The institutional review process for the doctoral programme in Archaeology at the University of Buenos Aires includes a series of sequential steps, each requiring an increasingly detailed development of the ongoing research leading up to the final thesis submission: a pre-project, a project, a plan, and finally the thesis itself. When I submitted my thesis project in 2023, it was returned to me with some harsh criticisms that shook the foundation of my proposal centered in working with soils as a means to develop an archaeology of agriculture.

Although I—along with my supervisors—agreed with some of the reviewers' comments, which ultimately contributed significantly to refining the justification of my thesis, a substantial portion of their feedback targeted the very idea of approaching soils through archaeology. In fact, in an informal comment, one of my supervisors was told: “She should study materialities, structures, artifacts… soil is not archaeology.”

At that point, I had already made significant progress in both fieldwork and lab analyses, and had presented several preliminary writings and conference papers on the topic, generally receiving positive feedback.

These comments by the reviewers led me to reconsider what became a fundamental decision for continuing the writing process, according to their anonymous recommendation at that stage: either “switch to a Ph.D. in Social Anthropology (given the ethnopedological focus shown in this first project) or reformulate the ideas to ensure the resulting thesis clearly qualifies as archaeological.”

By then, I had spent around 10 years working on soil studies in archaeology and had been teaching the course “Soils in Archaeology” at the Faculty of Natural Sciences, National University of Tucumán, for a few years. So, together with my supervisors, we decided to revise the project in certain aspects and continue within the Archaeology Ph.D. programme. At the time, I saw this both as a challenge and an opportunity to propose, justify, and “defend” soils as an archaeological materiality—one that is different from classical ones and “more complex”, but material nonetheless.

In this article, I present and elaborate on my perspective for developing an Archaeology of Soils as a means to approach peasant agricultural landscapes.

In its most classical conception, archaeology can be defined as the discipline dedicated to the study and interpretation of human behavior through material remains of past human societies (Renfrew and Bahn, 2007), or as the study of past sociocultural systems and processes through artefactual assemblages and their contextual relationships, with the goal of contributing to the clarification and explanation of the spatial-temporal dimension of human existence (Binford, 1962, 1988). On that basis, and as part of ongoing debates about the scope and meaning of the discipline, within the Interdisciplinary and Intercultural Collective from the High Valleys of Catamarca (CIIVAC for its Spanish acronym), in which I have participated for over 10 years, we define archaeology as a social science that problematizes social processes based on the study of material culture across time, in order to understand the present and imagine the future (Quesada et al., 2014; Korstanje et al., 2016, 2017, 2025).

This definition entails two significant implications: first, that archaeology should intertwine the study of the past with historical and contemporary social issues—a position already familiar and debated within the discipline (Politis and Pérez Gollán, 2004; Mannasse and Arenas, 2010; Meskell, 2012; Hamilakis, 2016). Second, it encourages us to rethink archaeological materiality—a debate that remains actual, especially since the ontological turn in archaeology (Nielsen, 2022; Alberti, 2016; Fahlander, 2017; Preucel, 2021).

Throughout its history, the discipline has approached artifacts from a wide range of perspectives: from their composition, addressing technological and archaeometric aspects, decorations, symbolism, and more (Knappett, 2005; Hodder, 1990); from chaînes opératoires, exploring their origin, use, disposal (and reuse, reclaiming, etc.), and their active role in social life (Ingold, 2007). More recently, other epistemological approaches have emerged, allowing us to explore the agency of things, materials, and the non-human in shaping human trajectories (Nielsen, 2006; Scattolin et al., 2009; Gastaldi, 2010, 2008; Di Salvia, 2016).

Building on these contributions, here I propose a perspective that considers the role of soils not only as matter but also as agents—incorporating both human and non-human interventions in their constitution as the foundation of peasant landscapes, understood across a past–present continuum. To do so, I draw on my experience in three highland valleys of north-western argentine's Andes: El Bolsón, Rodeo Gerván, and Los Morteritos–Las Cuevas.

