How does it feel to flow? I think of frictionless, liquid movement, of moving onward without effort, my limbs melting into the surrounding momentum, my body fluently aligning with the bodies next to me as we are carried along. Flowing seems to be not so much an act as a state of being, a state that requires some surrender to another energy, like nature, collectivity, or gravity. Some things are compelled to flow, like blood, time, oxygen, rivers, tears. But while some things appear to flow naturally, flowing is never straightforward. Its ease rests on labors, past and present; its current contains fractures and frictions in the form of turbulences. Take the Mississippi River, for example. It flows in specific ways because of geological formations, weather events, and ongoing interventions by the US Army Corps of Engineers, including dredging and damming, that have altered the river’s basin. The flow continues among eddies and interventions, but all the time it changes, too.

I fell into this journey and went with the flow. I had wanted to write about maize—maize as a “model” organism for industrial agriculture as well as genomic sciences and, more generally, for data-intensive evidence practices. I think it makes for an opportune companion for thinking about this time that some are calling the Anthropocene, because maize has developed with people and lands, and these lands and people have developed with maize. Enormous swaths of land in the Midwestern United States along the Mississippi River have been reconfigured to accommodate the industrial production of maize. This region known as the Corn Belt comprises 150 million acres that are almost exclusively planted in a corn-soybean rotation. Depending on one’s viewpoint, it can be a devastating sight. One of the world’s most diverse plants reduced to a handful of varieties planted in monocultured row after row after row. Not much lives in these fields apart from the cultivated plant. Landscapes without insects, without nematodes, spiders, poppies, grasses, people.

Food activist and journalist Michael Pollan calls these landscapes “clean fields.”11Michael Pollan, “Playing God in the Garden,” New York Times Magazine, October 26, 1998, https://www.nytimes.com/1998/10/25/magazine/playing-god-in-the-garden.html During my trips to view various parts of the Mississippi River Basin, at times all I could see out the window of the van was corn stretching all the way to the horizon. This sight could be read as “food security,” and perhaps induce a sense of comfort. For the local farmers, it’s equally existential. But it’s not that they have a choice. The image of White smallholding farmers living and working the land in ecological harmony perpetuated by popular imagery and Big Agriculture belies the reality of an extractive and punishing industry that saps the life out of people and plants alike. Rising demand for food, fuel, and feed is forcing farmers to maximize productivity, which is increasingly tied to data: measuring and quantifying nutrients, seed viability, weather conditions, planting depth or disease history are meant to ensure the most precise and economic distribution of input, be it seeds, water, fertilizer or other chemicals. A flow of data guides a flow of matter that secures the flow of food.

My interest in maize stems from my fascination with the plant geneticist Barbara McClintock, or more precisely, from reading Evelyn Fox Keller’s biography of McClintock, A Feeling for the Organism.22Evelyn Fox Keller, A Feeling for the Organism: The Life and Work of Barbara McClintock. San Francisco: W. H. Freeman, 1983. Keller, a feminist historian of biology, has pioneered a critical history of genetics in the twentieth century, attending to how social and cultural aspects have shaped (and in turn were shaped by) the “century of the gene.” McClintock revolutionized genetics. She showed that there is more to the genome than just genes. Rather, she figured out, the genome contains so called transposable elements (TEs)—“jumping genes”—which make up 50 percent of the human genome, 75 percent of the maize genome, and 98 percent of the iris genome, to name just a few.33See, for example, Barbara McClintock, “The Origin and Behavior of Mutable Loci in Maize,” Proceedings of the National Academy of Sciences, vol. 36, no. 6 (June 1 1950): pp. 344–55, https://doi.org/10.1073/pnas.36.6.344

TEs are vital for generating diversity, as their actions increase the frequency of mutations. Human geographer Kathryn Yusoff deems the jump to be a good model of movement for reckoning with the inhuman timescales of the Anthropocene.44Kathryn Yusoff, “Geologic Subjects: Nonhuman Origins, Geomorphic Aesthetics and the Art of Becoming Inhuman,” Cultural Geographies, vol. 22, no. 3 (July 2015): pp. 383–407, https://doi.org/10.1177/1474474014545301 Though genes are us, there is something equally inhuman about their scale. So jumping as an epistemic move works well in both directions—upstream and downstream, backward and forward. Jumping is flowing with turbulences. It lets me connect maize, the Mississippi, friends, and data.

