« Gardner, A. 2007. An Archaeology of Identity: Soldiers & Society in Late Roman Britain, Walnut Creek: Left Coast Press. | Main | Michael Shanks' intervention into Tara 2009 »

‘Epistemography’ and Archaeological Assembling. A Manifesto for Media.

Posted by Timothy Webmoor

Archaeology, Science and Technology Studies, University of Oxford

In 1922 the Mexican scholar Arreola published a study of maps and images which he had recovered from archives in Mexico City. Much of the material that he presented had not been studied before. Much of it was quite old, some of it dating to the initial conquest and consolidation of Mexico by the Spanish. One of these images was very old, even for being a copy of a lost original. It was called the ‘Mazapan Map’ and the original was estimated to have been rendered around 1560. It was part of 16th century records of farmlands and land ownership. The use of Nahuatl glyphs, the pictographic-ideographic language of the Aztecs, designating holdings and landowners, suggests it was most likely commissioned by the Spanish as part of reconnoitering their newly expanded empire.

'Mazapan Map'

The ruins of Teotihuacan cover the bottom portion of the ‘map’. The Pyramid of the Moon is at bottom left and the Pyramid of the Sun is at the bottom center. To the bottom right, the large, open rectangular shape of the ciudadela (1).

A ‘map’ complementary to the Mazapan was published two decades later in 1580. It was part of the relación geográfica de San Juan Teotihuacán. This map, rather than landholdings, emphasizes imperial infrastructure: the Spanish grafted over the Aztec. The road network emanates from the Aztec (and later Spanish) administrative center (Tenochtítlan) to the regions on the north of the Valley of Mexico. Like the Mazapan map, it also orients north to the left for the map-reader. Tenochtítlan can be seen at the crossroads (center right). The ruins of Teotihuacan are shown (highlighted in box) near the center, approximating the correct geographical relationship to the administrative capital, with the distinctive layout of the Pyramids of the Sun and Moon and the avenida de los muertos (avenue of the dead) outlined by smaller structures.

'Map' from the relación geográfica de San Juan Teotihuacán

Rather than Nahuatl glyphs, the map’s labels are presented entirely in Spanish. A testament to the rapid changes in the intervening twenty years of Spanish domination of the region. The ruins of Teotihuacan are labeled Moctezuma’s oracle, which the accompanying text of the relación explains is a place of pilgrimage for the Aztec ruler and his priests. A place to offer sacrifices. The Aztec oracle is mentioned in other post-Conquest texts, and this map suggests that Teotihuacan was the place of the imperial prognosticator (2).

Now, fascinating as both of these maps may be, contemporary archaeologists at the site do not unfurl them when they lay out new excavation trenches. Nor would they use them for navigating the complex and monumental ruins. Indeed, art historians show more interest in them than site archaeologists. Simply put, they are not good maps.

In this short paper I want to pursue why they are not considered proper maps. This could take us down the road of theories of correspondence, art historical analyses of single-point perspective, the development of the astrolabe and other instruments or even information design - all great pursuits. I want to lodge their consideration, however, with why quotation marks bookend these maps: epistemology. To do this I am going to unpack several closely related propositions.


1) Epistemology has been wrongly scapegoated.
2) An archaeological commitment to things, to ontology, resuscitates its alter ego as ‘epistemography’.
3) Knowing the past is assembling past.
4) Epistemography is the past made durable.

To unpack this series of propositions I will come back to these proto-historic, historic and contemporary maps of Teotihuacan. These media cascades aid in manifesting the practice of archaeological assembling and the principle of ‘epistemography’.


Epistemology is as troubling as it sounds. Perhaps this is why the term has dropped out of common parlance amongst archaeologists – and for that matter, amongst social scientists. It is just too contentious and awkward – a new twist on the mantra: ‘an inconvenient truth’. And for good reason. If we have not personally and professionally come through the ‘science wars’ of archaeology – many of us are too early in our careers – then at the least most of us are familiar with the drama, posturing and exchanges in the theoretical literature. We’ve had our initiations into the ‘isms’. Optimistically we might say that these theory wars refined our respective outlooks on how to best engage with things from the past.