The CIIVAC has been studying the social processes that shaped human groups in these valleys for over 30 years, from what is regionally known as the Formative Period (ca. 1,000 BCE−1,000 CE) to the present, using various indicators and always through an interdisciplinary approach (Korstanje, 2005, 2010; Mondini, 2021; Quesada and Korstanje, 2010; Maloberti, 2014, 2020; Brown, 2018). A key line of inquiry within the Collective—and one that remains active—is agricultural production and its role in social reproduction, as this region has seen sustained agriculture, with changes, ruptures, and continuities, over the last 1,200 years (Korstanje, 2010; Taddei Salinas et al., 2023). Two theoretical concepts have been central to this effort: the peasantry, as key agents in our analysis (Korstanje, 2010; Brown, 2018), and the past–present as a continuum—a framing that avoids the arbitrary temporal cuts typical of academic disciplines and instead embraces long-term perspectives (Lazzari and Korstanje, 2013).

In this process, as well as in our search for ways to link agriculture with other social spheres in the configuration of more complex landscapes, soils have played an important role as a complementary tool—particularly in the development of multi-microfossil analysis methodologies (Coil et al., 2003; Korstanje and Cuenya, 2008, 2010). However, to date, soils in our study area have not been examined specifically in relation to their role in the web of relationships that make up peasant agricultural landscapes.

In order to construct the logical argument that leads to the theoretical proposal at the core of this paper, I will begin by reviewing some archaeological perspectives on soil.

Before introducing this section, it is necessary to clarify the concept of soil, its differences from sediment, and the scope of this article.

Sometimes “soil” and “sediment” are used interchangeably, and sometimes there is confusion as to which term is correct (Holliday, 1990, 2004; Tchilinguirian et al., 2016).

From a strictly technical point of view, without room for nuance, all soils can be sediments, but not all sediments can be soils. Soils are formed from sediments that have undergone weathering and alteration in situ or have been transported and deposited. However, unlike sediments, soils require a sufficiently long period of environmental stability so that formative factors converge and give rise to particular pedological processes that enable pedogenesis.

From a practical standpoint, soil is often defined as the medium in which plants grow. This definition offers other nuances: the peasant, rural, or indigenous farmer does not stop to contemplate whether the substrate in which he sows constitutes soil or sediment. His practical knowledge enables him to recognize its potential as a planting bed.

From a phenomenological point of view, soil is recognized as a lived territory and as a place of mutual belonging with the people who inhabit it (Kusch, 1976, 1978).

Therefore, what follows are definitions and observations on soils that are not necessarily exclusive of sediments, but that serve to conceptualize soils as materiality.

Archaeological practice involves an inevitable relationship with soil: when surveying, we consider its various aspects and properties in order to identify sites, assess their state of preservation, and evaluate the possibilities it offers for working with and within it. When excavating, we directly intervene in its body in order to interpret it. Through excavation, we progressively remove it in situ to reach the materials that archaeology has classically focused on, such as ceramics or lithic artifacts (Salisbury et al., 2022). Our interpretations are built upon what is “un-covered” through the act of excavation. Profile cleaning becomes a tedious task, often perceived as obstructive or time-consuming in the excavation process. Similarly, sieving or screening produces piles of soil that we discard, while we recover from the mesh only those materials deemed worth taking to the laboratory for further analysis (sometimes “just in case…”). In our daily laboratory work, we also dispose of the soil or sediment that clings to the archaeological objects we will study. We clean, brush, wash, and scrape them. In microfossil processing from soils (an example I am somewhat familiar with), the sample is washed as many times as necessary to remove the sediment, retaining only the microfossils (comparable in size to the silt fraction) from the supernatant so that they can be observed under a microscope. Except in cases where soils or sediments are studied as a complement to other categories of archaeological materials, it seems they are always in the way, always a hindrance.

And yet, soil constitutes the broadest and most ubiquitous material category we work with.

In the few cases where archaeological literature explicitly addresses soil (sometimes also referred to as sediment), it is generally considered a matrix—that is, the container in which archaeological remains are found (Sampietro Vattuone, 2007, 2009). When soil samples are taken in archaeological contexts, they are processed and analyzed precisely as established by this matrix, complementing other studies.

Among the most comprehensive syntheses on soils in archaeology is the pioneering work “Soils in Archaeological Research” by (Holliday 2004), and, in the argentine national context—a fundamental reference for archaeologists and geoarchaeologists across South America—the work of (Tchilinguirian et al. 2016). Both texts highlight applications such as stratigraphy, site formation processes, paleoenvironmental reconstructions, dating, soil chemistry as an indicator of human activity, micromorphology, microstructure, and microstratigraphy of soils, along with the potential of concepts such as anthrosols and anthropogenic soils.