McClintock figured all of this out by asking why the kernels of flint maize, also known as “Indian corn,” are spotted. She planted maize in her experimental fields in Upstate New York near Cold Spring Harbour Laboratory and Cornell University, the two research institutions that would (barely) tolerate a woman scientist in the 1940s. Keller’s account, which is partly based on interviews with McClintock, describes the systemic discrimination McClintock had to face every step of the way. She also documents her principled resolve and her extraordinary understanding and intimate connection with her plants, her “feeling for the organism.” It took forty years for her research to be validated and recognized—because she was a woman, but also because no one could quite replicate her technique of “seeing” chromosomes, the entities in the cell nucleus that contain the genes.

Historian of science Bruno J. Strasser has argued that the corn genetics community that was established in the early twentieth century at Harvard University laid the foundation of what would become data-driven biology.55Bruno J. Strasser, Collecting Experiments: Making Big Data Biology. Chicago: University of Chicago Press, 2019. Its data practices, recoding practices, observing practices, collecting and archiving practices, as well as exchange of seeds and data, created not only lasting protocols and infrastructures but data science. I like to think that the colorful kernels of flint corn are the beginning of big data. They are nonbinary, carrying many shades of color. I hadn’t quite realized just how many until Michael Swierz, poet, translator and “participant ecologist,” while clearing the back seat of his car to make room for journalist and author Fritz Habekuss, MPI research scholar Christoph Rosol, and me, handed me an enormous mason jar filled with flint corn kernels. It was beautiful—small pebbles of magnolias, violets, yellows, whites, and blues. At first, I didn’t really know what I was looking at.

What would it mean to think about data through maize, to think about maize through data? For one, it’d make data more tangible and material, add a temporal, developmental dimension to its analysis. Concurrently, diffracting maize through data can render visible how variant forms of maize, like patented hybrids, are structured through heterogenous networks involving copyright and patent laws, international development policies, population statistics, and logistical standards. In my work as a sociologist and sociocultural anthropologist, I’m concerned with the datafication of nature and its social, political, and cultural ramifications. To me, the datafication of nature refers to a wide-ranging project that seeks to translate biotic and abiotic life into governable measures and values. It involves diverse practices, from digitizing natural-history specimens, to sequencing and barcoding environmental samples, to modeling vegetation types using remote sensing. Despite their obvious differences, all these practices represent attempts to abstract and reduce complex organisms and interactions into discrete data points. Yet, these processes are shot through with turbulences that don’t reduce but displace complexity. In going on this journey to trace flows of data and maize along the Mississippi, I wanted to recover some of this complexity.

With the help of McClintock, maize initiated a radical shift in Western science in thinking about the relation between bodies and their environments. According to Darwinian evolutionary theory, adaptions (evolution) of organisms occur over extremely long periods of time and no rapid mutations happen in “nature.” To be sure, this point has been a long-standing battleground in natural history as well as its successor sciences. The issue was settled when McClintock’s work, which found the genome to be much more responsive than previously thought, became accepted. This motivated the development of epigenetics, which proved that environmental factors can lead to changes on the genetic level, thus returning scientific consensus to the previously discredited idea of inheritance of acquired traits. So, flint maize kernels changed the understanding of interaction between bodies and environments, making bodies coextensive with their environments.

What I like about the story of McClintock seeing chromosomes that transform into jumping genes that fuel data-driven genomics is what it says about data. Data are the product of embodied labors, sometimes far away, but never entirely absent. In my work, I’m concerned with recovering or figuring these labors that make data work, that make data global, and, possibly, that make data accountable. My colleague at the Museum für Naturkunde Berlin, Christian Bölling, recently stressed the importance of “metadata provenance” for ensuring responsible sharing and reuse of data as well as the longevity of the resource. Whereas metadata describe the properties of an object, such as its size, language, or title, provenance metadata relates to an object’s creation and gestation: when it was made, modified, issued and who created it, contributed to it and retains the rights and ownership. This means that data models—the structures that determine the entities that data describes and how they relate—need to provide for data histories or, more precisely, genealogies. If we, for example, digitize a herbarium sheet containing (parts of) a pressed and dried maize plant, then we include relevant metadata concerning not just the physical object—its scientific name, where and when it was collected, by whom—but also concerning the entities and processes that contributed to its creation as a resource and digital object. But, naturally, there will always be data missing because they were lost or deemed not relevant or because the interface didn’t allow for their capture. Data provenance makes evident the many steps and protagonists involved in making data. It also points to the decisions behind systems and infrastructures, decisions that often manifest as gaps and absences.

Corn and soy constitute our planet’s most important grain crops, used for human food, animal feed, and ethanol production (biofuel). Eating constitutes an interesting flow between bodies and environments, where mutual ingestions eat away at species boundaries. Eating, my colleague Filippo Bertoni writes, is an “ecological relationship.”66Filippo Bertoni, “Soil and Worm: On Eating as Relating,” Science as Culture, vol. 22, no. 1 (March 1 2013): pp. 61–85, https://doi.org/10.1080/09505431.2013.776365 On my journey along the Mississippi, which took me from Minneapolis to just south of St. Louis, we did many kinds of eating. We ate processed bread and processed cheese as part of a parking-lot picnic in front of a bleak supermarket in a desolate-looking part of Hannibal, Missouri. We learned later from Faye Dant, founder of the Jim’s Journey: The Huck Finn Freedom Center, that this had been a thriving African American community that was destroyed in the course of the racist “urban renewal” policies of the 1950s and ’60s. Later that day back at our camp site, we ate a delicious hot pot cooked by our resident chefs and fellow River Travelers John Kim and Steven Diehl, made inside an enormous pumpkin procured by HKW’s Neli Wagner from a petrol station somewhere near the Quad Cities. They stuffed the decorative Halloween pumpkin with mushrooms, pumpkin flesh, and cheese and placed it on a log fire, where we watched it blubber for hours. In Burlington, Illinois, we were treated to a collective feast by the members of the Aldo Leopold Foundation, who kindly and generously shared their delicious dishes with us. I ate mac and cheese for the first time.

As Bertoni writes, “eating does not occur alone,” meaning that it always involves others, even if only distant and microscopic ones—the gut bacteria that help digestion and all the human and nonhuman participants of the food chain that deliver produce. Nowhere was the flow of relations that eating enacts as present as at the dinner that was prepared and shared at the Black Hawk Museum and Lodge in Rock Island, Illinois. The women who cooked the food described how their dishes combined food traditions from the local First Nations and from Indigenous Peoples in Brazil who have been fighting to protect their lands and right to nourishment. Traces of maize have been found in the remains of ceramic vessels in Ecuador, Peru, Brazil, and the Caribbean, some dating back to 8000 BCE. Imagine the social relationships that have all this time been flowing through the sharing of food and drink!

In relation to food production, food security has come to be linked to ensuring a contiguous data flow with no gaps. Here, the provenance of foodstuff is secured through establishing traceability. This means issuing unique identifications to elements in the supply chain, from farm to storage location, suppliers, logistical units, and companies. I started thinking about traceability after fellow John Kim mentioned its impact on small-scale organic farming. We were driving through the Driftless Area of Wisconsin, which is home to many organic farms, including the biggest producer in the US, Organic Valley. Traceability becomes an existential issue once it is seen as a prerequisite for entering markets or, indeed, for being considered “organic.” One problem for small-scale operations is the time and effort spent on record-keeping and data management, which becomes all the more complex the greater the diversity of plants a farm grows. I imagine that tracing and tracking heirloom varieties and seeds that have been kept and circulated in communities across generations is quite an impossible task. Yet, like all life, maize retains traces of its close and distant relatives in its genome: of the wild grass (teosintes) in Mexico, Guatemala, Honduras, and Nicaragua; of rice, from which it diverged 50 million years ago; and of sorghum, with which it parted ways only 9 million years ago.

