The Non-Adult Cohort from Le Morne
Cemetery, Mauritius: A Snap Shot of
Early Life and Death after Abolition
J. APPLEBY,
a
T. K. SEETAH,
a,c,d,g,h
* D. CALAON,
b
S.
CAVAL,
c
A. PLUSKOWSKI,
d
J. F. LAFLEUR,
e,f
A. JANOO
f
AND V. TEELOCK
f,g
a
McDonald Institute for Archaeological Research, Cambridge, United Kingdom
b
Università Ca’ Foscari, Venezia, Italy
c
ZRC, SAZU, Ljubljana, Slovenia
d
Department of Archaeology, University of Reading, Reading, UK
e
Le Morne Heritage Trust Fund, Le Morne, Mauritius
f
University of Mauritius, Reduit, Mauritius
g
Truth and Justice Commission, Port Louis, Mauritius
h
School of Forensic and Investigative Science, University of Central Lancashire, Preston, United Kingdom
ABSTRACT The cemetery of Le Morne in Mauritius dates from the 1830s and is thought to contain the remains of slaves,
freed slaves or potentially free Madagascans, which in itself has economic and social implications and makes
the cemetery all the more intriguing. During 2010, excavations recovered the remains of 11 individuals, of
which six were children. Although a small sample, the burials of the non-adults show several interesting
features. Two neonates were buried contemporaneously and may have been twins, while a late term foetus
in a grave with a young woman may represent an incident of coffin birth. This sample has the potential to cast
light on the burial practices of the slave/ex-slave community in Mauritius, about which little is currently known.
Copyright © 2012 John Wiley & Sons, Ltd.
Key words: diet and health; emancipation; Mauritius; neonatal; non-adult; slavery
Introduction
The region of Le Morne forms a peninsula on the most
south-western tip of Mauritius (Figure 1). It is most
famous for the ‘Brabant’, an inselburg that rises to some
545 m directly from the shoreline, with an outlook,
and sheer drop, to the Indian Ocean on its western facet.
The inselburg itself, noted to have one very precarious
access point, forms a plateau at its summit. This combin-
ation of a single entrance and an area of flat terrain made
the Brabant an ideal retreat for runaway slaves and came
to be known as the Maroon Republic. The maroonage
movement, exemplifying the spirit of resistance to
slavery, was commemorated with the inscription of
Le Morne as a UNESCO World Heritage Site in 2008,
firmly establishing the role of the island as a crucial
region for investigations of slavery and maroonage.
While historical records have revealed much regard-
ing slavery and resistance on Mauritius (Allen 1999,
2001; Teelock 1998, 2009; Vaughan 2005), archaeo-
logical research has lagged behind. For numerous
reasons, most notable being the dearth of material
cultures deriving from slave communities and the very
real need for systematic archaeological investigations
(which to date have been lacking within the context
of slavery and indenture) (Seetah 2010), history has
provided the main narrative, relying on careful investi-
gation of existing archives. However, given the types
of records kept on slaves, only specific aspects can be
studied from this source of evidence, with little chance
of gaining a true understanding of life ways or ‘resistance’
from archival records alone. For this reason, the inscrip-
tion of Le Morne as a World Heritage Site was unique
in that it was the first site to be dedicated based largely
on intangible evidence.
The aforementioned situation is effectively reversed
with emancipation. Freed slaves literally disappear from
the historical record (Teelock, 2005); however, these
* Corr espondence to: School of Forensic and Investigative Science,
University of Central La ncashire, Preston. PR1 2HE, United Kingdom.
e-mail: KSeetah@uclan.ac.uk
Copyright © 2012 John Wiley & Sons, Ltd. Received 11 October 2011
Revised 6 June 2012
Accepted 17 June 2012
International Journal of Osteoarchaeology
Int. J. Osteoarchaeol. 24: 737–746 (2014)
Published online 25 July 2012 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/oa.2259
communities start to develop their own material culture.
While, as slaves, they would have had material goods
within a limited capacity, the scale of expression within
the context of identity would have greatly increased
once free. In Mauritius, emancipation resulted in a migra-
tion away from inland sugar estates to coastal fishing
communities, and the post-emancipation settlement that
developed at Le Morne typifies this move. Within the
shadow of the Brabant, a small cemetery, which within
local oral tradition had been known as the Malagasy
Cemetery, revealed a first for Mauritius: the final resting
place of a potential post-emancipation slave community.
The article focuses on the first generation of ‘free born’
descendents in Mauritius, with an exploratory study
presented here that centres on burial practice and disease
in order to reveal aspects of early life for a small group of
individuals that did not enjoy the freedom their ancestors
had fought for.