The interdisciplinary work between earth sciences and archaeology gave rise to a field known as geoarchaeology, which has taken shape as a sub-discipline since the 1970s. From within this framework, archaeological interpretation has shifted from a perspective centered solely on material culture to one that incorporates the study of the deposits where finds are located—not as a mere addition of geological information to archaeological problems, but as a distinctly archaeological way of thinking about contexts. Thus, geoarchaeology integrates within the archaeological record the material objects, the containing matrix (soils and/or sediments), and the geoenvironmental context (Favier Dubois, 2023). However, within this discipline, a wide variety of archaeological studies are carried out with a geological, geomorphological, sedimentary, Quaternary geology, and/or soil science imprint, applied to contexts composed of both sedimentary (Favier Dubois, 2006) and pedological deposits (Zárate et al., 2000–2002).

Beyond geoarchaeological approaches, soil physics and especially soil chemistry have been instrumental in understanding the integrity of the archaeological record (in the absence or low frequency of artifacts/ecofacts in primary contexts), in taphonomic analyses, and in delineating different activity areas—as seen in the pioneering works of (Ortiz and Barba 1993) in Mexico and, more recently, those of (Roos and Wells 2017), the contributions of (Obregón et al. 2011) for the Andean region, and others including (Ortiz 2003), (Cuenya and García Azcárate 2004), and (Gianfrancisco 2005) for north-western Argentina.

Despite these contributions, we can say that only a handful of works place soils at the center of theoretical reflection for solving archaeological problems. That is why the development of an Archaeology of Soils remains an incipient endeavor (Linderholm, 2010; Salisbury et al., 2022). Aware of this situation, (Sampietro Vattuone 2007, 2009) proposes to approach soil as an artifact—a classic archaeological category that is virtually undisputed. According to the author, this concept encompasses any trace (physical object, feature, or landscape) made or altered by humans. By treating soil as an artifact, it can be studied in technological, chronological, and socio-functional terms (Sampietro Vattuone, 2009). From a different but related perspective—also appealing to its artefactual nature and to landscape definitions as a web of things (based on Ingold, 2007, 2015—conceptualizations), Álvarez (2021) defines soil as an assemblage of heterogeneous elements, a gathering of threads of life. That is, as a thing.

While these perspectives are insightful, I argue that conceptualizing soil as an artifact leads to a certain reification, overlooking the agentic aspects of its ontic nature, as will be discussed further on. At the same time, soil retains the particularity of also being addressed as an ecofact (especially in cases where there is no direct human intervention in its formation), insofar as it provides information on the long-term relationship between people and their environments. Because soils play important roles in site formation processes (Schiffer, 1991), they are also approachable from taphonomic perspectives (Behrensmeyer, 1978). Finally, the limitations of classical categories for understanding archaeological material culture allow us to conceptualize soil as a feature—in its non-portable, artefactual character.

Swedish archaeologist (Linderholm 2010) defines soil as an archive, in the sense that it preserves and reflects the environmental history of humans, plants, and animals from the past to the present. His definition resembles that of an ecofact, while also emphasizing its non-discrete structural character. Consistently, he highlights that soils (on a long temporal scale and broad spatial scale) may preserve our cultural heritage even better than museum collections (Linderholm, 2010, p. 1). He also recognizes soil's behavior as both matrix and system—constantly changing—and concludes that the multiple coexisting rates of change within it make it very difficult to compare to the biography of an artifact (Linderholm, 2010, p. 12).

One recurring way to define soils within archaeological discourse is through the metaphor of the palimpsest. In their study of changes in the use of agricultural soils (drawing examples from both North America and the Andean region), (Sandor and Homburg 2017) describe soils as a palimpsest of traces from different environments that succeeded one another through multiple forms of use and/or transformation. They argue that soil operates at multiple temporal scales and intensities, processing everything produced through its relationship with the environment—including human activity. (Wells 2006) also introduces the idea of soil as a palimpsest, but in this case of historically constructed and contingent ideas, beliefs, and practices. He presents soil as a reservoir of shared ecological knowledge, manifested in the pedological record and resembling an archive (Wells, 2006, p. 126), echoing Linderholm's (2010) framing.

