The Olmec and the Origins of Mesoamerican Civilisation
Neff, Hector, Antiquity
In a recent editorial, Martin Carver mentioned a series of papers that purportedly debate whether the Olmec of Gulf Coastal Mexico were Mesoamerica's 'mother culture' or one of several Early Formative 'sister cultures.' The papers in question concern a recent project that explored ceramic provenance using INAA (Instrumental Neutron Activation Analysis). It found a strong indication that ceramics were exported from the Olmec centre of San Lorenzo on the Gulf Coast, but no indication of imports coming in to San Lorenzo and no evidence that Early Formative ceramics moved between other centres (Blomster et al. 2005). Critics of this study (Flannery et al. 2005; Stoltman et al. 2005) claimed that the sampling design in the INAA study was deliberately biased toward finding Gulf Coast exports while concealing exports from other regions, that the INAA data themselves belie the conclusions reached, and that petrographic data 'overturn' the INAA results. These charges were rebutted at length in an article and a subsequent comment in Latin American Antiquity (Neff et al. 2006a, 2006b), which Professor Carver mentions. Unfortunately, however, by giving significantly more space to the claims of the critics, Antiquity's editor may have left the impression that the results of the INAA study had been decisively refuted.
The 'mother culture' and 'sister culture' metaphors are useful to the extent that they have stimulated proponents of both sides to make statements about the Early Formative archaeological record that, at least in principle, are testable. For instance, 'mother culture' advocates (e.g. Coe and Diehl 1980) have been suggesting for some time that carved-incised gray pottery with 'Olmec motifs' and fine white pottery were exported from the vicinity of San Lorenzo, while 'sister culture' advocates (e.g. Marcus 1989; Flannery and Marcus 2000) have been arguing for some time that San Lorenzo played no greater role than other centres in Early Formative ceramic exchange. The surprising pattern revealed by the INAA study clearly favours the statements made by 'mother culture' advocates. Because it is based on nearly 1000 ceramic and over 600 raw clay samples from key Early Formative centres throughout Mesoamerica, this finding also qualifies as an exceptionally strong empirical pattern.
The major product of the INAA study--the hypothesis that San Lorenzo exported lots of pots and other centres exported next to none--would require a considerable investment of new analytical effort for falsification. Flannery, Stoltman, and their colleagues, however, have employed several short cuts. First, they conducted a statistical 'reanalysis' of the INAA data (which were published on-line). Fallacies in their analysis are blatant and are detailed elsewhere (Neff et al. 2006a, 2006b); suffice it to say here that if one applied their approach to a study of twenty-first century automobile import and export, the Toyotas, BMWs, Fords, Chews, VWs, Nissans, Hyundais, Hondas--and maybe even the old Triumph Spitfires--that one sees on the streets of Los Angeles would be sampled randomly in order to define local automobile production in Los Angeles; comparison of automobiles from other places to this highly variable 'local LA group' would then suggest, erroneously, that a wide variety of cars were produced in and exported from LA.
Stoltman et al. (2005) also analysed a small sample of thin sections (around 20) petrographically and claimed to have identified several imports into San Lorenzo. Again, our detailed response appears elsewhere (Neff et al. 2006a, 2006b). In this case, the automotive analogy to their approach would be to define the late-twentieth century automobile industry in Detroit by a single Ford Maverick. More importantly, finding a few imports to San Lorenzo (by petrographic or chemical characterisation) would not 'overturn' the major pattern revealed by the INAA study, namely that San Lorenzo exported so many more pots than its contemporaries that only the San Lorenzo exports were detected in a sample of nearly 1000 Early Formative ceramics analysed by INAA. …