Ancient History & Civilisation



The middle of the nineteenth century was marked by a discernible shift in attitude to the human skeletal remains that had been exhumed from Pompeii. Instead of merely functioning as props for either literary or physical reconstructions, this material was now recognized as having value as a scientific resource. The first scholarly examination of the Pompeian human skeletal material was published a little over a hundred years after the first official excavation of the site.

The impetus for the initial investigations was the establishment of the Commission for the Reform of the Royal Bourbon Museum and the Excavations of Antiquities of the Kingdom in 1848. This commission was set up by Ferdinand II under political pressure, as there were serious problems with the management of archaeological sites in the region around Pompeii. Raphaele d’Ambra reported to the Commission later in that year that bones and other finds had been neglected and remained in deposits without any attempt to ensure their preservation. Further, permission had been denied to a French chemist, Jean Pierre Joseph d’Arcet, to conduct research on the human remains. The Commission made 11 proposals to facilitate the reopening and subsequent protection of the archaeological area of Pompeii. One of these proposals was to open a gallery of Pompeian skeletons, which would involve the donation of skulls and other skeletal material to the Royal University of Studies in Naples. This was the first real acknowledgement that the human skeletal remains were of anthropological significance. Unfortunately, the liberal political climate was altered by the restoration of the monarchy, which meant that the proposals of the Commission were never put into practice.1

Early investigation

Stefano Delle Chiaie revived interest in the issues associated with storage and research on the Pompeian skeletons in 1853. He was responsible for the establishment of a sizable collection of skeletal material in the Anatomical Museum of the Royal University of Naples and undertook the first study of the bones that were available.

The first major publication of the human skeletal remains from Pompeii that included raw data and systematically presented results appeared in 1882. It involved the research of Giustiniano Nicolucci, the founder of the Institute of Anthropology of the University of Naples.2 Nicolucci summarized and reviewed the studies on the human bones from Pompeii that preceded his work.3 His literature review is revealing as Nicolucci was very much a man of his era. To him craniology, and more specifically craniometry, was paramount in anthropological studies and works that were not devoted to measurement and discussion of skull form were not considered important. He was fairly dismissive of the 1854 publication of Delle Chiaie. This was partly because its main focus was on a description of the pathological changes that he observed. It also contained two analyses of the chemical composition of Pompeian bones, which were undertaken by Lehman for comparison with the chemical composition of modern bones. Nicolucci was apparently not interested in the pathology and lamented that craniology scarcely received a mention in this work, there merely being a short note stating that some skulls were globular, others ovoid and that a few were oblong in form. The final type of skull observed in this work was interpreted as African, probably representing slaves who were in the service of the wealthy citizens of Pompeii.4 Nicolucci was very critical of this work and stated that he thought that this brief description did not shed much light on ‘the natural history of Pompeian man’, especially since it did not include measurements. He stated that all Delle Chiaie’s work revealed was that there was no uniform skull shape in the sample and that one could divide the skulls into specific types. Nicolucci, however, did not appear to be convinced that enough evidence was produced to establish these assertions, especially with regard to the so-called oblong or African crania.5 Delle Chiaie’s publication certainly tends towards being data-free, which does limit its scientific value.

Nicolucci also expressed reservations about the Pompeian skull that was illustrated in the work of Sandifort as it also had not been measured and there had been no attempt to compare it to other crania. Conversely, he considered that the publication of another Pompeian skull by Vrolik and Van der Hoeven provided an accurate description. This skull had been affected by osteo-sclerosis and displayed extreme brachycephaly, with a cephalic index of 87.3. These two scholars interpreted this skull as Greek in form, based on the work of the nineteenth-century Swedish anatomist Retzius, who defined and popularized the cranial or cephalic index. This is the ratio of maximum width to maximum length of the cranium and it was applied in the nineteenth and the first half of the twentieth centuries as an indicator of ‘race’. Skulls that were described as long-headed, or dolicocephalic, had an index of no more than 75, whilst those that were more short headed were classified as brachycephalic and had an index of, at least, 80. Retzius classified the Greeks to be among the brachycephalic populations in Europe. Nicolucci, as a typical nineteenth-century anthropologist, accepted the cephalic index as a reliable criterion for ‘race’ determination. However, he chose to disagree with the famous Swedish anthropologist’s pronouncements as Retzius did not examine the skeletal remains of ancient Greeks. Retzius only used modern skulls, whereas Nicolucci found from his experience that the skulls of ancient Greeks were generally dolicocephalic. While Nicolucci considered that the judgement of the nationality of the Pompeian skull by Vrolik and Van der Hoeven was not exactly correct, he stated that he was certain that it did not represent one of the Pompeian types. He considered that its excessive brachycephalism was also partially due to pathological change.6

Presuhn published an article on Pompeian anthropology in 1881. This work was based on a minimal number of skulls. All that was stated was that they were big and robust with a strong and protuberant occiput, the facial angle was open, the face was full and the nose large. He considered that the stature of the Pompeians was medium, like other South Italians, and the hair was generally brown or almost black.7

