Key Pages- |
Changes [Sep 16, 2015]Home
Dispelling representation - mediation as practice
A pragmatic sensibility, which as I argue already pervades archaeology albeit in a non-explicitly acknowledged form, suspends the irresolvable epistemological worry about mirroring reality by shifting evaluation outside - to the end or outcome – of the representational equation. Pragmatists do this by affirming, phrased within traditional nomenclature, a ‘natural’ or ‘direct realism’. But I need to specify why, beyond the evident dead-end avenue of relativism and objectivity, we may be reassured that in discarding epistemology and taking a third, less traveled path we are on the right track. Pragmatic thinkers, emphasize practice and useful outcome. But what does this mean for archaeology? For this we must turn from pragmatic thought to a complementary group of thinkers working in the trenches of the science wars – a war itself which, much like archaeology, spins endlessly around the dichotomous fulcrum of humanity-science, realism-antirealism, subject-object.
Emerging from the break-up of Philosophy of Science into more ‘applied studies’ of what scientists in various disciplines actually do, Actor-Network-Theory (Callon 1986, 1997, Latour 2005, Law 1999) more generally and Bruno Latour (1987, 1993, 1999, 2004) most prominently attend to ‘thick description’, ethnographic approaches to scientific practice. His revolutionary insight is that modernist epistemology must indeed be suspended as the constituent elements upon which is founded – knowing subjects and known world – are in fact erroneous. They are not in fact there; no autonomous, independent subjects building epistemological equations to ford the divide to objects ‘out there’. Complementing the pragmatic thinkers historiography of a misguided, ill conceived modern epistemology, Latour argues that the inheritance of modernist ontology is equally unwelcome. The ‘modern constitution’, developed as it was by the history of dichotomizing thought made infamous by Descartes, established the division of Nature from Society as the two purified building blocks of the world (Latour 1993:13-15) Figure 3.5.
Beginning and building from these distinct, basic blocks of existence, Latour traces how such a modern constitution gave rise to a host of derivative distinctions, particularly that between humans and things. Such ontological categories were slotted into one pole or the other; into the Nature realm or into the Society realm, or a 'hybrid' somewhere in between these rarefied realms. No wonder, for Latour, that such a constitution quickly developed the need for an appropriate modernist epistemology. Such an epistemology needed, for the moderns, to establish connections between the realm of humanism to the realm of naturalism Figure 3.6.
Individuals on the society side of the constitution made representation of what was across the divide on the nature side of the constitution. So that for the moderns, representing is knowing. As we have already seen, such a modern constitution and its modernist epistemology subsequently gave rise to a host of settlements for ‘polishing the mirror of representation’ via logical principles.
Such a historiography of modernist thought is in no sense unique. But where Latour distinguishes himself is drawing upon his attention to scientific practice, whether historical (Latour 1988) or contemporary (Latour 1987, 1999:ch.2, 1979), to argue that such a deeply entrenched constitution for how we think and act in the world is deeply ironic. As an anthropologist, Latour suggests that what we say we do as moderns is quite different from what we actually do in practice. This is because modern science, in practice, begins with mixtures or hybrids of nature-culture and, far from purifying and entrenching the constitutional division, proliferates these ‘quasi-objects’ which only artificially can be separated or sifted-out into nature and society (Latour 1993:1-3,52). In a sense, he feels we live under a false ontological consciousness.
