Craniofacial Anthropometry of some Julio-Claudian Portraits (Summary of investigation) William Storage and Laura Maish  Email us about this page

Home Roman Imperial Portraits

Hairstyle has long been the principle means of identifying the subjects of Julio-Claudian and later imperial portraits. Recently, a few scholars have proposed that we may be overly reliant on this approach, the Lockenzahlmethode, as named by the Romische Herrscherbild series of books. In order to investigate the possibility that physiognomic characteristics might be undervalued in the process of identifying uncertain portraits, we conducted a preliminary study, collecting measurements of marble portraits, normalized by calculating ratios of various measurements. This technique is in common usage in the analysis of human craniofacial anthropometry for reconstructive and aesthetic plastic surgery.

We used fifty eight of the proportion indices identified by Farkas and Munro (1986), selected for the ability of their facial landmarks to be located in photographs of frontal and profile views of the sculptures. a few of these still required approximations because of the presence of hair or at landmarks that in a human would be better located by palpation. We included sculptures with restored noses in cases where enough of the original nose was present to reasonably estimate vertical position of the pronasale where enough original marble existed to establish nasofrontal and nasolabial angles. Most proportions involving ears were excluded, since few sculptures retain original ears. proportions involving measurements involving facial contours were also excluded, since they cannot be accurately derived from photographs. laser-scanning of the portraits would provide more and better data; the technique is in common usage both plastic surgery and in character animation for movie special effects (Aung 1995, Enciso 2003). We also measured the nasofacial and nasolabial angles in the portraits.

We wrote a computer program to allow us to display photographs on screen and to manually mark the facial landmarks on images (figure 1). The coordinates of the facial landmarks (listed in Table 2) were then recorded in a database, from which facial proportions and statistical calculations could be made and reported. Identification of certain facial landmarks from photos has been demonstrated in human craniofacial anthropometry, provided that the photos meet certain criteria (e.g. clarity, facial plane alignment, large subject-camera distance compared to head height); we adhered to these criteria.

Our analysis of the 25 portraits of Augustus, Caligula, Claudius and Germanicus indicates that craniofacial anthropometry is in fact of very little use in portrait identification. a comparison of facial proportions of Augustus and Caligula illustrate the problem. Including Augustus heads thought to have been recarved from heads of Caligula and heads of Caligula, few statistically significant differences in average measurements of craniofacial proportions and facial angles can be seen (Table 3). Generally, the standard deviation for a given proportion of one subject is comparatively large when compared to the difference between the mean  value of that proportion for the other subjects. Excluding the proposed Caligula/Augustus does not help the problem at all. surprisingly, an analysis of the difference between Augustus heads and proposed Caligula/august heads shows that for most of the proportions where a difference exists, it is in the wrong direction. For example, the average mandibulo-upper face height index is lower for Caligula/Augustus heads than for Augustus heads, yet the index is higher for Caligula heads than for original Augustus heads. similarly, the mean mandibulo-lower face height index of Caligula/Augustus heads is lower than that of original Augustus heads, yet the index is higher for Caligula than for Augustus. Of the 58 proportions analyzed, the 21 listed below were found to be most significant. These proportions are listed below (Table 1).

Table 1: Most significant facial proportions as determined by statistical analysis of 25 Julio-Claudian portraits examined.

 
Figure 1. The Getty Caligula/Augustus showing facial landmarks identified in frontal and profile photos
  
  

Table 2: Description of facial landmarks.