Full Terms & Conditions of access and use can be found at
Slavery & Abolition
A Journal of Slave and Post-Slave Studies
ISSN: 0144-039X (Print) 1743-9523 (Online) Journal homepage: https://www.tandfonline.com/loi/fsla20
The West African Ethnicity of the Enslaved in
Simon P. Newman , Michael L. Deason , Yannis P. Pitsiladis , Antonio Salas &
Vincent A. Macaulay
To cite this article: Simon P. Newman , Michael L. Deason , Yannis P. Pitsiladis , Antonio Salas
& Vincent A. Macaulay (2013) The West African Ethnicity of the Enslaved in Jamaica, Slavery &
Abolition, 34:3, 376-400, DOI: 10.1080/0144039X.2012.734054
To link to this article: https://doi.org/10.1080/0144039X.2012.734054
Published online: 14 Nov 2012.
Submit your article to this journal
Article views: 613
Citing articles: 2 View citing articles
The West African Ethnicity of the
Enslaved in Jamaica
Simon P. Newman, Michael L. Deason, Yannis P.
Pitsiladis, Antonio Salas and Vincent A. Macaulay
The African ethnicity of New World slaves was highly signiﬁcant for the transmission of
African social, cultural and religious beliefs and practices. This study employs the mito-
chondrial DNA (mtDNA) analysis of present-day Jamaicans in order to assess the
ethnic origins of their enslaved female ancestors (males, including white overseers and
masters, do not contribute to mtDNA). The evidence suggests that the Gold Coast was
the largest single source of Jamaican slaves who arrived, remained and survived in
Jamaica. While this ﬁnding ﬁts with some historical evidence, it reﬁnes the data contained
within the Voyages: Transatlantic Slave Trade Database, which indicates that the Bight of
Biafra provided the most enslaved Africans to Jamaica.
Olaudah Equiano’s autobiography begins with a richly textured account of the life and
society of Igbo peoples, ‘a nation of dancers, musicians, and poets’ in what is today
southeastern Nigeria, who in their tens of thousands were enslaved and transported
from ports in the Bight of Biafra to plantations in the New World. In 1766, at
about the age of 21, Equiano secured his freedom, but he continued to travel
around the North Atlantic world, and in 1771 while serving as steward on board
the Jamaica he sailed to the island of the same name. ‘There were a vast number of
negroes’ on Britain’s largest Caribbean island, and in Kingston, Equiano was fascinated
by the large numbers of Africans who gathered together on Sundays and holidays:
Slavery & Abolition, 2013
Vol. 34, No. 3, 376 – 400, http://dx.doi.org/10.1080/0144039X.2012.734054
Simon P. Newman is at the Department of History, School of Humanities, University of Glasgow G12 8QQ, UK.
Michael L. Deason is at the Institute of Cardiovascular and Medical Sciences, University of Glasgow, G12 8QQ,
UK. Email: email@example.com
Yannis P. Pitsiladis is at the Institute of Cardiovascular and Medical Sciences, University of Glasgow, G12 8QQ,
UK. Email: firstname.lastname@example.org
Antonio Salas is at the Universidade de Xene
´tica, Departamento de Anatomı
´xica e Ciencias Forenses, and
Instituto de Medicina Legal, Facultade de Medicina, Universidad de Santiago de Compostela, Santiago de Com-
postela, Galicia, Spain. Email: email@example.com
Vincent A. Macaulay is at the Department of Statistics, University of Glasgow G12 8QQ, UK. Email:
#2013 Taylor & Francis
Here each different nation of Africa meet and dance, after the manner of their own
country. They still retain most of their native customs: they bury their dead, and put
victuals, pipes, and tobacco, and other things in the grave with the corpse, in the
same manner as in Africa.
An ‘Atlantic creole’, Equiano had cultural roots in Africa and in British America, and
better than most he understood the signiﬁcance of the persistence of African cultures
and folkways in the New World. His detailed narrative is, however, all but unique, and
historians have struggled to comprehend the ethnic identity and cultural retention of
hundreds of thousands of African slaves and their descendants. During the seventeenth
and eighteenth centuries, a great many more Africans than Europeans travelled across
the Atlantic to the colonies of the New World, yet by and large we know a great deal
more about the histories of the far less numerous European migrants. Historians of
British colonies in the Caribbean and on the North American mainland have traced
the regional origins, the religious beliefs and the political practices and ideologies of
the Britons who established American colonies, while they remain largely ignorant
of the lives and backgrounds of the Africans who vastly outnumbered the British colo-
nists. However, as we learn more and more about the regional origins, culture and
´sof the Africans who populated the New World, it has become not just poss-
ible but imperative to consider African identity and agency in the Transatlantic Slave
Trade and the new societies that it created.
