Can the scholarly debate on ancient colonization benefit from a specifically settler-colonial perspective? And could settler colonial studies, even though developed for the early modern and modern world, also gain insights from the ancient classical world? The essays in this volume explore the value of examining settler colonialism as a structural phenomenon and, together, make a compelling case for adopting a long-term perspective to better understand its dynamics. While I fully share this approach— unsurprisingly, as a co-founder of the broader research project SECOPS, from which this book emerged—I would like to use this response to offer some additional reflections on the unique contributions that an ancient historical and archaeological perspective can bring to this discourse. In this brief reflection, I aim to highlight the intrinsic entanglement and deep interconnectedness of scholarly debates on ancient and modern (settler) colonial contexts. It is difficult to overstate this influence, which, intriguingly, constitutes demonstrable two-way traffic. A brief discussion of some key ancient historical models for understanding ancient—and more specifically Roman—colonization will reveal striking analogies with modern European expansion and colonialism. These connections are so profound that we cannot fully understand one without reference to the other. The ways in which these ideas travelled across time and space deserve particular attention, as several contributions in this volume demonstrate. Perhaps unsurprisingly for an archaeologist, I will argue that, in addition to texts, the material world plays a crucial role.
The feedback loop
Fig. 1. Conventional understandings of Roman colonization (After G. Moscara in Settis 1984, 150 fig. 129).
What we need to be aware of is the potential feedback loop in which assumptions about the ancient past shape the construction of the present world—a present that, in turn, informs historical and archaeological interpretation. This dynamic creates a loop and, indeed, risks circular reasoning. Ancient colonization, and Roman colonization in particular, provides an excellent case study of this mechanism. It spans an exceptionally long period, and in the Western world we possess relatively abundant information about it—precisely because it has never lost its appeal as a classical example. Until relatively recently, there was broad scholarly consensus among ancient historians and archaeologists about the essential character of Roman colonization (e.g., Salmon 1969; Brown 1980). Drawing on both ancient written sources and material remains, these scholars portrayed Roman colonization as the establishment of new, Roman-looking towns with neatly organized hinterlands designed for agricultural production by settler-farmers. These new city-states, functioning not only as military strongholds in newly conquered territories but also as key vehicles for the spread of Roman culture, were seen as central to the formation and consolidation of the Roman Empire—a
strategy combining both carrot and stick (See Fig. 1.).
Over the past two decades, cracks have begun to appear in the conventional model of Roman colonization, and they are now significant enough to warrant closer investigation. Archaeologists have identified
puzzling chronological gaps between the official foundation dates of colonies and several structural features typically associated with them. Historians have noted that a large portion of the textual sources postdate the actual foundation of the colonies and may reflect a mindset quite different from that of the early phase of Roman expansion (e.g. Bispham 2006). It has thus become clear that a new approach to reconstructing
ancient Roman colonization is required. To deliberately pave the way for such an approach, it is crucial to make explicit the formation of the earlier consensus model and its ideological genealogy.
Teasing out the historiography of certain ideas or concepts about the ancient world is often an all-consuming enterprise. It is certainly more efficient—and, admittedly, more appealing—to propose fashionable new perspectives on old data, or better still (though less common), to present fresh field data. In the case of ancient colonization, and Roman colonization in particular, the weight of historical associations makes it, in my view, imperative to situate ourselves within the long loop of historical thought, which is saturated with strong ideas and ideologies about colonization, colonialism, and imperialism. The writing of early Roman colonization, as noted, began only after the fact, within the socio-political context of a vast and already consolidated Roman Empire at the end of the first millennium BCE. It became part of a long intellectual tradition that, in various ways—sometimes explicitly, sometimes implicitly—sought to explain the remarkable success of Roman imperialism. This tradition runs visibly from the works of Velleius Paterculus and Livy through Machiavelli and Lipsius to the studies of Beloch and Pais, and into more recent scholarship (Pelgrom & Stek 2014). Yet the highly complex and varied ways in which this process unfolded across different scholarly debates have been massively compressed and simplified—and, surprisingly, have never been given full attention, not even in the most recent studies of Roman colonization (Stek 2017).
This is evidently not the place to embark on such an enquiry in full, but we may begin by considering two key themes that pervade both classical and modern scholarship. These themes run through scholarly debates not only about Roman colonization but also, more broadly, about societal progress and political science; they frequently appear in the so-called ‘big history’ genre as well as in popular culture. They are therefore apt indicators of the transhistorical process of referencing Roman colonial values—a process that, as I will argue, can generate a dangerous feedback loop. This issue needs to be addressed not only in archaeological and ancient historical debates but also in discussions of settler colonialism in the early modern and modern periods, where clear echoes of these two Roman colonial role models can often be found.
The Lipsius model for ancient colonization
The first model, which I refer to as the Lipsius model, focuses on culture. In the 16th century, Justus Lipsius articulated a clear vision of both the character and the specific role of colonies within the broader imperial strategy. According to this model, colonies functioned primarily as instruments of civilization. Roman towns were conceived as miniature copies of Rome itself, complete with forums, central Capitolium temples, and so-called comitia and curia serving as the principal political institutions of the colonial community.
“The colonies resembled Rome, their mother city, in all respects. There came to be fora, capitols, temples
and senate buildings to resemble Rome […] For when colonies were set up like this, their cities were
populated, cultivated and even polished in the best possible way through the arrival of new inhabitants
and the importation of the arts. […] The utmost beautiful region of Europe owes its entire culture to [the
Romans].”
(Lipsius 1598, Admiranda sive de magnitudine Romana, Chapter VI, de coloniis).
Lipsius’ model in the cultural sense has unleashed a massive number of ‘romanization’ studies. These flourished in the late 19th and early 20th centuries, often reflecting overtly colonialist perspectives. Roman expansion was presented as the automatic, straightforward, and inherently positive spread of Roman material and intangible culture, eagerly embraced by conquered regions and polities. The last half century has been spent criticizing such conventional views. The main takeaway is that (Roman) material culture is open to multiple interpretations, and the experiences of conquered groups could vary significantly, allowing for deviant or even ‘resistant’ responses too (discussion in Stek 2013). Interestingly, however, the Lipsian model is hardly found in the actual ancient sources. Retrospectively, scholarship’s fascination with the link between culture and power likely owes as much to the context of a decolonizing world as to contemporary anxieties about culture, imperialism, and totalitarianism. Wishful thinking can sometimes be suspected in the works of the 1990s and the early 21th century. Moreover, it is important to distinguish between Rome’s possible cultural-imperial intentions (nowadays almost completely downplayed) and both its actual success rate and the ways in which it unfolded locally. Surely, there is a notable difference over time, where the emperorship of Augustus starting in the late 1st century BC has often been taken as convenient breaking point, as Daniels in this volume also discusses.
In any case, also for the earlier, Republican period, the Lipsian model has typically been projected onto early Roman towns such as the colonial settlement of Cosa in South Etruria (founded 273 BCE), where, in the 1950s, the American excavator readily identified what he interpreted as a true small-scale copy of Rome (Brown 1980).
The Cincinnatus model for ancient colonization
The second model, which I call the Cincinnatus model, focuses on morale. In contrast to the culturally oriented Lipsius model, it emphasizes the agricultural hinterland. The Cincinnatus model highlights the socio-economic foundation of the colonial community: agriculture and autarky. It is named after the general from Rome’s early mythical-historical period. As the ideal veteran, Cincinnatus was famously ploughing his own plot of land when the Senate, panicked by an incursion of the neighbouring Aequi, called upon him. He exchanged the plough for the sword, but, after defeating the enemy, returned to his plough only a fortnight later. Civic and military duty and austerity emerge here as key Roman virtues, but equality is also part of the ideal. The emphasis on the man’s own plot of land is crucial: agriculture was not only one of the foundational markers of civilization but also one of the most tangible ways of claiming land.
Fig. 2. Illustration in Willem Goes’ book on land division and colonization (Goes 1674).