However, as with the definition of soil as artifact proposed by (Sampietro 2007, 2009), these ideas are not always developed in depth. In contrast, more exhaustive discussions appear around the concept of the palimpsest, particularly as applied to landscapes, as in Bailey's (2007) work. He offers an in-depth reading of the term, from its linguistic and palaeographic origins to its application in studies of memory construction and socio-cultural heritage—including Freudian and performative uses. Bailey distinguishes between two divergent interpretations of palimpsests: the cumulative palimpsest—the successive overlaying of activities (the most common archaeological usage), and the true palimpsest, in which all previous information is erased except the most recent inscription (Bailey, 2007, p. 203–204). He uses this distinction to illustrate the fragmented nature of the archaeological record and argues that palimpsests can help overcome both linear, sequential views of time and the spatial abstraction of discrete “places.”

(Ingold 2021), for his part, notes that the palaeographic palimpsest analogy entered archaeology in the 1950s (Crawford, 1953) to describe land repeatedly used over time. Ingold continually draws analogies between the palaeographic palimpsest and soil—emphasizing its depth dimension rather than its position as the lower boundary of the landscape. He distinguishes two types of erasure that palimpsests involve: striking through and erasing. The former is sudden, violent, explosive—yet the line that strikes through also preserves what it hides, sometimes even emphasizing it. Erasing is slower and more diffuse: like an eraser on a page (both degrading together) or water washing over soil, uncovering plant roots or erasing footprints. As Bailey also suggests, a palimpsest is not formed by layering but by erasing—allowing the old to re-emerge. This is Ingold's anti-stratigraphic principle: the past surfaces as the present digs deeply, like ink sinking into parchment (Ingold, 2021, p. 5).

If we apply this reading to soil, we find that erasure of the old by the new is not always inevitable. Evidence of multiple activities—synchronous or asynchronous, at different scales—can coexist, in line with Bailey's cumulative palimpsest. At other times, this evidence is erased by human action (e.g., stone clearing, cleaning, digging, intentional or unintentional disturbance), or by soil itself (through pedological processes such as leaching, percolation, translocation).

Pedoarchaeological stratigraphy is highly complex, combining principles of archaeological superposition (“what lies deeper was deposited earlier”), pedological development (“horizons form from the surface downward”—Porta Casanellas et al., 2003), and archaeological stratigraphy, which further complicates both by including human agency (Harris, 1991).

Thus, the analogy of soil as a palimpsest, though widely used in archaeology, cannot be applied directly without nuance. I draw here on both the cumulative and preservative capacities described by (Linderholm 2010) and (Wells 2006), as well as Ingold's (2021) anti-stratigraphic logic, alongside the many life cycles that occur within and through soil in constant relation with all it touches.

Finally, (Salisbury 2012) defines soils as part of material culture. Starting from the critical observation that lithics and ceramics have traditionally enabled the study of symbolism, identity, exchange, power, migration, and other social phenomena, while soil has been relegated to a mere indicator of agricultural potential, Salisbury offers a comparative discussion. He reviews various definitions of “material culture” and matches them with the nature of soils, ultimately concluding that soils “[...] play an essential role in the formation of memory, identity, community, and worldview” (Salisbury, 2012, p. 24)—a proposition with which I fully agree, and which provides a foundation for what follows.

My proposal is to study agricultural landscapes, conceived as taskscapes, with an emphasis on soils, drawing from a framework of vital materialism. This perspective allows me to shift the archaeological gaze toward an “in-between” or an “across”: that is, not to focus exclusively on agricultural structures as discrete units, but rather to approach them within a broader web of relations, where connections and movements among them can be contemplated—conceiving of a landscape continuum of soils entangled with people.

The previous paragraph calls for, at the very least, a review of certain key theoretical concepts necessary for developing this proposal. The concept of landscape already has a long trajectory in the history of our discipline (Criado Boado, 1993; Anschuetz et al., 2001; Tilley, 2004), but precisely for this reason it has been used as a theoretical notion to refer to different ideas. It is therefore important to revisit the term and clarify what we mean by landscape, at least for the purposes of this paper.

