'Le Newerk of Maydeston' - Excavation of a Medieval Hospital Site at St. Peter's Wharf Maidstone

115 ‘le newerk of maydeston’ – excavation of a medieval hospital site at st peter’s wharf, maidstone michael henderson and heather knight with a contribution on the documentary history of the hospital by Sheila Sweetinburgh T his report presents the results of work undertaken between 2006 and 2009 on a site at St Peter’s Wharf, Maidstone, by Museum of London Archaeology (MOLAmola) concentrating on those remains, masonry walls and drain, and cemetery burials, that can be identified as forming parts of the hospital of SS Peter, Paul and Thomas the Martyr. This was probably founded between 1244 and 1260 but disused by the end of the fourteenth century, though its chapel still survives. The cemetery burials were concentrated on the western side of the site, those to the east having been disturbed in the nineteenth century. Detailed osteological analysis of the cemetery population demonstrated a high number of adult males, with women and children also represented. This was comparable to evidence recorded from previous hospital infirmary assemblages. T he site at St Peter’s Wharf, St Peter’s Street, is in the centre of the town and bounded by St Peter’s Street to the west and the River Medway to the east (Fig. 1). The approximate centre of the site is at NGR 575642 155670. F ig. 1 Site location. MICHAEL HENDERSON AND HEATHERheather KNI GHT 116 T he various phases of investigation (shown on Fig. 2), all undertaken by MOLAmola, consisted of a watching brief in 2006 (Ferguson 2007) leading to an evaluation in June/July 2006 (Rahmatova 2008a) and an excavation F ig. 2 Areas of archaeological investigation. ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 117 in July/August 2008 (Rahmatova 2008b). A watching brief in May-July 2009 (Knight 2009) monitored drain trenches. The evaluation identified the masonry drain of the thirteenth-century hospital whilst the excavation phase focused on the inhumation burials in the hospital cemetery, which survived in the north-western part of the site. Much of the rest of the site had been heavily disturbed by the construction of the Maidstone Gas Company’s works in 1848 and there was, for example, no archaeological survival in Trenches 1 and 2 (Fig. 2). The hospital of ‘le Newerk’ by Sheila Sweetinburgh Among the documents now held at Hatfield House is a letter from Archbishop Boniface, dated 1258, to the bishop of Lincoln that quotes an indulgence of Pope Alexander IViv (elected 1254) to be given to donors who contribute towards the building of a hospital for the poor and infirm in the archiepiscopal manor of Maidstone (Hatfield House [HH]: Deed 110/19). Exactly when this building work began is unclear but presumably soon after 1244 when Boniface first came to England, although it is possible it post-dates his enthronement in 1249. Nevertheless, work on the archbishop’s new hospital seems to have continued for some years because Nicholas, son of Robert le Pylere (HHhh: Deed 109/21), gave three ‘deywercas’ of land near the way that led to Maidstone bridge towards the building of the hospital at ‘Petereshell’ [in the west borough]. Even though the deed is undated, it has been assigned to 1270, as have a further two deeds (HHhh: Deeds 58/14; 56/10) that refer to donations towards the hospital’s construction from Nicholas le Pikere and William de Mywelle, both local men. In part this may explain the two dates of 1244 and 1260 given for the hospital’s foundation in antiquarian sources such as Lewis (1844, 216) and Russell (1881, reprinted 1978, 22) in the absence of a foundation charter. Yet, even if the building work was slow, the hospital seems to have been functioning prior to 1259/60 because by that time a master had been appointed. As master of the hospital Robert de Bradegare was the recipient of two pieces of land from Hugh a local weaver, the larger of the two areas again being close to the river (HHhh: Deed 110/31). Indeed much of the land the hospital received in the earliest charters seems to have been concentrated close to the river which was presumably advantageous in terms of its management. H ugh’s charter also refers to the poor maintained in the hospital, and like the other early grants there is no mention of pilgrims nor the saint to whom the hospital was dedicated. However another local donor, Alexander the son of Ralph de Wik, did call it the hospital of the novi operis which he said was for the sustenance of the poor (HHhh: Deed 180/14). Again in the absence of a foundation charter it is unclear how MICHAEL HENDERSON AND HEATHERheather KNI GHT 118 many poor people were to be aided but an inquisition a century later (1375) reported that Archbishop Boniface’s hospital