Pulling Up Holes, Pulling Down Hills

### How People Who Actually Work the Land Understand the Landscape on Which They Work

*Please note that this is the version of the paper from which I worked when presenting at the annual meeting of the American Folklore Society: I did not, for example, actually read the conclusion. I would like to thank the members of the audience — all those folks from North Carolina — who gave such rich feedback. I made notes of your questions, but the richness and clarifications you demanded are not yet in this paper. Thank you for your generosity.*

#### Introduction

The cognitive anthropologist Charles Frake once observed that psychologists have a long, historical fascination with navigation because the task is clearly defined and has easily measured goals. By and large, psychologists prefer the confines of investigator-defined tasks and contexts so that they can control for variables. The problem with such an approach, as psychologists themselves discovered (and yet never seem to learn), is that contexts are themselves culturally-defined, or as one of my UL colleagues described it: how does one make claims for universality based on tests administered to a several dozen university undergraduates? One option has been to establish an infrastructure for the replication of psychological research results; another has been to find “performances that can be seen as displays of cognitive ability” that exist in the workaday world in which most of us find ourselves (Frakes 255). While such a description, “performances as displays of cognitive ability,” is redolent for folklorists, it is not entirely sufficient for psychologists who have worried that, in seeking out cognition in the world people actually live in, a task cannot possess sufficient enough precision such that its successful accomplishment is not open to negotiation. (I know: we regard this as precisely the kind of “deep” space for investigation. Here, humanistic depth must be shallowed for the sake of scientific precision, from the point of view of cognitive psychology.)

Navigation, as Frake noted, “provides an especially nice display of cognitive performance” if for no other reason than failure is clear: one either runs aground, is lost at sea, or arrives safely in port (255). Much of the work in psychology has focused on Micronesian navigation, and, as Frake states:

> The lesson to be drawn from these studies is that the islanders’ seafaring exploits do not depend on some uncanny intuitive powers, nor on personality quirks driving people to seek danger, nor on the luck of lost sailors adrift at sea, nor even on rote-learned “local knowledge”. Instead these navigational abilities depend on a profound general knowledge of the sea, the sky and the wind; on a superb understanding of the principles of boat-building and sailing; and on cognitive devices—all in the head—for recording and processing vast quantities of ever changing information. (256)

What Micronesian navigators present is a community of practitioners who engage in a clearly delineated task that is itself based on traditional knowledge. That knowledge is usually conceived of as a model by psychologists (Oatley 538).

Having spent the past five years working with fabricators and farmers in a largely agricultural ecology (but one still with boats in the picture), I have been a regular witness to a wide variety of tasks which have fairly fixed notions of a successful outcome: either a tool or machine works, or it does not; either a field produces a crop, or it does not. Well, it’s not entirely so cut and dry, to use an agricultural metaphor: after all, machines can work well or simply well enough. Cuts in rice fields can produce a range of yields, and those yields themselves are averaged across both a field as well as a farmer’s total acreage in production that year. That noted, it is my belief that there is probably an equal range of sufficient results in Micronesian navigation that allow for a greater range of results than the literature suggests.

Would it be possible, I wondered, to discover a task within my own ethnographic area of inquiry, define its parameters, and determine what it has to contribute to our understanding of human cognition? In this paper, I would like to focus on a single task, water leveling a rice field, to see just what it can tell us. [Not a great transition, but it will have to do to keep things short.]

The following study is based on two different kinds of data. The first consists of dozens of conversations with farmers, usually in impromptu situations, where I either listened to a conversation about sensing or where I was able to follow-up a passing comment about sensing with a question or two to ground the utterance in a fuller, albeit analytical, context. In some instances, when the conversation was taking place already among a group, I was able to step back from the question and follow as the group considered the matter largely among themselves: while they were always aware of my presence — they are not impolite men — they often warmed so much to the discussion that it became theirs. The other kind of data involved my embarking upon a series of what can only be called agricultural ride-alongs. I was able to do so because in most instances farmers use the largest, most powerful tractors for this kind of work, and the larger cabs of these tractors often have a jump seat of at least space for an observer to perch, although not necessarily a comfortable space for the hours one sometimes spends sitting there.

I should also note that because this work is so important and the machines involved so large, powerful, and expensive, the farmers mostly do this work themselves. It is not left to hired hands nor junior family members There is considerable skill involved.

A farmer depends upon his skill to level a field, and it is also the case that his skill is subject to a wider, often quite objective evaluation. Internally, a farmer who does not how to “feel the seat” — as an acute awareness of one’s machinery and the task to which it is addressed is called — spends too much time either looking at gauges in front of him or looking behind him to double-check the state of his plow. Looking at those things distracts him from seeing where he is going, resulting either in inefficient plowing or in very slow going. Externally, any field can be, and usually is, observed by other farmers. Uneven fields reveal themselves by changes in color and height or rice. A mottled field raises questions and comments in nearby equipment sheds, in agricultural supply stores, and after church. Its opposite, a field uniform in color and height, receives appreciative nods and comments.

Such an outcome depends on a profound knowledge of the topography of the landscape, including a sense of the underlying geology, as well as a highly attuned sense of one’s equipment. Mediating the relationship between the equipment and the landscape is a function of a collection of abilities and sensitivities, what Frake terms “a high order of cognitive ability,” that makes it possible for a farmer to know how much dirt he has in his plow, where he is in a field, and how far he has to go from a high spot, a hill, to a low spot, a hole, calculating the overall loss of dirt for the distance that must be travelled between those two points.

#### Defining the Task

The gentle topography and relatively thin layer of topsoil of south Louisiana make it a good terrain for rice, if, unfortunately, not good for much else. Farmers work the land to the best of their abilities, using the tools they have, many of which are, as I have detailed elsewhere (and look for the forthcoming book!), hand-made in nearby fabrication and equipment shops to their specification. One of those pieces of equipment is the water plow, a large blade pulled behind a tractor to grade a field level.

Each portion of a field, a step in a very shallow terrace, is called a cut, and the goal is for each cut to be as level as possible. Should one portion of a cut be higher or lower than the rest, the cut cannot be reliably flooded or drained when the time comes. The difference in height between cuts can be as little as a few inches or as great as a few feet. The difference allows one cut to drain into another until the water reaches the bottom of the field where it drains into a ditch, a coulee, or sometimes back into a canal, where it will get recycled. (During dry periods, water is pumped up from either wells or canals into the top cut.) The flow of water is managed individually either through drains that are opened or closed by hand or through drains that allow a farmer to set how much water to hold on a field. Fields will be filled or drained at various moments in the growing season.

Level cuts within a field are all important to the field’s overall functioning. The traditional way to level a field is to water plow it. Like its name suggests, water plowing is done with the cut “flooded up.” With water anywhere from calf-high to thigh-high, a farmer will drive into a field with a water plow attached to his tractor. Plows run from twenty to thirty feet in width, with the larger plows now possessing wings that can be raised and lowered for transport along area roads. Once in the field, a farmer will make a few rounds to establish the overall nature , or feel, of the cut, where it is high and where it is low. Before the arrival of the laser level, this was done entirely by feeling the way the tractor pulled the plow through the field and then turning to see how much mud and water was in the plow.

In the present moment, most farmers attend to the difference between a stationary laser transmitter stationed on the side of the field and spinning out an invisible, but level, plane of light and a receiver attached to the plow. A console in the cab of the tractor reveals how high or low the plow is to the norm set by the transmitter.

“Zeroed in”, either by feel in the past or a combination of feel and gauge in the present, a farmer begins the job of leveling a field which he cannot see. Already his circuits around have muddied the water. Now he drops the water plow blade into the water and proceeds to pull it this way and that. Sometimes he moves across the width of a cut, and sometimes up and down its length. Sometimes he moves diagonally and sometimes he goes around and around. The entire time he is, yes, keeping an eye on the laser level readout, but he is also feeling his way around the field. The goal of this exercise is to “pull down” unseen hills and “pull up” unseen holes.

The question before us is what cognitive devices are in use as a farmer navigates his way through cut after cut, slowly resolving a natural landscape into a series of artificial and thus highly productive planes. The answer, in our case, is that cognition is not “all in the head.”

[Expansion point here for fuller essay: review of Oatley’s scheme and its application here. Transition from cognitive model to cognitive devices needs to be made.]

#### Embodied Cognition

Viewing the enclosed cab of an eight-wheeled, articulated tractor from the road, one imagines that the operator is, if not quite a disembodied mind dully driving this way and that, then at least so alienated from the interaction between machine and landscape as to rely mostly on visual cues and the scant few sounds that make it past the roar of the engine and the insulation of the glass windows. Nothing could be further from the truth. Having ridden extensively both in these giant tractors while farmers plowed as well as in combines while they harvested rice or soybeans, I can safely attest to the fact of how little they actually pay attention to any and all gauges and readouts that report engine RPM, grain flow, or the height of grain in a hopper. Instead, farmers are constantly “feeling” and “listening” to the machines in which they ensconce themselves in order to get work done.