From (Foucault 1989), who emphasized the social and historical nature of what we call the environment, to the somatic phenomenology of (Merleau-Ponty 1984), who worked on awareness of corporeality and embeddedness in the inhabited world, and including definitions developed specifically within archaeology, the concept of landscape has taken on different nuances. (Criado Boado 1993) used it to refer to everyday contexts, socially constructed and dynamic; (Gosden and Head 1994) focused on how people's relationships with their surroundings over time generate feelings of belonging and spatial continuity. Nonetheless, for quite some time, landscape functioned largely as a backdrop for the subjects and objects of the inhabited world (Tilley, 2004).

Seeking to overcome this limitation, archaeology turned to Social Anthropology, and particularly to the ideas Tim Ingold began to propose in the 1990s—ideas I find particularly helpful in illustrating the point I aim to make, and to which I will return throughout the following pages.

For Ingold, space is the most abstract, empty, and indifferent concept we use to refer to the inhabited world (Ingold, 2015, p. 9): people live on the earth, on the ground, in the countryside—not in space. We dwell in the world through, around, from, and toward; we inhabit it through movements in which we trace paths and tracks that intertwine, entangle, and cross over one another, generating knots. The world, then, is formed archi-texturally by the mesh of these knots in the trajectories of dwelling. Each thread is a form of life, and each knot a place (Ingold, 2015, p. 19).

(Ingold 2000) defines landscape as the product of dwelling, of people's being-in-the-world, who leave traces as records or testimonies of their lives (the threads described above), emphasizing its historicity. Landscape, then, is a relational field, where its forms emerge as condensations of beings' activity within that field. This perspective acknowledges the historic character of landscapes, highlighting both their changes and continuities, as well as how they are perceived and engaged by people.

Landscape is thus understood as a commitment to the inhabited world, with a material basis (land) and a projected and meaningful totality (scape), constituting a unit characterized by the symmetry and mutual belonging of human and non-human beings. To better grasp the concept—which, even with theoretical advances, has often remained a mere backdrop—(Ingold 1993) introduces the concept of taskscape, imbued with deep phenomenological meaning. It can be understood as the landscape of practices, of agency; a space where perception and action converge, since the practice of the various beings inhabiting the landscape is itself involved in understanding and shaping it. Through dwelling, humans establish relationships with different corporeal others.

As a world of practices in motion and in temporal becoming, the taskscape is highly dynamic and never complete (Ingold, 1993, p. 194–195). In these embodied practices, people establish concrete forms of social bonding in and with the world, which enable, on one hand, the emergence of common sense through the daily reproduction of social relations—which in turn sustains those very relations (Padawer, 2013); and, on the other, as a consequence, a savoir-faire or practical knowledge that stems from the ability to perceive, to judge what is perceived (Chevallier and Chiva, 1996), and to act accordingly.

Practical knowledge arises from the convergence of inherited and practiced common sense with the personal, embodied experiences of each social subject in relation to the material world. People exercise and modify that practical knowledge every time they use it—thereby transforming the world (Chaiklin and Lave, 2001): a world that is lived, material, and sensuous.

All that has been outlined so far helps to justify the inclusion of soil depth in the material configuration of inhabited taskscapes—particularly agricultural landscapes, where people and soils are intimately linked through a world of practices and tasks supported by practical knowledge. However, to understand soils within landscapes as part of material culture, we need an additional conceptual shift—one I address by turning to the philosophical contributions of Jane Bennett's vital materialism, which is particularly relevant for its commitment to the world through the vitality of bodies.

(Bennett 2022) understands vitality as that intrinsic and singular quality that gives something the capacity to intercept the trajectories of other things, making things happen in context. Her vital materialism departs from historical materialism (where things seem to gain agency only when activated by human action or gaze), challenging the passive conception of the world around us. Instead, she proposes a view of matter as active, dynamic, unpredictable, and capable of agency (Manccioni and Jorge, 2022), offering a new perspective that sees humanity as a complex and rich assemblage of vibrant materials (the minerals in our bones, the metals in our blood, the electricity in our neurons), entangled with an equally material world.

The concept of assemblage, developed with clear influence from the rhizomatic model of Deleuze and Guattari (2004), emphasizes the vital potentials of matter: assemblages are not governed by a central head, but each member contributes its own vitality to allow the assemblage to generate things—while simultaneously retaining some of its own vitality, which could potentially form new assemblages (Bennett, 2022, p. 75). Thus, assemblages have irregular topographies, and their vibrant vitalities are not evenly distributed (Bennett, 2022, p. 74).

