Shrinking the Mystery of Tsantsa Identification: Using Qualitative and Quantitative Measurements to Distinguish Natural and Imitation Shrunken Heads

During the 17 th through mid‐20 th centuries, the Jivaro tribes of South America participated in ritual head shrinking as part of a religious ceremony following battle. In the late 19 th and early 20 th centuries, travelers to South America took interest in these shrunken heads, known as tsantsas, and brought them back home as curios, inspiring the production of fake shrunken heads for the tourist trade. Numerous tsantsas, both real and fake, have ended up in museums or private collections worldwide. As many museums have very specific rules about repatriation of human remains, understanding differences between authentic and imitation tsantsas is imperative. GOALS The purpose of this study was to qualitatively and quantitatively describe the five tsantsa specimens held in the Anthropology Collection curated at the Denver Museum of Nature and Science (DMNS). METHODS First, each specimen was scored independently by two observers for 19 qualitative features typically associated with authentic tsantsas. Each observer also collected 29 standard facial measurements from four key locations (eye = 6, nose = 9, mouth = 4, ear = 10). Each facial measurement was size‐corrected using the geometric mean of all measurements for the specimen. The size‐corrected facial measurements were analyzed using principle component analysis (PCA). RESULTS Inter‐observer agreement on scoring of the qualitative features was high, and analysis of the facial measurements revealed no significant differences between raters. The qualitative features suggested the presence of two distinct groups within the DMNS sample: A Probably‐Authentic Group (three specimens) exhibiting 15 or more out of the 19 authentic characteristics, and a Probably‐Imitation Group (two specimens) exhibiting 8 or fewer of the 19 characteristics. PCA of all size‐corrected variables identified two factors accounting for 78% of the variance in the sample (factor 1 = 47%; factor 2 = 31%). The PCA revealed that the five specimens clustered within two distinct groups, distinguished primarily along factor 1. The group composition determined by PCA was concurrent with the two groups indicated by the qualitative traits. CONCLUSIONS Agreement between analysis of qualitative tsantsa characteristics and facial morphometric measurements suggests the presence of three authentic and two imitation tsantsas in the DMNS collection. To affirm these findings, microscopic hair analysis is also performed to establish the presence of human hair. However, DNA analysis is needed to determine with confidence whether the DMNS tsantsas are human or fabricated. This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .