Project 2

Origins of Technology

Seeking the origins of technology in rare middle Pliocene sediments in eastern Africa
Searching for evidence of early technology in Hadar, Ethiopia

Principal Investigators

Project 2 investigates the environmental contexts of the earliest steps by hominins toward cognitively complex behaviors and technological advances.

This project builds on Project 1’s examination of early stone-tool use to document the origin of persistent stone tool manufacture, one of the hallmarks of human cognitive complexity. Understanding the context in which this behavior arose, and the paleobiology of the possible stone tool makers, will shed light on the origin of cultural accumulation.

Detailed Project Description

Although eastern Africa currently holds the earliest evidence of the Homo lineage (ca. 2.3-2.4 Ma) and Oldowan lithic technology (ca. 2.6 Ma), the strategically critical time period between 3.0 and 2.5 Ma remains poorly documented in this part of the continent. The decline of generalized Australopithecus species, the origin of at least two adaptively distinct daughter lineages, Homo and Paranthropus (“robust” australopiths), and the first appearance of deliberately shaped stone tools are all thought to have occurred during this time period. These events are broadly correlated with profound, astronomically forced changes in global climate that led, after 3.0 Ma, to drier and more seasonal African environments.

Any site in eastern Africa that documents this temporal interval deserves special scrutiny because of its potential to shed light on the early evolution of enhanced cognitive capability, technological innovation (including stone-tool use leading to manufacture), and their environmental contexts. The Afar Ledi-Geraru research area (ALG), located in the Afar Rift Valley of Ethiopia, is one such site.

While the ALG overlaps parts of the time span of the neighboring paleoanthropological sites of Hadar, Dikika, and Woranso-Mille, it provides a unique and continuous window on the interval between 2.9-2.6 Ma. The results of the winter 2013 field season (funded by NSF) in the ALG encourages further research due to the preliminary findings in three separate geological basins, Lee Adoyta, Markaytole, and Ali Toita. Discoveries in these localities potentially offer insight on interactions between environmental variation and the advent of novel hominin behavioral adaptations. Our proposal calls for expanded exploration of these deposits with the goal of further identifying localities for fossil collection and archaeological excavation.

The Lee Adoyta Basin contains dateable volcanic marker beds that, combined with results of paleomagnetic sampling, demonstrate a preliminary age of ~2.9-<2.6 Ma. Fossils of numerous mammalian species have been recovered from across the basin; combined with geological studies, these will be important in reconstructing environmental shifts across this time interval that record both fluvial and near-shore environments. A potential Oldowan archaeological locality in the northern part of this basin, as yet undated, has also been identified.

The Markaytoli Basin yielded an array of mammals that are completely different than those recovered from the Lee Adoyta Basin, which is about 5 km to the north. A prominent volcanic horizon outcrops in the region, but its geochemistry does not match any other tephra in the region (and thus far no dateable crystals have been recovered from it). However, fossil monkeys, bovids, and suids tentatively indicate that the Markaytole sediments are as much as 1 myr younger than those of Lee Adoyta. Both Acheulean and Oldowan stone tools have been found in the basin, although no material has yet been recovered in situ.

In the Ali Toita region, we have so far identified three archaeological localities with Oldowan tools and fossils in situ. These localities overlie a tephra that is currently being dated. As much of the area encompasses 2.9–2.6 Ma, these localities could provide information on some of the earliest evidence of hominin behavior, perhaps as old as the ca. 2.6 myr-old sites at Gona.


Following field research strategies we have employed elsewhere (e.g., Hadar), field methods include collecting all taxonomically diagnostic elements of all macrofauna, including well-preserved long bone shaft fragments, which may record evidence of hominin behavior or other accumulating agents (e.g., carnivores). In addition, postcranial articular elements will be collected for macromammals, as these are used in functional studies of locomotion and associated habitat reconstructions.

All collected fossils will be identified by members of the project team, who are experts in the appropriate fields. In addition to basic systematics, data will be collected to facilitate analysis of functional-adaptive traits. Analyses of both phylogenetic and trait structures of communities will be performed using previously applied methods. We plan to conduct carbon-isotopic analyses of enamel of specific fossil taxa to provide information on spatial and temporal variation in diet as it relates to habitat. We have already begun to integrate the ALG faunal data with the Hadar Geoinformatics Project database, which integrates spatial and temporal datasets from multiple fossil research areas to advance fundamental paleoanthropological research, education, and conservation in the greater Hadar sedimentary basin. One of the key issues to be investigated in this context is faunal turnover and environmental change across the 3.0 Ma mark, which spans the uppermost part of the Hadar Formation at ALG and neighboring Hadar.

Archaeological evaluation of stone artifacts will focus on the technological sophistication associated with tool-using behaviors in the Pliocene. Studies of raw material selectivity and transport behaviors, important monitors of cognitive sophistication, will be augmented by the extensive geological mapping of local conglomerates, which will be sampled and geochemically analyzed as a comparison with rock types incorporated in these early industries. Additional analyses will focus on the technological capacity of the earliest toolmakers. We will investigate the stone artifacts employing a host of techniques (e.g., refitting, morphometric analysis, reduction sequence modeling, chaine operatoire) currently applied to Oldowan sites. The large (ca. 5,000), 2.3 myr-old sample of Oldowan artifacts from the Busidima Formation previously excavated by the IHO team at Hadar will form an important comparative reference for any archaeological materials recovered from LGA sediments.

Bones with traces of surface modifications will be investigated. As outlined in the Project 1 proposal, recent investigations suggest that it is possible to quantitatively distinguish between different agents of bone-surface modification using variables collected at microscopic scales. We will integrate analyses of any modified bone recovered from LG sediments with the Hadar bone-modification study described in Project 1.

There are three levels of sampling beyond walking surveys to acquire information about taphonomy and paleoecology (crawling, surface transects, and excavation). These sampling procedures, in conjunction with sedimentological sampling and study, will allow us to analyze how fossil material is associated with various depositional environments and subsequent post-depositional alteration. Crawling consists of moving across the surface, shoulder to shoulder on hands and knees for ~100 percent recovery of any fossil specimens. For surface transect samples, we define narrow corridors (1–5 m in width and 100–300 m in length) over an area chosen to avoid slope wash from overlying sedimentary units. Every bone encountered will be collected and examined for surface modification. Excavation may be done in areas that reveal a concentration of material during the surface transect sampling. As excavation provides the best stratigraphic control, we will use this technique at sites that exhibit any evidence of concentrated material or hominin behavior (archaeology).

Collection and excavation procedures will employ an on-site Trimble Nomad data collector (2-m accuracy) and preprinted barcodes recording details of provenience, taxonomy, and element information, synched to a laptop-based master ArcGIS/Access database system for all collected/excavated faunal and archaeological materials. A Trimble handheld receiver (decimeter accuracy) is used to map geologic marker beds and features.