We can observe certain similarities between Bennett's notion of assemblages and (Ingold 2015); Ingold's (2021) concept of place: life unfolds along lines that tie into one another (places), much like agency flows within assemblages. The more entwined the lines, the denser the knot; the more densely assembled the materials, the greater their vitality.¹

In her theoretical proposal, Bennett questions the nature of materials based on their inherent vitality: she revalues waste, debris, decay, and rot—materials that seem biographically finished, and which are, in fact, the very stuff archaeology works with. All of this provides tools for socializing matter, while also allowing social relations to be materialized for archaeological purposes.

In light of this, an Archaeology of Soils for addressing agricultural landscapes becomes an invitation to inhabit what appears empty—as seen from a Cartesian spatial lens, considering that archaeological agricultural structures and interstructural spaces often appear “clean” of archaeological materiality, in the classic sense of the term. Yet, by conceiving of a landscape continuum where matter and vital agencies are assembled through time, articulated through a world of moving practices, we begin to question those voids—reframing them as full of vibrant materials assembled in context, even if they merely constitute an obstacle (like a stone in the path), a discomfort (like a stone in a shoe), or a resistance (like wind against movement).

It is also an invitation to (re)think agricultural practices—not merely as something that is done, but as complex processes of acquiring savoir-faire, savoir-percevoir, and savoir-relier: learning to act, perceive, and relate to the vital materials of agricultural landscapes within a particular assemblage—the soilscapes.

The concept of soilscape comes from soil cartography and is generally used to refer to geomorphological units associated with soil types across broad areas (Buol et al., 1973; Zinck, 2012). As a conjunction of soil and landscape, they are defined as soil landscapes and soils in the landscape—or, put another way, as the assemblage of soil-bodies in a landscape.

Now, if we consider, based on everything outlined in this section, that landscapes are materially and vitally constituted—and deep (a dimension granted by the bodies of soil as soilscapes)—and finally, as taskscapes, we can then conceive of agricultural landscapes as in motion, formed by the continual generation of contact and mutual perception among vibrant, vital matters. These enable corporeal experience and make things happen—and the material consequences of such events are accessible through archaeology.

Finally, I would like to discuss the concept of Hybrid Archaeologies. To talk about something hybrid can be complicated. While some emphasize that the hybrid cannot reproduce itself, others highlight its fruitful, resilient, and resistant character (Liebmann, 2013, p. 30). A hybrid is formed by assembling diverse things, becoming something different while maintaining the identity of its origins. Rather than categorizing and ordering, the hybrid inhabits the interstitial and the liminal (Silliman, 2015).

Thus, the hybridization of archaeological practice involves generating knowledge that reflects multiple, intersectional and contradictory identities in relationship to one another (while inhabiting the resulting tensions). It means claiming heterogeneity and avoiding standardized, pure content/state classifications with set boundaries. Instead, we move between our categories of analysis (Stockhammer, 2013).

In this sense, Hybrid Archaeologies include archaeological practices that bridge different fields by appealing to the transversality of knowledge, practices, discourses, theories, and methodologies (Meskell, 2012; Silliman, 2009; Lazzari, 2011; Lazzari and Korstanje, 2013). They also incorporate a social and political questioning of archaeological work to empower local communities (Meskell, 2012; Hamilakis, 2016). Rather than diminishing the legitimacy of the archaeological discipline, this approach enriches it by providing interpretive depth and directing it toward contemporary societies, thereby deepening its social role (Meskell, 2012, p. 135).

In this case, the most appropriate pragmatic framework is Hybrid Archaeologies, as it allows me to address agricultural landscapes in a temporal and spatial continuum by articulating and balancing all previous concepts.

Following the theoretical framework developed so far, I now propose to approach as an archaeological object of study what I define as the Soil–Earth–Territory Complex (SET), emphasizing its validity as an archaeological materiality.

At this point, I would like to introduce some reflections drawn from the work I have been conducting over the past several years in the high valleys of western Catamarca (Northwestern Argentina), which serve to exemplify my proposal.

Rodeo Gerván, El Bolsón, and Los Morteritos–Las Cuevas (Figure 4) are three high-altitude valleys (~3,000 masl), characterized by a semi-arid, extreme, and highly contrasting climate (both daily and seasonally), a highly dynamic geomorphology, and a rural population that has depended on agriculture and livestock raising for the past 1,200 years—with shifts and continuities, of course—as established by over three decades of research by CIIVAC (Korstanje, 2005; Korstanje et al., 2017; Kulemeyer et al., 2013; Taddei Salinas et al., 2023).