Archaeological investigations at Le Morne
Since 2009, archaeological work has focused on the
exploration of a number of sites that fall within the
buffer zone of the Le Morne UNESCO WHS. An initial
survey of the cemetery itself revealed approximately five
structures, delineated by basalt blocks, and clearly
identified as graves; geophysical survey uncovered a
further 16 structures. In 2010, a concerted clearance of a
larger area surrounding that investigated in 2009 identified
45 structures in total and revealed that two smaller satellite
burial areas ostensib ly orbited the main cemetery.
During the 2010 campaign, a total of 11 skeletons
were recovered from eight graves (Figure 2). Of this
skeletal sample, six individuals were non-adults: three
perinates and three young children. The perinates were
all in double graves: one was buried in Structure 1, the
grave of a young adult female (perhaps its mother), and
two were placed within a single wooden coffinina
grave of their own (Structure 6).
Although the precise date of the cemetery has not yet
been determined by
14
C dating, a relative date in the
1830s is suggested by coins associated with one of the
adult graves. This period encompasses the last years of
slavery and the first few years after abolition, which took
place in Mauritius in 1835. There is the potential that
the cemetery was in use before this; however, we can
be reasonably confident of this date as the cemetery first
appears on regional maps in 1880 (Descubes, 1880), and
the region itself only became a stronghold for runaway
slaves in the latter part of the 18
th
century (Allen 1999).
Materials and methods
The non-adult skeletal remainswereanalysedinDecember
2010 by JA according to the recommendations of
Figure 1. Upper inset: Mauritius and surrounding islands in relation to eastern Africa; lower inset: Mauritius, capital starred, and location of Le Morne.
Main: outline of Le Morne and regions mentioned in the text.
738 J. Appleby et al.
Copyright © 2012 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 24: 737–746 (2014)
Buikstra & Ubelaker (1994). Meas urements were taken
according to the criteria of Fazekas & Kósa (1978).
Long bones and dentit ions were X-rayed in order to
assess the presence of periods of growth interruption
and to assist in ageing. Because the biological background
of this population is not clearly understood, no attempt
was made to sex the non-adult remains (for example by
measurement of the emergent permanent dentition).
Age categories are those suggested by Lewis (2008).
Preservation was recorded according to the criteria set
out in McKinley (2004), with grades 0–1 recorded as
‘good’,grades2–3as‘moderate’ and grades 4–5+ as ‘poor’.
Fragmentation was recorded as ‘moderate’ where at least
25% of surviving long bones were fragmented and as
‘significant’ when at least 50% of surviving long bones
were fragmented.
Results
Preservation
Bone surface preservation and fragmentation showed
considerable variability. In general, the perinate skele-
tons were better preserved than were those of the
children (Table 1). All three perinate skeletons were
at least 75% complete with generally good surface
preservation, although in some cases, the long bones
and in all cases the crania were highly fragmented.
In contrast, the skeletons of the three slightly older
non-adults were at best moderately well preserved
and two were substantially incomplete. Surface damage
was often more marked on the posterior surfaces of
the bones, and the ends of long bones were frequently
Legend
N
Subadult,
3-5 years
Late term Foetus,
37-41 weeks
Subadult,
3-4 years
Subadult,
2-4 years
Late term Foetus,
36-40 weeks
Late term Foetus,
33-35 weeks
Figure 2. Site plan with position of excavated burials. This figure is available in colour online at wileyonlinelibrary.com/journal/oa.
Table 1. Preservation, fragmentation and completeness of non-adult skeletons from the Le Morne cemetery
Structure Age group Preservation Fragmentation Completeness (%)
1 Perinate Good Moderate 75+
6/7 Perinate Good Moderate 75+
6/8 Perinate Moderate-good Significant 75+
25 Child Moderate Significant 50–75
29 Child Moderate Moderate 100
30 Child Poor Significant 25–50
739The Non-Adult Cohort from Le Morne Cemetery, Mauritius
Copyright © 2012 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 24: 737–746 (2014)
damaged. Together with frequent breakage of the long
bones, this limited the measurements that could be
taken, which has implications for the calculation of
skeletal age.
The most poorly preserved skeleton was from
Structure 30. While most graves had a depth of around
90 cm, Structure 30 was 120 cm deep. This meant that
the skeleton was below the level of the present day high
water at spring tides, and it seems likely that this had
contributed to the destruction of the bone.
Perinatal individuals
Ageing techniques were designed to maximise the
information that could be gleaned from the sample
given the fragmentary nature of most of the skeletons.
For the perinates, long bone diaphysis lengths were
compared to the charts produced by Fazekas & Kósa
(1978). In addition, both linear and logarithmic regres-
sion equations were used, according to the recommen-
dations of Scheuer et al.. (1980; Table 2).
