In 2010, Sujit Sivasundaram published a groundbreaking essay titled ‘Sciences and the global: on methods, questions, and theory’ in the journal Isis.Footnote ² Reflecting on the challenges posed to the ‘global history of science’ by what critics supposed to be a scarcity of sources from beyond Europe, Sivasundaram proposed a method of ‘cross-contextualization’, whereby scarce and unorthodox sources would be read alongside more plentiful, conventional ones. This approach extended beyond merely reading against the grain; it also promised to show how materials produced in disparate cultural contexts could, when studied together, mutually thicken and complicate one another’s interpretations.

Sivasundaram’s article – and his broader ‘Global histories of science’ Isis Focus section to which it belonged, featuring field-defining essays by Marwa Elshakry, Helen Tilley, Shruti Kapila, and Neil Safier – has had a significant impact on the history of science over the past decade and a half.Footnote ³ Yet, since its publication, much has changed in the discipline, with an ever-growing array of cultures and practices of knowledge from disparate geographies and chronologies falling under its purview.Footnote ⁴ Through a series of dialogues, this roundtable explores not only the continuing importance of Sivasundaram’s intervention in global histories of the sciences, but also draws attention to how historically oriented scholars across disciplinary boundaries are experimenting with cross-contextual approaches to illuminate otherwise marginalized, overlooked, and suppressed voices, worlds, and stories. Included among the new directions envisioned by this roundtable are taking cross-contextualization seriously as an actor’s – as well as an analyst’s – practice, and juxtaposing settings removed from the colonial/colonized binary. The conversations in the roundtable, while by no means prescriptive, offer a set of heuristics for scholars and students seeking to engage with asymmetrical archives.

That historians – or for that matter, scholars from other humanistic disciplines – read across different archives and genres as part of their practice is hardly surprising. This roundtable offered practitioners an opportunity to critically reflect on what distinguishes cross-contextualization from such routine triangulation. As we understand it, cross-contextualization differs from kindred approaches by reorienting scholarship away from the search for explicit connection between contexts, and rather towards resonances that emerge between them through generative, surprising, or uncomfortable juxtaposition. This technique raises questions about what kinds of evidentiary claims we can make about contexts that may otherwise have been overlooked in mainstream historiographies. In this way, cross-contextual methodologies can take us beyond Sivasundaram’s initial aim of showing that non-Western cultures possessed and practised something we can analytically treat as ‘science’. Rather, these approaches serve as guides towards recognizing resonances between different contexts, whose connections may be indirect. For instance, geographically distant and culturally disconnected parts of the world may be affected by the same tsunami, and cross-contextualizing their responses – political, epistemic, cultural – may proffer a story worth exploring.Footnote ⁵

Amid contemporary historical scholarship, environmental history has perhaps most boldly brought together drastically different conceptual categories and genres of sources into synthetic narratives. Scholars within the field have begun to read not only different cultural contexts against each other, but interweave human pasts with natural histories too.Footnote ⁶ As Dipesh Chakrabarty urges, historians in the Anthropocene ‘need to connect deep and recorded histories and put geological time and the biological time of evolution in conversation with the time of human history and experience’.Footnote ⁷ At first reading, this programmatic statement appears to call for both historians and natural scientists to embrace something akin to an expanded cross-contextual methodology. Yet – as is made clear across the following five dialogues – just how such a project is pursued is inevitably political. It is both epistemically and ethically problematic to treat non-humans, from rocks to ice-cores to plants, as uncontroversial proxies (or contexts) with which to reconstruct – especially marginalized – human worlds of knowledge.Footnote ⁸ This roundtable thus seeks also to probe the ways in which granting agency to an array of non-human entities can serve to limit and distort rather than bolster the original, emancipatory horizons of cross-contextualization.Footnote ⁹