The limited pedological development in the area makes it difficult to define soils in the strictest sense; the most common profiles are of the A/C or A/C/2C types. Nevertheless, people have cultivated—and continue to cultivate—in these poorly developed soils. This has been possible because, over time, they have established relationships with soil–earth in agricultural contexts through a range of practices aimed at stimulating and enhancing its vital potentials for the creation of spaces of crianza-nurture, care and cultivation- (Grillo, 1994; Lema, 2014). In particular, irrigation and fertilization involve material additions (animal dung, plant remains, ashes, water) that leave traces within these soil bodies—traces that are measurable and quantifiable from both archaeological and pedological perspectives (such as variations in particle size distribution, structure, organic matter content, or pH).

Moreover, among the highland farmers, what takes place within the spaces of agricultural crianza (the plots) is not solely aimed at food production—or at least not in strictly economic terms from a Western capitalist standpoint. Beyond that, the inhabitants of the high valleys raise plants, animals, and soils—while simultaneously raising themselves through the process—enacting that which has been socially constructed, maintained, and inherited from their ancestors, in combination with personal experience. This heritage provides a strong sense of security in their practices, as it is supported and safeguarded by the savoir-faire received and learned from their immediate forebears (Cáceres, 2000). There is also a revalidation of ties to these ancestors, and a continuous commitment to and with Pachamama, who continues to offer protection. Each individual establishes personal and particular bonds with the STT, while collectively sustaining practices that create cohesion within the group and reaffirm their shared belonging to a territory.

For instance, every August, at the beginning of the agricultural cycle, the practices of fertilizing, feeding, and sharing/treating (convidar⁴) the earth converge. The first of these practices, which is associated with an agronomical sense, involves replenishing nutrients so the soil can produce throughout the agricultural cycle, which will soon restart. Feeding the land is a family and sometimes community practice. This is done through a deep offering pit—far deeper than what is removed to prepare planting beds in fields, which is reused year after year as the locus of the practice. Convidar is a private, intimate act performed in an equally intimate place. Each person offers what little value they have, and the offerings are generally superficial or shallow. These three practices coincide in their contemplative character, and occur at a time (August) of the earth's primordial reactivation as a provider and support of life, as Pacha-todo. During this period in the Global South, particularly in the Andean region, the earth is considered thin, open, and hungry, and requires nourishment (Van Kessel and Condorí Cruz, 1992, p. 86). Through these practices, people seek to restore balance.

Other moments of vital connection with the soil-earth, as Pacha, occur during the livestock birthing season (March to May). These practices differ from previous ones in that they are festive, expressing gratitude for the fruits the earth provides at that time.

There is a shared awareness of a mutual nurturing network: A well-fed soil-earth is a healthy one, that will produce healthy seeds and good pastures for fattening the livestock, which will also remain healthy and safe.

Although these practices differ in terms of time, materials offered, and who promotes them, they all have one thing in common: they are performed on the material dimension (earth), affecting the corporeal depth (soil), but with a component associated with the socio-symbolic dimension (territory) of the SET.

In the El Bolsón valley, two archaeological agricultural sites—Vaca Vizcana and Yerba Buena—revealed evidence of quinoa cultivation in pre-Hispanic times, based on both pedological properties and the presence of diagnostic microfossils (Maloberti, 2020). If we understand “quinoa cultivation” as an assemblage, we might say that it was made possible by the convergence of sufficiently fertile earth, a relatively stable environment (Meléndez, 2017), the availability of irrigation water and fertilizing guano (represented by diagnostic microfossils such as diatoms and spherulites, respectively, which the soil managed to preserve over time), among other factors.

Some components of this assemblage became available again—due to their vital potentials and affordances—to be re-assembled with other vibrant matter, enabling quinoa cultivation once more in the last 15 years, though under different conditions and circumstances. These include quinoa reintroduction projects promoted by national government agencies, whose local representative—an agronomist deeply committed to the savoir-faire of the highland inhabitants—was able to engage in meaningful dialogue with them, leading to the planting of quinoa on family plots.