At the Genoa National Fish Hatchery in Wisconsin, we were able to bear witness to how traceability is being integrated into restoration efforts. Lake sturgeon populations in North America (some scientists estimate globally) have been declining due to overfishing and habitat destruction, and so the hatchery is producing sturgeon to restock the populations in lakes around the country.77 T.J. Haxton & M.J. Friday (2018). Are we overestimating recovery of sturgeon populations using mark/recapture surveys? Journal of Applied Ichthyology. https://doi.org/10.1111/jai.13795 We arrived on a sunny day and entered a low building crammed with big plastic vats, each teeming with thousands of lake sturgeon fingerlings. They were no bigger than twelve centimeters but they looked their age: somewhere between 208 and 245 million years old. At the back of the building, a school class had gathered around an assembly line of volunteers who were tagging each fish using coded wire micro-tags. This is done so that, once released into the wild, hatchery-reared fish can be distinguished from wild populations (for the calculation of survival rates and population trends) and to distinguish between different lots of fish. At Genoa, they tag about 60,000 to 80,000 fingerlings per summer.

The spectacle upset me. The ancientness of the sturgeon, so present and visible in its astonishing body, was affecting, especially in contrast to the plastic-based hubris of the breeding program. Like maize, sturgeons have come a long way. In fact, they are almost 200 million years older than corn. Like all living organisms, they are “of all times.”88Michel Serres, “The Origin of Language: Biology, Information Theory, and Thermodynamics,” in Hermes: Literature, Science, Philosophy, ed. Josué V. Harari und David F. Bell (Baltimore: Johns Hopkins University Press, 1982), p. 75. It’s just that, with sturgeon, you can really see this. I’ve joined fellow paddlers Audrey Buturian Larson and Steven Diehl in the sun outside the shed where the lake sturgeons are being tagged. We are in a somber mood, somewhat upset by the goings-on in the building behind us. A long conversation spins out that weaves our discomfort about the industrial hatching of fish in the name of conservation with our dismay at the “industrial production” of students that sees them reduced to standardized assets on the balance sheets of corporate education. To our minds, fish and human bodies become abstract, commodified objects in the calculative logic geared toward the optimization of stocks.

A couple of days later, we amble around the John Deere Pavilion in Moline, Illinois. A large display is making the moral case for John Deere–powered industrial agriculture by bellowing exploding population statistics. Large percentages are printed all over the wall, an LED ticker provides the current number of humans on Earth, and a see-through plastic box is fed with a steady trickle of maize kernels. Each emits a click as it lands in an already vast sea of kernels, reminding visitors that another hundred, another thousand, another ten thousand humans have joined the planet, threatening its limited resources. KERNEL PANIC!

The data that are imagined to reside in maize are vast indeed. With its turn to big data, Big Ag has set its sights on data as a new cash crop. The movement of crops, from farms to silos, barges, and processing plants to plates and troughs is nowadays accompanied by a thick data flow that is feeding more and more distributaries. Farm machinery is equipped with sensors that continuously send soil, land, and weather data and numbers on fertilizer use and water levels. These data can be aggregated to set crop prices or predict insurance claims or develop fertilizer. They can also be combined with data on land registration and credit use to drive the dispossession and privatization of common lands. Geographer Alistair Fraser argues that data-based precision agriculture changes the production of food, altering agrarian relations—our relations, really, since we are never not inside the flow.99Alistair Fraser, “Land Grab/Data Grab: Precision Agriculture and Its New Horizons,” Journal of Peasant Studies, vol. 46, no. 5 (July 29 2019): pp. 893–912, https://doi.org/10.1080/03066150.2017.1415887 This is hard to see, although I was looking out for it from the canoe, from the window of the van, from the airplane flying over the Corn Belt. It remains a challenge, to see (the work of) data, to see organisms and (in)human history.

Data keep flowing from and through maize. And as they move, infrastructures emerge, equipment and humans transform, lands change—and so does maize, because the plant is engineered in response to the models that are predicted out of data streams. Many labors remain. Stories, too. They’re right there, in the code, even though splicing and cutting have left their marks. But the ghosts of teosintes, of Cahokia, of the Mississippi River, and of the sturgeons are patient. They are waiting for their moment to jump. McClintock has told us already.

To maintain matters in a state of flow.