Reviewer #4 (Public Review):
Berger et al. 2023a argues that Homo naledi intentionally buried their dead within the Rising Star cave system by digging pits and covering the bodies with infilled sediment. The authors identified two burials: Dinaledi Feature 1 from the Dinaledi Chamber, and the Hill Antechamber Feature from the Hill Antechamber. The evolutionary and behavioral implications for such behavior are highly significant and would be the first instance of a relatively small-brained hominin engaging is complex behavior that is often found in association with Homo sapiens and Homo neanderthalensis. Thus, the scientific rigor to validate these findings should be of the highest quality, and thus, provide clear documentation of intentional burial. In an attempt to meet these standards, the authors stated a series of tests that would support their hypothesis of intentional burials in the Rising Star Cave system:
"The key observations are (1) the difference in sediment composition within the feature compared to surrounding sediment; (2) the disruption of stratigraphy; (3) the anatomical coherence of the skeletal remains; (4) the matrix-supported position of some skeletal elements; and (5) the compatibility of non-articulated material with decomposition and subsequent collapse." (page 5)
To find support for the first (1) test, the authors collected sediment samples from various locations within the Rising Star Cave system, including sediment from within and outside Dinaledi Feature 1. However:
• The authors did not select sediment samples from within the Hill Antechamber Feature, so this test was only used to assess Dinaledi Feature 1.
• The sediment samples were analyzed using x-ray diffraction (XRD) and x-ray fluorescence (XRF) to test the mineralogy and chemistry of the samples from within and outside the feature. The XRF results were presented as weighted percentages (not intensities) with no control source reported. The weighted percentages were analyzed using a principal components analysis (PCA) while the particle-size distribution was analyzed using GRADISTAT statistics package and the Folk and Ward Method to summarize "mean grain size, sorting, skewness and kurtosis in addition to the percentages of clay, silt and sand in each sample." (page 28).
• The PCA results were reported solely as a biplot without showing the PC scores projected into the loading space, which is unusual and does not present the data accurately. Instead, the authors present the scores of a single component (PC2, figure 3) because the authors interpreted this component as "distinctly delineates fossil-bearing sediments from sterile sediments based on the positive loadings of P and S" (Page 6). However, the supplementary table that reports XRF bulk chemistry results as a weighted percentage of minerals within each sample (SI Table 1) shows mostly an absence of data for both Na and S. Since Na is at the lower end of detection limits for the method, and S seems to just be absent from the list, the intentions of the authors for showing the inclusion of these elements in their PCA results is unclear. Given that this is the author's primary method for demonstrating a burial, this issue is particularly concerning and requires additional attention.
• Regardless of the missing data, this reviewer attempted to replicate the XRF PCA results using the data provided in SI Table 1 and was unsuccessful. The samples that were collected from within the feature (SB) cluster with samples collected from sterile sediments and other locations around the cave system. Thus, these results are not replicable as currently reported.
• Visual comparisons of sediment grain size, shape, and composition were qualitatively summarized. Grain size was plotted as a line graph and is buried as supplemental Figure S13 showing sample by color and area, but these results do not distinguish samples from WITHIN the burial compared to OUTSIDE the burial as the authors state in the methods as a primary goal.
To test the second (2) aim, the "stratigraphy" was primarily described in text.
• For Dinaledi Feature 1, the authors state that the layer around Feature 1 "is continuous in the profile immediately to the east of the feature; it is disrupted in the sediment profile at the southern extent of the feature (fig. 3b)." Upon examination of figure 3b, the image shows an incredibly small depiction of the south (?) profile view with an extremely large black box overlaying a large portion of the photograph containing a small 5 cm scale. Visually, there is no difference in the profile that would suggest a disruption in the form of a pit. The LORM (orange-red mud layer) does seem to become fragmentary, but no micromorphological analysis was conducted on this section to provide an evaluation of stratigraphic composition. Also, by only excavating a portion of the feature, the authors were unable to adequately demonstrate the full extent of this feature.
• The authors attempt to describe "a bowl-shaped concave layer of clasts and sediment-free voids make up the bottom of the feature" (page 13) and refer to figures and supplementary information that do not depict any stratigraphic profile. Moreover, the authors state that "the leg, foot, and adjacent [skeletal?] material cut across stratigraphy" indicating that the skeleton is orientated on a flat plane against the surrounding stratigraphy that is "30{degree sign} slope of floor and underlying strata" (page 51, fig. 10c captions). There is no mention of infilled sediment from a pit and how this relates to the skeleton or the slope of the floor. It is therefore extremely unclear what the authors are meaning to describe without any visual or micromorphological supplementation to demonstrate a "bowl-shaped concave layer".