Recent research in psychology into haptic interfaces has mostly focused on how adding various kinds of vibrotactile cues can aid operators of cars or deep sea divers in processing diverse flows of information. For many, the ability to add a vibrotactile device to a car or to a watch is a way to overcome the visual overload many operators navigating complex environments feel. Ensconced within the tractor cab, we can delineate two distinct kinds of information:

The first has to do with the ground which passes underneath them unseen. Previous plowings of the field or the running of a crawfish boat can often result in ruts being left behind. From the farmer’s perspective, these ruts are undesirable in a rice field, since they can mean low spots where water may get trapped or they may, if long enough, drain the field inappropriately. In either instance, the ruts disrupt the farmer’s ability to control the water level in a field with the kind of granularity preferred. These ruts are felt as small, sudden drops in the body of tractor, and their width is gauged by a concomitant jolt. Most farmers have a very acute sense of the speed of their vehicle and thus typically a fairly good idea of the distance traveled between two moments in time. It also helps that they have had this ability to gauge distances and dimensions reinforced by knowing the width of a rut created by a crawfish boat wheel or by another kind of plow: these two kinds of information, one visual, but in memory, and one tactile, in the present, are combined in the moment of water leveling to afford them a high degree of precision. Depending upon the depth of the rut and the overall fit of the tractor, there may be a concomitant sound made by the tractor, which might also be felt. A tractor with a somewhat loose fitting somewhere, for example, will make a distinctive clunk, which many farmers will listen for, often knowing that the clunk is only prompted by changes in depth of a certain size or kind.

The second set of vibrotactile cues, which are also accompanied by a sonic cue, are produced by the tractor’s engine and reveal to the operator the degree to which the engine is under a load. Farmers typically describe this as feeling or hearing the engine strain, and it is, I confess, one of the more nuanced moments of perception that I have come across in my years of research: there is little to no obvious change in the pitch or the volume of sound these large, diesel engines make. At three hundred fifty horsepower or better, the engines in these tractors are capable of pulling a water plow through the water with relative ease, and it is not unusual for them to be doing so at extremely slow speeds. Because the plows can push so much water in front of them, farmers must work at slow speeds in order to make certain that they do not spill, or slop, water over the small levees that outline a field. During leveling, the water is rich in topsoil, and given the thinness of the layer of topsoil, sometimes only a matter of inches, in the area, any loss is considered needlessly wasteful. Thus, the larger engines are run at what almost seems an idle, heard and felt as a low rumbling. As the plow being pulled picks up water and mud, however, the engine begins to work a bit harder, and farmers listen and feel for that moment when, perhaps, the engine will need to be fed a bit more fuel.

In both instances, the farmer is highly attuned to the tractor. They described this process in two different stages. The first stage occurs when a farmer is just starting off, just learning how to farm, how to work with equipment. As a teenager working with an older family member or friend, typically fathers and sons but sometimes uncles and nephews, a farmer has to learn to “feel the seat” as one young farmer told me. It is a matter of learning how to feel the bottom of a field with the tractors’ tire, the young man noted, and in doing so reached out and down with his arms and spread out his fingers, as if he were imagining himself crawling through the water, feeling with his hands to determine how the land lay.

A farmer learns these things on a particular piece of equipment, and so the second stage occurs when he transitions from one piece of equipment to another, because each piece of equipment has its own feel, not only as a piece of machinery but also as a sensing device. Another farmer who had recently purchased a John Deere tractor after using nothing but Case tractors for twenty years noted that it was going to take a great deal of getting used to, “[the John Deere] tractor runs different, works different.” The same observation occurs when a farmer has gotten used to the feel of a particular brand of equipment and that manufacturer makes a significant change to the drivetrain, the suspension, or some other facet of the machine that requires the farmer to “re-calibrate” their senses.

#### Possible Conclusion

This way of thinking about the relationship between the thinking we do and the things with which we think is described by Edwin Hutchins as “distributed cognition.” An anthropologist, Hutchins has sought to bridge the gap between his own field and psychology, between culture and cognition as objects of study. Conventionally, of course, the two are considered distinct areas of inquiry, but only, as Hutchins observes, because the boundary between inside and outside have been so firmly drawn, which “creates the impression that individual minds operate in isolation and encourages us to mistake the properties of complex sociocultural systems for the properties of individual minds” (355). Hutchins’ argument is that cognitive sciences have over-allocated intelligence to the inside of human subjectivity. The problem with such a view from his, and we can now also say from Heidegger’s, point of view is that it mistakes, potentially, one dimension of a larger system for the system itself.

