use (an instrument created by Estatuto da Cidade, in 2001) which increased property taxes. When the economic climate is favorable, the land is finally sold or used to realize its construction potential.

The German-style social function of property was, in practice, ignored in 1946. Article 147 mentions the social function. Nevertheless, it was linked to eminent domain as the only way to take property devoid of its function: "Article 147-The use of property will be conditioned to social well-being. The law may, by observing the provisions of article 141, § 16, promote the fair distribution of property, with equal opportunity for all" (Brasil, 1946)³. Eminent domain, in turn, was limited by the same article 141, § 16, to indemnity in cash. Thus, although it provided for eminent domain for social interest, it tied it to full indemnity. The restrictive text of the article was a victory of the ruling elite, who, in a close relationship with the gal. Dutra's government imposed a significant defeat to the base reform agendas (^{Bercovici, 2020}, pp. 199-200).

Eminent domain for social interest was the result of a restrictive vision of the social function of property and a way to hinder agrarian and urban reforms. This restrictive vision would reverberate even in the Estatuto da Cidade (Cities Statute) 2001, and even the progressive sectors will not easily dissociate themselves from it. Thus, even before Law no. 4,132/1962, the basic elements of the concept of eminent domain for social interest were established: expropriation, for social housing construction, of property that violates the social function of the property (understood as underutilization), upon prior compensation. João Goulart would be responsible for trying to take a step further, given the constitutional limitations, towards the realization of the base reforms, using exactly this concept, only substituting the compensation in cash for compensation in public debt bonds. The Goulart government, which had been reelected vice-president, and which in its first popular vote had more voters even than popular president Juscelino Kubitschek, would be grounded on base reforms and aversion to foreign capital:

With the country's return to presidentialism in January 1963, João Goulart acquired full powers to try to promote the Base Reforms. Celso ^{Furtado (2013)} was charged with elaborating a Development Plan, called the Triennial Plan. According to the Three-Year Plan: "The country's current agrarian structure thus becomes a serious obstacle to the acceleration of the national economy's development, making it necessary to adjust it to the demands and needs of Brazilian society's progress". The Triennial Plan identified the origin of the relative backwardness of Brazilian agriculture (the low productivity and poverty of the rural population) with the deficient agrarian structure existing in the country. The striking feature was the absurd and uneconomical distribution of land, situated between two extremes. (^{Bercovici, 2014}, p. 100)

Likewise, with urban reform on the agenda, Act n. 4,132/1962 finally regulated article 147 of the 1946 Constitution, focusing, regarding the urban environment, on the “utilization of all unproductive property or exploited without correspondence with the housing, work and consumption needs of the population centers which it should or may supply through its economic destination," the "maintenance of squatters on urban land where, with the express or tacit tolerance of the owner, they have built their dwellings, forming residential nuclei of more than 10 (ten) families," on "the construction of popular housing" (Brasil, 1962, art. 2, clauses I, IV and V).

The agrarian reform, on the other hand, was proposed by an amendment to the constitution, regulating the long-term payment of indemnity, and the creation of national institutions. In 1962, João Goulart created supra, Superintendência de Política Agrária, "with special powers to promote land takings" (Memoria da Democracia, 1962). In March 1964, at the down of Jango's government, he would make a discourse to the "Comício das Reformas" (Reforms' Rally) indicating the approval of a decree by supra, authorizing the taking of land near highways, railroads, dams, and sewage public works. The style of the discourse was characteristic of the developmental era, although pointing to the proposal's limitations:

Workers, I have just signed the Supra decree. I signed it with my mind on the tragedy of our Brazilian brothers suffering in the countryside of our homeland. It is not yet the agrarian reform for which we fought. It is not yet the reformulation of our impoverished rural landscape. It is not yet the freedom card of the abandoned peasant. But it is the first step: a door that opens to the definitive solution of the Brazilian agrarian problem.

At this point, it is necessary to analyze the role of the Alliance for Progress, a United States development program for Latin America, administered by USAID. At the time the development program was launched, a high-ranking official (Assistant to the Secretary, Department of Agriculture), during a 1961 congress in Santiago de Chile, referred to the central importance of base reforms in terms of human, economic, and social development for the region, linking them to the fight against communism:

For that reason, land reform in its broadest meaning, including improvement of all social and economic relations among men engaged in agricultural production, is and must be a basic principle for which the United States stands in world affairs.

Our purpose is to serve the cause of democracy, and thus overcome the conditions which offer communism and other totalitarian forms of society their only chance to advance.

Our concern with land tenure problems elsewhere in the world reflects not only our opposition to communism and other totalitarian ways of life, but our positive belief in Democracy as a way of life recognizing inherent human rights, individual dignity, and brotherhood of the human family. (Waters, 1961, p. i)

Such a discourse was predominant at the time, being an essential part of the post-war development context. The most obvious comparison is with the Italian land reform, largely linked to the Marshall Plan (1944-1950). The purpose of the reform law, called Legge Gullo, of 1944, was to try to reduce the differences, including the economic model, between this more rural part of the country and the highly industrialized North (in particular Piedmont and Lombardy) through a wide redistribution of land in the Mezzogiorno (a generic name for the southern provinces of Italy).

The measures taken until the new law of 1950, which reformed basic protections present in the previous law, included allowing the 1187 cooperatives with 250 000 members to legally occupy unproductive plots of land, and 165 000 hectares were granted in their favor. In addition, benefits were instituted for peasant cooperatives to maintain productivity at competitive levels and better prices than larger-scale individual owners. The advances being considerable, were halted in 1950 after the conservative Catholic wing of the Italian Parliament gained a majority and allied itself with the decaying agrarian elites from South Italy.