Ageing by both linear and logarithmic regression
equations produced consistent results, both for differ-
ent bones using the same technique and for the same
bone using the alternative technique. The Fazekas
and Kosa method tended to overage in comparison to
the regression techniques by a week or two, a differ-
ence that is not significant given the degree of variation
encompassed by each.
Most of the applied age calculations for the perinate
from structure 1 gave an age of between 33 and 35
weeks gestation, making it approximately one month
premature (Table 2). At this stage, it is possible that
it could have been born alive, although to survive it
would have required very careful care, something that
was probably an impossibility in rural 19
th
-century
Mauritius. A stillbirth is the most likely explanation,
but it is also possible that the foetus underwent coffin
birth due to the decomposition process of the mother.
This might explain its position between the legs of the
female skeleton (Figure 3), which exactly mirrors a case
of coffin birth from 1930s Poland (Panning 1940).
The perinates from Structure 6 both represent
almost full-term or full-term pregnancies (different aging
methods gave results between 36 and 40+ weeks, with
the majority falling between 38 and 40 weeks; Table 2).
Their placement in the same grave and within a single
coffin gives rise to the interesting possibility that they
may have been twins. Twin burials have only rarely
been reported within the archaeological record: a twin
burial dating to the Upper Palaeolithic has been iden-
tified at Krems-Wachtberg in Austria (Einwöger et al.
2006), while later prehistoric examples have recently
Table 2. Ages in gestational weeks of perinatal skeletons calculated from the linear and logarithmic regression equations of Scheuer et al. (1980) and from long bone
lengths according to the criteria of Fazekas & Kósa (1978). All measurements were taken from the left skeletal element, except those marked with *, which were taken
from the right.
Structure 1/C Structure 6/7 Structure 6/8
Linear
regression
Logarithmic
regression Length
Linear
regression
Logarithmic
regression Length
Linear
regression
Logarithmic
regression Length
Humerus 33.74* 2.33 33.86* 2.26 34* 39.05* 2.33 38.68* 2.26 >40*
Radius 33.56 2.29 33.70 2.24 34–36 39.12 2.29 38.70 2.24 >40 38.07* 2.29 37.82* 2.24 40*
Ulna 38.51* 2.20 38.27* 2.10 >40*
Femur 34.73 2.08 34.88 2.04 36–38 37.74* 2.08 37.50* 2.04 40* 38.93 2.08 38.45 2.04 >40
Tibia 34.44* 2.12 34.60* 2.11 36 38.02 2.12 37.66 2.11 38–40 39.03 2.12 38.46 2.11 40
Est. age Foetus Perinate Perinate
740 J. Appleby et al.
Copyright © 2012 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 24: 737–746 (2014)
been reported from Olèrdola in Spain (Crespo et al.
2011) and Ochtendung in Germany (Flohr 2012). In
order t o determine whether the two infants from Le
Morne we re also a twin burial, or m er ely repres ented
the burial of two unrelated infants, it was necessary
to investigate both the skeletons t hemse lves, and their
positioning within the grave (cf. Duday, 2009).
Measurements of the two foetuses were compared
to determine the amount of difference between them
(Table 3). In order to provide a direct comparison,
these were only included where bones fr om the s ame
side were available from both individuals. There was
generally very little difference between the two indi-
viduals. Individual 6/8 was slightly larger in all of t he
postcranial measurements taken, but very slightly
smaller on the size of the occipital . This latter differ-
ence is so small (0.4 mm in width and 0.1 mm in
length)thatitisunlikelytobemeaningful.Inthe
postcranial elements, the smallest difference was
0.9 mm (both femoral an d tibi al dia meters) and the
largest 2.4 mm (tibia length). There is an apparent
relationship between the size of the measurement
and t he degree of difference. Thus, the largest
differences ar e between the l ongest measurement s
(femur and tibia lengths).
Table 3. Comparative measurements in mm for the perinates
from Structure 6. * = slight imprecision caused by either erosion
or the need to reconstruct the bone not considered sufficiently
severe to prevent the measurement from being taken.
Structure 6/7 Structure 6/8 Difference
Length of basilar part
of occipital
12.0 11.9 0.1
Width of basilar part
of occipital
14.5 14.1 0.4
Ilium width (left) 28.4 30.2 1.8
Humerus width (right) 15.1 16.5 1.4
Humerus diameter (right) 5.1 6.1 1
Femur length (right) 73.2* 74.6 1.4
Femur diameter (right) 6.3* 7.2 0.9
Tibia length (left) 63.1 65.5 2.4
Tibia diameter (left) 5.8* 6.7 0.9
Figure 3. Inset: recovered foetal bones from Structure 1. Main: skeletons in situ. This figure is available in colour online at wileyonlinelibrary. com/jo urnal/o a.
741The Non-Adult Cohort from Le Morne Cemetery, Mauritius
Copyright © 2012 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 24: 737–746 (2014)