Though it was apparent that Nicolucci was unimpressed with the fact that Presuhn formulated his description of the ancient Pompeians from virtually no skeletal evidence, he did not comment on the reconstruction of features that cannot be determined from the skeletal record, like hair colour. At best, such attributions could only have been based on extrapolation from ancient paintings or observations of the modern population. There was a long tradition for such reconstructions, which extended well into the twentieth century. Angel, in his work on skeletal material from Attica, for example, demonstrated a remarkable ability to detect not only hair but also eye colour from skulls. This can be seen in his description of people he classified as Nordic-Iranian from their skulls as ‘probably … tawny haired, blue-or-green eyed blondes as well as brunettes’.8 Such an approach negates the value of any study since the conclusions are drawn prior to commencement from preconceived ‘types’, presumably recognizable in modern populations.9

It is possible that Nicolucci ’s preference for things cranial explains why he neglected to mention the examination of a single right femur by Amabile, which was found to have an untreated compound fracture that had healed at right angles.10

Despite his obvious bias, Nicolucci presented a critical summary of the first studies of the Pompeian skeletal collection. It is apparent that he perceived the need for a more systematic study based on a large sample as compared to the essentially antiquarian approach of his predecessors.

The work of Nicolucci

Nicolucci ’s work provides a very good case study of nineteenth-century physical anthropology. It deserves detailed consideration as it was the major publication on the Pompeian skeletons in the nineteenth century and continued to be considered the seminal work on this material until this type of craniometric approach was challenged in the latter part of the twentieth century.

The skeletal remains that Nicolucci used for his study were either displayed in cabinets or stored in special rooms in the anatomical museums in the University of Naples. Nicolucci did not state whether these collections comprised complete skeletons or merely crania but it is notable that he considered that the main purpose of this material was to provide information about the cranial form, and, therefore, enable scholars to determine the exact ‘type’ of the Pompeian skulls. These could in turn be compared with ancient Italian crania from known populations to establish the relationship of the Pompeian victims with the other ancient populations in the region. Nicolucci also considered that this work would be of value in the determination of the cranial and physiognomic variability of ancient and contemporary Campanian people. He stated that this aspect of the study could be guided by observations from Pompeian wall paintings.11

Nicolucci described the history and ethnology of Pompeii to provide a context for his craniological study. His ethnological description was based on the ancient literary sources, like Strabo and Pliny the Elder (Chapter 4). He did not use the skeletal record to test the literary sources.12 Their accounts were accepted as fact. This approach to written evidence in relation to physical evidence has been typical until comparatively recent times. Nicolucci also subscribed to the notion that the old, the very young, the infirm and women made up the majority of the victims, even though it was unencumbered by evidence.13

Sex and age-at-death of the sample

Nicolucci studied a sample of 100 skulls, 55 of which he determined to be male and 45 female. He considered that the entire range of ages was represented in his sample, with the majority being between the ages of 60 and 90 (Chapters 6 and 7).

Non-metric traits

Nicolucci only commented on the presence of three non-metric traits associated with the skulls he examined. These were metopism, wormian bones and inca bones (Chapter 9). It is significant that Nicolucci related these findings to cephalic index and cranial capacity, which were considered to be the most important population indicators in the nineteenth century. From his conclusions from the data on metopism, it was obvious that he considered that non-metric traits also provided a contribution as population descriptors. Nonetheless, his presentation of the cases of wormian and inca bones in the sample was more antiquarian in nature, where incidence was apparently noted for curiosity value and no conclusions or comparisons were drawn, though it is possible that comparative data were available at that time.14

Metric observations

Nicolucci stated that his classi fication of Pompeian skulls was based on the work of Linnaeus. In the mid-eighteenth century, Carl Linnaeus constructed the first formal classification scheme of different human groups as part of his greater taxonomy of living organisms. He identified four separate types: Europaeus albus (European white), Asiaticus luridus (Asiatic yellow), Americanus rufus (American red) and Africanus niger (African black). This work, like others of its period, incorporated perceived cultural characteristics with the physical traits.

Nicolucci emphasized the importance of the cephalic index in relation to the other metric data he collected by discussing it in a separate section. He determined that 14 per cent of the sample he studied was dolicocephalic, 43 per cent mesocephalic and 43 per cent brachycephalic, with a mean index of 77.7, which is in the mesocephalic, or moderate-headed, range. He noted that there were differences in the proportions of this index between the males and females in the sample, with males having a relatively higher incidence of dolicocephaly, whilst the females tended to be more mesocephalic and brachycephalic. The mean cephalic index for the female sample was 78.1 whilst that for the males was 77.3. He concluded that the variation he observed in the cranial index in this sample did not reflect, as some scholars before him had suggested, foreign forms. He considered that all the Pompeian skulls conformed to a common type, which he called the ‘Pompeian’ or ‘osco-campano’ type. This information, in conjunction with the rest of the large set of metric data that Nicolucci collected, was used to describe the features of the ‘Pompeian’ type.15

