A model-based approach to the tempo of "collapse": The case of Rapa Nui (Easter Island). (April 2020)

• Record Type: Journal Article
• Authors: DiNapoli, Robert J.; Rieth, Timothy M.; Lipo, Carl P.; Hunt, Terry L.
• Is Part Of: Journal of archaeological science. Volume 116(2020)
• Publication Date: 2020-04
• Subjects: Collapse -- Bayesian analysis -- East Polynesia -- Monumental architecture -- Tempo plots; Archaeology -- Periodicals; Archéologie -- Périodiques; 930.1
• DOI: 10.1016/j.jas.2020.105094
• Languages: English
• ISSNs: 0305-4403

Abstract

Abstract: Rapa Nui (Easter Island, Chile) presents a quintessential case where the tempo of investment in monumentality is central to debates regarding societal collapse, with the common narrative positing that statue platform ( ahu ) construction ceased sometime around AD 1600 following an ecological, cultural, and demographic catastrophe. This narrative remains especially popular in fields outside archaeology that treat collapse as historical fact and use Rapa Nui as a model for collapse more generally. Resolving the tempo of "collapse" events, however, is often fraught with ambiguity given a lack of formal modeling, uncritical use of radiocarbon estimates, and inattention to information embedded in stratigraphic features. Here, we use a Bayesian model-based approach to examine the tempo of events associated with arguments about collapse on Rapa Nui. We integrate radiocarbon dates, relative architectural stratigraphy, and ethnohistoric accounts to quantify the onset, rate, and end of monument construction as a means of testing the collapse hypothesis. We demonstrate that ahu construction began soon after colonization and increased rapidly, sometime between the early-14th and mid-15th centuries AD, with a steady rate of construction events that continued beyond European contact in 1722. Our results demonstrate a lack of evidence for a pre-contact 'collapse' and instead offer strong support for a new emerging model of resilient communities that continued their long-term traditions despite the impacts of European arrival. Methodologically, our model-based approach to testing hypotheses regarding the chronology of collapse can be extended to other case studies around the world where similar debates remain difficult to resolve. Graphical abstract: Image 1 Highlights: Collapse hypothesis for Easter Island tested using Bayesian models. Tempo plots quantify the onset, rate, and end of monument construction. Results contradict 'collapse' scenario for Rapa Nui. Tempo plots provide means for evaluating collapse hypotheses.