Positive no doubt. We could indeed trace trajectories of how these skirmishes, these family feuds –most vociferous in the states and UK - have been passed on to us as our disciplinary inheritance. And many provoking questions might be asked. For instance, was the inconclusiveness of the debates largely the result of an expanding academy? Did the rising power of the ‘pc’ movement muffle academic disagreement. While these questions would certainly touch upon the history and contemporary practice of archaeology, a tangible outcome of the theory wars has to due to with our discipline’s boundaries and policing. Something most savvy archaeologists are more than aware of with work engaging contemporary ruins that transgress canonical categories.

My proposition is this: epistemological considerations were scapegoated. And a less acknowledged consequence has been to drive epistemology from explicit consideration. To force it underground. Backfilling over it as an issue that had been dealt with, dismissed as unhelpful – even boring.

So why would I resurrect such a tedious topic. I am not going to defend it. At least not as it was adopted from the ‘handmaidens’ to the sciences. I am urging a reconfiguring of epistemology to fit matters of common concern for archaeology.

As it is, archaeologists appropriated epistemological concerns from the estate of philosophers of science, only to closet this new explicit (reasoning) identity within a closed architecture.

Consequently, these adopted values of ‘what is to count’ continue to exert real influence. From editorial boards, funding agency boardrooms, departmental offices, publishing houses, social relationships. Judgments informed by long-standing epistemological criteria are still very much a part of archaeology’s mundane practices of governance. Even if they are fuzzier for not being articulated in print and debate. Lodged in a bureaucratic pathology of sorts. And there remain real tangible consequences for practitioners.

Now there have been, to be sure, a few efforts at manifesting the operation of epistemology in the more recent literature. Upgrades for the archaeological process to bring it current for a more philosophically diverse, savvy and cosmopolitan archaeology (3). However, it is a commitment to ontology, to matters of concern, which offers archaeologists the best tool for exhuming epistemology. But this Lazarus must take on new form.


Recently there has been a call to turn (or return) to things. Some may suggest this comes from Continental currents of thought as analytic orientations abate. Yet there are good reasons to dismiss this claim. Instead, spread roughly across the same academic topology as the earlier wranglings, disciplines ranging from anthropology and literary studies to philosophy and science and technology studies have pushed the idea of ‘taking things seriously’. This has been a much needed corrective to the postmodern paralysis. A hyper-interpretivism that stranded many areas of inquiry in a semiotic exile in the wake of epistemological grounding.

To be sure, there are subtle distinctions that manifold deep differences with respect to how the terms things, materials, materiality, material culture and even 'stuff' - for 'stuff it' theory - are used by archaeologists.

For my purposes, archaeology’s commitment to things, far from denuding archaeological explanations, is integrally involved in the transformation of ontology’s alter ego.

Epistemology has been deflated. In its stead, epistemography has been a more or less explicit component of the return to things. The close attention to scientific practice, launched by those students of science studying how work is accomplished, drew attention to the real content of scientific knowledge. Mundane media and practices involving their enrollment for making statements. ‘Circulating reference’, ‘lateral or serial relations’ or the ‘crafting of resemblances’ are rich descriptive accounts of how we move as archaeologists amongst sites, features, landscapes, artifacts, and our archives and media. With media made amenable to publication, presentation and calculation – to media made mobile. Through this process we make knowledge claims. The stuff epistemology was meant to study.

As it turns out, attentive description to this process re-oriented the idea of epistemology. It was all so close-to-hand that epistemology lost the detail (of knowing the past) for the forest. It took its 'eyes' off the action. Attention to non-lofty workings of archaeology and science may strike some as too simple. But I believe this has been beneficially humbling. Very small steps do the heavy epistemological lifting for science. This is a crucial Archimedean point for the discipline of things.


Epistemography is attention to these humble actions of people and things. Following how their relations form assemblages that endure - for the short or long-term. It is engineering. Let me be careful with this statement as I do not want to be confused for suggesting we construct knowledge – or even worse, that we ‘socially’ construct knowledge (an unhelpful adverb).


Digital Media Work: ratios of mediation

Transforming the dynamic and materially complex ruins in the landscape into media artifacts gathers certain qualities while sieving away others. This is a pragmatist perspective. It is not about fidelity or faithfulness. It is about accuracy and precision. Necessary to get the past to work. From this perspective, ‘epistemology’ (now bookended by quotation marks) is at the core of what archaeologists do everyday: we assemble the past to understand and engage it.