The Pompeian type

From his cranial analysis, Nicolucci considered that he was able to identify a Pompeian ‘type’. This ‘type’ was considered to be specifically Southern Italian and was comparable to other regional ‘types’, like the Oscan or Samnitic ‘type’. It presented under the various cranial forms, dolicocephalic, mesocephalic and brachycephalic, though, in general, mesocephaly was found to be predominant, followed by brachycephaly and then dolicocephaly. Nicolucci considered that brachycephaly was more frequently found in the female than in the male cranial series, whilst there was a higher incidence of dolicocephaly in the male series. The mean cranial capacity for males was calculated at 1500 cc for the males and 1323 cc for the females. The majority of the males were found to have a cranial capacity higher than the overall mean of 1412 cc, whilst most of the females were observed to have a cranial capacity below this figure. The forehead was not found to be very broad and apparently was less than that observed on Roman crania. The breadth of the anterior part of the skull was said to be a distinguishing feature of Roman skulls. The top of the frontal bone was observed to generally take a more or less ogival or pointed arch form, a characteristic that was considered to be a regional constant through time.

The form of the face was described as more or less oval, with cheekbones that only projected minimally. The jaw of the Pompeian type was considered to be rather large, heavy and high, the net result of which was a lengthened chin. The nose was characterized as high, large and narrow, or leptorrhine, with a nasal index of less than 48. The orbits were observed to be rounded and slightly slanted at the external edge. They were of moderate aperture. The orbits were found to be proportionately larger in the female sample.16

It is worth noting that though Nicolucci claimed that the population was mostly mesocephalic, there actually were equal numbers of brachycephalic and mesocephalic skulls in his sample. The sample mean was mesocephalic. Perhaps the reason for Nicolucci’s tendency to present the population as more mesocephalic than the evidence actually suggested was that he had a preconceived idea of the features of the people from this region. The socalled Mediterranean ‘race’ that was first fully described by Ripley in 1899 supposedly included the people who occupied the area around Pompeii. This ‘race’ was characterized as relatively ‘longheaded’. Kroeber noted that despite the fact that this would mean dolicocephaly, most European ‘racial’ groups described as ‘longheaded’ were, in fact, mesocephalic. It is likely that these notions about regional differences predated Ripley, which may explain Nicolucci’s attempts to massage the data.17

The contribution of art

Nicolucci claimed that the wall paintings that had been revealed during the excavations in Pompeii provided an additional source of information about Pompeian anthropology and physiognomy. He stated that the different indigenous types were well illustrated in these paintings. He argued that even though most of the paintings showed depictions of Greek mythological scenes, local models were used. He presented examples from the Italian Renaissance to demonstrate his point. He considered that Raphael’s Madonna was an example of Umbrian beauty and that Titian’s voluptuous Venus was a reproduction of the form of a typical Venetian woman. He further claimed that the indigenous Pompeian type could still be observed in the modern populations in the region around Pompeii. He included a plate of lithographs in his monograph, which he described as ‘faithful reproductions’ of faces from Pompeian wall paintings. He used these to illustrate the faces of the local Pompeian types, which, not surprisingly, were noted as mostly mesocephalic (Figure 3.1).18

The idea that the wall paintings re flect the features of the ancient Pompeians has persisted, as exemplified in the writing of Conticello, a previous Superintendent of Pompeii, who described the various indigenous

Figure 3.1 Illustration of the Pompeian Type as identified by Nicolucci from Pompeian wall paintings (Nicolucci, 1882, 24–25)

components that he considered were detectable in the wall paintings from the Villa of P. Fannius Synistor in Boscoreale: ‘The Campanian element is seen in the soft, flabby physiognomies, the large dark eyes, the familiar facial features and the indolence of the heavy, thick-skinned bodies that lack inner tension.’19

There has been considerable discussion about whether Pompeian paintings comprise portraits or idealized images.20 It has been suggested that whilst portraits can be identified amongst Pompeian wall paintings, for example that erroneously identified as Paquius Proculus and his wife, the term cannot be used in the modern sense. Current wisdom is that portraits were supposed to be read, first, in terms of the context in which they were found and, second, in terms of the style and the attributes associated with the subject. These attributes provided information about the profession and status of an individual, as can be seen in the presumed portrait of Paquius Proculus, which shows the subject holding a writing implement, waxed tablets and a scroll, thus indicating literacy.21 It is difficult to determine whether these portraits are actually representational. It does, however, seem unlikely that the illustrations Nicolucci chose for his work represent specific individuals or even generalized images of members of a population. The suggestion that the works of Renaissance artists reflect regional types can also be questioned. A comparison, for example, between the work of Leonardo da Vinci and Botticelli, two fifteenth-century artists who were both trained in Florence under the same master, Verrocchio, demonstrates that they were not painting the same ‘types’ of individuals. The features of Leonardo’s so-called Mona Lisa, which is purported to be a portrait of a Florentine woman, are demonstrably different to those of the women in Botticelli’s Primavera.22 It is rather simplistic to assume that idealized models of women owe more to regional form than the idiosyncratic personal preferences of the artist. As can be seen from Conticello’s description, the interpretation of indigenous types from paintings appears to be based on preconceived ideas, like the notion that South Italians are dark and swarthy.

Evidence of foreign cranial types in Pompeii

Nicolucci considered that four of the hundred skulls he studied presented a type, which he described as very close to that of the ancient Romans. All four were male, two being mesocephalic and two brachycephalic. According to Nicolucci, there were a number of characteristics that suggested a Roman origin for these crania. He determined that these four crania were fuller, anteriorly wider and more flattened than the other Pompeian heads. The brows and orbits were larger, and the jaws almost circular in form. The cranial capacity was equal to the mean of the Roman crania, which was calculated at about 1525 cc. He also observed what he considered to be Pompeian traits on these skulls, such as the lack or minimal protrusion of the frontal sinuses, the lack of or slight depression of the nose at its root and ‘the singular delicacy of all the contours of the skull’. He speculated that these people were not purely Roman but instead were the result of intermarriage between Romans and the indigenous Pompeian population. He also proposed as an alternative explanation, that fusion of types he observed might relate to the common origin of the Roman and Campanian populations.

One skull in particular claimed Nicolucci ’s attention. It was interpreted as that of a young male, notable in Nicolucci’s eyes for its considerable length and laterally protuberant zygomatic bone as well as for its smooth temples, prognathic maxilla and large and heavy mandible. The facial angle was found to be about 70 degrees. He calculated its cephalic index to be 68.8 and its cranial capacity to be just below 1351 cc. Nicolucci considered that this skull had no European parallel and that its characteristics could only be seen in the so-called negroid type associated with people from the African continent. The principal measurements of this skull were compared with the mean measurements of the dolicocephalic male Pompeian skulls. Nicolucci found the proportions of this skull to diverge significantly from those of the other Pompeian dolicocephalic skulls and all other Italian dolicocephalic skulls.

Nicolucci primarily based his classi fication of this skull on a figure in Retzius’ Ethnologische Schriften as well as Hartman’s Die Nigritier since he did not have any African skulls for comparison. The figure Nicolucci consulted in Retzius’ publication was of an old Abyssinian who had been in the service of a European family. He considered that the proportions of the different parts of the head in Retzius’ publication corresponded very approximately with those of the skull from Pompeii. He suggested that there was, therefore, a high probability that the cranium in the Pompeian sample belonged to an individual from this region. Nicolucci did not consider the discovery of a ‘negro’ skull in an ancient Roman city at all remarkable since slaves from conquered countries were a well-known element of the Roman economy. In addition, he cited the identification of ‘negro’ features on a plaster cast of a victim stored in the museum in Pompeii.

Nicolucci concluded that the population of Pompeii in AD 79 was heterogeneous, incorporating an indigenous population along with people from other provinces, such as Rome, and from countries beyond Italy, as evidenced by the skull he identified as ‘negroid’. He noted that though there were a number of cranial forms as reflected by the presence of the three types of cephalic index, they combined to form a specific Pompeian ‘type’ which was comparable to that identified as the ancient Oscan ‘type’ in other parts of Southern Italy.23

Nicolucci’s contribution to knowledge of the ancient Pompeians

Nicolucci ’s work differed from that of his predecessors in that he chose to work on a large sample of material. The work that preceded him was often antiquarian in nature, generally involving only a few examples from which conclusions were drawn or single specimens that were thought to be of interest for some specific feature, such as pathological change. Like most other nineteenth-century physical anthropologists, Nicolucci considered that it was possible to characterize a population solely on the basis of skull shape (see below). He made a battery of measurements on an apparently statistically valid sample. His statistical treatment of the data was very basic, mainly involving the determination of means for specific measurements and indices.

The most signi ficant contribution Nicolucci made to physical anthropology was the publication of the raw data he collected so that it could be compared to and incorporated into contemporary skeletal studies (see below and Chapters 6, 7 and 9). The interpretations he made from this data are generally not relevant in terms of modern skeletal studies as most of the notions he held about the value of craniometry for the determination of European population types are no longer considered valid. For example, the majority of the differences that Nicolucci invoked to separate the Pompeian from the Roman ‘type’ are almost certainly too superficial to be more than artefacts.

A point to consider when assessing a nineteenth-century analysis of skulls is the demonstrated tendency for such works to be used to reinforce commonly held beliefs about the status of different ‘races’ and sections of society. Nicolucci’s work is distinguished by the absence of value judgements, especially when compared to the work of Linnaeus, which provided the basis for his study.24

It is notable that the cranial capacity of the skull Nicolucci interpreted as a ‘negro’ male is much lower than the male mean for the Pompeian sample. In fact, it is barely higher than the mean that he calculated for the female sample. Most nineteenth-century scholars believed that cranial capacity was a reflection of the size of the brain and, hence, intellectual capability. Groups were ranked hierarchically by the means of their cranial capacity. Such an approach can be seen in the work of Morton. This notion has since been discredited as no evidence has been found to link brain function with size.25 In the same vein, the difference between the male and female means for cranial capacity is so great that, even though it is not stated in Nicolucci’s paper, it is possible that this feature was used as a major criterion for sex determination, as it was commonly assumed that female cranial capacity is significantly lower than that of males (Chapter 6).

Nicolucci did not state what he used to measure cranial capacity. It has been demonstrated that there can be considerable variation in this measurement as a result of the use of different materials; for example in 1841 Morton revealed that white mustard seeds produced more variable results than lead shot. Morton observed differences of up to four cubic inches on remeasurement of skulls with mustard seeds and a margin of error of only one cubic inch with one eighth inch diameter lead shot.

Gould considered that it was possible that Morton may have exercised unconscious bias in the interpretation of measurements made before he forsook mustard seeds. It is possible that Morton unintentionally favoured results which reinforced his preconceived notions of what the cranial capacity ranking of the skulls should have been. That any bias on his part was unconscious was borne out by the fact that Morton presented both his raw data and described his methods.26 Nicolucci’s work is harder to scrutinize as he did not document his materials and methods. While it is likely that the interpretations of the skulls were directly related to the cranial capacities he calculated, consideration should be given to the possibility that Nicolucci’s method of measurement could have unconsciously provided him with the results he expected for both sex and ‘race’.

Historical overview of population studies

To appreciate Nicolucci ’s contribution to our understanding of the Pompeian population, his work needs to be viewed in the context of the development of the discipline of physical anthropology. From the eighteenth to the second half of the twentieth century, the main focus of population studies, based on human skeletal remains was taxonomic, where human groups were separated into so-called ‘races’ by anthropologists, including Nicolucci, Angel and Fürst. The word ‘race’ was initially introduced into scientific literature as a zoological term by Buffon in 1749. ‘Race’ was first used in a classificatory sense for humans in 1775 by Johann Friedrich Blumenbach, a German doctor and anatomist who is credited as a pioneer in physical anthropology and founder of craniometry. The impetus for the taxonomic approach to describe human variation was the work of Linnaeus. Traditionally craniometric analysis of the skull was used to provide this information, though non-metric traits could also be used as population markers.27

Blumenbach considered race to be a useful tool for classi fication. He did not invest too much meaning into his system as he considered it to be arbitrary. This appears to have been a reasonable approach, which is borne out by recent studies. In general, he described human variation as continuous, with relatively trivial differences between groups. The classification of people into specific groups is dependent on the criteria that are selected. Various criteria, such as retention of the enzyme lactase into adulthood, the presence of the gene for sickle cell anaemia or different fingerprint patterns all produce sets of groups which are composed of totally different collections of people. Despite Blumenbach’s attitude to race, the concept became politicized. In the nineteenth century, there was a tendency to use racial classification to rank different human populations, usually with the group with which the investigator was affiliated at the apex. The Parisian Société d’Anthropologie, founded by Paul Broca in 1859, institutionalized craniology as the basis of anthropological research.28

Nineteenth-century racial classification systems were based on several assumptions; most importantly, that skeletal traits were immutable and that the divisions between races were hierarchical. The notion of immutability was challenged by Boas in 1911 when he published his findings of cephalic index measurements, a popular race descriptor, on the children of European immigrants to America. He discovered that environment played a significant role in the determination of this index, as it varied between the children and their parents. Subsequent studies in different parts of the world have confirmed the plasticity of certain traits as a result of altered environment.29

The whole concept of racial classi fication was reassessed at the conclusion of World War II and in the following years, when it became apparent that it had been used as a justification for genocide. The agenda of bioanthropological studies was reconsidered and the majority of scholars abandoned racial studies. The issue of the validity of this form of classification, especially for the so-called European races, was discussed at length. Some anthropologists, like Coon and D’Amore et al., continued to use the racial classification systems, even though most scholars would no longer consider them appropriate, especially for the description of European populations.30

Biologically, different races are categorized on the basis that they display a tendency to become separate species. One of the features of human populations is that there is no evidence that speciation is occurring. This can be demonstrated by the fact that individuals from populations that have been geographically and temporally separated for many millennia can reproduce and produce fertile offspring. It is now recognized that there are no human ‘races’ and that population differences are not discrete but continuous. The use of the term ‘race’ is no longer considered acceptable by most scholars as it is biologically inappropriate. In addition most researchers would prefer to distance themselves from the nefarious applications of racial studies in the past.31 Nonetheless, there are, albeit superficial, differences between populations that can provide clues to origin. These result from various processes, including adaptation, genetic drift, small foundation populations and inbreeding groups. While it may seem precious to use other descriptors, like ancestry or population affinities, for current studies that attempt to identify different human groups, they do reflect a more appropriate scientific terminology.

The twentieth century

Barnicot and Brothwell

Barnicot and Brothwell32 used the data collected by Nicolucci for the males in his sample as comparative material for a statistical study of ancient and modern bones from various regions in an attempt to characterize the ancient Etruscans and to establish their origins. The Penrose statistic was used to determine the distance between populations in terms of size and shape, based on the measurements for maximum cranial length, breadth and height. They found that the Pompeian male sample was close to the Etruscan and modern Roman samples, as well as those of modern Basques and Iron Age Greeks and Britons. When the set of measurements was increased to ten, the Pompeian, along with Roman, Basque, British and Greek Iron Age skulls were found to be further removed from the Etruscan skulls in terms of shape. In contrast, the addition of more characters had the general effect of reducing the distance between the samples in terms of size.

Consideration should be given to the exclusion of female skulls in this population study. Traditionally, skeletal population studies were based on males because, as Barnicot and Brothwell state, ‘there are … systematic differences in size and shape between male and female skulls’.33 Barnicott and Brothwell accepted Nicolucci’s sex attributions without question. This is not surprising in the light of the fact that their only interest in Nicolucci’s work was that it provided a data set of an ancient Italian population that could be used to compare with skeletal material of ancient Etruscans. Since Nicolucci did not publish his criteria for sex determination, it is impossible to assess the degree of confidence one could have in his sexing. This means that even if one accepts the assumption that population differences are best represented by one sex, the validity of this work can be questioned on the basis of uncertainty as to the accuracy of sex separation.

The work of D’Amore, Mallegni and Schiano di Zenise

D ’Amore, Mallegni and Schiano di Zenise made an examination of the human skeletal remains in Pompeii to mark the 1900th anniversary of the eruption of Mt Vesuvius in 1979. Given the status of such studies by the latter part of the twentieth century, it seems remarkable that the main aim of their study was the ‘racial’ classification of the ancient Pompeian population. They based their work on an examination of 123 skulls. They concentrated their efforts on crania for convenience because the skull has traditionally formed the basis of most anthropological studies. The crania which formed their sample were in relatively good condition, though they lacked mandibles. Skulls were selected from the skeletal deposit in the Sarno Baths. They were cleaned and then moved to the Forum Baths. Each skull was arbitrarily assigned a number.

Numerical and qualitative observations were made for each skull in the sample. Non-metric or epigenetic traits were also described in accordance with the definitions of Testut, which date to 1917. The epigenetic results have not yet been published.34

This study was undertaken as a companion piece to the work of Nicolucci which they described as ‘very accurate and detailed’.35 One of the aims of D’Amore et al. was to check their results against those obtained by Nicolucci. Despite a concerted effort, involving searches in both the Istituti biologici della Facoltà di Scienze di Napoli and in the areas of bone storage in Pompeii, they were not able to locate the skulls that Nicolucci examined. They also applied for but were denied access to the skeletal collection in the Istituto di Antropologia in Naples where they thought Nicolucci’s sample might be found.

Some of the skulls in the sample they collected from the Sarno Baths had large numbers drawn on the frontal bone. They thought it was possible that these bones had been studied by Nicolucci. However, when they checked the measurements against those recorded by Nicolucci they found that there was no concordance. They concluded that the excavators of the skeletons were responsible for these numbers and that these numbered skulls had never been studied.36

Sex attribution and determination of age-at-death

D ’Amore et al.’s first article involved the examination of the sample of 123 skulls to determine their sex and age-at-death (Chapters 6 and 7). They classified 43 skulls as female and 80 as male and calculated that 35 per cent of the sample they studied were female and 65 per cent were male. They concluded that there was probably considerable overlap between the sexes for the features that they chose for sex separation in their sample of Pompeian skulls. They interpreted two skulls as juvenile, 24 males and 25 females as adult, 45 males and 17 females as mature adults and nine males and one female as senile. Pooling the two genders the percentages were: 1.62 per cent juvenile, 39.84 per cent adult, 50.4 per cent mature and 8.3 per cent senile.37

‘Racial’ classification

In their second paper, D ’Amore et al. presented the preliminary results of their ‘racial’ typology of the Pompeians.38 They did not consider that their conclusions were radically different to those reached by Nicolucci nearly a century earlier when he described the ‘Pompeian type’. Their aim was to examine some aspects of the Pompeian crania in greater depth to build on Nicolucci’s original study.

D ’Amore et al. described the Pompeian sample they studied in terms of four indices, namely, horizontal or cephalic, vertico-longitudinal, superior facial (or frontal) and nasal. They considered that these provided the basis for an initial ‘racial’ diagnosis. They defined the ‘types’ for males and females from the means they obtained for each index. The males were described as mesocephalic, orthocranial, mesosemial and mesorrhine, while the females were described as brachycepalic, orthocranial, mesosemial and leptorrhine. The modes produced the same results. The main difference between the males and females was that the females had rounder heads and finer noses.

They presented their transformed data in a series of four tables. Unfortunately, they did not exclude the two juvenile crania from their study. Juveniles are not usually used for population studies as measurements and proportions alter during the period of growth.

The first table summarized the results obtained for the four indices they calculated. The number of male and female cases for each division of an index were published, along with their percentages, means and standard deviations. The number of cases for certain measurements did not justify this treatment; for example, the provision of a percentage, mean and standard deviation for two to three females for several classifications is of questionable statistical value. No statistical procedures were undertaken to test significance.