It is also, morally, among the fairest ways of claiming land—but only in one specific manifestation: when the portion of land is conceived as the domain of a single farmer-colonist and his nuclear family (as opposed to, for example, slave-labour-based large estates owned by wealthy, urbanbased landowners). The cultivated plot is bounded by the practical need to sustain a livelihood and equally limited by the amount of labour a single family unit can provide. In many ancient colonization narratives, including mythological ones, there is a strong emphasis on equality among members of the civic community, often expressed in the form of equally sized plots for housing and agriculture. By an interesting twist in the historiography—one that warrants further discussion—the sum of these equal portions of land, and especially the very process of laying out such grids in conquered territories, has been linked to a more imperialist notion of imposing order on wild nature and perceived primitiveness.
A special role was played by a distinct genre of ancient Roman literature: the texts of the socalled Agrimensores or “land surveyors” (Campbell 2000). Composed of various works by different authors, mostly writing in the (late) imperial period, this corpus discusses the many practical issues land surveyors encountered when measuring land for taxation and documentation purposes. Its emphasis on territorial claims and property is striking, and both the texts and their accompanying technical drawings (such as Fig. 2.) were instrumental in shaping an idealized image of the colonial hinterland as an orderly and majestic territory, standing in sharp
contrast to the wild, untamed nature and populations surrounding it.
Historical loopings: the present resonating in the past
These two models clearly stem from different ideological backgrounds and convey distinct meanings. Nevertheless, they have intersected and, over time, been entirely conflated in the creation of the standard idealized image of the Roman colony—as presented in foundational 20th-century works such as Roman Colonization under the Republic by E.T. Salmon and Cosa: The Making of a Roman Town by Frank Brown. Elements from both the Cincinnatus and Lipsius models have been merged into a one-size-fitsall paradigm that emphasizes both Lipsian notions of cultural supremacy and the Cincinnatean ideals of equality and virtue within a new socio-political order.
The problem is that the image of the Roman colony—so explicitly articulated by Lipsius and deeply embedded in Western memory—also influenced the archaeologists who first excavated Roman colonies and mapped their territories. Early excavations focused heavily on the urban aspects of the colony and their resemblance to Rome, as exemplified by the case of Cosa in Etruria. The enthusiasm leaps off the page when the American excavators not only uncovered a forum but also identified a Capitoline temple and a comitium—just like in Rome. Here, the feedback loop is visibly at work: the 20th-century excavators were shaped by a pervasive, centuries-old ideal model of Roman towns and colonies, which in turn led to a biased interpretation of what were assumed to be ‘hard’ archaeological facts (Fentress 2000).
The territories of Roman colonies, too, were studied. Especially through field surveys, one of the most established methods in landscape archaeology. In these surveys, teams of archaeologists systematically searched the ploughed fields of a region and recorded all concentrations of surface finds. These findspots then appeared as dots on a map. In this discipline as well, the results initially seemed to confirm the traditional model: surveys were thought to have identified the remains of colonists’ farms, and the dots were reconstructed into a neatly organized hinterland scattered with such farmsteads. However, this reconstruction of the rural landscape – based on the ideal of the Cincinnatus-style soldierfarmer – may itself be biased (Pelgrom 2018). This bias partly arises from the methodology of field survey, which by definition concentrates on arable land. It may also result from an overly automatic association between surface scatters of ceramic finds and a narrowly defined typology of the colonial farmstead.
Together with a team of scholars from the Landscapes of Early Roman Colonization project (2013–2019), we subjected this dilemma to a thorough test through intensive fieldwork in two Roman colonies, complemented by data analysis from several other colonial territories. We identified patterns of colonial-period sites in the two case study areas in central and southern Italy that differed markedly from conventional expectations—no Republican Cincinnatus-style farmers in the colonies of Aesernia (263 BCE) or Venusia (291 BCE). Instead, we observed a more adaptive mode of land occupation, not unlike the practices already in place among pre-existing communities. Rather than the expected regularly spaced, isolated colonial farms, we found a more organic pattern of settlement, with hamlets and clusters of sites alternating with empty zones (Stek et al. 2015; Pelgrom et al. 2015; Casarotto et al. 2016). In other early colonial territories too, we may see non-urban, village type settlement organization rather than Romanized agro-towns. In some cases, such as the colony of Alba Fucens, this deviant pattern can be related to other-than-agricultural economies, such as those based on lacustrine food production and transhumance (Stek 2018). A form of agrarianism looms large in the ancient Roman texts, but may well hide different realities in the field.
