The central argument in Gary Urton's new book and its multiple ramifications unwrap with the elegance and precision of a khipu, the knotted-cord device at the core of the art of Inka record keeping and the main subject of this contribution. For 25 years, Urton has carefully studied the structural features and conventions of more than half of the 923 specimens currently inventoried at Harvard's Khipu Database Project, in what amounts to the most important effort to date to decipher Inka and early colonial-era knotted strings and interpret the histories contained therein. This study, built around a series of readings of museum and archaeological samples, sums up the main findings of Urton's two-and-a-half-decade scholarly endeavour.
With one exception, Urton carries out his study to a great degree independently from the two other major traditions of khipu studies: the analysis of ‘paper-khipu’ transcribed in colonial sources and ethnographic study of modern or patrimonial khipu. Nonetheless, Urton's decision seems warranted by the ultimate aim of his study: to become the first step towards a history of the Inka Empire that uses surviving khipu as primary sources. Urton contends that the history captured in khipu is fundamentally a history of ‘organizational structures’, not unlike the slow-changing structures that inspired the Annales school's ‘total histories’, with their emphasis on the intertwined analysis of three types of times: the longue, moyenne and courte durée. In that sense, History in Knots revisits the problematic relationship between structure and history that inspired some of the seminal works of the Andean ethnohistorical tradition.
Urton begins his examination of Inka khipu by pointing out that most specimens in the database come from the coastal plains and valleys, where preservation is easier, rather than the highlands. Further, it is impossible to attain absolute dates for pre-Hispanic and early colonial khipu (14C analysis places them within a c. 1450–1650 ad timeframe). These two caveats notwithstanding, archaeological and historical evidence do allow for an identification of many of the khipu sets and archives studied by Urton as ‘Inka’, that is, as pertaining to the roughly 150 years during which the lords of Cuzco ruled over large swaths of the Andes (c. 1400–1532).
Like other ancient empire builders, Urton reminds us, the Inka depended on reliable statistical data, which accounting specialists generated at local and state levels and recorded on khipu, in order to rule over Tawantinsuyu, the Empire of the Four Parts Bound Together. Beyond their obvious relevance for the development of the Inka state, numerical khipu (as opposed to the as-yet-undeciphered ‘narrative’ ones) also served a central purpose in the creation and maintenance of Inka hegemonic power. As they recorded population figures and assessed tribute-labour in administrative capitals and local villages, khipu specialists deployed, performed and even formalized a series of organizational principles embedded in khipu. These principles, in turn, reflected ‘the mayor institutions, structures, and practices of Inka state governance’ (p. 33). Khipus encoded basic principles of Andean socio-political organization, such as dualism, complementarity and hierarchy; it was this fundamental ‘compatibility’ of dualistic and hierarchical structures between imperial centres and local communities that helped extract the necessary consent from subjugated populations, making their cooperation with Inka rule appear beneficial and voluntary. The Inka achieved this complementarity through the production of matching or closely matching ‘copies’ (khipus that worked akin to double-entry bookkeeping) and the creation of an ‘accounting hierarchy’, comprised of lower-, intermediate- and higher-level cord masters who busied themselves with the related tasks of ‘summing up’ (aggregating) and ‘dividing down’ (apportioning) quantities, moving information up and down the khipu hierarchy of the Empire.
Nevertheless, like other record-keeping devices and practices before and since, khipu categories and patterning informed and ordered the social realities that cords were meant simply to record. Through a series of insightful, meticulous and mostly persuasive readings of khipu numerical contents, formatting patterns and structural properties, Urton suggests the ways by which quantitative khipu such as the ones found in Inka sites like Atarco, Puruchuco, or Inkawasi (all on the south-central coast of Peru) could have recorded semasiographic (that is, not tied to any spoken language) information. That information included quantities, but also ethnic identifiers and intra-group affiliation and organization about particular Andean ayllu (nested-like, corporate kin groups), moieties (anansaya and lurinsaya) and rounded-up decimal units of tributary households, such as pachaca (100s), guaranga (1000s) and hunu (10,000s). Moreover, the biennial calendar-tributary khipu of Laguna de los Cóndores (north-central Andes) and the pair of ‘matching khipu’ with opposite directionality, both studied by Urton in the book, also suggest the ways in which more complex Andean socio-political formations (a segment of the Chachapoya confederation and the ceque system of Cuzco, respectively) could have been encoded in knotted cords.