It is the authors’ contention that a greater proportion of Jamaicans today are des-
cended from people who lived in the Gold Coast region than any other. While there
is evidence to suggest that more enslaved Jamaicans had disembarked from the
Bight of Biafra than from the Gold Coast, our research suggests that disembarkation
did not necessarily reﬂect region of origin, and that in all probability more Africans
from the Bight of Biafra were sold off the island than was true of Africans from the
Gold Coast. Historians have detected a signiﬁcant increase in slave importation
from the Bight of Biafra and West-Central Africa during the waning years of the
slave trade, when the greatest number of enslaved Africans arrived in Jamaica. There
is evidence, too, suggesting the presence of a higher proportion of women arriving
from the Bight of Biafra than from the Gold Coast. However, our research into the
DNA of contemporary Jamaicans and West Africans reveals that the maternal ancestry
of more Jamaicans was from the Gold Coast than from any other part of West Africa,
and it was women from this region who gave birth to the greater proportion of chil-
dren born into slavery on Jamaica. As a consequence, more Jamaicans today trace their
matrilineal origins to the Gold Coast than to any other region of Africa (Figure 1).
African ethnicity within Jamaican slavery is, as Trevor Burnard has observed, ‘an
enduringly important historiographical issue’, with signiﬁcant implications for our
understanding of the transmission of African culture to the island, the formation of
Afro-Jamaican families and the nature of creolization. From the very beginning of
the English governance of the island, planters in Jamaica valued Gold Coast slaves
in Jamaica because of what white Jamaicans interpreted as these slaves’ fortitude
and capacity for arduous labour, while at the same time the independence and poten-
tial rebelliousness of Gold Coast men and women struck fear in the hearts of their
Slavery & Abolition 377
owners. Between 1655 and 1739, Gold Coast slaves were held responsible for at least six
armed insurrections and innumerable instances of insubordination, ﬂight and suicide.
One of the leading historians of slave rebellions in the British Caribbean has suggested
that African-born rebels played a leading role in many early rebellions, and that their
leaders invariably ‘dedicated themselves to the total eradication of their white oppres-
sors and to the founding of an Akan-style autocracy in place of the toppled plantoc-
racy’. Gold Coast society and culture informed not just rebellion against plantation
society, but also the dream of a society to replace it.
The predominance of Gold Coast peoples amongst Jamaica’s enslaved population
does not, at ﬁrst sight, sit well with statistics drawn from the remarkable Voyages:
Transatlantic Slave Trade Database, which has signiﬁcantly expanded our understand-
ing of the scale and the characteristics of the forced African migration to the Americas,
and thus enabled scholars to assess the likely African ethnicity of enslaved peoples in
different parts of the New World. The database website includes a sophisticated ‘Esti-
mates’ function, which employs algorithms based on patterns in the surviving data for
the slave trade in order to ﬁll in gaps in the historical record, thus allowing more com-
plete and rounded estimates of the total number of enslaved Africans who crossed the
Atlantic. Employing this software, an estimated total of 1,205,000 Africans arrived
alive in Jamaica between 1661 and 1807 (between 12 and 13 per cent of the total
number who departed the African coast bound for Jamaica are likely to have died
before arrival at the island). Of these, some 1,178,100 (98 per cent) were carried on
English and later British slave ships. Both the surviving hard data and the rounded
estimates of the database indicate that with an estimated total of 355,600 (30 per
cent), the Bight of Biafra was the place of embarkation for the largest number of
Africans arriving in Jamaica, closely followed by the Gold Coast, from which an esti-
mated 350,400 (29 per cent) of the enslaved departed. About 162,300 (13 per cent) are
Figure 1. Three primary West African regions of origin for enslaved Jamaicans and their
descendants: The Gold Coast, the Bight of Benin and the Bight of Biafra.
378 Simon P. Newman et al.
judged to have come from the Bight of Benin. These three regions of Guinea are judged
to have accounted for 72 per cent of the enslaved Africans who disembarked in
Jamaica, and this information has given historians, anthropologists, archaeologists
and other scholars a great deal of information about the African ethnicity of the Jamai-
can population and their descendants.
The ﬁgures for the entire duration of the legal transatlantic slave trade conceal shifts
over time. During Jamaica’s formative ﬁrst century (1650 –1750), many more enslaved
Africans appear to have arrived from the Gold Coast than from the other regions: an
estimated 154,126 arrived in Jamaica from the Gold Coast, compared with 105,291
from the Bight of Benin, and 99,930 from the Bight of Biafra. During this period,
large sections of the island were cleared for the ﬁrst time, and converted to plantation
agriculture. It was brutally hard work, which along with the disease environment took
a harsh toll on the enslaved work force, and the formation of coherent slave families
and communities that were able to reproduce were slow to appear. Moreover, there
were signiﬁcant gender imbalances amongst the newly arrived Africans. In the surviv-
ing records for this ﬁrst century, we can see that only 38.6 per cent of those arriving
from the Gold Coast whose gender was identiﬁed were female, compared with 39.2
per cent females from the Bight of Benin. In stark contrast 50.5 per cent of the enslaved
arriving from the Bight of Biafra were female. Applying these percentages to the
Voyages estimates of total numbers for 1650 – 1750, one can suggest that roughly
59,500 of the enslaved arriving from the Gold Coast, 50,500 from the Bight of
Biafra, and 41,250 from the Bight of Benin were female. Thus, despite the fact that
more Africans arrived in seventeenth-century Jamaica from the Gold Coast than
other regions, a lower proportion of Gold Coast females to say nothing of high
death rates amongst an enslaved population that was far from self-reproducing all
meant that it is unlikely that any one ethnic group, including the Gold Coast, domi-
nated the Jamaican enslaved population at the turn of the eighteenth century.
Between 1750 and 1807, slave imports rose dramatically, and survival rates amongst
the enslaved on Jamaica began slowly to improve. Jamaica was ﬁrmly established as
Britain’s largest and most successful Caribbean colony, and with growing levels of
reproduction a more creolized slave society was emerging, although it is unlikely
that natural reproduction exceeded mortality rates before the early nineteenth
century. The ethnic composition of the African arrivals shifted during this period,
with an estimated 255,683 enslaved Africans arriving in Jamaica from the Bight of
Biafra, 196,241 from the Gold Coast and 56,977 from the Bight of Benin. Once
again, the Gold Coast appears to have been exporting the lowest proportion of
enslaved females, who amounted to 34.1 per cent, with 43.9 per cent females from
the Bight of Benin, and 42.5 per cent from the Bight of Biafra. Applying these percen-
tages to the overall estimates for 1750–1807, one can extrapolate that in the region
67,000 enslaved females arrived from the Gold Coast, 25,000 from the Bight of
Benin and a massive 108,500 from the Bight of Biafra.
A clear pattern thus emerges from both the actual data and the rounded estimates of
the Voyages database. Although the Gold Coast furnished more of Jamaica’s enslaved
population during the seventeenth century, a smaller proportion of these migrants
Slavery & Abolition 379
were female, and survival and birth rates remained low. Over the course of the eight-
eenth century, the Bight of Biafra sent a growing proportion of the Africans arriving in
Jamaica. Between 1750 and 1807, approximately 92,000 enslaved African women came
from the Gold Coast and the Bight of Benin, but the Bight of Biafra clearly dominated,
accounting for roughly 108,500 of the enslaved women who arrived in Jamaica. Over
the entire period from 1650 to 1807, the Bight of Biafra region accounted for more of
the island’s enslaved population than any other region of Africa. Moreover, the data
suggest that the proportion of enslaved women arriving from this region was even
higher, with signiﬁcant implications for the formation of families and the transmission
of African ethnicity and culture to subsequent generations. Between 1650 and 1807, an
estimated 159,000 enslaved female Africans came to Jamaica from the Bight of Biafra.
Of these, over two-thirds arrived between 1750 and 1807, by which time plantation
society was somewhat more settled, a creolized population was beginning to
emerge, and family formation and reproduction amongst the enslaved was rising. In
contrast, approximately 126,500 enslaved females had arrived in Jamaica from the
Gold Coast, and a further 66,250 had arrived from the Bight of Benin.
However, analysis of the DNA evidence from both Jamaica and West Africa compli-
cates the extrapolated results from the Voyages database. The results enhance our
understanding of the ethnic roots of Jamaicans, increasing the likely signiﬁcance of
Gold Coast social and cultural forms in the creation of an Afro-Jamaican creole
society and culture. Our investigation suggests that the Gold Coast was in fact the
single largest region of origin of the enslaved Africans who arrived and remained in
Jamaica, and of their descendants on the island today. These ﬁndings do not necess-
arily cast doubt on the statistical information contained within the Voyages database.
Rather, they point to two factors that the database cannot accommodate. First, the
Voyages database reveals only the coastal port or community from which slave ships
received their enslaved cargo, but this does not necessarily coincide with the actual
regional origins and identities of these enslaved men, women and children, who
often came from communities in areas quite distant from the ports of departure.