Historical loopings: the (imagined) past resonating in the present
The Lipsius-Cincinnatus model resonates strongly with more recent historical events and processes. Roman colonies, viewed through this lens, offer clear examples of what we would now call settler colonialism. The depiction of the colony of Savannah, Georgia, established by the British in 1733, serves as a striking mirror of the contrast between colonial order and perceived wilderness. More recent examples also echo ideas already familiar and debated in antiquity. One such instance is the use of lots for assigning plots of land, as in the founding of what would become Tel Aviv in 1909, where seashells were used as ballots. Similarly, debates over unrest among the lower classes and their demands for land from the elite loom large in Roman Republican history—strikingly reminiscent of modern cases, such as those raised in Dinah’s paper on the VOC (Dinah Wouters in this volume).
A large number of Roman colonial towns were already inhabited settlements prior to colonization, although this fact has often been omitted or downplayed. In the case of the colony of Venusia, we were able to demonstrate pre-Roman occupation through archaeological surface material recovered from an area of the ancient town that is currently uninhabited. However, the majority of colonial sites have been continuously occupied, which makes systematic proof of earlier phases more difficult. This highlights a common feature of colonial settings and discourse: land is easily declared free, empty, and uninhabited—from the colonizers’ perspective. This reminds us of the ‘swampification’ discussed by Zwiers for Dutch ‘internal’ colonization in the 19th century, to portray land as useless and wild in order to heighten the positive impact and righteousness of colonial intervention: in the Roman colonial period too, considerable marshy areas were turned into agricultural land, such as the Pontine plains south of Rome, and we can expect similar biases in the representation of the previous landscape and its positive transformation by Roman colonial force. The debate over the right to colonize supposedly ‘new’ land continues to resonate in contemporary contexts. Important to underscore, is that Classical Roman references are by no means the only ones invoked in this regard; for instance, the biblical and politically charged phrase “to make the desert bloom” has been used repeatedly by Ursula von der Leyen in reference to Israel’s supposed merits in regard, even if this portrayal might not be historically accurate and ethically problematic (Von der Leyen 2022).
The presence of the same or similar topoi in ancient colonial discourse and in (early) modern contexts is, therefore, both exciting and frustrating. As with narrative analysis, we rarely find direct evidence of influence. Even in scholarly work, where one would expect precise documentation of intellectual lineages, such references are often left unacknowledged. In fact, the flattening and simplification of scholarly discourse in the mid-20th century caused significant damage by failing to cite original sources in bibliographies.
Transmission is the crux
The question of transmission – how, why, and in which contexts certain colonial topoi reappear – is a central issue in current debates. This is evident in several contributions, but perhaps most clearly in the work and paper by Dinah Wouters. In her project, she studies a Latin commentary on the Agrimensores authored by Willem Goes, an official working for the VOC. Wouters makes a compelling case that the Agrimensores corpus was actively put to use in debates on colonial strategy in the 17th-century Dutch East Indies. The very fact that a VOC administrator devoted time to the study of Roman land surveying is arresting in itself. Wouters places this engagement within the broader context of the popular Ragion di stato genre. In fact, one could argue that Greek and Roman texts on colonization already function as early prototypes of the Ragion di stato tradition, since the search for the (secret or not-so-secret) key to Roman imperial success lies just beneath the surface of many of these works, as Polybius’s canonical text already makes clear. As several contributions in this volume show, it is worthwhile to examine practical texts, such as strategic and technical treatises, as closely as theoretical or historical writing. In the context of ancient Roman colonial landscapes in Italy, I was personally shocked to discover how clearly, and indeed how ‘untouched’, Roman ideal models reappear in quite ‘practical’ texts on Italian Fascist colonization in Africa.
In the context of the Fascist Italian state’s efforts to build a colonial empire, debates emerged about which forms of colonization would be most effective. Given the scale of Italian migration to the Americas, various ideas were proposed to prevent the depletion of the Italian population. One variant, the so-called colonizzazione demografica, was seen as particularly attractive, not least because it would produce more soldiers and workers, rather than losing them to the Americas or elsewhere.
In sometimes highly practical guidelines, the two variants of ancient Roman colonization reappear. For instance, Carlo Giglio described the colony as a “centre of radiation of Fascist and Italian civilization in Africa” (Giglio 1939), an almost literal echo of the Lipsian model. At the same time, we also recognize features of the civil and military ethos of the Cincinnatus model: “If it is the plough which traces the furrow it will be the sword which must defend it. The workers are soldiers and the soldiers workers. Legionaries are those who conquered the Empire, legionaries will be those who with their toil will render it fertile” (Fossa 1938, echoing Mussolini’s speech at the inauguration of the new province of Littoria, modern Latina, on 18 December 1934).
Perhaps the most striking aspect of this material is that we are not reading theoretical historical studies but practical manuals on how best to colonize a given territory. In fact, it seems that Italian colonial aspirations have centered more on settler colonization than other contemporary imperial powers, both in the Italian late 19th century and the Fascist variant (Tekeste 1987). The fact that the practical advice of such manuals often failed in implementation, and that colonial efforts of this kind largely collapsed, with many settlers returning home disappointed or worse, is another story (Larebo 1994). Of course, there are no footnotes or other explicit references to the ancient Roman roots of these ideas. Yet their presence nonetheless reveals the enduring power of such concepts along a transhistorical trajectory.
I would like to conclude, however, by turning to another medium through which ancient colonial values continue to trickle down to us, and inevitably shape our historical interpretations. To best grasp this example, please have a look at Figs. 3 and 4 which I took with our drone when working in the area of Venosa. These concrete-built farms dot the landscape around the modern town of Venosa. They are laid out at regular intervals; each is separated by an equal plot of land. The farms typically consist of three to four rooms, with an outdoor oven in the garden. Today, some are temporarily occupied by African migrants working in the local agricultural sector. A few have been enlarged and are now used regularly by Italian families, while many others lie abandoned. The farms clearly reflect a larger settler planning scheme in which the nuclear family and agricultural production appear to have played a central role, an example of ‘internal colonisation’. The project dates from the 1950s, following the Riforma Fondiaria in Basilicata, which was part of a wider land reform in especially South Italy in the post-Second World War period (Prinzi 1957; Mecca 2012). Venosa is the main town in the area today. As mentioned earlier, the site of Venosa (Venusia in Latin) was colonized by settlers sent out by Rome in 291 BCE, and was included in our fieldwork project to test the conventional model of Roman colonial settlement organization. For the area in antiquity, we did not encounter an early, orderly, and monumental Roman colonial landscape at all. Instead, we found a territory divided among different village communities.
There is certainly an irony in the fact that the ideal Roman colonial landscape never existed here in antiquity, yet briefly came to life in the 1950s. The humbling lesson for us as archaeologists is that we cannot afford to avoid comparisons between antiquity and modernity. On the contrary, we should approach these key issues directly—through interdisciplinary and transhistorical perspectives – these are very necessary comparisons.
Figs. 3 and 4. A colonial landscape of the mid- 20th century AD near Venosa, Basilicata (photos by T.D. Stek)
Bibliography
Bispham, E. (2006). Coloniam deducere: How Roman was Roman colonization during the Middle Republic? In G. J. Bradley & J.-P. Wilson (Eds.), Greek and Roman colonization: Origins, ideologies and interactions (pp. 73–160). Classical Press of Wales.
Brown, F.E. (1980). The making of a Roman town: Cosa and urbanization in Republican Italy. University of Michigan Press.
Campbell, B. (2000). The writings of the Roman land surveyors: Introduction, text, translation and commentary. Society for the Promotion of Roman Studies.
Casarotto, A., Pelgrom, J., & Stek, T.D. (2016). Testing settlement models in the early Roman colonial landscapes of Venusia (291 B.C.), Cosa (273 B.C.), and Aesernia (263 B.C.). Journal of Field Archaeology, 41(5), 568–586.
Fentress, E. (Ed.). (2000). Romanization and the city: Creation, transformations, and failures (Journal of Roman Archaeology, Supplementary Series 38). Journal of Roman Archaeology.
Fossa, D. (1938). Lavoro italiano nell’impero. Mondadori.
Giglio, C. (1939). La colonizzazione demografica dell’Impero. Edizioni Rassegna economica dell’Africa italiana.
Goes, W. (1674). Antiquitatum agrariarum liber singularis. Amsterdam.
Larebo, H.M. (1994). The building of an empire: Italian land policy and practice in Ethiopia 1935–1941. Oxford University Press.
Lipsius, J. (1598). Admiranda sive de magnitudine Romana. Jan Moretus.
Mecca, I. (2012). Borghi e paesaggi rurali della Basilicata: Tipologie edilizie e techniche costruttive. Esempi di Architettura, 1(1), 1–10.
Pelgrom, J., & Stek, T.D. (2014). Roman colonization under the Republic: Historiographical contextualization of a paradigm. In T.D. Stek & J. Pelgrom (Eds.), Roman colonization under the Republic: Towards a new
interpretative framework (pp. 10–44). Palombi.
Pelgrom, J., Marchi, M.-L., Cantoro, G., Casarotto, A., Hamel, A., Lecce, L., García Sánchez, J., & Stek, T.D. (2015). New approaches to the study of village sites in the territory of Venosa in the Classical and
Hellenistic periods. Agri Centuriati, 11, 31–59.
Pelgrom, J. (2018). The Roman rural exceptionality thesis revisited. Mélanges de l’École française de Rome – Antiquité, 130(1). https://doi.org/10.4000/mefra.4770
Prinzi, D. (1957). La riforma agraria in Puglia, Lucania e Molise. Arti Grafiche Laterza.
Salmon, E. T. (1969). Roman colonization under the Republic. Thames & Hudson.
Settis, S. (1984) (ed.), Misurare la terra: centuriazione e coloni nel mondo romano, Modena.
Stek, T.D., & Pelgrom, J. (Eds.). (2014). Roman Republican colonization: New perspectives from archaeology and ancient history (Papers of the Royal Netherlands Institute in Rome, vol. 62). Palombi.
Stek, T.D. (2017). The impact of Roman expansion and colonization on ancient Italy in the Republican period: From diffusionism to networks of opportunity. In G. Bradley, E. Isayev, & C. Riva (Eds.), The peoples of
ancient Italy (pp. 269–294). De Gruyter.
Stek, T.D. (2018). Early Roman colonization beyond the Romanizing agro town: Village patterns of settlement and highland exploitation in central Italy. In T. D. Stek & J. F. Osborne (Eds.), The archaeology of
imperial landscapes: A comparative study of empires in the ancient Near East and Mediterranean world (pp. 145–172). Cambridge University Press.
Stek, T.D., Modrall, E. B., Kalkers, R. A. A., Van Otterloo, R. H., & Sevink, J. E. (2015). An early Roman colonial landscape in the Apennine Mountains: Landscape archaeological research in the territory of
Aesernia (Central Southern Italy). Analysis Archaologica, 1, 229–281.
Stek, T.D. (2013). Material culture, Italic identities and the Romanization of Italy. In J. W. Richardson & F. Santangelo (Eds.), The Blackwell companion to the archaeology of the Roman Republican period (pp. 337–
353). Wiley-Blackwell.
Tekeste, N. (1987). Italian colonialism in Eritrea, 1882–1941: Policies, praxis and impact. Uppsala University Press.
Von der Leyen, U. (2022). Speech by President von der Leyen at the Ben-Gurion University of the Negev, June 14, 2022. European Commission. https://ec.europa.eu/commission/presscorner/detail/hu/speech_22_3685.