Let’s turn our attention back to the series of maps that have rendered the same ruins for nearly 450 years.

Reconnaissance survey and mapping of this archaeological zone took the better part of the 1960’s. The Teotihuacan Mapping Project’s (TMP) map was finally published in 1973. 147 maps at a scale of 1:2000. It covered the entirety of the architectural and artifact concentration - well over 20 square km. Map #1, or the ‘Millon map,’ became a rock star poster for archaeological mapping and cartography. Now, we can approach this map from a static, atemporal perspective. The TMP map as representation, removed from its relations. An independent referent or snap shot of the site – a static slice of cerca 1968.

Now epistemology would incline us to regard it as just such a stand-alone output. But a commitment to things and their relations with other participants – understood without an asymmetrical and arbitrary suture between bearers of ‘agency’ – reveals the map as gathering. Such a concern opens the black box and reveals a more messy middle. Not just a flat, two-dimensional analog of the site in analogue.

Mixed-maps of Teotihuacan (4)

Epistemography follows the relations that pass through the map like circuitry. It would ask not whether it is verisimilitude. But how well are these relations bound together? How sturdy is this feat of engineering? In fact, the TMP is formidable media architecture. It is built for future generations. It aligns not just future archaeological media, which hang upon its scaffolding, but also contemporary and future engagements with Teotihuacan. Since 1973, all archeological excavations and surveys have been conducted with relation to the TMP. From Sugiyama and Cabrera’s excavation within the Pyramid of the Moon, to Cowgill and Sugiyama’s exploration of the Temple of the Feathered Serpent, to Cabrera and Gomez’s revealing of the apartment compound of La Ventilla. All inserted in the extensible media of the TMP map. More importantly, action, people, institutions and instruments – actants – distant in time and place are nonetheless all gathered up with the TMP map.

In terms of the (problematic) notion of cultural heritage, the map continues to coordinate non-archaeological engagement with the ruins. Beginning with establishing boundaries of the site. A collection of project archaeologists, local municipal leaders, Instituto Nacional de Antropología e Historia (INAH) officials, construction crews, fencing materials, heavy equipment, survey stakes, legal decretos or decrees and the map, expropriated well over a hundred residents from their land and pushed agricultural tenancy off the 263 hectares. And the map continues to coordinate how tourists tread, when vendedores sell their wares, and where toltec shamans with L.A. hippies bribe skeptical guards. Thirty years after its publication, this map was enrolled in even stranger relations. Sam Walton and Wal-mart, Jerry Yang and Yahoo!, Carl Sagan and laser time capsules. Eclectic collectives bound up with the TMP map at different times, for different reasons, and to different effects. An ecology of relations; complexly inter-dependent, overlapping in space and time. Yet all anchored to this archaeological ‘habitat’. A ‘heritage ecology’ involving all. And subsequent to 1973, all predicated upon, acted with, and aligned to this media artifact. This product of an archaeological intervention in the 1960’s.

We could, with more historical documents, follow the cascade of actions which ripple through these media. To see how well earlier maps gathered relations together. As we’ve previously considered, the 16th century maps would not have been considered maps. Not from an epistemic tradition that wants mimetic media. Unpacked by epistemography, however, these maps were enrolled in a host of actions. Territorial disputes, the determination of Aztec pilgrimages, the building of the Spanish Empire’s infrastructure, the guiding of its imperial governance. They also had a role in coordinating future engagements by antiquarian explorers. Explorers who grafted subsequent renderings of Teotihuacan into this growing network of visualizations. Romantic figures such as von Humboldt, Brantz Mayer, Désiré Charnay and Ramón Almaraz.

1844 Brantz Mayer map

1857 Désiré Charnay map

1865 Ramón Almaraz map

Teotihuacan Media Cascades

Throughout the 19th century, more and more maps were produced of the ruins. Each one assembled for different purposes. Drawing together different sets of relations and involved in heterogeneous actions.

We could consider any one. Marquina’s map of 1919 for instance. This map, perhaps more than any other, was integral to the trajectory of land expropriation of locals and set up the site’s management. Conspiring with an already dilapidated livestock fence to restrict inscription of Teotihuacan to the central area of the site. The central ceremonial district. A action that reverberates today through a network of actants extending to Bodega Aurrerá (Wal-mart Mexico) headquarters in Mexico City or Yahoo!'s Sunnyvale California campus.

Unlike the TMP, however, these other maps were not similarly engineered to align future media. Was this because they were deficient as 'representations'? Classic correspondence theory would say yes. They were certainly not Euclidean. And they were assembled without many of the qualities noted above. No modularity, universality, compatibility. They were not maps without quotations marks. This conclusion would close down the other sets of relations that these maps gather as an archive of Teotihuacan. Such an epistemic stance would sever the blood flow that innervates heritage at a site such as Teotihuacan. It would turn these maps into dead media.


To conclude, I want to highlight why I am a fan of the TMP. Why these other maps, with or without quotation marks, are not the same for an important reason. For contemporary archaeologists, locals of the valley, heritage managers, shamans and UNESCO delegates, they are simply not as durable.

Less durable, they sustain less action. Epistemography moves our attention to this action of assembling people and things from the past and present. This is how archaeologists are taken seriously. The more heterogeneous, ‘heavy’ (with ‘actants’ and their relations) and useful the assemblage, the more (epistemic) weight it carries.

Epistemography is the Lazarus of epistemology. An epistemology centrifugally centered upon fidelity, coherence and correspondence. A troublesome taskmaster indeed. Asking archaeologists to search in abstraction for suitable principles. Criteria which were endlessly debated, never sufficient-in-themselves, and were finally dismissed – or at least closeted.

With a care for things, there is no divide between knowing and doing. These are heuristic, therapeutic terms: ontology, epistemography, pragmatology. And we shouldn’t set them up as new principles. But they bring our archaeological attention back to the ground. Back to mundane matters, back to the relational action of people and things. Gathering up relations, removed in time and space, these assemblages – like maps – are ‘living’. That is, they have dispersed, distributed, but nonetheless definite action. Essence performed not a background condition. An anthropocentric existentialism reconfigured for all.

Epistemology’s Lazarus is livelier for the resurrection.


(1) The glyph next to the upside down, stepped image of the Pyramid of the Sun (bottom center), does not identify it as the “tower or hill of the sun” (whereas a glyph next to the stepped image at the lower left identifies it as the “tower or hill of the moon” [ytzacual metzli]), but rather indicates that the monument serves as the boundary of agricultural fields. Similarly, the ciudadela is identified as the “place of burials in honor of the sun” [tonali itlaltiloyan], possibly indicating an awareness of the mass, dedicatory burials beneath the Temple of the Feathered Serpent. The large personage in the center of the ciudadela (lower right) is labeled as the current landowner of the ciudadela. Who possibly owned the majority of the site in 1560.

(2) I always find this connection reassuring somehow: there were 'new agers' at Teotihuacan 450 years before I studied their engagements as forming one 'niche' of the overall heritage ecology of the site.

(3) Binford's "relative objectivity," Shanks and Tilley's "partial and contingent objectivity," Hodder's "guarded objectivity," and Wylie's "mitigated objectivism." These hybrids of post-analytic and continental currents of thought have been instructive. But they are little discussed outside cliques of philosophically inclined archaeologists. More importantly, they rely upon, reproduce and reify a presumptive split between people and things.

(4) The TMP or 'Millon map' with mixed maps of subsequent projects. Scales are relative to each map. Sources for adapted maps:

Millon, René, Bruce Drewitt and George Cowgill. 1973. The Teotihuacán Map. Vol. 1 part 2. Austin: University of Texas Press.

Serrano Sánchez, Carlos. Editor. 2003. Contextos Arqueológicos y Osteoloía del Barrio de La Ventilla (Teotihuacan 1992-1994). México, D.F.: Universidad Nacional Autónoma de México.

Sugiyama, Saburo. 2005. Human Sacrifice, Militarism, and Rulership: materialization of state ideology at the Feathered Serpent Pyramid, Teotihuacan. Cambridge, UK: Cambridge University Press.

Sugiyama, Saburo and Ruben Cabrera. 1999. Proyecto Arqueológico de la Pyrámide de la Luna. Arqueología 21:19-34.

____ 2007. The Moon Pyramid Project and the Teotihuacan state polity: A brief summary of the 1998–2004 excavations Ancient Mesoamerica 18:109-25.