of ‘le Newerk’ had been founded for ten persons and a chaplain who was to celebrate for the souls of the king and his progenitors, the founder and his successors and all the departed faithful (Calendar Inquisitions Miscellaneous, IIIiii, no. 959). The master was to be resident, which suggests that in total at the hospital there were twelve people (if the master and chaplain were not the same) and it seems highly likely that this was an all-male establishment. Whether the allowance was the same for all is unclear, but each of the brethren was to receive weekly seven loaves of different types of bread and twenty-one quarts of ale, with a daily dish of pittance worth a penny, which presumably varied according to the liturgical season (Cal. Inq. Misc., IIIiii, no. 959). A s noted above, the hospital was located on the west bank of the Medway and close to the bridge, and even though the town was primarily across the river it was not a great distance from the main urban area. This presumably explains a charter of 1261 where the archbishop procured the right to hold a market at the hospital. Yet this attempt to expand Maidstone’s commercial life to the west of the Medway appears to have been unsuccessful, the area remaining underdeveloped (Clark and Murfin 1995, 25). Notwithstanding this apparent lack of local support, the hospital’s position vis-à-vis the road and bridge (c.100m) probably enhanced the community’s ability to collect casual alms from passing travellers and perhaps people from the town’s hinterland en route to the east bank (Sweetinburgh 2004, 98, 99). Early support for the hospital was not confined to members of the laity, most of whom were tenants of the archbishop’s manor of Maidstone, because the second master, William de Sele, was the recipient of a tene-ment given by the prior of Leeds Priory in 1286 (HHhh: Deed 160/4). By this time the hospital’s holdings included land and pasture in East Farleigh and also a fulling mill in Loose, but it only seems to have received a couple of tenements (HHhh: Deeds 110/20; 110/21; 57/22; 56/10). This pattern of receiving land rather than property continued into the early fourteenth century, the hospital amassing further lands in Maidstone and East Farleigh, and woodland from the prior at Rochester (HHhh: Deeds 160/2; 109/27; 57/1; 110/21; 199/27). R egarding the hospital’s dedication to SSss Peter and Paul, the fourteenth-century benefactors do not mention this either, although a few do continue to refer to it as the hospital of the New Work. Yet it did have a chapel but whether the provision of ten beds was in part of the chapel (as at St Mary’s hospital, Dover; St Mary’s hospital, Chichester) or in a separate building (St John’s hospital, Canterbury) is unknown from the documentary sources (Cal. Inq. Misc., IIIiii, no. 959). T he arrival of the Black Death and subsequent plague outbreaks may ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 119 have had a disastrous effect on the hospital’s community because at Rochester the chronicle and rental evidence suggests that the city and priory experienced high levels of mortality. By this time, too, gifts in the form of land grants to the Maidstone hospital had almost ceased, although this is not unusual because the hospital had been in existence for over a century. Nevertheless, a reduction in personnel may have helped to dissuade potential benefactors, and instead the house’s archive seems to reflect a policy of leasing out part of its holdings (HH: Deeds 58/12; 214/9; 117/9). Yet, even if the hospital may have experienced some difficulties in the later fourteenth century, it may have sought to extend its charitable remit by providing overnight accommodation for pilgrims. This remains total conjecture but there is a strong tradition that the hospital did provide succour for pilgrims and, if this was not the case at its inception, the 1360s and 1370s seem to be a viable alternative. Reactions to plague apparently included pilgrimage to St Thomas of Canterbury if the amounts collected at the various stations linked to the saint at Canterbury Cathedral are indicative (Nilson 1998, 214). In addition, St Thomas’ jubilee of 1370 was apparently a great success in terms of pilgrim numbers and the first mention of the (in)famous Roode of Boxley also dates from the same period (Elizabeth Eastlake, pers. comm.). Alms given at Boxley Abbey attest to an immediate rise in its popularity as a site of veneration (Elizabeth Eastlake, pers. comm.), many of the pilgrims passing through nearby Maidstone where some of the poorer ones may have stayed at SSss Peter and Paul’s hospital, as they did at hospitals on the main routes from London and the coast (Sweetinburgh 2004, 96; Webb 2000, 224). However it is possible the