Many were victims of regional conﬂicts, and had travelled far before they were
loaded onto slave ships on the coast. Second, and most signiﬁcantly, Jamaica was
both an importer and exporter of enslaved Africans. Decisions over which slaves
should be retained for Jamaican plantation work, and which would be selected for
resale off the island were informed by planters’ preferences for certain perceived
types of Africans, and this likely had an impact on the ethnic composition up of the
African population of the island. Seventeenth- and eighteenth-century accounts
reveal white Jamaicans’ recognition of the reality of African ethnicity, but ignorance
of the actual nature and the social signiﬁcance of such identities was profound. The
majority of slaves in West Africa were female, and amongst free and enslaved alike
women played a signiﬁcant role in child-rearing, family organisation and thus in
the transfer of cultural belief systems and practices to their offspring. Perhaps as
many as half of the Ashanti population, for example, may have lived in matrilineal
households headed by women.
In West Africa, and later on in the African societies
of the New World, fathers could and did exercise power and authority over children,
380 Simon P. Newman et al.
yet on both sides of the Atlantic mothers played highly signiﬁcant roles in kinship
communities, roles that white observers seldom understood or appreciated. The
Voyages database indicates that more of Jamaica’s slaves came from the Bight of
Biafra than from the Gold Coast, that a greater proportion of the Biafran slaves
were female, and that more female Biafra slaves arrived in the eighteenth century
when conditions were sufﬁciently improved to allow for gradual improvement in
living conditions, life expectancy and reproduction. However, DNA evidence suggests
that more of the maternal ancestors of modern Jamaicans were from the Gold Coast
than were from the Bight of Biafra, which suggests that more enslaved children in
Jamaica were born to Gold Coast women and raised, to varying degrees, in those tra-
ditions from Gold Coast societies that survived the Middle Passage (Figure 2).
DNA analysis has great potential for historical research into the origins of deﬁned
groups of individuals. One of the most effective methods of DNA analysis assesses
the maternal lineages present in target populations and compares these with their
likely sources. During reproduction, the genetic material from the mother and
father shufﬂe to create a new genome, with one distinct exception. Thousands of mito-
chondria can be found in every human cell and produce the chemical energy for life.
As a result of a symbiotic event billions of years ago, they have their own distinct
genome. Mitochondria are used to propel the sperm towards the egg during reproduc-
tion. The tail, and subsequently any mitochondrial DNA (mtDNA) from the father, is
shed once the sperm reaches the egg. Consequently, mtDNA is transferred directly
between mother and offspring without any contribution from the father. Coupling
the predictable rate of mutation with meticulous ﬁeld research has enabled the cre-
ation of an evolutionary tree detailing global mitochondrial diversity, with major
branches largely corresponding to particular continental regions.
are called haplotypes, and each population has a distinct composition of haplotypes.
A group of closely related haplotypes is referred to as a haplogroup. Due to its uni-
directional inheritance, mtDNA distribution is naturally sensitive to demographic
Figure 2. Major ports of departure, West Africa. Most of the sources employed by histor-
ians record the ports from which enslaved Africans were transported, but reveal little about
the actual regional origins of the enslaved. Many had never before seen the coast, or the
port from which they set sail.
Slavery & Abolition 381
pressures such as population migrations and genetic drift. Every human population
shows a mixture of haplotypes, and although these haplotypes may be found in
other groups, the distribution of haplotypes is always unique. Using this mitochon-
drial evolutionary tree, genealogical information from discrete groups of people can
be compared with relative ease, giving useful information about the geographic
origins of individuals as revealed by their DNA.
As TV crime dramas and criminal investigations and trials illustrate, contamination
or degradation of samples, to say nothing of human error, can compromise forensic
DNA analysis. However, such problems are far rarer in academic laboratories in
which groups, rather than individuals, are studied in the most rigorous fashion.
Trained academics collect samples from living individuals, which are then processed
with great scientiﬁcally rigour, and the data generated are then analysed according
to set procedure. While the DNA identity of a single individual may be compromised
in any number of ways, examination of larger groups and broad population trends
means that even if an occasional sample is mislabelled or inadvertently exchanged
with another, the anomalous results would be subsumed into the overall results of a
broad-based cohort study.