Urton's proposed matches between khipus and socio-political structures, though convincing and statistically likely, are not always free of anomalies, discrepancies and variations, as the author himself admits. Historians of the colonial era, for instance, will probably find some of Urton's equivalences between social and tributary categories (i.e., ayllu and pachaca), as well as his observations about the aims and methods of routine inspections (visitas), partially problematic or contradicted by other sources. Nevertheless, relatively minor numerical discrepancies between ‘matching khipu’, along with Urton's well-informed historical assumptions and hypothetical formulations, seem a reasonable price to pay for the possibility of interpreting extant khipu specimens in social, political and religious contexts, re-embedding them, sometimes for the first time in centuries, in their original historical setting. This is no small feat, one that, until recently, seemed virtually impossible with the available evidence.
Of particular interest is Urton's methodology, which correlates published early colonial-era censuses to extant specimens. Based on the distribution of the range and frequency of unit groupings (families/households) recorded in these colonial censuses, Urton elaborates a series of hypothetical khipu exemplars, which he then compares to the larger database, identifying 50–60 possible khipu analogues that share these features. This is a remarkable finding. Although multiple sources identify census-recording as one of the main functions of khipu (if not the capital one), no census khipu—with the likely exception of khipu UR6 from Chachapoyas—has yet been positively identified. A large number of censuses and tributary rolls (padrones), conducted by both civil and ecclesiastic inspectors and stored in colonial archives, await the type of painstaking analysis pioneered by Urton.
Equally important is Urton's proposed identification of the Santa Valley ‘khipu archive’, a group of six interrelated specimens looted from an unknown location on the north-central coast and then acquired by khipu scholar Carlo Radicati, with a tribute reassessment (visita) carried out in the village of Recuay in 1670. Urton posits a possible match between, on the on the hand, the alphabetic record of tributaries (organized by ayllu/pachaca) and their tribute dues in specie, and, on the other, the numerical values recorded in the knotted strings of Radicati's collection. Although Urton acknowledges that certain problems arise with the identification of the khipu with this particular location (the ‘Santa Valley’), if his correlation were correct, it would be the first known ‘match’ ever between an alphabetical document and its corresponding khipu transcription. Such a match would grant the Santa Valley cords the status of a ‘Rosetta Khipu’. In this section of the book, as well as in others, Urton argues convincingly that khipu structural and formatting features, such as cord grouping, colour coding, cord attachment type and knot direction, signalled ayllu/pachaca affiliation. Thus, each of the six khipu can be read as a record of tributary obligations owed by each of the pachaca (unit of 100 tributaries) of Recuay.
Interestingly, Urton finds that, although the Recuay visita stated that each tributary was to pay close to 3 pesos, values recorded in the cords which, in his interpretation, likely referred to specie paid per tributary vary from 0 to 13. Urton offers a clever solution: while the written tribute assessment (tasa) recorded the theoretical contribution of each tributary (an equal share of the total for the repartimiento or fiscal district), khipus were internal records meant to aid local khipukamayoq in keeping track of these individual payments. Therefore, they encoded the actual contributions of tributaries of disparate wealth and status. While Urton's hypothesis of divergent contributions captured in khipu is suggestive—colonial-era historians have offered copious examples of social differentiation and class conflict within indigenous communities—it remains to be empirically proven. While I agree that knotted cords recorded types of information that were only internally relevant (i.e., meaningful to communal tax collectors and administrators, but not necessarily to Spanish officials and priests), Urton's analysis offers little in the form of historical context to assess whether such wide distinctions of wealth had emerged in Recuay or the Santa Valley around 1670 and why. Only a few among the privileged cacique (noble) class could have been able to afford 13 pesos worth of annual tribute (let alone an equivalent amount every six months) and caciques were, de jure and de facto, exempt from these fiscal burdens. Another possible explanation for the 0–13 range variation would be that it reflected some other arithmetical operation resulting from converting tribute owed in specie to its monetary equivalent.
Anthropologist and khipu scholar Frank Salomon wrote in 2004 that the literature on khipu resembled ‘a bridge half-built from two piers’, with elaborate khipu specimens on one side and, on the other, testimonies about Andean domains of activity or thought said to have been encoded in cords (Salomon Reference Salomon2004, 18). Urton's carefully attained matches between khipus and alphabetic accounts, along with the many other findings revealed in this work, are starting to fill the seemingly unsurmountable gap between the two sides. This book will herald a new era in khipu studies.