bridging of the Medway at Aylesford, probably in the later fourteenth century, could also have led to pilgrimage traffic bypassing Maidstone (Knight 2009, 15). Nevertheless if this did happen it was apparently insufficient to maintain the hospital’s community because the inquisition of 1375 states that there were presently five men living at the house (Cal. Inq. Misc., IIIiii, no. 959). This shortfall seems to coincide roughly with the early plague outbreaks and perhaps as a means of compensating for the decline in numbers several archbishops had been giving places or corrodies to their (aged) household and manorial servants. Again this occurred at other archiepiscopal hospitals in Kent (the three Canterbury hospitals of St John, St Nicholas and St Thomas). Yet the hospital was seemingly able to provide a very reasonable level of support of 14d. per week for each brother (Cal. Inq. Misc., IIIiii, no. 959), which suggests that the house was far from destitute. Instead its incorporation by Archbishop Courtenay into his new college of All Saints on the opposite side of the river in 1395 (HHhh: Deed 199/31. Hasted 1798, 308; vch 1926, 232) may indicate its value to the fledgling institution. Clark and Murfin have come to the same conclusion because, as they note, the hospital’s income included the MICHAEL HENDERSON AND HEATHERheather KNI GHT 120 F ig. 3 Medieval stone drain. ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 121 profits of several churches, land rents and other temporal sources (1995, 25). An indication of the hospital’s assets can be gleaned from a charter of 1397 in which the master and college were the recipients of a messuage, a hundred acres of land, ten acres of pasture, an acre of woodland and 10s. rent from Maidstone (HHhh: Deed 56/12). Nonetheless, even if the hospital of SSss Peter and Paul had ceased to exist as a charitable institution before 1400, a remnant of the old hospital still stands: its chapel forms the chancel of the church of St Peter immediately west of the site (see Fig. 2) Excavation Findings T he hospital drain A substantial drain (Fig. 3), built entirely in mortared Kentish ragstone and running east from the hospital complex towards the river Medway, was first recorded in 2006 (Ferguson 2007) and subject to detailed excavation in 2008 (Rahmatova 2008b). The first element of the hospital to be identified, it would have carried foul and storm water and waste to the river. Its superstructure of randomly coursed blocks rose to an arched capping and its internal height from floor to apex was 1.6m (Fig. 3). The base of the drain, c.1.15m wide, abutted the internal faces of the walls and must have been inserted after they were built. In all c.33m of the length of the drain was recorded (see Fig. 4). The central section of the drain had been cut through by a later service trench and the drain had been blocked, almost certainly when the gasworks were being built. T he hospital buildings S outh of the drain, a series of medieval ragstone walls (Fig. 5) were recorded in 2009 in a limited area of investigation on the western edge of the site, c.30m to the north-east of St Peter’s church (Knight 2009). Two parallel east-west walls and a north-south wall defined the eastern side of a rectangular room, c.1.9m wide internally and at least 2.9m long. There was no evidence for the room’s function. The east-west walls were over 0.6m wide and a probable buttress on the northern side of the northern-most wall suggests that this was an external face. The north-south wall was much narrower, at only 0.36m wide, and was most likely a partition wall within an east-west aligned building. The floor of the room, constructed from post-medieval brick, indicates that the building was in use several centuries after the hospital closed at the end of the 14th century. To the south of the room was a third east-west aligned ragstone wall, in this case 0.5m wide, which may have formed the south side of a c.1.3m wide corridor or cloister walk running along the south side of the building. T he southernmost of the recorded walls, a 1.3m long section of ragstone MICHAEL HENDERSON AND HEATHERheather KNI GHT 122 wall either turning south or buttressed at its western end, was built over an early, undated pit. Although it appears to be contemporary with the three walls to its north, it does not fit easily with them nor is on quite the same alignment. F ig. 4 Plan of graves and location of drain and medieval walls. limit of excavation internal surface: found and conjectured wall/foundation: found and conjectured drain: found and conjectured (arrow indicates direction of flow where known) ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 123 T he cemetery T he cemetery was on the northern side of the hospital