A signiﬁcant obvious beneﬁt of researching mtDNA is that it changes through
mutations that happen at a predictable rate when passed from mother to child as
opposed to autosomal DNA, which not only incurs varying mutation rates but is
also shufﬂed at each generation. Global mtDNA variation can be plotted as a tree,
with different branches corresponding to populations in different regions of the
world. The one signiﬁcant disadvantage of using mtDNA is that, hypothetically,
someone could be almost entirely European autosomally with African mtDNA
because of a chance encounter dozens of generations ago. For example, the descendant
of a Greek merchant and a sub-Saharan African in Egypt could have been enslaved and
transported to Jamaica, bringing – in this case – European mtDNA into the Jamaican
gene pool. To combat such samples having an overly strong inﬂuence on results,
sample sizes are always kept as large as possible and ﬁndings are tested against pre-
viously published results to check for correspondence whenever possible. Thus, dra-
matic variations are ironed out by the identiﬁable trends of large samples.
Earlier mtDNA research undertaken in Jamaica has revealed a nearly pure West
African matrilineal origin for the modern population, with few genetic inroads by
European, Asian or indigenous American matrilines. This is not surprising, given
the historical presence of an overwhelming proportion of African slaves and their
descendants, a relatively small number of European women and the almost complete
obliteration of the original indigenous population. Furthermore, different West
African ethnic communities made clear and distinct genetic contributions with each
group upon arrival in Jamaica having a distinct haplogroup proﬁle distribution, repre-
sentative of their larger ancestral African identity. Using a variety of statistical
methods, it is hence possible to trace the ancestry of modern Jamaicans to particular
areas of West Africa.
The process of collecting DNA samples was relatively simple. Volunteers from
around Jamaica were asked to complete a simple questionnaire stating their birthplace,
382 Simon P. Newman et al.
their parents’ birthplaces and – if known – the birthplaces of both sets of grandpar-
ents. Individuals born outside of Jamaica or with any reported maternal relatives born
outside of Jamaica were excluded from any further analysis. Although Kingston and St
Andrew’s parish are over-represented in the sample, by and large Jamaicans from all 13
of the island’s parishes are represented in terms that roughly correspond with each
parish’s share of the island’s population. For example, residents of the parish of Man-
chester constituted 7 per cent of the Jamaican population in 2010, while just under 8.5
per cent of the volunteers whose parish of birth was known came from this parish.
Similarly, 7 per cent of Jamaicans live in Clarendon Parish, and 9 per cent of the vol-
unteers were born there.
A sterile, specially designed swab was used to collect cheek cells from each of the
volunteers for DNA extraction. After applying the aforementioned exclusion criteria,
samples from 400 individuals were then included in further mtDNA analysis. A data-
base of 9265 individual mitochondrial sequences was compiled from the genetic
anthropological literature to facilitate a comparison between roughly 180 African
ethnic groups from sub-Saharan Africa, and from those collected in Jamaica. These
ethnic groups were then partitioned into discrete historically signiﬁcant geographic
regions. Senegambia describes the area north of the Nun
˜ez River; Sierra Leone com-
prises the Nun
˜ez River near Boke
´, Guinea, up to and including the Assini River in
ˆte d’Ivoire; the Gold Coast runs east of here up to and including the Volta River;
the Bight of Benin covers the Volta River in Ghana to the Nun River in Nigeria; the
Bight of Biafra covers the area east of the Nun River to Cape Lopez in modern
Gabon; West-Central Africa was deﬁned as the rest of the western coast of the conti-
nent south of this point; ﬁnally, South East Africa included anywhere east of the Cape
of Good Hope. The Windward coast was combined with Sierra Leone due to poor rep-
resentation in the mtDNA literature. Employing advanced statistical analysis, the most
likely maternal contribution of particular regions of coastal Africa to Jamaica can be
determined by comparing both haplogroup distributions and haplotype similarities.
Seven of the Jamaicans were halogrouped as non-sub-Saharan. All of the remainder
had West African roots. In order to estimate the proportion of maternal ancestry
from each major African slave-trading region present in the population of modern
Jamaican mtDNA, an admixture model of the halogroup proﬁles from the various
African regions were ﬁtted to the proﬁle of the sampled Jamaicans.
The results of our investigation are summarized in Tables 1 and 2.
Without access to the information available through these forms of DNA analysis,
historians, anthropologists, archaeologists and other scholars have employed more tra-
ditional sources and approaches in order to assess the nature and persistence of African
ethnicity amongst enslaved populations and their descendants in the New World. Scho-
lars such as Orlando Patterson, Sidney Mintz and Richard Price stressed that the trauma
of the Middle Passage and the familial and social dislocations occasioned by plantation
slavery effectively ‘randomized’ the enslaved populations of plantations, making it all
but impossible for African cultural practices to be transmitted to the New World in
viable forms. Afro-Caribbean and Afro-American cultures were, they concluded, the
result of innovative adaptations by isolated, traumatized Africans adapting to new
Slavery & Abolition 383
lives in the absence of family and friends from their African past. Edward Braithwaite’s
key study of Jamaica works along these lines: he suggests that African inﬂuences had
little if any real effect in Jamaica. In Braithwaite’s Jamaica creolized slaves played a
pivotal role in the development of a new society and culture, but it was a New World
culture in which the traditions of different African ethnic groups played no role.
However, evidence indicating the relative coherence and persistence of a particular
ethnic group may have militated against such randomization and facilitated greater
social and cultural continuities between West Africa and Jamaica. John Thornton is
one of a number of historians who have argued that because slave ships tended to
draw their cargos from only one or two points on the Guinea coast, the enslaved
cargo often came from a common region: while the enslaved may have travelled some
Table 1 African ancestry of all Jamaicans in sample – haplotype method.
P(0) (per cent) P(1) (per cent) P(2) (per cent)
Senegambia 11.59 12.39 13.99
Sierra Leone 14.31 14.53 14.79
Gold Coast 19.56 17.24 16.42
Bight of Benin 17.04 17.45 16.23
Bight of Biafra 15.01 14.08 13.35
West-Central Africa 10.99 11.16 11.24
South East Africa 5.80 7.68 8.65
East Africa 5.71 5.47 5.33
Senegambia 10.46 9.77 12.15
Sierra Leone 13.97 15.44 15.76
Gold Coast 19.48 17.52 16.76
Bight of Benin 18.78 19.04 16.95
Bight of Biafra 15.65 14.49 13.41
West-Central Africa 10.71 10.81 11.03
South East Africa 5.55 7.61 8.59
East Africa 5.40 5.33 5.34
Senegambia 11.49 12.12 13.71
Sierra Leone 13.91 14.04 14.36
Gold Coast 19.13 17.12 16.13
Bight of Benin 16.68 16.90 15.75
Bight of Biafra 16.48 15.43 14.55
West-Central Africa 11.12 11.37 11.62
South East Africa 5.62 7.55 8.51
East Africa 5.56 5.48 5.37
Notes: A haplotype refers to the unique combination of differences from the internationally standardized mtDNA
reference sequence (the revised Cambridge Reference Sequence, rCRS, provides a base reference point for all
mtDNA research). This table has been created employing the higher resolution haplotype method, in order to
assess the number of shared haplotypes between the Jamaicans on the one hand, and the Africans from each
region on the other hand. P(0) shows perfect matches, P(1) shows matches with just one difference and P(2)
shows matches with just two differences. Sahelian refers to the interior region of north central Africa, below the
Sahara Desert, and today comprising parts of Senegal, Mauritania, Mali, Niger and Chad. Pygmies are people of
very short stature (with an adult height of less than 150 cm) from equatorial Africa. Tables 1 and 2 show the
results when the mtDNA of Sahelian and pygmy peoples are excluded.
384 Simon P. Newman et al.
distance to the port from which they left Africa, these men, women and children were
often drawn from particular areas: it was relatively uncommon for a European
trading castle to simultaneously house slaves from many different areas. Thornton con-
cludes that often there were only relatively minor cultural differences between people on
board given slave ships, heading towards destinations such as Jamaica.
Without doubt, the processes of the Middle Passage and enslavement in the New
World forced many people from distinct regions, communities and cultures in
Table 2 African ancestry of all Jamaicans in sample – haglogroup method.
Senegambia 4.90 per cent
Sierra Leone 9.65 per cent
Gold Coast 47.67 per cent
Bight of Benin 12.25 per cent
Bight of Biafra 6.43 per cent
West-Central Africa 8.92 per cent
South East Africa 9.21 per cent
East Africa 0.96 per cent
Senegambia 7.48 per cent
Sierra Leone 9.16 per cent
Gold Coast 33.56 per cent
Bight of Benin 26.07 per cent
Bight of Biafra 5.98 per cent
West-Central Africa 8.08 per cent
South East Africa 8.77 per cent
East Africa 0.89 per cent
Senegambia 4.82 per cent
Sierra Leone 9.23 per cent
Gold Coast 45.62 per cent
Bight of Benin 10.54 per cent
Bight of Biafra 9.54 per cent
West-Central Africa 10.94 per cent
South East Africa 8.45 per cent
East Africa 0.87 per cent
Without Sahelian or Pygmies
Senegambia 7.23 per cent
Sierra Leone 9.21 per cent
Gold Coast 34.28 per cent
Bight of Benin 21.39 per cent
Bight of Biafra 9.07 per cent
West-Central Africa 9.70 per cent
South East Africa 8.29 per cent
East Africa 0.82 per cent
Notes: The haplogroup method provides a lower resolution, more rounded picture by grouping together similar
haplotypes. A rather less sharply deﬁned group of haplotype proﬁles for a speciﬁc African region are thus
compiled into a haplogroup, and then the percentage of similar haplogroups present amongst the Jamaican
sample are assessed. Using the haplogroup percentage as a template, the model then seeks to recreate the
haplogroup numbers present in the Jamaican sample. The result is a rather more rounded assessment of the
African ancestry of the Jamaican test group.
