New Challenges and Perspectives on the Long‐Term Study and Curation of the Collection of Identified Human Skeletons Housed at the National Museum of Natural History and Science, Lisbon, Portugal

Almost 20 years ago, one of us (HC) published a paper in this journal raising international attention to a vast collection of identified human skeletons curated at the National Museum of Natural History, University of Lisbon, Portugal (Cardoso 2006a). The collection had enormous research potential due to its size and the quality of the documentary information associated with the skeletons. It was amassed between the late 1980s and 2004 by the late Luis Lopes and the first author, who was the de facto curator between 1998 and 2011, and who further increased its potential when the juvenile component was expanded and developed (Cardoso 2006b). The collection grew from about 1500 individuals curated in 1998 to about 1800 individuals, with 1674 fully accessioned, by 2011 (Cardoso 2014). The skeletons originated from unclaimed temporary secondary interments in the modern managed cemeteries of the city of Lisbon and were obtained with permission and in collaboration with municipality authorities. All individuals can be personally identified through a combination of burial, death, and birth records, but by 2011 this process had only been completed or completed partially for 912 of them (Cardoso 2014). Most of the individuals in the collection lived and/or died in the city of Lisbon between 1850 and 1950. Due to its origins and background, this and other similar skeletal series can be described as cemetery-based skeletal reference collections, which are curated in institutions throughout Central and Southern Europe and Latin America (Cardoso 2021). At the time of the publication of the 2006 paper (Cardoso 2006a), the collection as a whole still experienced significant challenges with conservation, and the National Museum of Natural History and Science in Lisbon undertook a substantial effort to address the most critical issues over the years. These issues with long-term conservation were made particularly urgent as the collection was increasingly sought after by both Portuguese and international researchers (Cardoso 2014). For example, between 2001 and 2011, the collection was featured in more than 40 papers published in peer-reviewed journals, with about 80% being published since 2006. Two conservation and curation challenges that have had an impact on the integrity of the collection have become increasingly evident in recent years. The first of these concerns is the problem of dissociation. The second deals with conservation, particularly with the handling of juveniles. These challenges have resulted in a partial loss of information and a loss of skeletal elements, raising concerns about the collection's long-term future and ethical stewardship. In this letter, we provide recommendations for dealing with the dissociation and conservation problems for any data already collected and for future data collection, and we present suggestions for the long-term conservation of a very sought-after collection. In the 1980s, during the early efforts to amass the collection, only a section of it had been partly labeled and/or properly accessioned. That labeling process was halted when Luis Lopes, then curator of anthropology, retired in the early 1990s. A new effort was initiated by one of us (HC) in the early 2000s to carry on with the accessioning process and the labeling of the skeletons. As in many other museums, resources were limited, and the process would take several years before it was completed. Skeletons with accession numbers 1 through 759 were partly accessioned and labeled in the 1980s, and then the labeling was completed during the mid- to late 2000s. The remaining accession numbers, 760–1674, were labeled at a second stage initiated in 2011. While efforts were taken to establish a cautious workflow during labeling—including devising a detailed labeling protocol—a dissociation problem has since been identified in accession numbers 760–1674. Dissociation is the separation between the specimen and its accession number (Graham 2018; Newton and Cook 2018) and is one of the most important agents of loss of scientific information in museum collections. While labeling is done in a manner to minimize the risk of dissociation in museum collections (Waller and Cato 2011), in this case, a labeling error was likely the source of dissociation. A problem of this nature results in a permanent loss of scientific data, as any single skeleton can be associated with an incorrect accession number and, thus, associated with the biographic information of an unknown other individual. The issue of dissociation may affect entire individuals or individual bones. The dissociation problem was detected in the mid-2010s by the first author when discrepancies were observed between several individual skeletons and their associated biographical information, such as recorded sex and age. The discrepancies were so pronounced that they could not be due to normal variation where males follow a female pattern or vice versa, or where skeletal age appears to be inconsistent with chronological age. Additionally, visiting researchers also reported several suspected mismatches. There are very few cases where this dissociation occurred at the time of amassing the collection, due to burial practices that resulted in the commingling of two individuals (see below). Isolated errors of this type are known from a number of collections and with a careful review of skeletal and documentary data, they can be assessed and excluded from research when necessary (e.g., see Albanese 2018). However, there is now evidence of a very large number of new cases impacted by disassociation, potentially affecting almost two-thirds of the collection. The workflow process and the location of the problems suggest there was a switch of identities that occurred mostly between two or possibly three adjacent accession numbers. While the exact extent of the problem is unknown, it has not affected the skeletons with accession numbers 1–759. These individuals had been partially labeled in the 1980s and 1990s, minimizing any dissociation problems, and were labeled through completion by museum technical staff under the direct supervision of one of us (HC). This extensive disassociation also did not affect the entire juvenile segment of the collection, including individuals with accession numbers below and above 759, as these individuals were labeled following a separate process. This labeling took place as a component of the doctoral and post-doctoral projects of the first author and by 2011 was practically complete. The process was given priority by museum senior administration because the foci of these projects were the juvenile component of the collection, and much time and effort were dedicated to its conservation. The labeling was completed by students, research assistants, and volunteers, under the direct supervision of the first author and using a protocol similar to that used for the remainder of the collection. Later in the mid-2010s, a detailed dental conservation project was completed by another of us (YQ), which was meant to prevent loss and dissociation. Until 2011, the conservation of the juvenile segment of the collection was never managed by museum technical staff. A recent publication (Magno 2022) confirms the dissociation issue and the unknown extent of the problem. To ascertain the dissociation, we compared the biographical information provided in Magno's (2022) Doctoral dissertation for one of their published case studies with data associated with the same accession number from a file curated before 2011. This comparison revealed discrepancies between the two datasets. For example, accession# MB61-1022 is described by Magno (2022) as a 50-year-old male, whereas in the file mentioned above the same accession# identifies a 57-year-old female. We are certain that this author was unaware of the dissociation, but this example highlights the greater impact on research of this problem. This example is not meant to single out any one researcher. In fact, Magno's thorough approach to presenting a case study allowed for confirmation of dissociation. To address the dissociation issue, we outline recommendations below for researchers who have used the collection since 2011 and have already collected and generated data. We also provide some guidelines to those who have yet to collect data from this collection and are planning to do so. A summary of these problems and recommendations can be found in Table 1. Juveniles (ages 0–20 years) various accession numbers First, data collected from skeletons with accession numbers 1 through 759 are unaffected. The labeling of this section was completed before 2011, with a different workflow, where skeletons were already partly labeled, and subsequent labeling was personally supervised by the first author. As per above, juvenile skeletons in the entire collection are also unaffected, as they were subjected to a separate labeling process. Second, because the extent of the actual dissociation problem is unknown, it is advisable to avoid using the data collected from skeletons with accession numbers 760–1674. Any data collected from these skeletons should be taken with caution, particularly if the researcher has encountered cases where the skeletal sex and age are not consistent with the documented sex and age. One possible solution is to seek identity confirmation by reaching out to other researchers who may have collected some of the same data—preferably metric, morphological, or even paleopathological—prior to 2011, and to which the data can be compared. Third, if analysis of data collected from these skeletons is underway or has been published, we highly recommend authors reassess their data by reaching out to other researchers and comparing at least some data. The publication of article errata may be necessary. The second set of recommendations below is directed at researchers who have yet to collect data and are planning to do so. The safest approach is simply to avoid collecting data from any individual with accession numbers 760–1674 and to focus exclusively on the individuals with numbers 759 and under. As noted above, there are cases of dissociation that stem from the burial and exhumation process at the cemetery, resulting in the comingling of several individuals. Such cases are easily identified from the collection records and can be excluded from research, as they have the same secondary interment plot (ossário or compartimento) number, followed by the A and B letters for each individual (Table 1). These numbers are available in the collection records and can also be found in the tag associated with each skeleton. These dissociation cases occur throughout the collection, and the museum can provide a list of known cases. For sampling purposes or other reasons, researchers may be compelled to collect data from skeletons with numbers 760 and higher. If that is the case, great care should be paid to the matching of osteological observations with biographic data (matching age and sex, for example). Researchers must be aware that this approach may not necessarily detect all or even most of the cases of switched identity—as a 45-year-old male might be skeletally indistinguishable from a 65-year-old male. While isolated errors like this can exist and be assessed and excluded by researchers (Albanese 2018), the extent of the problem in this collection is systemic. The recommendation provided here is to rely on similar or identical data—preferably metric, morphological, or paleopathological—which may have been collected by a colleague before 2011 and to which data from these skeletons can be compared/contrasted. Another issue that researchers should be aware of is that even contrasting measurements or observations with data collected prior to 2011 may be unable to identify problems of dissociation which only affect specific bones and not the whole skeleton. The dissociation issue may be more complex, as it is also unclear whether dissociation affected individual bones rather than complete skeletons. While this is now impossible to reliably ascertain, there are a few rare cases throughout the collection where one individual is associated with a duplication of the same bone labeled with the same accession number. These cases can still be identified in individuals with accession numbers 1 through 759—which have not been impacted by the dissociation problem. This situation is not of particular concern because this duplication of bones is a consequence of burial practices. Individuals in the collection were buried and then exhumed from temporary burial plots before being transported to the museum (Cardoso 2006b). During exhumation, cemetery staff often miss small bones and leave them behind, which will then become commingled later with a new individual when they are buried and subsequently exhumed from the same temporary grave. These problems and possible solutions have been described for other documented collections that were not derived from cemeteries (see Albanese 2018). In addition to the problem of dissociation, in the last decade, part of the collection has suffered a progressive deterioration in its conservation and preservation status due to intensive use. In this case, the juvenile individuals have been the most affected. In 2019, one of us (YQ) completed a bone preservation assessment of the 92 juvenile skeletons identified in Cardoso (2006a). This assessment was completed by simply quantifying the number of bones that were present in each skeleton. This 2019 assessment was then compared with a similar study completed in 2011 when the same count was done. A comparison between these two assessments showed that 46% of the bones present in 2006 and 2011 in each skeleton had been completely lost by 2019. From an average of 98% completeness, the juvenile skeletons went down to about 55%. Younger individuals (60% of bones lost) were more affected than older individuals (27% of bones lost). These bones are no longer preserved, are not available, or are permanently lost. The perspective we offer here serves as a foundation for recommendations to mitigate further damage or loss. Based on our observations, one contributing factor to the conservation issue may be the limited supervision during data collection, which could inadvertently lead to inconsistencies in the handling of materials. Unrestricted access is a consequence of structural problems common to many museums, such as limited funding to have staff overseeing or supporting research. This may have resulted in some researchers having abused their access privileges. Furthermore, we are unaware of any procedures in place that assist the curator in tracking the conservation status of the collection over time, particularly given the high demand for the juvenile component. Regardless of research level of experience, providing unrestricted and unsupervised access may also have resulted in a sense of lack of accountability for misplacement, inadvertent destruction, removal, or simply loss of skeletal material. Greater scrutiny in access requests, where proven experience with juveniles remains, for example, is required from the student and/or their supervisor, will also help prevent future loss. The bone preservation assessment comparing counts of bones of juveniles in 2011 and 2019 was submitted as a report to the Director of the Museum in 2019. Subsequent to this report, the museum implemented a policy to address the conservation of juvenile individuals by restricting student and researcher access to this component of the collection. One of the ways to address the concern with permanent loss of material is that of increased supervision of work by the curator and/or museum staff members. While we are quite aware of the financial constraints of museums, this does not mean that curators or staff are required to stand behind each researcher and follow each step they undertake. Rather, it is to establish a policy of accountability, where a researcher is given specific rules about what can or cannot be done and is held accountable by periodic visits by museum staff to the laboratory. It would also be useful to make use of the skeletal inventories which were initiated in the 2000s and have only been partially completed. Their completion would allow for a detailed monitoring of the conservation status of each individual. Each new visit would require, for example, a complete review of the inventory, or at least of those skeletal elements that are of interest to the researchers. Such inventory could be managed in a digital database and then serve as a basis for ongoing conservation checks. More cumbersome measures could involve, for example, requesting each researcher to complete and sign an inventory before and after data collection. Additionally, initiating the digitization of the collection and creating high-quality 3D models of each skeleton can also assist in the long-term curation of both juvenile and adult individuals. While such an undertaking would likely take a very long time and be costly, the gains would surpass the risks of further irreparable loss in the long run. The technology is now much cheaper and easier to use, while the labor cost associated with a trained technician to capture and store scans may still make this option prohibitive. In the early stages of such a large undertaking, digitizing can be strategically focused on the most sought-after skeletons and elements or those facing conservation concerns. The museum has previously shown capacity for developing and strengthening the skills and resources associated with 3D digitization and can draw on support from students and even visiting researchers. A summary of the conservation problems and recommendations to address them can also be found in Table 1. As biological anthropologists, we have obligations to our scholarly discipline and the wider society in which it operates. Our primary ethical obligation is to the people with whom we work, or in this case—the skeletons of the people with whom we work or have worked. We are committed to ensuring their integrity, dignity, and privacy, and avoiding their undue exploitation. We also have an ethical responsibility for the integrity of research findings that are disseminated to both the scientific community and society at large, and for preserving the collection for future scholars. Given the burial practices and privacy laws in Portugal, there are no immediate implications for living relatives and direct descendants, but there is certainly a larger community that deserves this issue be addressed properly. Both the problems of dissociation and conservation are very concerning given that collections like this one are not the property of any organization. Instead, the collection is held in trust by the Nacional Museum of Natural History and Science for the Municipality of Lisbon, its cemeteries, and ultimately for the Lisbon community. Within this framework, the museum has legal, ethical, and professional responsibilities to the Municipality of Lisbon and the community which it represents and from which these individuals originate. More importantly, by receiving these individuals into a collection, the museum has made an implicit promise to protect these skeletonized individuals from undue harm and destruction. Based on discussions with colleagues and information shared by reviewers of this manuscript, the problems discussed here may be common at other institutions that house similar collections. The lack of open communication and transparency does not help to address the problems with this collection or with other collections. It is an ethical imperative that we have an open dialogue about these issues and for institutions to share with researchers all information about possible and potential problems with collections. Academic careers can be stifled or destroyed by circumstances that result in the collection of erroneous data, even if unknowingly. These problems with data could profoundly impact the accuracy and validity of numerous publications, including numerous osteological methods (such as for age and sex estimation) developed using the collection. Because these methods can then be applied in a of the possible impact recent research potentially those It that it is also imperative that in these curation issues and in problems no long it There is still much research potential in the segment of the collection that has the disassociation problem. The of the collection that is now is still of and may to support research the time and effort in its curation may not the future ethical risks This the by the curator of the collection to concerns of and associated with the scientific study of human skeletal we to see in this data curation project administration data curation project administration data curation We the many colleagues and with whom we discussed the conservation challenges of the collection at the National Museum of Natural History and Science in Lisbon and other similar collections. shared and and that of the three and have been in the of this The authors no of The authors have to

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.