The outcomes of the “quinoa cultivation” assemblage—both in the past (pre-Hispanic times) and in the present—show clear parallels, even if they were not composed of the exact same components.

At some point between those two moments, quinoa seeds scattered across the landscape, buried in the soil like a seed bank, remained latent—awaiting suitable environmental conditions for a wild variant of the plant to emerge on the hillsides. This may constitute a new assemblage: known locally as paico or quinoa wascha.⁵

If we adopt the SET Complex I propose—in which materials, agencies, and practices are assembled—then soil constitutes not only a dynamic seed bank in ecological terms, but also in social terms: a place where seeds are in contact with pests that may affect them, with the soil that will nourish them once germination becomes possible, and, upon sprouting, with people who recognize them as quinoa wascha, who will care for, nurture, and reproduce them. The seeds are, in the context of the multiple assemblages that make up agrarian landscapes—along with the vital knowledge and materials that enable those landscapes. In this way, seeds in the soil—understood as a dynamic bank—reflect historical trajectories embedded within a world of productive social relationships.

Assemblages, writes (Bennett 2022), come into being so that “things may happen” (and the plural here is key). Based on the archaeological evidence of cultivation at the two aforementioned sites, we can take this reasoning even further.

The assemblage “cultivation” (earth producing), for instance, must remain active for a period without depleting the soil's fertility. This means that, before reaching exhaustion, the assemblage must disassemble itself and reassemble into a state of “rest” (fallow, in agronomic terms) until the soil's vitality is restored. Then, when conditions permit, a new “cultivation” assemblage can be activated.

At the same time, while the “cultivation” assemblage becomes inactive and “rest” is activated here, a new “cultivation” may emerge a bit farther on, in a place previously held in “rest.” This ongoing activation and reactivation of assemblages of vibrant matter generates topographic variation in terms of vitality—which can be interpreted as an agrarian landscape that breathes and continually reinvents itself, grounded in the STT.

If we return to the graphical scheme used to represent soilscapes earlier (Figure 3), we might say that some of these assemblages last longer than others and unfold across the landscape (Figure 5). This is the concept of a material, vital, and agentive soilscape that I propose—one whose corporeal, physical, and socio-symbolic foundation lies in the notion of the SET Complex.

We can now begin to view the constituents of the world as vibrant matter's assemblages (Bennett, 2022): the SET itself (complex, multidimensional, and hybrid); the seed that it shelters but that also grows through it over time; the manures that contribute their share of nutrients; the offering or ritual, which renews a socio-symbolic commitment to it; the animal that today plows the earth, tomorrow feeds on the stubble, and later contributes its dung as fertilizer; the person in assemblage with the shovel, and later in assemblage with the hoe—which are not the same, and which make different things happen.

We thus also come to understand that these assemblages inherently carry a temporal dimension—or rather, multiple temporalities. In this way, we can consider that the physical, material, and agentive assemblages that constitute what I have called the Soil–Earth–Territory Complex are themselves interconnected in the shaping of agricultural landscapes, within a past–present continuum.

The concept of the SET Complex is a proposal for approaching an object of study that is difficult to fit within the classical categories of analysis in our discipline—yet which is, at the same time, undeniable and valid as archaeological materiality. Far from being merely soil in agronomic terms, it possesses additional material, social, and symbolic dimensions that render it more complex. An Archaeology of Agriculture that incorporates this concept necessarily includes temporal and spatial depth, as well as its many constituent vital materials—all of which are in dynamic relationship with what occurs on and through them, without imposing arbitrary temporal boundaries.

When I received feedback on my dissertation project—suggesting either that I reformulate the topic or transfer to a doctoral programme in Anthropology—I realized that there was something problematic about the archaeology I was proposing. Was it really archaeology? Is it possible to study agriculture by thinking from and through soil alone? Why, for instance, does landscape—a potentially more abstract or even metaphysical concept—seem more widely accepted than soil, which is material, tangible, and quantifiable?

We are already “squeezing” information from soils and sediments to lend greater weight to our archaeological interpretations—but we have yet to pause and ask what this type of complex materiality truly entails.

After more than 10 years working on agricultural soils in the high valleys with CIIVAC, I have only recently started to realize that doing archaeology by thinking from soils is uncomfortable. That is why I have chosen to fully dwell in these uncomfortable disciplinary spaces—a choice that ultimately led to the need of writing this article.