buildings and drain. A total of 55 full or partially articulated individuals were excavated in 2008 and subsequent archaeological phases (Rahmatova 2008b; Knight 2009) of which 31 were found in 30 clearly defined grave cuts. However, the large volume of disarticulated human bone, most likely disturbed by the industrial development of the site in the post-medieval period, indicates that the cemetery population was originally at least double this figure. All the graves were aligned east-west with the skull of each skeleton at the west end of the grave (Fig. 6 and Fig. 4). Where identified, bodies were supine and extended with the hands on or beside the pelvis and so entirely consistent with medieval Christian burial practice (Gilchrist and Sloane 2005, 152; Roberts and Cox 2003, 222). T here are at least four phases of burial on the site with some later graves cutting through and disturbing earlier ones. Some graves contained more than one skeleton but none contained evidence for coffins. Cloth shrouds, pinned or sewn together, were the commonest form of burial wrapping (Daniell 1997, 156; Gilchrist and Sloane 2005, 106-110). An absence of pins from the St Peter’s Wharf part of the cemetery indicates that here the shrouds were sewn. However, ground conditions were such that no evidence of the shrouds material had survived. F ig. 5 Plan of Ragstone walls. MICHAEL HENDERSON AND HEATHERheather KNI GHT 124 F ig. 6 Plan of burials. ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 125 N o evidence for grave markers was apparent, though this is not unusual. The lack of permanent grave marker does not mean, however, that the graves were not marked in the short-term by either the mound itself, a hearse cloth or by the laying of flowers (Daniell 1997, 146). As the hospital was founded 1244-1260, the first burials most probably date to the later thirteenth century. Most of the graves were sealed by a layer of cemetery soil which represented the relevelling and/or reorganisation of the burial ground and which included disarticulated human bone. There was, unfortunately, no dating evidence obtained from this deposit though it did contain residual pottery fragments dating from c.ad 900-1100 and 1125-1250 which would suggest activity on the site prior to the founding of the hospital. (A pit found beneath the medieval structures was the only clear evidence of this earlier phase of occupation.) However, as only a few graves postdated this relevelling it is likely to have occurred towards the end of the life of the hospital. Nor were there datable grave goods within the graves themselves. This again is typical of burial practice during the medieval period. W here pottery contemporary with the use of the hospital was encount-ered, it was as residual material in later, disturbed contexts. For example, a sherd of pottery from a lid or base of a flaring dish probably manufact-ured in Maidstone c.1250-1400 came from a nineteenth-century indust-rial context (Blackmore 2010). The cemetery population by Michael Henderson T he analysed assemblage comprised 55 individuals. A large quantity of disarticulated human bone from the site, which emphasises that the burial ground was severely truncated during the subsequent uses of the site, was not subject to further analysis because of its highly mixed and fragmented character. However, assessment indicated that it represented a minimum number of 120 additional burials of which 85 (70.8%) were adults and 35 (29.2%) subadults. It is apparent that the analysed sample represents a minority of the buried population and that the cemetery as a whole could have held a higher proportion of subadults. T he analysis of skeletal assemblages from monastic burial grounds and lay populations within parish cemeteries has previously revealed significant patterns of spatial zoning by sex and age, though these vary according to the function and order of the institution involved (Gilchrist and Sloane 2005, 203-205). It was not possible to determine the full extent and layout of the original hospital burial ground partly excavated at St Peter’s Wharf and the extent to which the demography and other characteristics of the analysed sample are representative of the original cemetery population as a whole is unclear. O f the 55 analysed individuals from St Peter’s Wharf, 46 (83.6%) were MICHAEL HENDERSON AND HEATHERheather KNI GHT 126 adults (≥18 years old) and nine (16.4%) subadults. It was possible to determine the sex of 31 (31/46: 67.4%) of the adults within which there was a pronounced bias towards males who outnumbered women by 2.9:1. Other hospital populations such as St Mary Spital, London, St Nicholas, Lewes and St Leonard, Newark, have also all shown predominantly male populations (ranging from 45-75%) and low proportions of subadults. These results may indicate high levels of young, adult males, possibly rural migrants, drawn to larger towns and cities. Pregnant women were often excluded from some types of hospital and were forced to seek alternative refuge: for example higher proportions of female burials and infants have been found at leper hospitals (Gilchrist and Sloane 2005, 205-06). The age at death ranged from children aged 1-5 years to adults aged ≥46 years. Overall most adults died aged 36-45 but half of the women were ≥46 years old at death and poor preservation of aging elements within the osteological sample may indicate that this age profile actually understates the longevity of the hospital population. The stature of 13 of the sexed adults could be determined. The mean stature height for men was 1.720m (5ft 7½in.), and 1.566m (5ft 1½in.) for women. These values are quite typical of late medieval populations examined elsewhere (Roberts and Cox 2003, 269). Bacterial and viral infections would have accounted for high mortality in the medieval period and prior to the advent of antibiotics approximately half the population would not have survived past childhood (Ortner 2003). However, overall pathological prevalence rates at St Peter’s Wharf were low with infectious bone changes limited to three individuals and no evidence of specific infections such as tuberculosis or leprosy. Many pathological conditions only affect the soft tissues of the body and death could often occur long before any disease reached the bone. Any T able 1. summary of the cemetery population F ull/partially articulated skeletons Defined grave cuts D isarticulated skeletons – no. of contexts D isarticulated skeletons – estimated minimum no. of individuals 2006 evaluation/ 2008 excavation 30 17 20 ≥89 2009 excavation 21 14 7 ≥31 T otal 51 31 27 ≥120 ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 127 bone changes observed therefore may reflect the survival of healthier individuals who overcame illness and not reflect the true rates of disease in a population (Wood et al. 1992). T he St Peter’s Wharf population showed lower prevalence rates of dental disease than those typically recorded for the period (Waldron, 2007, 117-119; Roberts and Cox 2003, 265). All dental pathology, with the exception of enamel hypoplasia (a defect in enamel tooth formation), displayed an increase with age before a decline into the older age categories. Males suffered a greater number of carious erosions (tooth decay), calculus (hardened plaque), periodontal (gum) disease and tooth loss during their lifetimes than females, though this might indicate a difference in diet between the sexes, the small size of the female dentition sample, particularly in the oldest ≥46 year age range limits the scope for direct comparison. T he most dramatic example of pathological bone change recorded in the St Peter’s Wharf assemblage was that of male [3028], of undetermined age. There was evidence of at least four distinct blade wounds to the cranium from an edged weapon, possibly a sword, and it is clear that the victim was not wearing head protection or armour. Conclusions T he development of towns throughout England in the eleventh and twelfth centuries led to the founding of many hospitals around urban areas during the thirteenth century (Schofield 1999, 210; Magilton et al. 2008, 36-37). Although hospitals fall into three principal categories – infirmaries, leper hospitals and almshouses – all were essentially flexible and frequently changed their primary role (Gilchrist and Sloane 2005, 205). The demographic profiles of those buried at such sites may provide evidence of specialisation by age or gender. Burial grounds associated with the accommodation of travellers may contain a higher number of young adults whereas an infirmary may reveal a wider distribution of ages and evidence of chronic pathological bone changes (Magilton et al. 2008, 33). A nalysis of the human skeletal remains recovered from St Peter’s Wharf presented an opportunity to study a previously unknown hospital population, dating from the later thirteenth century and after, from the south-east of England and to compare it with contemporary burial grounds. While analysis was limited by the fragmentary and incomplete nature of many of the skeletal remains, this assemblage provided important demographic and health evidence regarding the local population, contributing to the body of data collected from skeletal remains for this period. Whilst the osteological analysis could not determine the exact function of the hospital, the demographic profile revealed a high number MICHAEL HENDERSON AND HEATHERheather KNI GHT 128 of adults, including some females, who had survived into the older age