Although the law was reformed and the Marshall Plan abolished at the same time, land reform continued, and American banks and the World Bank fed the Fund for the Development of the South (Cassa per il Mezzogiorno), making possible projects in the countryside and industry related to the eradication of malaria, soil drainage projects, securing basic inputs such as electricity and steel, etc. What happened then during the 1950s was a large wave of expropriations, with land going to both individuals and cooperatives, amounting to another 700 000 hectares. In return, the landowners received financial incentives, mainly low interest rates, if they used the compensation for the expropriations in the productive activity carried out in the remaining area itself. Small farm owners were also granted, in addition to the real rights over the land, micro-financing provided by the Fund. Thus, the soil of the Mezzogiorno was divided (not without violent reactions by the former aristocrats) between smaller latifundia that supplied raw materials for industry and the small producers who ensured the country's food sufficiency. The second part of the plan turned out to be the most frustrating: The reform moved on to the implementation of urban infrastructure from the 1960s to the 1980s. However, the funding for public works ended up being appropriated on a large scale by corruption and Mafia groups, culminating in the Tangentopoli scandal (^{Cacciarru, 2020}).

In Latin America, on the other hand, international support for agrarian reform was much less effusive. Despite initial support for agrarian and urban reform, the Kennedy administration began to prioritize rapprochement with national and political elites over effective economic development. The program underwent a major overhaul, making the repayment of debts to the IMF through fiscal adjustment the main priority in the economic field. The effectiveness of Aliança para o Progresso on the economic side was not substantive either. The Marshall Plan would invest $ 13 billion in West Europe, 90 % of this money by grants. On the other hand, the annual aid to Latin America, which was initially $ 30 million, would never surpass $ 150 million. As the distribution was uneven in the Brazilian case, it helped enhance only some regional public policies, generally directed to education or health. The most important consequence of the program for Brazil was the creation of the Comissão Mista Brasil-Estados Unidos para o Desenvolvimento Económico (CMBEU) in 1950, which helped later in the creation of BNDES.

However, according to the design of the program for Brazil, the aid would be granted to the federal states and not to the Union, the us began to use the distribution of aid strategically, strengthening the opposition to the Goulart government and collaborating with the violent closure of popular democracy (^{Loureiro, 2020}, pp. 35-73, 183-188). João Goulart's project of executing the Base Reforms, quite squalid in relation even to the Italian agrarian reform of the 1950s, would be interrupted soon after the regime change. According to José Murilo de Carvalho (1987), land reform would be among the reasons for the military coup of 1964, and it is not illusory to say that also urban reform was a strong additional reason, as much as it supposed the same defies and was proposed in the same period (pp. 93-95). The military regime was made of apparent continuities in this area, with evident problems and a conservative turn, starting with the revocation of the land reform law (by the Estatuto da Terra, Land Statutes) and the targeting of urban reform by the Housing Financial System, which became a mere program of financialized housing production with problems of location, quality of materials, security of tenure, etc.

Conclusion

The Era Vargas was undoubtedly one of economic growth allied to a deep reformism based on Latin American developmentalist ideas, especially from the 1950s on (Furtado, CEPAL). Developmentalist/ Weberian ideas inserted directly into the Administrative Law were substantial gains granted until now. But several problems concerning the base reforms and the right of property could be enumerated, ending up revealing a context of frustrated spatial Keynesianism:

The most practical and despairing issue from those enumerated above is the problem of uneven growth. During the twentieth century, the city of São Paulo grew from 239000 to 10434252 inhabitants.⁴ The concentration of surplus led to a concentration of people and a demographic explosion in the major cities, without any planning. The favelas, originally a reality born in Rio de Janeiro in small dimensions, became the way of living for the majority of Brazilians during the 1980s crisis. The outskirts of Brazilian megacities to the present day do not have basic infrastructure such as minimum standards of housing, transportation, water supply, or sewage.

Still focusing on São Paulo city as we dispose of the data from Rede Nossa São Paulo (2021), seen as the best structured in Latin America, we can observe the inequalities are overwhelming and linked to a racial cleavage: The richer neighborhoods have 5 % to 10 % black people, and the poorest have approx. 60 °% of black people. In one of those neighborhoods, Jardim São Luis, 68,8 % of the people live in favelas. In these neighborhoods, the State does not collect service taxes: Iguatemi contributes only 0,013 % of the city's taxes on services, even with a population of 127000 inhabitants. Finally, 29 neighborhoods aren't even close to a train or metro station.

It is undeniable that the outskirts of Brazilian cities always guaranteed an inexpensive reproduction of the working force for capitalists, producing a surplus population ideal for informal jobs. Even after the pandemic crisis, domestic workers still represent 7 °% of the job market, and have a medium wage, 30 °% lower than the minimum wage.⁵ In the 90s, still 60 °% of Brazilian people did not have a formal job, a reality that has returned in the last few years. That is what Francisco de Oliveira calls the "platypus," as if the Brazilian cities were as ambiguous as this animal's appearance to us, for containing at the same time the realities of the richest and the poorest zones of the world (2003).

Brazil ever failed to centralize decisions, at the same time never giving economic autonomy to other scales apart from the national one. As a consequence, could not distribute spaces and rent, always giving full priority to agrarian and banking elites. But other legal factors are intertwined with general factors: