The Legend of Roman Absolute Dominium

Roman property is enveloped in a mist of mythology. In his 1885 lectures at the University of Rome, Professor Vittorio Scialoja (1856–1933), one of the most prominent Italian Roman law scholars of the time, introduced the topic by warning students that “the Roman theory of property has inspired the strangest legends not only among the populace but also, sometimes, among the educated.”Footnote ⁵⁷ Scialoja was onto something. Roman property has long struck the imagination not only of lawyers but also of historians, economists, and philosophers because of its supposedly “absolute,” “individualistic,” “unitary,” “extremely concentrated” nature. A quick glance at modern Roman law textbooks reveals the ubiquitous presence of the idea that Roman property is absolute. “The Roman law of classical times” – a leading textbook reads – “is dominated by what is commonly called the absolute conception of ownership,” which is defined as “the unrestricted right of control over a physical thing.”Footnote ⁵⁸ This unrestricted right of control includes the right to use (ius utendi), the right to draw fruit (ius fruendi), and the right to abuse (ius abutendi).Footnote ⁵⁹ The owner, we also learn, has very limited ability to parcel out to other individuals these three entitlements, in the way, in the Anglo-American common law, an owner can, for example, divide up ownership of land between a tenant for life and a reversioner. This limited ability in Roman law makes property a “unitary” or “concentrated” right.Footnote ⁶⁰

Such assertions of absoluteness and individualism come with either scathing criticism or impassioned praise. Critics denounce the ruthless individualism of Roman property as both dangerous and lacking realism. For example, Otto von Gierke (1841–1921), who opposed the idea of a “Romanist” private law for Germany, championing instead an indigenous German private law, was blunt in his criticism of Roman property. The Roman concept of absolute property, which had been adopted in the first draft of the German Civil Code, was a fantasy, and a dangerous one, because it encouraged the assumption that property rights are unrestricted and it ignored the limits required by ethics and public law.Footnote ⁶¹ Another critic, Sir Frederick Pollock (1845–1937), could not resist teasing his fellow jurists for embracing a Romanist concept of absolute property that “has no being for English lawyers except when they are trying to pose as economists.”Footnote ⁶² As to how Roman property came to be so strongly individualistic, a bewildered Pollock noted that “we are not aware of any answer to this, or even of any clue. Roman law makes its first historical appearance with its leading ideas already full blown.”Footnote ⁶³ Others, however, acclaimed Roman absolute property as a sound model that we should strive to revive. In his book In Defense of Property, Gottfried Dietze (1922–2006), a German-born American political theorist of anarcho-conservatist leaning, lauded Roman law for being “the most individualistic and property-conscious of all legal systems.”Footnote ⁶⁴ Writing at a time when America was suffering the consequences of a “growing disregard for property,” Dietze praised the Roman jurists for realizing that “the liberal principle demanded that ownership should be as unrestricted as possible and that the greatest possible latitude should be given to individual action and initiative.”Footnote ⁶⁵

Despite its pervasiveness, the idea that, in Roman law, property was an absolute right is, as Scialoja put it, a legend, concocted by Roman law and property law scholars in the nineteenth century. The actual Roman law of property was complex, diverse, and transformed significantly over time, spanning as it did a period of a thousand years, from the Twelve Tables (451–450 bc), an early statute that restated the rules of traditional customary law, to the Corpus Iuris Civilis, the wholesale compilation of Roman law carried out by the Emperor Justinian primarily in the second quarter of the sixth century ad.Footnote ⁶⁶ These thousand years saw the succession of radically different constitutional regimes: from the early Republic, where power was shared between a handful of magistrates, the senate, and the citizens’ assemblies; to the Principate, in which powers were concentrated in the hands of the Emperor, who shared legislative authority with the senate; to the Dominate, where the Princeps became a sort of Hellenistic monarch, first in Rome and then in Constantinople.Footnote ⁶⁷ These thousand years also saw drastic changes in Rome’s economic structure and class relations.Footnote ⁶⁸ From a small city-state based on agriculture and a relatively strong free peasantry; to the progressive concentration of land in aristocratic hands; to the huge territorial expansion of the later Republican era and the beginning of the Empire, with the concomitant accumulation of vast private fortunes by means of exploitation, corruption, and plunder; to the economic integration and growth of the early Empire; to the eventual solidification of social and economic identities starting in the third century ad; and to the economic turbulence and social fragmentation attending the fall of the Western Empire. Finally, these thousand years witnessed a rich and changing cultural landscape: from the appropriation and modification, to the point of originality, of Greek philosophy and political theory; to the “invention” of law as a specific form of social regulation; to the birth of Roman Christianity and the development of a systematic Christian theology.Footnote ⁶⁹ These political, economic, and cultural/ideological transformations were inevitably reflected in the intellectual development of property law; through their logical and systematic methodology, the Roman jurists developed a rich conceptual vocabulary of property, that is, a set of concepts, organizational schemes, and doctrines to describe and solve the constantly changing legal questions posed by property.

When they set out to develop modern property, the nineteenth-century jurists approached the juristic texts of Roman antiquity with a combination of sincere scholarly devotion and instrumental spirit. They regarded the writings of the Roman jurists as highly authoritative and deserving of rigorous historical and philological study. Yet they approached these writings with their own practical questions, ideological views, methodological queries, and professional ambitions. Hence, they selectively borrowed from the vast and diverse conceptual vocabulary of property developed over the centuries by the Roman jurists only the ideas and doctrines that spoke to the absolute character of property rights, overlooking all else. Readers accustomed to the legend of absolute dominium that pervades the nonspecialist literature on Roman property will be surprised at how much richer and more varied the Roman conceptual vocabulary of property actually was. Lost in the legend of absolute dominium is all that was pluralistic, amenable to being disaggregated, relative, and limited in Roman property law.

The literature that explores the complexity of Roman property law is wide and ambitious, and scholars have offered remarkably different narratives of Roman property’s development.Footnote ⁷⁰ The objective of this chapter is not to provide an exhaustive and detailed account of Roman property law but rather to offer an inevitably simplified overview of Roman property that will serve as a preamble to the story of the invention of Romanist-bourgeois property. In the pages that follow, we will look into the features of Roman law that attracted the nineteenth-century property modernizers: a scientific method and an inventory of concepts and doctrines that could easily be repurposed.

A Powerful Professional Role Model: The Roman Jurists and the Promise of an Impartial Legal Science

Who were the Roman jurists who captivated the imagination of their nineteenth-century heirs and shaped their professional ambitions? And what were the reasons for, and the scope of, their promise of a neutral property science? To fully understand what exactly the Roman jurists had to offer their nineteenth-century successors, we have to look back to the early days of Rome.

The jurists’ demand for recognition as a distinct professional class grew out of the fascinating development of the nature of their work, from religious rituality to scientific reason.Footnote ⁷¹ In early Rome, law was the domain of a specific priestly circle, the pontifices, who were the guardians of the mos maiorum, the religious and social custom of the ancestors. The pontifices were not priests in the sense of spiritual figures. Rather, they were men of patrician social standing who, because of their economic privilege, could undertake public duties without pecuniary remuneration. Their prescriptive pronouncements took the form of “responses” (responsa) to questions of the patres (the living male patriarchs of the community), who wanted to know what ritual should be used when performing specific operations of daily life, such as claiming power over a thing, conveying property, or entering into marriage.Footnote ⁷² The pontifices’ responses combined formalism and practical rationality, prescribing times, words, gestures, and the use of ritual objects, such as the wand, the piece of bronze, and the scales, to address practical questions of everyday life. As Aldo Schiavone has noted, the combination of formalistic ritualism and practical reason that we find in this early priestly jurisprudence would, in time, develop into a critical feature of legal science, both Roman and modern.Footnote ⁷³ This ability for both abstraction and case-based practical knowledge would be the basis of the jurists’ assertion of expertise and their claim to power.

Between the end of the fourth and the beginning of the third century bc, the locus of power shifted from religion toward politics and giving responsa became an aristocratic prerogative, rather than a priestly one.Footnote ⁷⁴ Responsa were now produced by experts drawn from the new “aristocracy of government,” which included not only the patrician elite but also the most important plebeian families.Footnote ⁷⁵ These aristocratic legal experts were not yet “jurists” in the modern understanding of the term, and their work was not yet “science” in the sense of constructing abstract concepts; rather, they engaged in practical legal analysis with the aim of solving individual cases.Footnote ⁷⁶ The “revolution” that transformed legal knowledge into a science and the jurists into a highly influential professional class of their own happened around the middle of the second century bc. The only surviving narrative account of the rise of Roman legal science is a short passage from the Enchiridion, a treatise written around ad 130–140 by the jurist Sextus Pomponius. Pomponius believes that three men, M. Manilius, M. Junus Brutus, and P. Mucius Scaevola, “established the civil law” and that “Quintus Mucius, son of Publius and pontifex maximus, was the first to set the civil law in order by arranging it generatim [by genera] into eighteen books.”Footnote ⁷⁷

Pomponius’ account is laconic and sheds no light on the reasons that led to the “establishment” of a science of law nor on the specific contributions of this new science. In his book The Rise of the Roman Jurists, Bruce Frier identifies the origins of Roman legal science in the “tacit bargain” between the newly emergent socioeconomic elite of the Late Republic, who, at a time of political instability and social ferment, demanded greater security of property and contract rules, and the jurists, eager to be recognized as a powerful professional class. The new elite included the established “equestrian” class of well-off citizens – who owned less property than did members of the senatorial class, often derived their income from their estates and from trade, and in many cases performed roles that were crucial to the functioning of the state – as well as a heterogeneous group of upwardly mobile municipal notables, shippers, and merchants. Largely comprising self-made men, this group had individualistic social values and was more rule-oriented in its understanding of private law than were the old citizens who had greater familiarity with Roman private law;Footnote ⁷⁸ consistent with this interest in private law as it developed from the middle of the second century bc is the fact that the equestrian class would produce some of the most influential of the Roman jurists. The rules of property and contract appeared to this new elite increasingly more unstable for a variety of reasons. Among these reasons were the significant increase in the number of disputes following the expansion of Roman citizenship after the Social War of 91–88 bc and the earlier introduction of a new procedural system, the formulary procedure, which gave the praetor, the magistrate who controlled the granting of remedies, more discretion to fashion legal actions and, hence, to shape new law. The jurists’ response to the new elite’s demand for greater stability in private law was to turn law into a written science with its own concepts and modes of reasoning, seemingly insulated from transient political changes.Footnote ⁷⁹

What was the essence of this new science? All Pomponius tells us is that Quintus Mucius Scaevola (140–82 bc) was the first to organize the civil law generatim, that is, by genera, or “categories.” Generations of historians from the nineteenth century to the present have identified the systematic organization of legal concepts as the unique contribution of this new science and hailed Quintus Mucius, and his colleague Servius Sulpicius Rufus (ca. 105–43 bc), as close contenders for the role of the father of Roman, and modern, legal science.Footnote ⁸⁰ Whether the Roman jurists themselves understood systematic legal analysis to be the essence of the science of Quintus Mucius and Servius is unclear, but certainly this seems to be the view of Cicero. Cicero, who, as a former student of Quintus Mucius and a friend of Servius, was close to both men, praised Quintus Mucius as the exemplary Roman hero and credited him with making good practical use of the law while also lauding Servius for transforming law into an ars of iuris scientia.

The supposedly systematic approach of Quintus Mucius and Servius was part of a larger intellectual movement to improve the organization of knowledge in different domains, from law to grammar, to eloquence, to architecture.Footnote ⁸² As Cicero described it in his De Oratore:

That the Roman jurists’ “scientific revolution” consisted of binding together into a system legal concepts that had been loose and divided, as Cicero nicely captured it, was also the understanding of one of the most prominent Romanists of nineteenth-century Germany, Georg Puchta (1798–1846). Puchta played a critical role in popularizing the legend of Quintus Mucius as a virtuous statesman and the father of the systematic approach to legal analysis. To be sure, Puchta may well have been anachronistically attributing to Quintus Mucius his own methodological creed, which focused on the task of system-building.

The truth is that while the legend of Quintus Mucius continues to thrive, we know very little about the actual nature of his work. None of his writings have survived intact, and we are left with sparse traces of his thought in later works, almost always in the form of quotations by later jurists.Footnote ⁸⁴ These later quotes can only give us a superficial sense of what Quintus Mucius’ systematic classifications may have looked like. For example, Gaius recounts that Quintus Mucius identified five types of tutorship:

Similarly, we learn from the jurist Paulus, active during the reign of the Emperors Septimius Severus and Caracalla, that Quintus Mucius’ classified possession into distinct types, or genera possessionis, although Paulus was not persuaded by this schema:

The method of classifying larger concepts into kinds that we see applied in these brief surviving quotes was not Quintus Mucius’ own invention; he borrowed it from Greek philosophy. Diairesis was a method of reaching a definition through the logical division of kinds (genera and species). This classification into kinds would then lead to the discovery of the principles governing the kinds and explaining individual cases. The diairetic method was first developed in Plato’s dialogues and later became a method for classification used in the Aristotelian and Stoic schools.Footnote ⁸⁷ Quintus Mucius was familiar with diairesis because it was a part of the intellectual culture of the Roman aristocracy.Footnote ⁸⁸ Quintus Mucius’ appropriation of diairesis for the purpose of legal analysis was the fundamental move that transformed law into a “science,” in the sense in which the term is used by Plato and Aristotle, and introduced Roman jurisprudence into the circle of the Hellenistic professional sciences.Footnote ⁸⁹ Through the diairetic method, law achieved what was required from a science: the logical classification of elements, their systematic integration, and the definition of their governing principles.

The methodology supposedly developed by Quintus Mucius would influence the nineteenth-century craftsmen of Romanist-bourgeois property in two critical ways. To begin with, the diairetic method combined two intellectual processes – formalistic abstraction and practical case-oriented analysis – which the modern jurists would continue to see as the essence of legal science and of their unique expertise.Footnote ⁹⁰ To analyze the facts of social life, such as belongings, exchanges, and obligations, into genera and species, the Roman jurists needed both a good grasp of the living reality they were studying as well as the ability to organize this living material in abstract concepts. In property, this meant organizing the myriad different types of “things” (res) that can be the object of ownership according both to their physical characteristics and to the interests and values they implicate; it also meant classifying the many possible ways in which something “belongs” to someone within the abstract concepts of dominium, bonitary ownership, possession, and a number of lesser real rights. This ability for abstraction and practical reasoning was sharpened by the use of diairesis but, as we have seen, it was already present in the early days of the priestly jurists who prescribed formal rituals for everyday legal acts.

Another aspect of Roman legal science would influence the creators of modern Romanist-bourgeois property. From Quintus Mucius onward, the Roman jurists saw their abstract concepts not merely as useful categories of thought but as real entities, as objective things that legal science would study and analyze but that had a life of their own, separate from, and preexistent to, the intellectual activity of the jurists.Footnote ⁹¹ As Aldo Schiavone notes, it is difficult to determine whether Platonic or Aristotelian ideas directly influenced the jurists’ belief in the objective existence of legal concepts, but certainly this epistemic commitment to the objectivity of legal concepts owes a lot to Greek philosophy and would feature prominently in nineteenth-century property science.Footnote ⁹²

Historians like Franz Wieacker have doubted that either Quintus Mucius or Servius could have developed anything like a systematic organization of the law since both were still inevitably hamstrung by the jumbled and casuistic framework of the Twelve Tables, the earliest and most influential compilation of Roman laws.Footnote ⁹³ The story of Quintus Mucius and Servius may well be yet another modern legend concocted by nineteenth-century Roman law scholars eager to bolster their own professional power.Footnote ⁹⁴ Yet its influence is still with us today. Just as the legend of absolute dominium lent prodigious power to the modern robustly individualistic idea of property, so too the legend of Quintus Mucius’ systematic science conferred tremendous strength on the abstract formalistic approach that enabled the fiction of absolute property.

The Elements of the Roman Conceptual Vocabulary of Property

To be sure, dominium ex iure Quiritium, that is, the right that gave Roman citizens exclusive and ample legal power over the things that, under the law, could be privately owned, was an important concept of Roman property law. However, the Roman conceptual vocabulary of property comprised a variety of doctrines and concepts that made property plural, flexible, and limited. These elements of pluralism and variability include: (1) a classification of the different things (res) that could be the object of property rights and the correspondent variety of res-specific property regimes; (2) a wide menu of lesser forms of “ownership,” conflated under the label of “possession,” that existed alongside full dominium; (3) a menu of lesser property rights over things owned by another (iura in re aliena) that allowed owners to parcel out entitlements and achieve a variety of social and economic goals; (4) a complex regulatory regime that limited owners’ entitlements and the broad catch-all concept of “abuse of rights”; and (5) the notion of lex agraria, a blueprint for a comprehensive plan to correct patterns of inequitable access to fundamental resources. In the pages that follow, we will focus on these five features of Roman property both because they present a stark contrast with the legend that Roman property was absolute and unitary, and because these five features would have pride of place in the making and remaking of modern property. Each of these features would be retrieved by subsequent generations of modern jurists who would draw inspiration from them while fiercely disagreeing over their meaning and applicability.

The Roman Law of Things: The Classification of the Various Types of “Res”

A reader searching for the definition of property in the Roman sources would be disappointed. After all we have said about the legends surrounding Roman “absolute” dominium, the absence of the definition, or of any definition, comes as a surprise. But offering one, univocal definition of property, good for all purposes, is not how Romans thought about property or law more generally.Footnote ⁹⁵ As noted by Pietro Bonfante (1864–1932), one of the most eminent Italian Romanists, “the attempt to define property is typical of modern jurisprudence. In the Roman sources, there is no trace of a definition. There are, indeed, many definitions that pass as textual and Roman, or at least as inspired by Roman texts. But this is an illusion of the exegetic fetishism of modern times.”Footnote ⁹⁶

Roman law textbooks, from Gaius’ Institutes to modern textbooks, begin their exposition of the law of property not with a definition of the concept of property but with a classification of the various types of res. For our purposes, we will translate res as “thing,” although the Roman meaning of res is broader and is the object of significant controversy among Roman law scholars.Footnote ⁹⁷ The idea that property is a “law of things” is a popular one today, nicely captured in the title of a highly-cited recent article by US property scholar Henry Smith.Footnote ⁹⁸ Roman property law was, quintessentially, a “law of things.” Before delving into a more detailed exploration of how the Roman jurists classified things, we need to pause and consider the fundamental relevance of “the law of things.” While obscured in the modern legend of Roman absolute dominium, this pragmatic focus on things is the single most important legacy of Roman property. The Roman classification of things is interesting intrinsically, as a mode of legal analysis, as well as for the influence it had on modern jurists.

As a mode of legal reasoning, the Roman “law of things” is a good example of the Roman jurists’ talent for combining practical reasoning and formal categorization. The Roman jurists did not base their classification of things on some abstract and intuitive understanding of the different types of things. Rather, they brought into play specific frames of positive or normative analysis, external to law and drawn, to a considerable extent, from religion and philosophy. This pragmatic, positive, and extralegal cognizance of things was then organized into formal, legal categories. The significance of the Roman “law of things” for modern property thinking is momentous albeit largely overlooked. Contrary to the legend of a monolithic and absolute dominium, the Roman classification of things speaks to the pluralism of property values and the consequent variability of property entitlements. As Pietro Bonfante noted, the Romans had the intuition that “property is not, and cannot be, a uniform, monolithic block; property is subject to varying limits and regulations, so that it assumed radically different configurations depending on the object that is owned.”Footnote ⁹⁹ In the hands of modern jurists, this intuition about the pluralism and variability of property would be a fertile source of theoretical and practical innovation.

Let’s now take a look at some of the specific ways in which the Roman jurists classified things, which is maddeningly complicated to our modern eyes. In modern property law, the most familiar and fundamental distinctions are that between movables and immovables, or real property and personal property, and between private property and public property. While some of the Roman distinctions mirror modern ones, much of the Roman world is opaque to us and we must guard against anachronism in our effort to understand its legal system. To understand the Roman classification of things we have to shed these familiar distinctions and look at Roman law and Roman society with fresh eyes.

The first and fundamental distinction between the various res, fundamental because it marks the very boundaries of the realm of property, is that between “corporeal things” (res corporales) and “incorporeal things” (res incorporales). This origin of this distinction is philosophical.Footnote ¹⁰⁰ The Roman jurists borrowed it from Roman philosophers who, in turn, got it from Aristotle. We can find a neat example of this slippage from the language of philosophy to that of law in Cicero’s Topica. Cicero starts by providing a philosophical definition of corporeal and incorporeal things but then draws his examples from law.

The transition from philosophical debates to legal textbooks seems to be complete by the time of Gaius, the jurist of the second century ad, who, in his Institutes, discusses the distinction between corporeal and incorporeal things in a manner that closely tracks Cicero’s and that would be literally reproduced in Justinian’s Institutes.Footnote ¹⁰² For Gaius, corporeal things are the material, tangible things that are the object of property, while the term “incorporeal things” has the broader sense of “legal relations.” In other words, the distinction between corporeal and incorporeal things seems to have the same meaning of another distinction that is recurrent in the sources, that between corpora and iura, that is, material, tangible things and rights. Both ways of distinguishing demarcate property as the realm of tangible things.

Another distinction that was central to the Roman social and legal imagination is that between res in commercio and res extra commercium. The former were things that could be privately owned and freely exchanged in the market; the latter were exempt from private property and market transactions because of either their religious significance (res divini iuris) or their public import (res humani iuris). Among the things that involved religious interests were the objects and buildings used to honor the gods (res sacrae), to remember the dead – such as graves and burying sites –(res religiosae), and the walls that encircled the city of Rome and its gates (res sanctae). Two aspects of this classification are fascinating for the modern reader. First, this distinction does not map neatly onto our familiar modern distinction between private property and public property. As Yan Thomas has suggested, by classifying things into res in commercio and res extra commercium, the Romans marked off a separate realm – the sacred – from that of the public, while using terms that indicated their contiguity and close relation.Footnote ¹⁰³ This fuzziness of the line that separates the sacred and the public is for us largely unfathomable, as it confounds our ideas about the separation between human and divine, state and church, law and religion.

A second aspect of this distinction that deserves mention is its constitutive role. By ascribing things to different classes, Roman law did not simply recognize and ratify the existing, shared, and plain cultural, social, or economic value of a thing. It also constituted new things, giving them new status, meaning and value.Footnote ¹⁰⁴ Take the case of “sacred things” (res sacrae), that is, the things used to honor the gods. A building or an object would become “sacred,” and hence exempt from private ownership and market transactions, through a procedure. Res sacrae were consecrated to the gods in a solemn ceremony. After previous authorization of the Roman people (through a lex, a plebiscitum, or later a senatusconsultum), a magistrate representing the Roman people transferred ownership of the res sacra to the gods, represented by the pontifex.Footnote ¹⁰⁵ Occasionally, the consecration of property to the gods could become a highly controversial affair involving fierce political rivalries and acts of retaliation, as happened in a case involving Cicero. During Cicero’s exile, his enemy P. Clodius Pulcher, consecrated Cicero’s house on the Palatine Hill to the goddess Libertas. Clodius’ consecration was clearly an act of political aggression against a rival. In a highly emotional speech entitled De Domo Sua, Cicero fiercely disputed the validity of the consecration and, ultimately, prevailed. Cicero’s speech deserves attention not only because it gives us a sense of the procedural requirements of the ritual of consecration, but also because it illuminates the constitutive role of the Roman classification of things. Through the legal and religious ritual of consecration, Cicero’s house, privately owned, was constituted into something different, a res sacra, thereby changing its status and value. A distressed and mystified Cicero decries the cruel and vengeful use of the legal category of res sacra:

Moving to distinctions that are more familiar to the modern reader, Roman law’s treatment of the distinction between private and public property deserves close attention. Things that were exempt from private property and market transactions because of the public interest were subdivided into “common things” (res communes), “public things” (res publicae), and “things belonging to the municipality” (res universitatis). The concept of “common things” has inspired the moral and legal imagination of generations of modern jurists in Europe and beyond, who have framed legal claims about access to natural resources, water, or artistic and historic treasures in the language of “common things.” Despite its modern appeal, the notion of res communes is controversial among Romanists because, while it appears in Justinian’s Institutes, it is absent from Gaius, whose Institutes inspired Justinian’s, and from virtually every other classical source. The only exception is Marcianus, a jurist of the time of the Severi, who wrote that:

Roman law scholars, from Mommsen to Bonfante, have dismissed the significance of this passage. For some, Marcianus was sloppy in his use of language, and by res communes he really meant res publicae.Footnote ¹⁰⁸ Others assert that Marcianus was simply translating in legalese what was, in fact, a philosophical concept.Footnote ¹⁰⁹ Once again, as in the case of corporeal and incorporeal things, the legal category of “common things” seems to be directly informed by philosophical ideas. In a passage of his moral treatise De Beneficiis, Seneca explains the philosophical concept of “common things” as the gifts that the gods have bestowed on all humans, regardless of merit, such as the sun, the alternation of seasons, natural fonts, and regularly blowing winds.Footnote ¹¹⁰ Over time, “common things” became a legal concept that applies to the air, the flowing water, the sea and the sea shores, as well rivers and river banks. In Justinian’s Institutes, the idea of an essential, innate gift from which no one can be excluded is no less vivid than in Seneca but is associated with a specific legal regime, the core element of which is that the public has the temporary and shared right to access and use common things.Footnote ¹¹¹

Turning to things that belonged to the realm of private property and market transactions, the fundamental modern distinction between real property and personal property exists in the Roman sources, but its actual relevance is unclear, at least in the classical period. For one thing, this distinction seems to be absent in the terminology of the classical jurists, who, instead of “real property,” used more specific expressions such as fundus (parcel of land), praedium (farm), or fundus et aedes (land and buildings). For another, the supposed differences in the way classical Roman law treated real property and personal property (employing different terms for usucapio/prescription, recognizing different remedies to protect possession, and indicating differences in the modalities for conveyance by traditio) are less meaningful than they appear.Footnote ¹¹²

Regardless of the merit of these arguments, Roman law scholars seem to agree that, in the classical period, the distinction between real property and personal property is obscured by another, more fundamental, distinction: that between res mancipi and res nec mancipi.Footnote ¹¹³ This latter way of classifying things is foreign to the modern way of thinking. It is rooted in the peculiar ritualistic formalism that characterized early Roman legal thought and it refers to the different modes of conveyance required for these two classes of things. Res mancipi could only be conveyed by mancipatio or in iure cessio. A simple delivery would not suffice. Mancipatio was a ritual ceremony involving a fictitious sale, described in detail by Gaius:

As Gaius’ passage suggests, mancipatio as a mode of conveyance was not available to just anyone. The limited availability of mancipatio reflected the inequities of the Roman law of persons. Mancipatio was initially limited to Roman citizens and only subsequently extended to Latins, both coloniarii (inhabitants of the colonies) and Iuniani (informally manumitted slaves). And women could perform mancipatio only with the authorization of their guardian. The other way for transferring ownership of res mancipi, in iure cessio, was no less formal or cumbersome. It simulated the action asserting ownership (vindicatio) and it required appearing before the praetor.

Ceremonies aside, what types of things were res mancipi, and how is this distinction related to that between movables and immovables? Pseudo-Ulpian informs us that:

The nature of the things on Pseudo-Ulpian’s list suggests that res mancipi were the resources that had significant value in the early Roman agricultural economy. In other words, the distinction between res mancipi and res nec mancipi was based on their socioeconomic significance.Footnote ¹¹⁶ The distinction partially overlaps with that between real property and personal property because most items on Ulpian’s list happen to be land or rights related to land. And the rationale of the two distinctions seems to be similar: certainty and publicity. Because of their socioeconomic significance, the transfer of ownership of res mancipi required specific, highly ritualized modalities capable of ensuring certainty in economic transactions. By the time of Diocletian (ad 284–305), the distinction between res mancipi and res nec mancipi had lost relevance and it was eventually abolished by Justinian. As the distinction between res mancipi and res nec mancipi declined in importance, that between movables and immovables gained more relevance.

Real property was the object of further classifications that reflected either the economic use of the resource or its location. Immovables were divided into rustici and urbani. The former were land and buildings used for agricultural purposes, the latter for residential purposes. The economic use, not the location, was the basis for the distinction, so that there was rustic real property in the city and urban real property outside the city. The location of the land was, at least initially, the basis for another distinction, that between Italic land, located in Italy, and provincial land. The distinction was a critical one: Italian land could be owned in dominium, that is, the full ownership reserved to Roman citizens. By contrast, as Gaius explains:

Because it belonged to the Roman people, provincial land could not be the object of dominium. It was owned by the Roman state and granted to private individuals who had entitlements similar to those of an owner but who were subject to an annual tax. These provincial landholders were not formally owners and were designated with formulas such as habere possidere or frui licere. However, as time passed, the distinction lost its initial connection with the location of the land and came to be based merely on the legal regime the land was subject to. Italic land came to designate any land subject to the ius Italicum, even if actually located outside of Italy. Hence, the ius Italicum is yet another example of how the Roman classification of things actually created or constituted new “things” with new value and legal meaning. The ius Italicum was a legal fiction, a privilege granted to certain communities in the Roman provinces whose land was treated as if it were in Italy. Lands that had the privilege of the ius Italicum were governed by Roman private law and could be the object of dominium.Footnote ¹¹⁸ The legal fiction of the ius Italicum may have been an important tool to negotiate the political complexities of Roman imperial expansion. This new, imperially minded use of the ius Italicum was formulated in the first years of Vespasian’s Principate, when Vespasian, pressed by the urgent need for new revenue, reorganized imperial finances, significantly raising taxes. The grant of the ius Italicum was a way to exempt from the highly unpopular tax reforms colonies established by Italians, who were overwhelmingly legionary veterans.Footnote ¹¹⁹ The distinction between Italic land and provincial land lost relevance with time and was eventually abolished by Justinian.

Finally, another set of classifications hinged on the physical nature of the thing itself. Roman law distinguished between fungible and non-fungible things. The former (res quae in genere suo functionem recipiunt) are things whose individual units are capable of mutual substitution, such as wine, olive oil, or grain of the same quality. Non-fungible things are unique in their individuality. A bemused Bonfante illustrates this distinction between fungible and non-fungible things with an anecdote. Today, Bonfante notes, we think of works of art as non-fungible. This was not the case for the Romans. Apparently, at the time of the conquest and sack of Corinth (146 bc), which entailed one of the largest hauls of artistic treasures in Rome’s history, the general Lucius Mummius had the shippers who brought back the bronzes to Rome promise that, if the bronzes were lost in transit, they would replace them with new ones.Footnote ¹²⁰ Probably, Bonfante took too seriously a joke about Mummius’ lack of sophistication that was popular among Mummius’ rivals.

Roman jurists also distinguished between things that can be consumed and things that cannot. The distinction was legally relevant mainly because the former could not be the object of a usufruct, that is, the right to use another’s property and to take its fruits, without impairing its substance. Another distinction, one that is partially opaque to us because rooted in Stoic physics, is that between “single things” (res unitae), “composite things” (res compositae), and “totalities of things” (universitates rerum). Let us hear Pomponius’ explanation:

Pomponius describes “unitary” as characterized by the fact that they contain one single spirit. His explanation is informed by the Stoic understanding of nature. For the Stoics, material things of the world hold together because of an internal flow of either one pneuma or numerous pneumata, the breath(es) of life, that produces cohesion, called hexis, in the matter.Footnote ¹²² Hence, a log or a seashell is held together by the “dynamic process” of hexis, which, in the case of a unitary thing, consists in one pneuma, translated by Pomponius as “spirit.”Footnote ¹²³ In property law, whether a thing is single or composite or a universitas rerum mattered mostly for the purpose of acquisition by usucapio (adverse possession). Does the adverse possessor of a composite thing acquire the whole, or are the distinct things that become part of composite thing acquired separately? Pomponius explains that:

While intricate and, at times, opaque, the significance of the Roman classification of the various kinds of res cannot be overstated because it impacts the very conceptual structure of property. Where the modern reader, familiar with the legend of Roman unitary and absolute dominium, expects to find a general, all-encompassing definition of property, we find instead a pluralistic structure, consisting of many res-specific property regimes.

A Critical Conceptual Pair: Absolute Dominium and Possession

The pluralism of the Roman classification of things did not guide the creative activities of nineteenth-century jurists, who, instead, embarked on the quest for the definition of property. They projected their own agendas and concerns on the few, short ancient passages that could suggest a definition of property, at times going so far as to drastically change or even invert what they found in the works of their Roman predecessors.

As it happens, the Roman jurists themselves had little to say about the concept of property. To begin with, the terminology varies and one would be tempted to conclude that, in the early and classical period, the Romans did not even have a term that unequivocally meant “property.” In fact, the words used for property seem to have had broader meanings. Mancipium, which seems to be the earliest term for property, also designated the authority of the king, of a magistrate, or of the pater familias over the members of his family. Dominium was often used in the wider sense of a “subjective right,” and dominus had the broader meaning of “right-holder,” not only the holder of property rights (dominus proprietatis) but also the holder of a usufruct (dominus usufructus) or the heir (domimus hereditatis).Footnote ¹²⁵ Only the word proprietas, which became more widely used in the later imperial era, seems univocally to have meant “property.”Footnote ¹²⁶

No less puzzling is the fact that, while Romanists have identified in the sources several “definitions” of property, all of which emphasize the broad scope of the owner’s powers, at closer inspection, none of these supposed definitions is really about property. Rather, these definitions are extrapolated from texts that discuss specific legal questions that bear only a loose relation to property. A first definition takes the form of a maxim inferred from a rescript of the Emperor Constantine.Footnote ¹²⁷ It reads “every man is the ruler and arbiter of his own property according to the ius civile.” The precise Latin wording of this definition is not found in any ancient text but is rather a product of later juristic reformulation.Footnote ¹²⁸ Neither this definition nor the rescript, however, includes the word dominus. Further, in order to construe the meaning of “one in complete control of his property,” the maxim had to take the opposite of the meaning of the relevant sentence of the rescript, which says explicitly that the “ruler and arbiter” cannot do as he pleases with regard to his property in all instances. So this first definition can be ruled out.

Another supposed definition that made an impression on the nineteenth-century jurists for its absolutist tone is that property is “the right to use and abuse one’s thing (ius utendi et abutendi re sua).” It is drawn from a passage of Ulpian on the claim for inheritance:

The reader will wonder what this passage has to do with property. The text addresses the problem related to a possessor in good faith of an inheritance, only indirectly casting some light on the question of the powers of an owner. Because the good faith possessors of the inheritance here believed that they were owners and hence were entitled to “abuse” their property, they should not be held liable for the acts listed, including “squandering” and “causing waste.” Nineteenth-century property writers would tirelessly play up these words as suggesting that Ulpian conceived of property as the absolute right to use and abuse one’s thing. However, the matter is more complicated since, as the translation suggests, the word abuti is closer in meaning to “consuming” than to “abusing.”Footnote ¹³⁰ In other words, the ius utendi and abutendi re sua suggests that the owner has the right to use the thing and to fully consume it to the point of exhausting it, rather than the right to abuse, waste, or capriciously destroy the thing.

A third famous “absolutist” definition of dominium says that “it is the right of complete disposal over a corporeal thing, as long as it is not prohibited by law.”Footnote ¹³¹ But this definition is not Roman. Its author is the medieval jurist Bartolus de Saxoferrato (1313–1357), who coined it in his comment on a passage of Ulpian that explains the difference between ownership and possession but says nothing about an owner’s absolute right over a thing.Footnote ¹³² Yet another supposed definition of property, one that would serve as a template for many of the modern civil codes, is taken from Florentinus, a jurist of the second century. Florentinus tells us that:

This sentiment, here concisely and powerfully expressed, would be critical to modern political and legal theory, that is, in the idea of negative freedom and its limits, but, once again, it is not about property.

The terminological and definitional vagueness that surrounds Roman property leaves us wondering whether the Roman jurists actually intended to be “absolute” and “unitary,” as the nineteenth-century architects of Roman-bourgeois property wanted us to believe. The answer is far from straightforward. Undoubtedly dominium ex iure Quiritium had unique features. It was the broadest form of ownership, reserved to the Quirites, the ancient Roman citizens, intrinsically connected to the (Roman) land and therefore charged with high political and symbolic meaning. As Pietro Bonfante argued, dominium was a right akin to territorial sovereignty and it belonged to its owner, the quiritary citizen and pater familias, “not only for the economic purpose of utilization, but also for the political purpose of preservation and defense.”Footnote ¹³⁴

Because of its social and political relevance, dominium was highly symbolic and its unique features were staged through elaborate rituals. Its intrinsic connection with sovereignty over land was performed through the ritual of limitatio. In this solemn ceremony, the boundaries of land owned by a Roman citizen were marked with termini, that is, posts of wood or stone. A free space, five-feet wide, was left to separate neighboring parcels. This free space, called the limes, was devoted to public use, ingress and egress, and ploughing. The limitatio could be of one of two types: centuriatio, which divided the land in square or rectangular parcels forming a regular grid, and scamnatio, which divided the land in rectangular plots irregularly arranged. A fascinating example of legal geography, the limitatio shaped the appearance of the physical landscape and determined the legal and social relations of its inhabitants.Footnote ¹³⁵ It marked, physically and symbolically, an enclosed space within which the sovereignty of the owner extended above and below the surface, was purportedly immune from the interference of neighbors and of the state, and was perpetual.Footnote ¹³⁶ In other words, the function of the limitatio was to eliminate, symbolically, from the property the reality of the inevitable conflicts and interrelations between neighbors by relegating these conflicts into a dedicated public buffer zone. The nineteenth-century architects of Roman-bourgeois property would successfully popularize it, making it a central image of the modern liberal legal and political imaginary. But the real authors of the image of property as separate spheres of absolute sovereignty are the Roman pontifices and jurists, who, through their ritualism, inscribed it, however unrealistically, onto the landscape of property.

Rituals aside, in what way was the dominus’ sphere of “sovereignty” truly absolute? The dominus’ partial immunity from state power was achieved by exempting the dominus from real estate taxes. The dominus paid taxes, of course, but personal taxes, not property taxes. Dominium was absolute also in another way: it was protected through a special action, the rei vindicatio. The rei vindicatio was an assertion of absolute title. It was the only action of this type. The dispossessed dominus sued the possessor of the thing for its recovery, asserting “ex iure Quiritium meum esse aio” (I claim this under quiritary law).Footnote ¹³⁷ The ability to make this assertion of title in a rei vindicatio was the essential feature of dominium. It was so essential that the jurist Celsus, not without some circularity, tells us that:

But, practically, the rei vindicatio was hardly a convenient remedy. If the defendant denied the plaintiff’s ownership, the plaintiff had to prove that he had acquired dominium over the thing from its previous dominus, who in turn had acquired it from the dominus before him. Proving the entire chain of title back until the first owner’s original acquisition must have been, in many cases, almost impossible. The possessory interdicts available for the protection of possession were a much more efficient option. Ritualism suggesting full sovereignty over a thing also characterized the transfer of dominium. For dominium to be successfully transferred, the parties had to use one of two specific forms, mancipatio or in iure cessio. Mancipatio, we have already learned, consisted in the transferee’s declaration that “this is mine” accompanied by a ritual with bronze and scales in the presence of five Roman citizens as witnesses. In iure cessio, as we have seen, also involved a resolute claim of ownership in the context of a fictitious dispute in which the transferee acted as a dispossessed owner.Footnote ¹³⁹

Despite these many rituals and formalities asserting the “absolute” character of dominium, dominium was, obviously, not absolute. It was certainly not absolute in the sense that the dominus had full enjoyment of the thing, free of any restrictions. Obviously, equal, unrestricted enjoyment for all owners is hardly conceivable. For an owner to have unfettered entitlements against all persons with respect to a thing would mean that others have no legitimate interests, meriting legal protection, in their things or in their person that may conflict with the owner’s exercise of his rights. Even the crudest legal system restricts the owner’s enjoyment to allow the similar enjoyment of other owners and to protect, even if minimally, the public interest. As we will see shortly, Roman law imposed a set of limits on owners that look very much like those of a modern legal system. The transformation from a small agrarian economy to an imperial market economy, the empire’s growing demand for revenue, and the increasing pressure for redistributive policies left their mark on the law of property. By Justinian’s time, the public policy limitations on property did not differ too much from the modern regime of limited property entitlements we are familiar with. Dominium could also have been absolute in the sense that the dominus was immune from losing ownership without his consent. But this was not the case in Roman law, which knew various forms of prescription (usucapio, longi temporis praescriptio, praescriptio quadriginta annorum), by which an owner could lose title to a possessor in good faith by the simple lapse of time.Footnote ¹⁴⁰

Finally, as Barry Nicholas suggests, dominium may have been absolute in the sense that it was the best, ultimate right and the only of its kind because there was no other formal right to “own” a thing. In other words, one was either a dominus or a mere de facto possessor. However, dominium was not the only right of its kind. There were two modes of owning that were, in substance, equivalent to dominium. The first was the case of the “bonitary owner,” the transferee who had received a res mancipi by mere conveyance (traditio) rather than through the prescribed forms. The second was the person whose title was defective because he or she had received the thing, in good faith, from someone who was not the owner. Technically, both the “bonitary owner” and the good faith possessor had no better title than a mere possessor, entitled to the limited remedy of the interdicts. However, in the late Republic, the praetor granted both protection equivalent to that of an owner (dominus), de facto creating two other forms of ownership that came to be known as “praetorian ownership” and differed from dominium only in name.Footnote ¹⁴¹ After a certain praetor, called Publicius, made available to the “bonitary owner” the Actio Publiciana, which was essentially a vindicatio in which the necessary lapse of time for usucapio was fictitiously presumed, the “bonitary owner” was protected against everyone, including the owner ex iure Quiritium, and hence was, for nearly all practical purposes, in the position of an owner. But the Romans, Nicholas notes, could not bring themselves to call the “bonitary owner” dominus and opted for the convoluted formula “in bonis esse,” which means “to have something in one’s estate.”Footnote ¹⁴² The modern jurist can only be baffled. Someone who is not an owner merely because he failed to perform an old cumbersome ritual is put in the same exact position of an owner, through a legal fiction, and yet denied the name “owner” and designated instead with a clumsy circumlocution. Preserving the symbolic and conceptual uniqueness of dominium required as much.

To conclude, the answer to the question of whether dominium was truly absolute is both no and yes. Dominium was not absolute in its actual daily operation. It was not the ultimate and unique right, as the owner was not entitled to unfettered enjoyment and was not wholly immune from loss. But dominium was absolute in the formalities and rituals it involved. The most important moments in the life of dominium – from the tracing of the physical boundaries of the parcel, to the transfer of ownership, to its protection – were marked by resolute assertions of absolute ownership. The echoes of such assertions would resonate in the overblown rhetoric of the treatises and monographs of the nineteenth-century architects of Roman-bourgeois property.

Dominium also had a twin concept, possession (possessio). Today, possession has largely slipped out of legal scholars’ sight, and it is often dismissed by Anglo-American jurists as “the vaguest of all vague terms.”Footnote ¹⁴³ Yet, possession was a critical concept in Roman law and, in turn, in Romanist-bourgeois property law. Dominium and possession were twins because, in very simplified terms, dominium was a right while possession was a fact. In other words, the dominus was entitled to have a thing while the possessor actually had a thing, meaning that the possessor exercised actual physical control over a thing and intended to exercise such control, albeit without having a formal legal entitlement.Footnote ¹⁴⁴

However, the notion that, in Roman law, possession is merely a fact and not a right is misleading and needs clarification. There were two situations in which the fact of possession did have legal consequences, regardless of formal title. The first was usucapio, that is, the acquisition of dominium by continuous possession. Present certain requirements, the person who possessed a thing for a protracted period of time became its owner. In other words, usucapio operated similarly to a familiar institution of Anglo-American property law that puzzles every first-year law student, that is, adverse possession. The second case in which possession had legal consequences is less familiar but truly fascinating. While the possessor had no formal entitlement over the thing, she or he still received some degree of protection from the law through remedies known as “possessory interdicts.” Interdicts were orders issued by the praetor at the request of the claimant and aimed at quickly solving the controversy. Through these interdicts, the possessor could restrain others from interfering with his possession and recover possession from anyone who had dispossessed him. Different types of interdicts applied to specific types of property and circumstances. The interdictum utrubi applied to movables while the interdictum uti possidetis was available for immovables; the interdictum de vi covered cases of dispossession through physical force and the interdictum de vi armata cases of dispossession with the assistance of armed persons.Footnote ¹⁴⁵ To enjoy the protection of the interdicts, the possessor had to hold the thing in the manner of an owner, that is, to have both sufficient physical control of the thing, either directly or through another person, and the intention to possess like an owner.Footnote ¹⁴⁶

Why the law would protect a possessor who is acting with no right was a question that fascinated the nineteenth-century architects of modern Romanist property law. In 1803, the German jurist Friedrich Karl von Savigny published a treatise On Possession that proved immensely influential and sparked a vivacious, and at times violently polemical, debate among jurists throughout Europe.Footnote ¹⁴⁷ For Savigny, Roman possession was a matter of protecting the peace and order of society against force and lawlessness; for others, possession was about protecting the will of the possessor, actualized in the exercise of control over a thing; still others stressed that safeguarding possession was an indirect and imperfect means of affording effective and temporary protection without the need to prove chain of title.

The Limits of Roman Property

The nineteenth-century proponents of modern, Romanist dominium were eager to emphasize that the very idea of limits was foreign to the way the Romans thought about property. As Bonfante put it:

Yet, there are good reasons to question Bonfante’s statement. Limits to property entitlements may have been “few and dissimulated” in the early days of Rome but, as Roman society developed and the Roman economy grew, the number, if not the nature, of the regulatory limits to private property came to resemble modern property law. For religious as well as public health reasons, owners were prohibited from burying the dead on their land within the city walls. In case of land located outside the city, when a holder of the ius sepulchri was not also owner of the land, the latter had a duty to grant the former access to the land for the purposes of mourning and honoring the deceased. If flooding prevented the use of a public road, neighboring owners had a duty to grant a temporary public right of passage over their land. By late imperial times, limitations were imposed on owners’ rights over the subsurface of their land.Footnote ¹⁴⁹ A constitution of the Emperors Gratian, Valentinian, and Theodosius on quarries prescribed the division of the materials excavated between the owner, the person who excavates and the state:

Detailed regulations also limited the ability of owners to develop and build on their land, whether urban or rural. Conservation regulations, aimed at securing the stability and the aesthetic value of buildings in the city, prohibited owners from separating valuable materials, for example, marbles, from buildings for the purpose of selling them. And a senatorial decree during the reign of the Emperor Claudius prohibited altogether the sale of buildings when the object of the transaction was not the building itself but rather its materials. Further, legislation in the imperial period began to prescribe the minimum distance between buildings and to regulate the height of buildings and the width of party walls, reflecting a concern present in modern legal systems as well. Also, under the Emperors Leo and Zeno the regulation of relations among neighbors made a qualitative and quantitative leap, becoming a comprehensive, detailed body of rules. A constitution issued by Zeno for the city of Constantinople gives us a good sense of what the new regulatory regime looked like. In addition to the new limits as to height and distance, Zeno introduced easements (δουλεία, in the Greek text of the constitution) for light and air and for the view of the sea. Another important limitation, this one originating in the Twelve Tables, was the actio aquae pluviae arcendae, an action given against the owner of neighboring land for having constructed improvements that changed the natural flow of rainwater such that it damaged the plaintiff’s property. By Justinian’s time, juristic debate over the actio aquae pluviae arcendae had expanded to cover interferences that resulted in a diminution of the natural flow of a source of water.

It was a common contention among the nineteenth-century proponents of modern dominium not only that limits to property entitlements were “few and dissimulated” but also that the Roman state had no eminent domain power to take private property for public use. Proponents of this view pointed to the absence of any legislative or juristic text expressly recognizing the principle and to a variety of texts from which it can, supposedly, be inferred that no such principle existed. For example, in his speech De Lege Agraria, Cicero, criticizing the demagogic and unrealistic nature of proposal of the tribune P. Servius Tullius that the state buy privately owned parcels of land and redistribute them to the landless, notes:

What Cicero is describing is the classic holdout problem, traditionally offered as a justification for eminent domain: when private owners refuse to sell their land, which the government needs for a public purpose, the government takes it by eminent domain and pays just compensation. Cicero’s ironic interrogation – “what will happen if owners refuse to sell?” – is taken to suggest that the Roman state had no power to take property in the case of holdouts. Further evidence of the absence of eminent domain in Rome was found in an anecdote recounted by the historian Livy (64 or 59 bc–ad 12 or 17). During their term of office (179–174 bc), the censors M. Aemilius Lepidus and M. Fulvius Nobilior, Livy tells us, had planned to build an aqueduct, but the realization of the project was frustrated by the opposition of one single owner, M. Licinius Crassus, who refused to sell his land.Footnote ¹⁵² However, this story does not prove much; it may speak more to the power of one man than to the question of eminent domain. Crassus was a member of one of the prominent families in Rome, the gens Licinia, and it is possible that he would have succeeded in his opposition even if eminent domain existed.

Were the Roman law scholars who, in the nineteenth and twentieth centuries, argued for the existence of eminent domain in Roman law all socialistic ideologues anxious to redeem Roman property from the accusation of being excessively individualistic and anti-social, as Bonfante seemed to believe? Certainly not. While the absence of explicit statements and discussion of eminent domain is surprising, there is an abundance of texts from which such a general principle can be inferred. The Romans may not have felt the need to articulate such a general principle, or to develop a full-fledge theory of takings, but they were certainly familiar with the state taking property for public purposes. To begin with, citizens’ property could be taken for religious reasons if it interfered with the Roman people’s relation with the supernatural. We have seen how entangled the public and the sacred were, and indeed Roman property can hardly be understood if one forgets their interrelation. Cicero tells a story that gives a sense of how religion could justify a loss of property:

The augures were a college of high priests whose distinct prerogative was to ritualistically interpret the auspices, which meant observing certain natural phenomena to determine whether or not the gods approved of an important public action about to be launched. What is not clear from this passage is whether the mandate to destroy Claudius’ house gave rise to any claim for compensation.

But the most significant traces of eminent domain in the Roman law of property can be found in the literature on aqueducts. Our informant is Frontinus, not a jurist but a high-ranking public official who was appointed water commissioner for the city of Rome in ad 97. In his De Aquis Urbis Romae, a treatise about the administration of the Roman aqueducts, Frontinus explained the process of planning and building an aqueduct. In the section devoted to the tracing of the aqueduct’s path, Frontinus cited a decree of the senate imposing a variety of limits on private owners whose land was located on the aqueduct’s route. Further, Frontinus described, as an example of admirable balancing of private and public interests, what to our modern eyes seems a typical case of taking with payment of just compensation:

This brief glance at Roman property’s limits leaves us with a puzzle: if the regulations limiting owners’ entitlements were vast and detailed, and the texts alluding to the state’s power of eminent domain are numerous, what justifies the idea that Roman property suffered no limit? Roman law scholars who are proponents of this idea have one last arrow in their quiver: the absence of a general principle prohibiting aemulatio or “abuse of rights,” that is, the owner’s abusive exercise of her rights. The principle was included in several of the nineteenth- and twentieth-century civil codes and was the object of enthusiastic endorsements and heated polemics in the debates between the proponents of modern absolute dominium and their social critics. The former sought to show that a general principle of “abuse of rights” was nowhere to be found in the Roman sources, while the latter were eager to affirm the Roman character of the doctrine of “abuse of rights.” As with eminent domain, the proponents of “abuse of rights” could not point to a text explicitly articulating a general principle but found numerous sparse allusions to the doctrine. Most of these were allusions to a prohibition of a spiteful, malicious, or purposeless exercise of one’s right in connection to a specific resource, namely water. For example, a passage from Ulpian seems to suggest a consensus among jurists that acts done with the intent to harm, animus nocendi, and not for useful purposes, are not within the scope of the owner’s right:

Another allusion to malice, considered important by Roman law scholars because it does not concern waters but rather seems a more general statement of a principle is contained in a passage from Celsus, a jurist of the classical era. Celsus, while discussing the rights of the good faith possessor who makes improvements on the owner’s land, states:

Bonfante may have been right that the Roman jurists aptly disguised the limits to property. They made it easy for their modern counterparts to downplay the relevance of these limits and to focus instead on the absence of general principles regarding eminent domain or the prohibition of abuse of rights. Yet, these limits were neither few nor insignificant.

Public Land and Provincial Land: The Lesser Forms of Resource-Specific “Ownership”

Dominium was the supreme property form but, alongside dominium and the two equivalent forms of “praetorian ownership,” there were other forms of “ownership.” These other forms were not called “property” but were rather designated with formulas such as possidere, uti, frui, or habere possidere frui, all of which evoke the idea of a full enjoyment of the thing. And, in fact, the scope of the owner’s entitlements was so broad as to make these forms hardly distinguishable from “property” as we think of it today. The existence of these lesser “properties” is one of the most important features of Roman property that the nineteenth-century legend of dominium has virtually erased. One aspect of these lesser forms of “ownership” is worth noting. These “properties” were, to a large extent, resource-specific. They were tailored to the specific interests and needs implicated by different types of land that were key to the Roman economy or to Roman geo-political designs, such as public land and provincial land. By variously tweaking the entitlements comprised in these lesser forms of ownership, the Roman state was able to effectively pursue a variety of political and economic goals with regards to public and provincial lands.

Not all land in Italy was held as private property by domini. Large swathes of land were public land, known as the ager publicus populi Romani.Footnote ¹⁵⁷ Public land, which the Roman state acquired by confiscation from defeated enemies, was owned by the state but was made available to private citizens through a variety of forms that closely resembled ownership. Initially, not all private citizens had access to public land. The sources for the early Republic suggest that plebeians, the lower class, were originally excluded from public land.Footnote ¹⁵⁸ The tenure regime for public land was, initially, very informal. Public land, to the extent it was not used by the state, could be freely “occupied” (ager occupatorius) by users who wished to work it. A passage by Appian of Alexandria, a Roman historian of Greek origin who wrote in the second century ad, gives us a good sense of the informality characterizing this early “ownership” regime for public land:

In this early stage, occupiers’ use rights were highly unsecure. Users held the land without a legal title and the state could take away the land from the occupier whenever it was needed, without any duty to pay compensation. How much public land users could occupy and who was allowed to occupy are highly contentious questions. From the agrimensores, the professional surveyors who codified the Roman land system, we gather that, at an earlier time, “each man did not occupy as much land as he could then cultivate, but he sought to gain as much as he had the hope of cultivating.”Footnote ¹⁶⁰ It also appears that in 367 bc, the Lex Licinia de modo agrorum capped the amount of land that could be occupied by one individual to 500 iugera. This limit, however, seems to have remained dead letter.

From the third century bc on, these informal use rights evolved into more stable and secure forms of “ownership” in response to larger structural changes in the Roman economy as well as in the Roman class structure. Over the course of the third century, what some historians have called a Roman market economy started taking shape.Footnote ¹⁶¹ As Rome rapidly grew into a Mediterranean empire, the acquisition of new territory and the influx of money and slaves made larger commercial agricultural production possible. Commercial agriculture required a significant investment in the forms of equipment, slaves, wages for free laborers, livestock, seeds or plants and transportation, but insecurity of tenure under this informal form of occupation made commercial farmers reluctant to invest. Under the pressure of this expanding economy, the inflexibility of a property system that offered either absolute dominium or informal, unsecure use rights became apparent, and new types of “ownership” were developed whereby land remained the property of the state but users were granted an official title, broad entitlements, and security of tenure.

The earliest of these new “ownerships” concerned public lands known as ager quaestorius.Footnote ¹⁶² This form of tenure derived its name from the fact that it was “sold” in fifty-iugera blocks at an auction, presumably at market rates, by the quaestores, elected officials who supervised the treasury and the finances of the state. The sources use the word vendere, which means to sell, but what was “sold” was, technically, use rights, not the land itself, which remained the property of the state. However, the “owners” had broad entitlements that made their position similar to ownership. Use rights over the parcels gave the “owner” full control over the use and management of the land, were freely transferable, and could be bequeathed to heirs. In the “law in the books,” “owners” had a duty to pay rent, but there is reason to doubt that, in practice, collection actually happened. As to security of tenure, it appears from the sources that the protection of “owners” against dispossession by third parties was gradually reinforced by making available the possessory interdicts. What is not clear is whether the state could take back the land. Some scholars have argued that the parcels could be bought back by the state only at the initiative of the “owner.” In any event, regardless of what the status of the “owner” vis-à-vis the state was formally, it seems that, in practice, tenure was made relatively secure by the fact that supervising, administering, and, possibly, taking back public lands was costly and complicated for the state.Footnote ¹⁶³

Another form of “ownership” for public lands, called ager in trientalibus, was developed as a way to finance the Second Punic War.Footnote ¹⁶⁴ Private citizens gave their gold, silver, and jewelry to the treasury for the purpose of financing the war effort, with the agreement that they would be paid back in three money installments. However, in 200, when the time came for the second payment and no money was available, the senate decided to repay citizens by granting use of parcels of the ager publicus. As Livy recounts,

Rich farmers interested in commercial agriculture found this compromise very attractive. The fact that the land was situated within a fifty-mile radius from Rome meant that it was in high demand among those wishing to produce for the market in the city. As in the case of the ager quaestorius, “owners” of the ager in trientalibus had the entitlements of an owner and virtually complete security of tenure. Land could only be taken back by the state at the initiative of the “owner” who preferred cash, and there is no evidence that anyone exchanged land for money. The ager in trientalibus was mentioned as a category of land in the lex agraria of 111 bc and, by this time, its “owners” had held their parcels undisturbed for almost ninety years.Footnote ¹⁶⁶

Yet another special “ownership” form was developed for public lands known as ager censorius because these were leased out by the censor, a high Roman magistrate, with an arrangement that resembled a long lease rather than ownership.Footnote ¹⁶⁷ However, in the sources, the recipients of such lands were at times called redemptores (buyers), which suggests that, in practice, their entitlements were not too different from those of an owner.Footnote ¹⁶⁸ Although the prevailing opinion was that recipients of ager censorius were lessees rather than owners, the potential perpetuity of the lease and the ability to transfer the land to one’s heirs made them de facto owners.

Public lands were a critical resource in the economic, social, and political life of the Roman Republic, and the Roman state made available these lesser forms of “ownership” for a number of reasons. Chief among these was the policy of supporting the material prosperity and, in turn, the demographic growth of Roman farmers as well as of Latin and Italian farmers, who were not Roman citizens but who provided manpower critical to Rome’s wars of expansion. These various forms of private “ownership” of public lands also made it possible for the state to finance its war efforts and to promote economic initiatives, such as the development of commercial farming.

Public land was not the only type of land for which the Roman state made available a smaller, customized form of “ownership.” Provincial land (solum provinciale) was exempt from Roman property rules and from dominium. However, because the majority of the inhabitants of the empire lived in the provinces, a type of “provincial ownership” had to be developed. As with the various “ownership” forms available for public lands, provincial ownership was not technically ownershi, but rather was described, in the classical sources, as possessio or usufructus. Roman jurists always treated provincial ownership as an exception. They emphasized the rules and doctrines that did not apply to it: ownership of provincial land could not be transferred through mancipatio or in iure cessio and it was exempted from acquisition by usucapio.Footnote ¹⁶⁹ As a result, we know very little about the positive regime of provincial ownership, beyond the fact that it was subject to a yearly tax and that it was protected through a vindicatio, not a rei vindicatio, which was reserved to dominium, but a vindicatio based on aequitas, that is, on considerations of equity. In recent years, historians of Rome have started exploring in greater depth provincial ownership but, because of the scant evidence, have found it difficult to develop a detailed schema of provincial landownership that is consistent across provinces. The picture these scholars trace is one of significant legal pluralism. As a general principle, Rome respected the local legal systems in the provinces and only intervened to clarify or supplement local property regimes.Footnote ¹⁷⁰ However, on the ground, a spectrum of different ownership bundles developed out of the interaction of the competing claims of different actors: local elites, local peasants, Romans in the provinces, and Rome’s political elite.Footnote ¹⁷¹

Egypt is the one province for which we have good evidence based on a rich stream of source material and hence has been the focus of this recent scholarship.Footnote ¹⁷² In Egypt, a variety of land regimes coexisted. To begin with, a special form of ownership was made available for provincial state land. In Egypt, Rome inherited a long tradition of state ownership of land and, under Roman rule, all land for which taxes had previously been paid directly to the Ptolemaic monarchs continued to be treated as public land. This was the case for formerly royal land and portions of the lands formerly belonging to temples. This public land was assigned to state farmers whose entitlements comprised use rights, a positive duty to cultivate the land, a duty to pay a yearly tax, the right to transfer or “sublet” their cultivation rights, and the right to pass their rights and duties to their successors.Footnote ¹⁷³ Other land was treated as “private,” including land already recognized as private under the Ptolemies, but also some temple land, the land of the Ptolemaic military settlers, and former state lands that had been privately cultivated for generations. “Owners” of this private land had a different set of ownership entitlements and duties. They had a duty to pay a tax, called the artabeia, which was, on average, only a third of the rate of the tax applied to public land. Also, because this land was originally public, transfer of this land could not take the form of a true sale but rather required special modalities. Yet another type of “ownership” characterized temple estates, where priesthoods administered the many economic functions – including farming, manufacturing, and grain distribution – that had made each estate “a system unto itself” since pharaonic times.Footnote ¹⁷⁴ These holdings could be inherited by successors, leased and pledged as security.

By recognizing these types of “ownership” with different scope and shape, the Roman state sought to achieve both economic and fiscal goals with regard to the province of Egypt.Footnote ¹⁷⁵ The Roman government promoted the free alienability of privately owned land and protected security of tenure in order to unleash private investment in large productive estates. With sufficient financial outlay, highly remunerative, but capital-intensive, forms of agriculture, such as viticulture, could be further developed. In fact, there is much evidence that, during the course of Roman rule, Egypt saw the creation of large commercial estates that had not existed before.Footnote ¹⁷⁶ Another reason for the state to retain direct control of land in Egypt was the desire to manage the productive targets for the province’s economy, deciding what crops would be planted and, for example, subsidizing the production of wheat by leasing wheat land on favorable terms.Footnote ¹⁷⁷ From a fiscal perspective, through these forms of private tenure, the Roman state could do much to ensure that revenues, which largely depended on the rents extracted from the production of small farmers on state lands, were stable over the long term. This stability was crucial not only to the governance of Egypt but to the administration of the entire empire.

The Iura in Re Aliena and the Disaggregation of Property Entitlements

The supposed “indivisibility” of Roman property is a recurrent theme in Roman law and comparative law scholarship. As we have seen, the classical jurists, we are told, “had an extremely concentrated notion of ownership”Footnote ¹⁷⁸ and were generally uncomfortable with the idea that the entitlements comprised in the owner’s “absolute sovereignty” could be split up between different subjects. Hence, Roman law knew no doctrine of estates, whereby ownership of land could be divided in time between a present owner and a future owner to whom the property would, at some point, “revert.”Footnote ¹⁷⁹ This unitary character of Roman property also precluded anything in the nature of a trust, whereby property interests could be split between a trustee, who would have the power to manage the property, and a beneficiary, who would have the right to enjoy the profits.Footnote ¹⁸⁰

This supposed indivisibility is hard to fathom for contemporary property lawyers. Our property culture is permeated by the idea that property is a legal device to create, capture, and retain certain kinds of economic value, of utility. Divisibility, that is, the ability to split up ownership entitlements, is critical to creating and capturing economic value. In modern property systems, owners have the ability to create and maximize property value by disaggregating property entitlements.Footnote ¹⁸¹ How could it have been that Roman property law, which was the central institution through which the Roman elite extracted value, did not permit divisibility?

Indivisibility may well be yet another aspect of the “legend” of Roman property law. In reality, Roman law offered owners a number of ways to parcel out single entitlements through a sizable menu of “rights over a thing owned by another” (iura in re aliena). It is worth taking a closer look at the list of available rights not only to get a sense of the type of social and economic goals owners were able to achieve by creating and parceling out these lesser real rights but also to gain insight into the conceptual dilemmas that these rights posed to the jurists. Servitudes were the oldest of the iura in re aliena, and the only ones that can safely be said to exist already in classical law. A servitude is the right of a person other than the owner to make a certain use of that owner’s land, for example by using a road for ingress and egress or drawing water for irrigation purposes; such a person was usually the owner of a neighboring piece of land. Max Kaser argued that servitudes were sharply distinguished from ownership and classified as “rights in a thing owned by another” only at a later time and were initially conceived as functionally divided ownership. The person entitled to the servitude was considered as the owner of that part of the land over which the road or the watercourse passed. But this ownership was confined to the function or purpose of the servitude, and the owner of the servient estate was owner for all other intents and purposes.Footnote ¹⁸²

As compared with contemporary servitudes, the Roman law of servitudes limited what the parties could achieve in two important ways. First, in classical Roman law, the concept of servitudes likely included only servitudes that benefited land and not servitudes that benefited a person. Personal servitudes were a later development that stirred up the passions of nineteenth-century Romanists. With his usual sharpness, Pietro Bonfante noted that so called personal servitudes have little in common with “praedial” servitudes and that the assimilation of the two categories was a product of the conceptual confusion and ignorance of the “Byzantine” jurists.Footnote ¹⁸³ A second significant limitation came from the rule that a servitude could not impose on the owner of the servient estate an affirmative duty to do something, with the sole exception of support rights, that is, the right to have one’s structure or building supported by the neighbor’s land. As this passage from book 17 of Ulpian’s Libri ad Edictum shows, even this narrow exception stirred disagreement among Roman jurists:

Servitudes were subdivided by Roman jurists into two subcategories: rustic and urban. Because the former were critical to the Roman agrarian economy, they were treated as res mancipi, created through the ritual of mancipatio to ensure certainty and publicity about their existence and scope. The rustic servitudes included three types of right of way: iter (passage foot or horse), actus (passage of herds and agricultural equipment), and via (the use of a traced path with the predetermined width of eight feet in straight sections and sixteen feet at turns). Other rustic servitudes included a variety of iura aquarium – rights to draw water for irrigation or drinking, or rights to expel run-off surface waters – the right to dig sand, to take clay, to burn lime, to graze cattle and to gather acorns. Urban servitudes included three big clusters of rights, namely rights to expel rainwater and sewage, rights relating to the construction and support of buildings and rights to view, light, and air.

This quick glance at the list of servitudes raises the question of whether the list was a closed one. In other words, could the parties create their own custom-made servitudes, or were they limited to the standard servitudes we have just examined? This question is an important one. A system with a closed menu of forms guarantees predictability and low information costs. Individuals who wish to acquire property entitlements know what options are available and what they can expect.Footnote ¹⁸⁵ By contrast, free customization resonates with liberal values and promotes individuals’ ability of self-authorship by allowing them to freely acquire and exchange resources and entitlements and to enlist one another in the pursuit of private goals and purposes.Footnote ¹⁸⁶ As we will see over the course of our journey through modern property, from the nineteenth century to the present, property lawyers on both sides of this debate have drawn arguments from Roman law, emphasizing either the supposed “closure” or the alleged flexibility of its property system.Footnote ¹⁸⁷ However, once again, the truth is that we know very little about the possibility to customize servitudes in Roman law. The classical sources are silent on this point, but sparse references to the parties’ ability to create new servitudes appear in Justinian’s legislation.Footnote ¹⁸⁸ Further, some Romanists suggest that even in classical times, the parties had the ability to modify and reshape by agreement the servitudes recognized by the law.Footnote ¹⁸⁹

Another way of dividing up property entitlements was for the owner to grant an usufruct, which the Roman jurist Paulus defined as “the right to use and enjoy the fruits of another’s property without impairing the substance of the thing.”Footnote ¹⁹⁰ How to properly conceptualize the usufruct was a question that generations of nineteenth-century Roman law scholars debated with passion. Once the post-classical jurists, in all likelihood, invented the concept of personal servitudes, usufruct seemed to fit easily into the new category. However, Bonfante, who, as we have seen, had doubts about the very category of personal servitudes and explained its creation as stemming from the lack of a solid conceptual grasp on the part of the Byzantine jurists, was not convinced. The classical jurists, with their conceptual acumen, Bonfante argued, had understood that servitude and usufruct are radically different concepts. A servitude is a mere limitation on the dominium of the servient owner, who has to tolerate the dominant owner’s exercise of a narrow entitlement, say passage by foot or horse, but is not deprived of the enjoyment of the land. By contrast, the effect of the usufruct is to transfer to the holder of the usufruct a portion of the owner’s dominium, specifically, the right to use the land and to appropriate its revenue, for a period of time that cannot exceed the latter’s life. In other words, the usufruct does not merely limit dominium but actually carves out a portion of dominium.Footnote ¹⁹¹ And yet even defining the usufruct as a portion of dominium seemed hardly satisfactory, because the transferred portion came with one important limitation: the holder of the usufruct could not allow the thing to deteriorate or diminish or change its current productive use. Ulpian provides us with a wealth of examples:

In other words, Bonfante explained, the holder of the usufruct had a right over the thing in its current form or quality (species rei), not a right over the “substance of the thing” (substantia rei), which remained with the owner. As the non-agricultural sectors of the Roman economy expanded and the state became more involved in the extraction of raw materials and minerals, the limits on the ability of the usufructuary to change the use of the land were further relaxed. By Justinian’s time, the prerogatives of the usufructuary were broadened, allowing for change in the use of the land. The subsequent passage in the Digest from Ulpian was likely interpolated to provide an explicit rule for mining.Footnote ¹⁹³ The rule omitted any reference to the substance of the property and instead allowed the usufructuary to engage in the most productive use of the land:

This expansion of the entitlements of the holder of the usufruct made the usufruct look even more like dominium and less like a personal servitude. Throughout the nineteenth century, scholars of Roman law and private law continued to question the fuzzy line between dominium and usufruct and to see usufruct as a lesser form of “ownership” that existed uneasily alongside dominium.

Similar to the usufruct but more limited in scope was the right of usus, which entitled its holder to use another’s property but not to appropriate the revenue or fruits of the thing. If no use of the thing was possible other than taking the fruits, for example, in the case of an orchard, the right holder could use the fruits for himself but not sell them. A particular variant of the right of usus was the right of habitatio, that is, the right to use another’s house for dwelling. Two other “rights over a thing owned by another” that were fully developed only at a later time were the right of superficies and the right of emphyteusis. The former was developed in the context of increasing urban intensification and land oligopoly in the city of Rome, as land came to be owned by a relatively small number of corporations and private owners. The right of superficies was a way of slicing the land. It made it possible for the developer who built on land owned by another to acquire a perpetual and hereditary right to use the surface of the land by paying an annual “rent” called solarium.Footnote ¹⁹⁵

Emphyteusis was yet another hybrid form that presented classificatory problems similar to the ones posed by the right of usufruct. Land was transferred to an emphyteutical possessor over very long periods of time, or even in perpetuity, in exchange for payment of an annual rent.Footnote ¹⁹⁶ Emphyteusis shared characteristics with both a contract of lease, which transferred possession for a term, and a sale, which transferred full ownership. A constitution of the Emperor Zeno clarified that emphyteusis could not be assimilated to either a sale or a lease but rather was a distinct form giving rise to a set of entitlements that was unique in kind:

Zeno’s constitution settled the matter only temporarily, and the question of the real nature and appropriate classification of emphyteusis would periodically resurface. While the duty to pay rent put the emphytecarius in the position of a lessee, the emphytecarius had full and secure use entitlements analogous to those of an owner and could sell, bequeath, or otherwise alienate his or her rights to others. The emphytecarius could only lose his or her rights for non-payment of the rent for three consecutive years.

To conclude, while these lesser real rights, or “rights over a thing owned by another,” differed in shape and scope, each speaks to the pluralism and variability of Roman property. By appropriately combining these smaller rights, owners were able to achieve a wide range of social and economic outcomes. They could multiply the number of subjects who benefited from the resource, design with relative precision the scope of each user’s entitlements, disaggregate control and management of the resource at any particular time, and slice up the resource into different assets. Further, over time, as the rules regulating these lesser rights were relaxed, owners acquired greater ability to customize forms and outcomes. Finally, the broadest of these rights, usufruct and emphyteusis, granted the right-holder entitlements comparable to those of an owner. They were, effectively, more limited forms of “ownership” that added to the already striking variety of modes of owning that we have explored in the previous sections.

Property Law and Redistribution: The Lex Agraria

The Lex Sempronia agraria was a law proposed by the tribune Tiberius Gracchus in 133 bc, and re-proposed by his brother Gaius Gracchus in 123, that sought to limit the maximum amount of public land individuals could possess and to redistribute the land possessed in excess of this limit to the indigent.Footnote ¹⁹⁸ Although hastily dismissed by the proponents of modern dominium as a radical and arbitrary political act, a redistributive lex agraria, as shown by the recurring fascination and legislative energy it has engendered, is nevertheless a critical element of the conceptual vocabulary of Roman property, one that figured prominently in the discourse of eighteenth-century revolutionaries and of the nineteenth-century social reformers. In fact, among the Roman property concepts, the lex agraria may be second only to dominium in the lasting and vivid impression it has made on generations of historians, politicians, and activists. The lex agraria has provided modern reformers with a vocabulary to discuss questions of land redistribution, a blueprint for reform, and a set of powerful rhetorical arguments about the relationship between property, equality, and freedom.

While the lex agraria became a centerpiece of modern reformers’ discourse about distributive justice, the social and economic situation that led to its proposal, the aims of the Gracchi, and the actual content of the law remain the object of controversy among historians of Rome. Throughout the eighteenth and nineteenth centuries, there was sustained debate as to whether the lex agraria applied to private as well as to public land, but scholars now agree that it only concerned the ager publicus. In the conventional narrative, told by Appian and Plutarch and by generations of modern historians, the lex agraria was proposed to address the accumulation of public land by the wealthy and the consequent proletarianization of the free Roman citizen.Footnote ¹⁹⁹ Plutarch describes this process of proletarianization and its consequences with great clarity:

The story Plutarch tells is one of greedy accumulation of land by the rich and related proletarianization of the small farmer. More recently, historians have revised elements of this narrative, contesting the idea of a demographic decline and downplaying the causal relevance of the accumulation of public lands in the hands of the elite.Footnote ²⁰¹ In this revised account, the Gracchian lex agraria was passed to tamp down the growing demand for the limited stock of land caused by technological and economic developments in central Italy.

Paradoxically, for a law that has fueled so much political controversy over the centuries, what we know about the actual content of the Lex Sempronia agraria is largely speculative and is in fact inferred from a subsequent agrarian law, passed in 111 bc. One thing about the law seems clear: the Lex Sempronia agraria marked a fundamental turning point. Limits on the amount of public land individuals could occupy (passed through leges de modo agrorum) were not new, but violators had seldom been fined and their excess possessions were left undisturbed. The Lex Sempronia agraria, for the first time, took back the land owned in excess of the limit. During the second century bc, “owners” of the ager publicus had treated their land as private property, even though they had no formal guarantees of secure tenure: they had sold it, mortgaged it and bequeathed it. Naturally, dispossession caused great mayhem. Appian tells us that:

Whether dispossessed “owners” received compensation is unclear. It seems that, initially, Tiberius Gracchus intended to buy back the excess land from the owners but that the final version of the bill did not provide for any compensation. Tiberius apparently argued that, for the rich, acquiring secure tenure over their remaining, legitimate holdings of public land was itself just compensation for the loss of the excess land.

The actual limit set by the Lex Sempronia agraria is a debated issue. From Appian, we learn that the limit was 500 iugera plus an additional amount for the main occupant’s children;Footnote ²⁰³ from Livy, we hear that the limit was 1,000 iugera.Footnote ²⁰⁴ Most Roman law scholars side with Appian, reasoning that, in order to make its passage more likely, Tiberius would have wanted to present his law as a mere repetition of earlier laws – as reinforcing the mos maiorum – rather than as a radical innovation, and that he would hardly have been able to do so if the limit proposed had been twice that of the earlier legislation. It is also unclear how much of the excess land repossessed by the state each of the indigent beneficiaries received. Based on a passage from the lex agraria of 111, it is often suggested that the lots amounted to thirty iugera.Footnote ²⁰⁵ However, some scholars cast doubt on this figure, noting that the amount seems large compared to the amount granted to colonists in earlier colonization programs.Footnote ²⁰⁶ Also, if each beneficiary were assigned thirty iugera, only a relatively small number of people would have benefited from the Lex Sempronia agraria.Footnote ²⁰⁷ Another aspect of the Lex Sempronia agraria that has not been fully elucidated is the nature and scope of the beneficiaries’ entitlements. Historians agree that the recipients of the land had a duty to pay a rent (vectigal) and no right to sell the land. These two basic facts raise the question of whether the land became the private property of the beneficiaries or remained in the public domain.

An alternative possibility is that the land given to beneficiaries by the Lex Sempronia agraria was in fact private property but of a more limited nature. This may be yet further evidence of the fact that, in Roman property law, alongside full dominium, there was a variety of lesser forms of ownership. In a somewhat anachronistic fashion, one could say that Tiberius appears to have shaped the entitlements of this limited form of private ownership to maximize the likelihood that the law would achieve the goal of improving the economic condition of the recipients in the long term. Owners’ inability to sell was designed to prevent them from transferring their smaller holdings to large landowners who would have been eager to purchase them; such sales had played a central role in swelling the landless urban population, which the lex Sempronia agraria was meant to aid by returning such citizens to the land. The possible solution that Tiberius had crafted with this limited ownership form designed to protect the long-term security of the beneficiaries, if this was what he devised, was not to last, as the lex agraria of 111 made the land distributed under the Lex Sempronia agraria fully private.Footnote ²⁰⁸

The Gracchan reform program involved more than the distribution of public lands. One of the reasons that the Gracchi have been hailed as visionaries by modern progressive reformers is that the lex agraria and the related measures could be seen, with contemporary eyes, as establishing a comprehensive set of welfare entitlements. The limited private property entitlements over the parcels of land were complemented by other entitlements. We know from Plutarch that Tiberius intended to grant the new owners a sum of money to stock their small farms. Further, when Caius Gracchus re-proposed the lex agraria ten years after the death of his brother, he also advocated that the government supply grain at half the market price to indigent Roman citizens who applied. This marked a radical departure in policy.Footnote ²⁰⁹ The Roman government had previously resorted to similar measures in years of crop failure and famine, but the grain law was the first permanent relief measure. The elites saw the passage of the law as a populist measure, an attempt to win the support of the proletariat. But they still took advantage of it. Cicero tells the story of Lucius Calpurnius Piso Frugi, a wealthy consul who opposed the grain law but was then found in line to get his share of grain.Footnote ²¹⁰ The story has also been taken to suggest that the law as actually enacted did not include a need-test for beneficiaries. Practically, the grain law failed: it was financially unsustainable and it defeated the very purposes of the agrarian law, namely benefiting small farmers in Italy, by putting enslaved labor in the grain-rich provinces in competition with the free laborers of Italy.

The lex agraria acquired a legendary status among modern reformers and revolutionaries and was hailed as the first comprehensive social reform designed to endow citizens with a minimum of fundamental material resources. If we pierce the veil of legend that envelopes the Gracchan reforms, the lex agraria appears as a much more complex effort to strengthen the Roman Republic – socially, politically, and economically – by expanding democratic participation, managing social tensions, and promoting a sense of belonging, and bolstering military power.

Conclusion

As this brief excursion through the intricacies of Roman property law suggests, Roman property was not “absolute.” Dominium was robust and its claim to be the supreme form of ownership was staged daily through elaborate rituals. But alongside dominium, Roman property law offered a rich set of forms and doctrines that made it pluralistic and variable. The unique feature of Roman property – and its enduring legacy to modern property – is not its alleged absolute character but its “law of things,” its focus on resources. Rather than offering abstract definitions of dominium, the Roman jurists pragmatically looked at the special characteristics of different resources and relied on external frameworks, drawn from philosophy or religion, to examine the distinct values and interests different resources implicated. The result was a plurality of variable and limited resource-specific property forms that facilitated and balanced different, and often conflicting, normative considerations.

Further, despite the Roman jurists’ promise of an autonomous property science, Roman property law was actually at the center of a complex governance project. The conceptual vocabulary of property developed by the jurists was key to sustaining a highly unequal class structure, in which a small social and economic elite derived its wealth and power from the exploitation of a widening social base of Roman, Italian, and provincial small landholders, urban poor, and slaves.Footnote ²¹¹ As the various iterations of the lex agraria show, when the social tensions generated by this highly unequal class structure threatened to destroy Rome’s political institutions, adjustments in the legal regime of property mitigated class conflict by achieving some realignment of Rome’s political structure with its social base.

In addition to supporting an unequal class structure, Roman property law also supported the needs of a complex Mediterranean economy. Property forms and doctrines buttressed the economic policies of the Roman state. As we have seen, by making available several forms of private “ownership” for public land, the government promoted the expansion of large-scale, capital-intensive commercial agriculture on estates that produced for the Mediterranean market. While important for a full understanding of Roman property, this renewed focus on the operation of market forces and on trade in the Roman world should not obfuscate the fundamental differences between the Roman economy and the capitalist mode of production that the nineteenth-century jurists were seeking to support through Romanist-bourgeois property law. The Roman conceptual vocabulary of property supported an economic order of owners of land and slaves and of merchants, not of producers. Rome knew commerce and markets but not capitalism. The Roman economy was an overwhelmingly slave-based economy in which private property seldom met free, wage labor. As Aldo Schiavone puts it, this contact between property and wage labor is the authentic crucible of modernity, one that never materialized in Rome.Footnote ²¹²

Finally, the Roman conceptual vocabulary of property needs to be understood against the background of the Roman project of imperial expansion. Rome’s control of territory was only partly due to its military might. It owed as much to a complex set of policies, including a land policy, to govern the territories over which it had extended its hegemony.Footnote ²¹³ Imperial stability, Clifford Ando has shown, required consensus regarding Rome’s right to govern and to establish a normative political culture. Land policy and property law were part of the official discourse of Rome’s imperial government, a discourse that turned on Rome’s ability to elicit its subjects’ obedience and their self-justifications for participating in their own subjugation.Footnote ²¹⁴ Rome’s pluralistic system of provincial property allowed the Roman state to balance two conflicting goals: building consensus among its provincial subjects by leaving largely intact preexisting local property systems, while also facilitating the plainly imperialistic exploitation of the provinces. War and territorial expansion had become Rome’s largest form of investment, generating colossal economic returns for the Roman governing class.