categories, as well a number of subadult burials. This was comparable to evidence recorded from previous hospital infirmary assemblages. A s towns and markets grew and expanded there was an increase in the quantity and range of foods available to eat. A diet high in grain products such as bread and ale would have formed the basic staple, although a more varied diet that included fresh meat, poultry, fish, dairy produce and vegetables may have been available to those in towns that could afford it. Increased rates of dental disease for this period may reflect a lack of dental hygiene practised as opposed to a high protein diet (Roberts and Cox 2003, 241, 244, 262). O ne of the more severe and interesting examples of pathological bone change observed was that of an adult male who had suffered multiple sharp force cranial trauma with evidence of a previous healed wound. These injuries would have been fatal: they are likely to be the result of interpersonal violence but it is not possible to conclusively say whether the individual was a professional solider or how he came to be buried at St Peter’s Wharf. acknowledgments MOLA mola would principally like to thank esn Limited/Citygrove Securities PLC for funding the fieldwork and commissioning this report. Adam Single of Kent County Council Heritage Conservation Group provided important input into the project. During the fieldwork phase, project manager Paul Abbott, of Deakin and Jones, helped the smooth running of the archaeological interventions considerably. The earlier phases of the fieldwork were supervised by Bruce Ferguson. The MOLAmola specialists other than the authors who were involved in the assessment or analysis phases of the project were Ian Betts (building material), Lyn Blackmore (pottery) and Jacqui Pearce (clay tobacco pipe). Liz Goodman undertook the conservation work. Thanks are also extended to all other MOLA staff involved in the fieldwork. This article was edited by Julian Hill. The graphics were prepared by Judit Peresztegi. [Editor’s note: the full osteological analysis of the cemetery population is published on the KAS website, kentarchaeology.ac ] bibliography Blackmore, L., 2010, ‘The pottery from St Peter’s Wharf, St Peter’s Street, Maidstone, Kent (KTkt-spw06)’, unpubl. mola report. Clark, P. and Murfin, L., 1995, The History of Maidstone. The making of a modern county town, Alan Sutton, Stroud. ’LE NEWERK OF MAYDESTONmaydeston’: ST PETER ’S WHARF excavation, maidstone 129 D aniell, C., 1997, Death and burial in medieval England 1066-1550, London. F erguson, B., 2007, ‘St Peter’s Wharf, St Peter’s Street, Maidstone: Archaeological watching brief report’, unpubl. MOLAmola report. Gilchrist, R. and Sloane, B., 2005, Requiem: the medieval monastic cemetery in Britain, MOLAmola London. H asted, E., 1798, The History and Topographical Survey of the County of Kent, Vol. 4. K night, H., 2009, ‘St Peter’s Wharf, St Peter’s Street, Maidstone, Kent: a supplementary post-excavation assessment and updated project design’, unpubl. MOLAmola report. L ewis, S., 1848, A Topographical Dictionary of England. M agilton, J., Lee, F. and Boylston, A., 2008, Lepers outside the gate: excavations at the cemetery of the hospital of St James and St Mary Magdalene, Chichester, 1986-87 and 1993, CBA Research Report, 158, London. N ilson, B., 1998, Cathedral Shrines of Medieval England, Woodbridge. O rtner, D. J., 2003 (1981), Identification of pathological conditions in human skeletal remains, 2nd edn, London. P age, W., 1926, Kent, Victoria County History, vol. ii. R ahmatova, N., 2008a, St Peter’s Wharf, St Peter’s Street, Maidstone: an archaeological evaluation report’, unpubl. MOLAmola report. R ahmatova, N., 2008b, ‘St Peter’s Wharf, St Peter’s Street, Maidstone: a post-excavation assessment and updated project design’, unpubl. MOLAmola report. R oberts, C.A. and Cox, M., 2003, Health and disease in Britain: from prehistory to the present day, Stroud. R ussell, J., 1881, The History of Maidstone (reprinted Rochester, 1978). Schofield, J., 1999, ‘Landscapes of the Middle Ages, Towns 1050-1500’, in Hunter, J. and Ralston, I. (eds), The Archaeology of Britain: an introduction from the Upper Palaeolithic to the Industrial Revolution, 210-227. S weetinburgh, S., 2004, The role of the hospital in medieval England, Dublin. S weetinburgh, S. (ed.), 2010, Later Medieval Kent, Woodbridge. W aldron, T., 2007, St Peter’s, Barton-upon-Humber, Lincolnshire: a parish church and its community: vol. 2 The Human Remains, Oxford. W ebb, D., 2000, Pilgrimage in Medieval England, London. W ood, J.W., Milner, G.R., Harpending, H.C. and Weiss, K.M., 1992, ‘The osteological paradox: problems of inferring prehistoric health from skeletal samples’, Curr. Anthropol., 33(4), 343-70.