The third (3) test was to evaluate the anatomical coherence of the skeletal remains using macro- and micro-CT (computed tomography) of the Hill Antechamber Feature that was removed during excavation. To visually assess the anatomy of the Dinaledi Feature 1 burial, the authors describe the spatial relationship of skeletal elements as they were being excavated but halted partway through the excavation.
• The authors do not provide any documentation (piece-plotting, 3D rendering of stages of excavation, etc.) of the elements that were removed from the Dinaledi Feature. Figure 4 and SI Fig. S22 show the spatial relationship between identifiable skeletal elements that remain in the Feature. However, in Fig. 4, it is unclear why the authors chose to plot 2023-2014 excavated material along with material reported here, and it's even more difficult to understand the anatomical positioning of the elements given their color and point size choices. Although, the authors do provide a 3D rendering of the unexcavated remains showing some skeletal cohesion, apart from the mandible and teeth being re-located near the pelvis (Fig. 9). That said, it is very difficult to visually confirm the elements from this model or understand the original placement of the skeleton.
• 3D renderings of the Hill Antechamber feature skeletal material is clearly shown in SI Fig. S26. Contrary to what the authors state in text, there is a rather wide dispersal and rearrangement of elements for a "burial" that is theoretically protected from scavengers and other agents that would aid in dispersing bone from the surface. The authors do not offer any alternatives to explain disturbance, such as human activity, which clearly took place.
• Moreover, there does not appear to be any intentional arrangement of limbs that may suggest symbolic orientation of the dead (another line of evidence often used to support intentional burial but omitted by the authors). Thus, skeletal cohesion is not enough evidence to support the hypothesis of an intentional burial.
The fourth (4) test was attempted by evaluating whether some elements were vertically aligned from 3D reconstructed models of Hill Antechamber Feature and a photogrammetric model of the Dinaledi Feature 1. The authors state that "the spatial arrangement of the skeletal remains is consistent with primary burial of the fleshed body" (page 8 in reference to Dinaledi Feature 1) without providing any evidence, qualitative or quantitative, that this is the case for either burial.
Since this reviewer was unable to understand the fifth (5) test as it was written by the authors, I am unable to comment on the evidence to support this test and will default to the other reviewers for evaluation of this claim.
In addition to a lack of evidence to support the claims of intentional burial, this paper was also written extremely poorly. For example, the authors often overused 'persuasive communication devices' (see eLife article, https://elifesciences.org/articles/88654) to mislead readers:
"During this excavation, we recognized that the developing evidence was suggestive of a burial, due to the spatial configuration of the feature and the evidence that the excavated material seemed to come from a single body." (page 5)
As an opening statement to introduce Dinaledi Feature 1, the authors state the interpretation and working hypothesis as fact before the authors present any evidence. This is known as "HARKing" and "gives the impression that a hypothesis was formulated before data were collected" (Corneille et al. 2023). This type of writing is pervasive throughout the manuscript and requires extensive editing. I recommend that the authors review the article provided by eLife (https://elifesciences.org/articles/88654) and carefully review the manuscript. Moreover, as this text demonstrates, the authors’ word choice is indicative of storytelling for a popular news article instead of a scientific paper. I highly suggest that the authors review the manuscript carefully and present the data prior to giving conclusions in a clear and concise manner.
Moreover, the writing structure is inconsistent. Information that should be included in results is included in the methods, text in the results should be in discussions, and so forth. This inconsistency is pervasive throughout the entire manuscript, making it incredibly difficult to adequately understand what the authors had done and how the results were interpreted.
Finally, the "artifact" that was described and visualized using CT models is just that - a digitally colored model. The object in question has not been analyzed. Until this object is removed from the dirt and physically analyzed, this information needs to be removed from the manuscript as there is nothing to report before the object is physically examined.
Overall, there is not enough evidence to support the claim that Homo naledi intentionally buried their dead inside the Rising Star Cave system. Unfortunately, the manuscript in its current condition is deemed incomplete and inadequate, and should not be viewed as finalized scholarship.