Hutchins notes that John Searle’s “Chinese Room” thought experiment offers a nice encapsulation of the larger problem. In the thought experiment, Searle sets out the following scenario: he is locked in a room where messages in Chinese are slid under a locked door. He himself has no knowledge of Chinese, but he does have a book which allows him to determine the character sequences and to respond with a correct sequence of characters that he then slips back under the locked door. The outside observer perceives a meaningful reaction, but, given Searle’s role in the communicative instance, was there really any meaning? Searle’s response is not, and he intends the thought experiment as a rejection of the idea that the Turing test could gauge actual intelligence.

Searle intends the thought experiment, Hutchins points out, “as a demonstration that syntax is not sufficient to produce semantics” (361). But in setting up his experiment, what Searle has done is encapsulate a “sociocultural cognitive system.” On his own Searle cannot communicate but as an ensemble, he and the book in the room, can. That is, “the cognitive properties of the person person in the room are not same as the cognitive properties of the room as a whole” (362). Hutchins argues that much of the work done in artificial intelligence and in cognitive psychology consistently focuses on socio-cultural systems but mistakes them for individual minds. He concludes that the attribution to an individual mind of an entire system effects a kind of surgery in which interaction, and our chief means of interacting, our bodies, are removed.

Reduced so, the unhooking of cognition from interaction becomes clearly absurd. Hutchins responds that what we need is to study more cognition as it occurs in the world and study cognition less as a limited set of responses from an individual isolated in a laboratory. He proposes the term “cognitive ethnography.” Returning to some of the language used by Frake in his own description of cognitive psychology experiments, I am struck by the occurrence of performance, not just the use of the word but that it is used in ways folklorists would easily recognize:

We are concerned here not with judgements about the mentality of an age or the wisdom of a culture, but with the cognitive abilities of individual human beings. For evidence we must turn away from assessments of the strangeness of a culture’s beliefs or the weirdness of its symbols to an examination of performances that can be seen as displays of cognitive ability. But what counts as such a performance? Probably most things a human being does should count. The problem for the investigator, and sometimes for the performers them- selves, is to know what the performance is. “What’s happening?” Or, in psychologists’ language: “What is the definition of the task?” (255)

Frake notes that psychologists prefer to define their own tasks and remain anxious about user-defined tasks as being vulnerable to collusion. Folklorists and others who are used to working from the inside out see this less as a vulnerability and more as a matter of openness.

Such an openness to the “task world” allows us to form different understandings of what people do with their minds. But, as we have seen, “mind” must be broadly understood. And, to my mind, we must also necessarily be more open to the disciplines with whom we collaborate. Richard Bauman once noted:

that the enduring importance of the intellectual problems that the philological synthesis was forged to address constitutes a productive basis on which we as folklorists might orient ourselves to our cognate fields and disciplines. In my view, any scholar who is interested in any of the dimensions of interrelationship that link language, literature, culture, society, politics, and history together is potentially my colleague, whatever our degrees and whatever academic departments provide us a home and a living. (17)

Bauman is, of course, referring to the work that was begun as the ethnography of speaking and was later consolidated under the rubric of performance. It was, by the accounts of some of its vanguard practitioners, an attempt to take ideas and issues raised by philosophers like Heidegger and others working in the middle of the twentieth century to re-ignite the investigation of human being and to apply those insights within fields who had traditionally focused on the “other” of modernity. Forty years later, the new philology now has the opportunity to re-join philosophy as it itself has been transformed by studies of cognition across a wide range of fields.

[Dell Hymes was also part of cognitive anthropology at various moments.]

#### Sources

Bauman, Richard. 1996. “Folklore as Transdisciplinary Dialogue.” _Journal of Folklore Research_ 33 (1) (January 1): 15–20.

Frake, Charles O. 1985. “Cognitive Maps of Time and Tide Among Medieval Seafarers.” _Man_ 20 (2). New Series (June 1): 254–270.

Hutchins, Edwin. 1995. _Cognition in the Wild_. Cambridge, MA: MIT Press.

Oatley, Keith G. 1977. “Inference, Navigation, and Cognitive Maps.” In _Thinking: Readings in Cognitive Science_, ed. P. N. Johnson-Laird and P. C. Wason, 537–547. Cambridge University Press.

Rizzolatti, Giacomo, and Luciano Fadiga. 1997. “The Space Around Us.” _Science_ 277 (5323) (July 11): 190.