The other three tables showed the relationship between the data for different indices. Relative and absolute frequencies were compared with the data obtained for the cephalic and vertico-longitudinal indices, the cephalic and frontal indices and the vertico-longitudinal and frontal indices. No comparisons were made with the nasal index due to the comparatively small number of measurements that could be made to the facial region as a result of post-mortem bone loss. D’Amore et al. concluded from their assessment of the tabulated associations between the three cranial indices that the ‘type’ they described for their Pompeian sample really did exist and was reflected by the coexistence of mesocephaly, orthocrany and mesosemy. It is worth noting that these are the features which, according to the authors, define the males in the sample.

The final section of this paper involved the ‘racial diagnosis’ of the Pompeian skulls in this sample. D’Amore et al. classified the skulls as Mediterranean, the ‘race’ that they thought represented Southern Italian populations, albeit with some variation. They suggested that the cephalic index, within certain limits, provided a general guide for ‘racial’ identification. More certain identification was supposedly provided by the other cranial features, whilst local variation was thought to be reflected in the features of the face. The population was considered to be essentially mesocephalic. The relatively high incidence of brachycephaly was attributed to an earlier indigenous type in the region rather than the result of contact with short or roundheaded populations from Eastern countries, like Anatolia, Mesopotamia and Palestine.39

This paper can be criticized on a number of levels. The most important of these is that the basic premises upon which this study was based had been called into question in the preceding decades and ‘racial’ studies, especially those concerning the so-called European ‘racial groups’ had been discredited. Further, D’Amore et al., like Nicolucci, interpreted the data so that they would yield the results they expected, namely that the Pompeian sample reflected a mesocephalic population consistent with a Mediterranean ‘racial’ attribution. While it was perfectly reasonable for a scholar like Nicolucci to make such a study in a nineteenth-century context, it is remarkable that one hundred years later a similar analysis could be conducted so uncritically, and without any reference to the considerable body of recent literature on this topic.

Pompeian skeletal studies from the latter part of the twentieth century to the present

While it took a century for human skeletons to be recognized as an anthropological resource, it required considerably more than an additional century for them to also be seen as a class of archaeological evidence. This meant that, where possible, skeletons would be examined in their excavation context and they would be interrogated in such a way that they would yield answers to questions that actually were of interest to archaeologists. Prior to that, research primarily involved classification. Most archaeologists did not consider this work to be of great relevance to their research and the anthropological studies of Nicolucci and D’Amore and her team were largely ignored in the archaeological literature.

By the time I commenced my work on the human skeletal remains in Pompeii in 1986, all attention had turned to the large collection of victims, which had been discovered four years previously on the beachfront in Herculaneum. These complete skeletons could be carefully excavated and studied in relation to their archaeological context, essentially eclipsing the badly stored piles of bones stored in Pompeian bathhouses (Chapter 5). There was certainly no competition to gain access to the Pompeian material and my work was, at best, considered virtuous, though in general I was just seen as a quaint character working on fairly inferior archaeological material. Sara Bisel told me that she wouldn’t touch the collection of Pompeian skeletons with a bargepole.

Ten years later the situation changed, and despite the compromised nature of the sample, the scientific potential of Pompeian skeletons was recognized and they were again considered worthy of examination. Since my investigation, a number of other scholars have studied these bones. Projects have included revisiting the sample of stored bones in the Forum and Sarno Bath complexes, work on new skeletal finds and previously unstudied skeletons, such as those from the House of Julius Polybius (IX, xiii, 1–3). In addition, there have been attempts to undertake histological examination of the bones and to apply techniques like DNA analysis.40


Only 32 human skeletons were discovered in Herculaneum prior to the 1980s. The lack of bodies was interpreted as evidence that the majority of the inhabitants managed to escape the eruption of Mount Vesuvius, either by sea or by travelling north to Naples. It was assumed that they had this opportunity as they were not exposed to the pumice and ash that covered Pompeii. This material was considered to have been responsible for the large number of victims that were found at the latter site. It was suggested that the people of Herculaneum fled so quickly that they were forced to abandon those that were vulnerable as evidenced by the fact that the few victims found in the urban section of Herculaneum included individuals who could not fend for themselves, such as a baby that was discovered in a wooden cradle in the House of the Gem (Insula orientalis II, n.10). The skeletons of these victims were generally left in situ.

In 1982, a number of skeletons were uncovered on the ancient beachfront and in nearby boat chambers. By 2002, an estimated total of 350 individuals had been discovered in this area. The management of human skeletal finds was completely revised with these new finds (Chapter 11). Giuseppe Maggi, the Director of Herculaneum initiated a collaborative project with the National Geographic Society, which included the provision of an American physical anthropologist, Sara Bisel, who was responsible for excavation, restoration, conservation and scientific investigation of the human skeletal remains. She was ultimately responsible for the curation and publication of 139 of the victims.41

The work of Bisel

Bisel ’s pioneering work on the human skeletons in Herculaneum was the first attempt to really integrate physical anthropological and archaeological research. There is no doubt that skeletal studies at Herculaneum benefited from the majority of human remains having been found at a time when the value of bones as an archaeological resource was appreciated. This meant that there was the potential for the application of a more rigorous approach to an examination of the remains of these victims.