Let us offer a brief example to illuminate both some interpretive virtues and possible methodological challenges of cross-contextualization. Between 1685 and 1686, Simon van der Stel, the first governor of the Dutch Cape Colony, led a trek into Namaqualand (modern-day north-west South Africa and southern Namibia) in search of copper ore to revitalize the colony’s struggling economy.Footnote ¹⁰ The expedition’s record-keeper, the Silesian apothecary Heinrich Claudius, interspersed his natural historical observations of Namaqualand’s plants, animals, and landscapes with ethnographic reflections about its local population. He reported that the Namaqua people ‘eat everything that comes their way, down to rats, dogs, cats, caterpillars, locusts, etc. For hares alone they have the greatest distaste and loathing so that they will not partake of any dish containing hare, whether boiled or roasted.’Footnote ¹¹ On its own, Claudius’s passage tells us little about Namaqua cosmologies, and rather more about early modern Dutch ethnographic assumptions and practices. It leaves us unsatisfied, wanting to know more about why the hare was taboo. Little under two centuries later, the |xam-speaking storyteller Dia!kwain recounted the foundational myth of ‘The moon and the hare’, of which variations circulated extensively among many southern African groups, to the English schoolteacher Lucy Lloyd in Cape Town.Footnote ¹² Together with her brother-in-law, the Prussian anthropologist Wilhelm Bleek, Lloyd compiled an extraordinary anthropological survey of southern Africa’s Indigenous peoples in the late nineteenth century.Footnote ¹³ As Dia!kwain told Lloyd, the Moon had once sent a hare down to Earth to promise humankind immortality. The hare, alas, garbled the message, instead dooming people to a single, mortal life. In anger, humans refused to eat the animal, transforming it into a taboo. An oral history told by a nineteenth-century |xam-speaker thus seems to help ‘thicken’ an interpretation of an otherwise ‘thin’ seventeenth-century colonial diary.

This exercise, however, poses interpretive, political, and methodological challenges. Dia!kwain was not a Namaqua but a |xam-speaking San person. Thus, cross-contextualizing his story with Claudius’s account risks flattening cultural and linguistic differences between the groups. Moreover, the juxtaposition appears to ‘upstream’, or retro-project meaning across centuries, problematically suggesting a fixed timelessness of Indigenous southern African cultures.Footnote ¹⁴ Such issues present real difficulties for historically oriented scholars grappling with cross-contextual methodologies. Yet, it is worth asking why such caution so often exclusively constrains work on non-Western contexts. Imagine, for instance, describing two superficially similar cross-shaped structures: one atop the thirteenth-century Chiesa di San Francesco d’Assisi in Palermo, the other on the twentieth-century Hallgrímskirkja in Reykjavík. Few of us trained as academic historians would hesitate to classify both structures as ancient Christian symbols that, despite manifold differences in their contexts, share some meaning. There are, in other words, perceptible resonances (from our standpoint) between such temporally, spatially, and culturally disparate contexts. It is these whispers, we maintain, that are worth listening for.

Another question this case raises is that of method. How does one identify and choose what archives and materials to juxtapose in the first place? How should we, practically speaking, teach cross-contextualization to interested students and colleagues? In other words, how should we reflect on our agency as historians in experimenting with this approach? Of course, we should not underplay the role of serendipity in landing upon productively entangled sources. Our recognition of the apparent relevance of the Bleek–Lloyd archive to thicken the context of the Van der Stel notebook did indeed unfold unexpectedly, and not by design. That said, an intentional cross-contextual approach could also have brought us to this same productive juxtaposition. By critically reflecting on the historical construction of the two southern African archives, it becomes evident that even though they engage, respectively, with different Indigenous communities, they were connected (albeit indirectly) by Indigenous information orders whose contours do not appear explicitly within colonial archives.Footnote ¹⁵ As Laura J. Mitchell recently argued, the task of ‘archival patchwork’ – a method akin to cross-contextualization – is the ‘result not just of patient detective work, [with] the archives simply revealing their secrets to a historian who spends enough time looking. Patchwork is the result of choices.’Footnote ¹⁶ One such choice has to do with the historian recognizing that they have reached the end of their ability to interpret one particular historical context – as in the case of the Van der Stel notebook offering no further commentary on why the Namaqua refuse to eat hare – and then actively begin their search for another context to thicken the first with. This is often an iterative process, but it need not be undirected speculation. By historicizing – indeed, recontextualizing – the first archive by reading it ‘against the grain’, clues may emerge regarding where to look next. Thus, while cross-contextualization differs from reading archives against or along the archival grain, the latter becomes a helpful strategy in practising the former.

One approach that often proves fruitful – and was indeed raised in Sivasundaram’s original article – is a careful engagement with historical anthropology. Much like our example above, Marcy Norton has productively drawn on ethnographies produced in the past 150 years to enrich her analysis of Amerindian animal familiarization in the sixteenth and seventeenth centuries.Footnote ¹⁷ As she puts it, ‘where there is independent evidence for similar phenomena across time, they should be considered comparatively and seen as support for the strength of a particular structure’ – in her case, Amerindian information orders – ‘rather than as evidence for some overall “timelessness”’.Footnote ¹⁸ Moreover, as she underlines, the existence of a ‘common core’ of cultural practices across a broad region over a substantial period of time does not preclude different cultural and chronological specificities.

Yet, we should resist the temptation to systematize it into a how-to manual for global history. While the dialogues in this roundtable constitute efforts to put cross-contextualization into practice, we ought to remember that the approach – like history as an academic discipline more broadly – is not a set method. Moreover, as historians of science, we also want to draw attention to the ways in which cross-contextualization and methods akin play an important role in the practices of natural scientists. Lorraine Daston reminds us that scientists too cross-contextualize their data to construct richer, temporally sensitive accounts of their areas of study.Footnote ¹⁹ Take for example the remarkable work of Cambridge volcanologist Clive Oppenheimer, who has drawn on and intricately interwoven oral histories of volcanic activity from across the world with non-human proxy evidence to enrich reconstructions of past geological events.Footnote ²⁰ For historically oriented scholars, this raises important questions: what, precisely, is at stake when we employ such methodologies? Do scientists and humanists practice cross-contextualization in similar ways, or do our approaches diverge? And what distinctions – whether conceptual, epistemic, or political – separate these different modes of inquiry?

Bruno Latour, writing several years before the publication of Sivasundaram’s paper, anticipated one of the risks deriving from an uncritical, depoliticized embrace of something akin to cross-contextualization. As he warned, modern science often claims to give voice to the non-human world, to the point that ‘we actually know how to consult nonhumans better than humans!’.Footnote ²¹ Natural scientists, Latour noted, take great care to elicit credible testimony from non-human entities, devising experiments of extraordinary complexity to ‘consult’ rocks, plants, or elephants. By contrast, when it comes to humans we are often far less cautious: self evidence from the dispossessed is readily dismissed.Footnote ²² The danger, then, is that in privileging the non-human as a neutral source of evidence, dominant narratives can reproduce the silencing of subaltern human perspectives.Footnote ²³ Making visible the politics of representation, both human and non-human, is thus essential if cross-contextualization is to avoid reproducing the very alienations it sets out to challenge.

So, what is to be done? Are historians doomed simply to fall back into contextual scholarship on conventional written sources, perhaps reading them against the archival grain? Or can we, instead, animate other types of evidence; sources not necessarily connected to one another but capable of enriching the contexts otherwise suppressed? The oeuvres of Natalie Zemon Davis and Carlo Ginzburg beautifully teach us that speculation based on clues can be a productive response to evidentiary silences, and that it is better to ask bold questions, even at the cost of conjecture, rather than avoiding asking them at all.Footnote ²⁴ Cross-contextualization and speculative (micro)histories are not the same, but they are kindred methods that can speak to each other in their respective projects to reimagine forgotten or suppressed voices and worlds.

The dialogues in this roundtable are based on conversations that took place at a one-day workshop in the Department of History and Philosophy of Science in Cambridge in June 2025. In addition to invited participants, we had a small, but very active audience. The conversations that took place around each dialogue were essential for the day’s discussions and the work that has taken place since. First, Jeremy Schneider and Siyakha Mguni explore the ways in which the lithosphere can be read as an archive and a storyteller. Questions of myths and dreams offer compelling avenues to bridge early modern European geologies and natural philosophies with South African San cosmologies of rocks, caves, and transformation. Next, Anna Simon-Stickley and Amelia Urry build on Schneider’s examination of cross-contextualization as an actor’s practice. They turn to climate sciences and geo-histories, tracing how disparate proxies – from Nile floods to ice cores, paintings, and novels – have been reinterpreted to make claims about unknown environments. Subsequently, James Poskett and Lynette Russell navigate the complex world of material culture collections in their dialogue. They explore how shifts in perspectives between Indigenous and metropolitan actors can recalibrate the meanings and interpretations of ‘material culture’ in the history of science and museums. Then, Mika Hyman and Sebestian Kroupa continue to follow questions around collections, but shift their focus towards plants. Together, they explore how plants can be understood as historical and cultural actors, shaping Indigenous sovereignties and global exchanges. In the final dialogue, Rabia Abba Omar and Theo Di Castri highlight the importance of creative practices, embodied performances, and collaborations with artists and activists to reckon with silence, memory, and violence in archives and beyond. Lastly, Sujit Sivasundaram re-examines the institutional, disciplinary, and methodological contexts that shaped ‘Sciences and the global’, and, reflecting on how these have since changed, explores how cross-contextualization as a method has changed more broadly.