To illustrate the ubiquity of networks of humans and nonhumans incapable of being purified into either the nature pole or society pole, Latour steps us through what scientists do in articulating knowledge. He does this to debunk the modernist constellation of ideas which begin from the assumption of divisions, whether ontological or epistemological. As he states, “words and worlds do not represent two statures facing one another and marking the respective territories of two kingdoms . . . rather, words and worlds mark possible and not very interesting extremities, end point of a complex set of practices, mediations, instruments, forms of life, engagements, involvements through which new associations are generated” (Latour 2003:39). In his ‘Circulating Reference’, Latour (1999:ch2) tackles both mutually supporting notions in a single example of fieldwork by botanists and pedologists in the Amazonian region of Roraima. To fulfill their project goal, the researchers must produce a report detailing the whether the savanna is encroaching upon the forest, or if the reverse is taking place. The report must accurately portray the state of savanna/forest battle in order that inferences as to cause may be generated and supported by the evidence. However, in creating the map replete with locations of botanical and soil samples, Latour notes that instead of mimesis of representation, of a report mirroring what is actually happening ‘out there’, what actually transpires in practice is a series of translations across tiny gaps of material to media. “In actual practice, however, one never travels directly from objects to words, from the referent to the sign, but always through a risky intermediary pathway” (ibid:40). So for instance of this intermediary pathway, the scientists taking soil samples must transubstantiate or translate a given piece of soil into a code on a Munsell soil chart. The soil’s moisture content, the illumination of the location, the condition of the chart itself, and of course the capability of the researcher to judge fine shades of color all inter-act to create, in the end, a color code. For Latour, this is not correspondence or mirroring of the world into representation. A series of actions on the part of researchers and savanna soil are coordinated, are articulated in order that the code will do more than resemble the soil: “it takes the place of the original situation” (ibid:67). It takes the place of the soil so that it can be integrated with other information collected while in the field – it must be inter-fungible with graphical notation, statistical nomenclature, and textual narrative. Instead of a correspondence of the code to soil, we have an actively manipulated transformation for specific purposes – for standardization Figure 3.7.
Latour describes this entire process of moving across a series of small gaps as circulating reference Figure 3.8. And, as we no longer are talking of society mirroring nature, but rather of a heterogenous series of operations which are enabled by considering the transformative capacity of the researcher and his/her media as well as the unique capacity of nonhumans to foster, inhibit or qualify this transformation, it is better termed mediation.
Mediation as the ability to transform via chains of reference a Brazilian landscape, or an archaeological feature or site, comprises more actors than the modernist constitution traditionally allows for; the entire process or event requires the negotiation of a host of actants (human and nonhumans) in order for the successful coordination of actions. Furthermore, to mediate, as opposed to ‘represent’ with its historical connotation, conveys the active, ongoing nature of such a process. As Latour summarizes the process, humans can no longer be assumed to be the prime mover of action, but instead a ‘distributed and nested series of practices whose sum may be possible to add up but only if we respect the mediating role of all the actants in the series’ (Latour 1999, 181).
Let’s take an example of closer relevance to the topic of this dissertation to illustrate how Latour’s chain of ‘actants’ span the divide of correspondence theories of representation. Mapping at Teotihuacan, Mexico had been a fundamental technique for organizing and visually registering knowledge of this early and complex urban center of Mesoamerica. Likewise, while the above example discussed soil samples and provenance, the map is unquestionably the most important visual medium for archaeology. Maps function intimately, as seamless extension of how humans, as spatial, embodied beings, are hard-wired to think and act spatially - as some have argued in cognitive psychology (Kitchin 2002). Yet they are also overarching, ‘meta-media’, or ‘architectural media’, in that they enable, generate and organize other information rendered in visual or textual media – such as stratigraphic profiles, feature drawings, transect locations, field diaries and any other type of archaeological information that requires spatial- or geo-referencing for it to be ‘decoded’ for maximum relevance. Given these attributes, it is no wonder there is a popular and professional fascination with maps (Black 2003).
Furthermore, Teotihuacan has a long tradition of map making which can be traced through a heterogeneity of pragmatic goals. This is a topic which will be brought up in the subsequent chapter. So let’s consider the most precise and optically consistent projection of Teotihuacan to date: the exemplary survey map of René Millon, Bruce Drewitt and George Cowgill (1973). Published in 1973, the ‘Millon map’, as it has come to be known, manifested a systematic coordinate system (the first for Teotihuacan) for the archaeological site. At a scale of 1:2000, the map divided the city into grid units of 500m2, extending precisely drawn grid lines from the concrete and steel ‘zero point’ marker along the ‘center line’ (ibid:12-13). The monumental architecture itself has a stake in this far from arbitrary imposition of Euclidean coordinates: the central avenue (avenida or calle de los muertos) cuts a swath of more than 2 km thorough the densely urbanized zone at 15 degrees, 25 minutes east of astronomical north. Millon (along with a host of surveyors, colleagues, instruments, and media) oriented his ‘center line’ along this avenue and hence tied his maps coordinates to this alignment (Millon 1964) (Figure 4.1). The immobile features of the site were translated into the coordinate system of the map; the precise lines of the map become a graphical translation of the (less precise) alignment of structures. Thereafter, the coordinate lines, the metrical basis for the map, have a stake in the combined action of archaeologists with other things. It provides a basis for subsequent archaeological engagements with the site. Twenty years on from the publication of the Millon map, the La Ventilla excavations (e.g. Serrano Sánchez 2003, Gómez Chávez 2000 for full discussion) of the National Institute of History and Anthropology (INAH) enroll the Millon map in the delineation of their excavation units. Here, Millon’s block 91 (S1W2) is further divided into 100m squares and the northeast corner of the block, designated as the crossing point of x- and y-axes, ‘N1W1/N1W2/S1W1/S1W2’, becomes the primary datum for the excavation (refer to Figure 1in Serrano Sánchez 2003:52). In the process of laying out the initial trenches in excavation areas 1-4, coordinate, orientation, and line are translated into stake, orientation, and string. In the movement between paperwork and trenchwork; in the movement between the ‘abstract’ and the ‘concrete’, roles shift, the coordinate lines are replaced by threads of cotton which orient and direct the articulation of excavation baulks. The former, by virtue of its materiality, is delegated the task of maintaining a Cartesian grid, x- and y-axes, and thereby locating the trench on a flat, 2-dimensional surface. The later having taken on properties of the former—orientation, linearity and, hence, translatability into 2-dimensions—comes to share the role of maintaining the grid among its many other tasks (the baulk will, for example, also contain a profile of stratigraphic relationships). While these cotton threads and baulks co-direct the movement of the excavators and implements, they are also a practical outcome of relations between varieties of entities. Outlines of the structures of ‘Palacios A’ and ‘B’ are uncovered, and these also have a say in how the excavation proceeds. At this point in the excavation, there is a complex network of actants, of things and people, who all play a role in the course-of-action. And when primary excavations are complete in 1994, the dirt of the walls of excavation trenches and areas, the outlines of prehispanic structures encountered at La Ventilla, and even the ‘blank’ areas of undisturbed grass and nopal cactus, transform into the lines of the La Ventilla barrio map, appended to the crisscrossed axes of the Millon map. A full cycle of translation, or a series of shifting associations constitute the practice and result of archaeological research. Just as the qualities of the Millon map facilitated both the layout and transportation of features encountered in the La Ventilla excavations, these new lines of La Ventilla on the Millon map will have a stake in future projects at Teotihuacan.
Having the map will enable, not only certain decisions (such as where to survey, where to excavate, where to predict subsoil deposits), but also will likely ‘suggest’ these possibilities. Indeed, our aim has been to distribute agency (and so the anthropocentric term ‘agent’) to include the capacity of nonhumans such as the map to ‘persuade’ and ‘enable’ a course-of-action. The map works like a cognitive and sensory prosthesis in the hands of a map user (literally ‘mapwork’ in Webmoor 2005). In practice, in action (the pragmatic measure of knowledge Webmoor 2007)), the many seemingly discreet entities - archaeologist and map; map, metal stakes and cotton tread; metal stakes, cotton thread and excavation trench - swap many properties, confounding the identification of a locus of exclusively human sovereignty (for full discussion see Webmoor and Witmore 2008).
Forward to Chapter 3 conclusions