Sara Bisel was well positioned to study this material for National Geographic, as she was comfortable using an approach that was appropriate for the popular magazine (Chapter 1). She was under considerable pressure to individualize the skeletons she studied, giving them names and investing them with personalities that they never had. This, unfortunately, was not limited to the articles that appeared in National Geographic. Her approach, which sometimes included extending the evidence to the realm of speculation, intruded into her scientific publications (Chapters 1 and 8). Nonetheless, the work of Bisel represents a significant shift in attitude from that of her predecessors in Campania, who were primarily interested in physical anthropology, especially craniometry, to an interest in the broader archaeological issues. Her work, in part, reflects the interests of her mentor J. Lawrence Angel who, after having worked on ‘racial’ typology of the ancient Greeks in his early career, became one of the pioneers of palaeodemography after World War II.42

As it was customary for the dead to be cremated in the first century AD in Italy, Bisel considered that these skeletons were of particular value, since they provided the first sizeable sample of an articulated Roman population from that period. In contrast, she argued that the Pompeian skeletal sample was of little academic value because the bones of individuals had become disarticulated over time from the poor storage techniques in the eighteenth and nineteenth centuries.43 She used traditional anthropological techniques to establish sex, age-at-death, stature, stress indicators, population affinity and evidence of pathology and also undertook trace element analysis. Her aim was to examine the Herculaneum bones to determine the health and nutritional status of the population, as well as to gain insight into the occupations and social status of individuals.44 The results of Bisel’s study and those of subsequent researchers at Herculaneum are presented in Chapters 6, 7, 8 and 9.

The work of Capasso

By 1985, the remains of 229 individuals had been recovered from the area around the ancient beachfront. Luigi Capasso and his team commenced a study of 162 of the Herculaneum victims in 1993, which was published in an enormous volume in 2001.45 An additional 54 victims were examined by Torino and Fornaciari.46 Capasso’s sample included the 139 skeletons studied by Bisel. He estimated that he examined about 30,000 bones over a period of seven years.
Like all scholars who have worked on the victims of the AD 79 eruption,

Capasso appreciated that they represented a large sample of people who died quickly of the same cause in a very short period of time. He saw this sample as a source, not unlike a census, which could be used as the basis for palaeoepidemiological studies as well as palaeodemography. Unlike Bisel, he considered the latter to correspond with the morbidity from pathology in the living population (Chapter 7). His work included the use of traditional anthropological macroscopic observations and measurements to characterize the sample as well as trace element analysis. He also undertook microscopic and radiological examination of the skeletons. One of his main aims was to reconstruct every single aspect of the lives and deaths of each individual in the sample available to him, including their illnesses, their daily activities and work. He considered that wealth of artefacts and literary evidence provided a wonderful opportunity for a multidisciplinary study. While his aims were worthy, he used circumstantial evidence in the form of associated artefacts to aid in the interpretation of the skeletons he studied and, in some cases, extended the evidence to produce interpretations about the lives and occupations of the victims that could be criticized in much the same way as those of Bisel or Bulwer-Lytton (Chapters 1 and 8).47 Capasso continues to publish the results of his research, with particular emphasis on pathological issues.48

The work of Petrone

Petrone and his colleagues published a preliminary study of 215 victims, including those that were examined by Bisel and Capasso. As with the previous work on this material, these studies involved the use of traditional techniques to establish, sex, age-at-death and dental health. Trace element analysis was also employed to obtain an indication of palaeonutrition.49 It should be noted that work on the Herculaneum skeletons has been interrupted, sometimes for years as a result of political and funding issues. As a result, there were several years where numerous skeletons languished in the boat chambers at the beachfront. These bones were removed early in 2008 when research recommenced, again under the stewardship of Petrone. In addition to a traditional physical anthropological study, Petrone has now broadened his research design to include an interdisciplinary study with vulcanologists to determine the impact of volcanic events on human occupation in this region over time.50

In conclusion, it is possible to identify a series of alterations in attitude to the human skeletal remains from Vesuvian sites. The value of these skeletons as an anthropological resource that should be curated and studied was initially recognized about one hundred years after excavation commenced in Pompeii. The publications that were produced over the next century had minimal impact on the interpretation of Pompeii as physical anthropologists did not address issues that were considered relevant to archaeology. The Vesuvian skeletons were first acknowledged as a class of archaeological evidence that should be studied in their context as part of an integrated multidisciplinary project when the large number of skeletons were revealed at Herculaneum in the latter part of the twentieth century. Since then, there has been increasing interest and a number of projects have been undertaken on all the available victims that have been recovered from the AD 79 eruption.

If you find an error or have any questions, please email us at Thank you!