Slavery & Abolition 385
Africa to think of themselves – because of the ways they were treated – as members of
a pan-African group, a new kind of identity. Yet in the formation of new communities
in the Americas, these enslaved Africans inevitably drew on common languages, cul-
tural and religious bonds and so forth in order to come to terms with their situation,
and to fashion new identities. In these circumstances, the members of numerically
dominant African ethnic groups likely exercised a greater inﬂuence over the nascent
New World societies of the enslaved. New ‘nations’ – a term used by whites to describe
discrete groups of Afro-Caribbeans and Afro-Americans – came into existence in the
New World, and often whites in British America used West African ethnic labels to
describe the developing communities of the enslaved in Jamaica, Barbados, Virginia
and elsewhere. The ethnic identities of the Africans who fashioned new societies
and cultures in Jamaica and throughout the Americas are signiﬁcant in assessing
the communities and cultures that they created.
Modern historians, employing both the stated preferences of seventeenth- and
eighteenth-century planters and currently available records of slave shipments and
sales, have tended to conclude that Akan-speaking peoples from the Gold Coast
were often favoured by English planters in Barbados, and it was these people who
played a leading role in the creation of Afro-Caribbean society on that island. From
the earliest days of English sugar plantations, ‘The negroes most in demand ...are
the gold coast, or as they call them, Cormantines’. Responsible for selling slaves to plan-
ters, Royal African Company agents in late-seventeenth-century Barbados wrote to
company ofﬁcials in London describing ‘the dislike People have here to any but
Gold Coast Negroes’, while Henry Drax’s instructions for the successful running of
his plantation included the observation ‘that the Cormante or gold Coast Negros
have always Stood and proved bestt in this plantation[,] theirefor you will doe welle
to buy of that Nation than any other’.
Barbadian planters believed Gold Coast Afri-
cans to be stronger and more capable of the arduous work involved in sugar agricul-
ture and manufacturing than men and women from other regions of West Africa.
The preference of mid-seventeenth-century Barbados planters for slaves from the
Gold Coast was apparently taken up by Jamaican planters, not least because in the
ﬁrst instance many planters and many more seasoned slaves were transferred from
Barbados to Jamaica. Amy Marie Johnson has estimated that approximately 25 per
cent of the slaves that entered Jamaica during the 1670s came from plantations on
British colonies in the eastern Caribbean, particularly Barbados while others suggest
that the proportion may have been as high as one-third: a great many Barbados
slaves had been purchased from Royal African Company ships that had sailed from
the company’s forts on the Gold Coast, bearing a heavy proportion of Gold Coast
slaves. This is one reason why the proportion of Gold Coast slaves in Jamaica may
have been higher than the statistics for slave exports from Africa directly to Jamaica
indicate. Perhaps on the basis of these Barbadian Gold Coast slaves, Jamaican planters
continued to articulate a clear preference for slaves from that region of West Africa.
The Jamaican merchant Francis Grant was typical in his claim that slaves from Old
Calabar on the Bight of Biafra were ‘not nearly in so much estimation with the general-
ity of people ... as [slaves] from the Gold Coast’. When the Princess Amelia arrived in
386 Simon P. Newman et al.
Jamaica in December 1803, it proved possible to sell the cargo of 305 slaves from the
Gold Coast with ease, while the four or ﬁve ships from other areas of the Guinea coast
could ‘hardly sell a negroe’. Most contemporaries appear to have agreed that British
merchants and planters considered ‘Gold Coast slaves ...the most valuable of any
slaves exported from Africa’.
As plantation slavery expanded in Jamaica, the Barbadian preference for Coroman-
tin slaves was appropriated by Jamaican planters, who bought more and more of the
slaves shipped from the Gold Coast. Well over two-thirds of all slaves leaving the Gold
Coast went to British colonies in the New World, with Barbados serving as the major
destination in the seventeenth century, and Jamaica taking over in the eighteenth
century. Johnson estimates that during the later 1670s, the Royal African Company
brought an average of 1120 Gold Coast slaves to Jamaica each year, rising to more
than 2700 per annum a decade later. Britain’s West African base of operations was
Cape Coast Castle on the Gold Coast, along with various smaller forts and trading
posts, and as many as one-third of early-eighteenth-century slave ships travelling
from this hub took their human cargo to Jamaica. By the middle years of the eight-
eenth century, as many as 38 per cent of the Africans who arrived in Jamaica had
come from the Gold Coast. Local conditions on the Gold Coast facilitated the rise
of this trade, as a series of wars culminated in the rise of the Asante state: with gold
production on the wane, Asante success during and after these wars was fuelled by
trade, with the sale of slaves to Europeans – and particularly the British – rising
throughout the eighteenth century, and peaking in the 1780s and 1790s.
The Voyages database has demonstrated that between 1662 and 1867 over four-ﬁfths
of the slaves who crossed the Atlantic from Africa departed from just four regions, the
Gold Coast, the Bight of Benin (known as the Slave Coast), the Bight of Biafra, and
West-Central Africa. The historiography of the British Transatlantic Slave Trade has
been dominated by the ﬁrst three of these regions, which contained the largest popu-
lation densities in early modern sub-Saharan Africa, the greatest urban development
and some of the most sophisticated state structures. David Eltis and David Richardson
have argued that the Gold Coast, the Bight of Benin and the Bight of Biafra drew upon
largely exclusive provenance zones, each with a fairly high degree of ethno-linguistic
homogeneity. In short, each of these three areas of the Guinea coast appear to have
drawn slaves from particular and relatively uniﬁed areas stretching from the coast
into the African interior.
All of this is, however, informed supposition, given that the Transatlantic Slave
Trade database reveals the port of departure for a good proportion of slaves, but
little else of their original home and ethnicity. It is as if records of European migrants
to the America revealed only their departure from key ports such as London, Bordeaux
and Cadiz, obscuring the fact that Scots, Irish, Germans, Scandinavians, Catalans, Por-
tuguese and many others may have departed from these ports. In fact, the very label
commonly applied to Gold Coast slaves, Coromantee, is not a term to describe a par-
ticular ethnic group, but rather is a word derived from Kormantin, a small ﬁshing
community that housed England’s ﬁrst trading outpost on the Gold Coast. Coroman-
tee were not a people in Africa, but the slaves who departed from Kormantin and other
Slavery & Abolition 387
Gold Coast ports became members of the Coromantee nation in the New World: in the
eyes of their white owners, and over time in their own eyes too, these new ‘nations’
assumed a social and cultural reality unknown in Africa.
Merchants and planters in Jamaica and the New World had some sense of the differ-
ent African ethnicities and cultures, and recognized that enslaved people arriving from
a single port on one ship often spoke different languages. In 1781, Jamaica’s Royal
Gazette carried an advertisement for the sale of ‘two hundred and sixty, choice Coro-
mantee, Fantee, Ashantee and Akim Slaves’: 12 years later, the same newspaper
reported the arrival of the Union in Jamaica bearing a cargo of ‘538 choice young
Coromantee, Ashantee and Fantee slaves’. Similarly, Edward Long observed of the
The Negroes who pass under this general designation are brought from the Gold
Coast; but we remain uncertain whether they are natives of that tract of Guiney,
or receive their several names of Akims, Fantinsm Ashantees, Quanboos, &c,
from the towns so called, at whose markets they are bought.
Such sources illustrate a degree of contemporary white awareness of African ethnic
differentiation. Such awareness was, however, at best incomplete and relatively
David Eltis and David Richardson, two of the historians who played a leading role in
the creation of this remarkable database, acknowledge that the ‘major single destina-
tion of Gold Coast slaves was Jamaica’. Thus, despite the fact that the Voyages database
indicates that more of the Jamaican enslaved had come from the Bight of Biafra, Afri-
cans from the Gold Coast nonetheless achieved a degree of ‘Akan cultural prominence’
(including Ahanta, Fanti, Akim and Asante peoples), premised upon white planter
preference for Gold Coast slaves, and the largest concentration of Gold Coast
peoples anywhere in the British Atlantic world.
However, Douglas B. Chambers has built upon data from the Voyages database to
challenge the idea that Akan slaves from the Gold Coast dominated Jamaican
society, arguing that Igbo slaves from the Bight of Biafra were at least as signiﬁcant.
Perhaps as many as 80 per cent of the slaves loaded on to European ships on the
coast of the Bight of Biafra were Igbo-speaking peoples from the Calabar backcountry,
‘a distinct ethno-historical group who shared a distinctive set of ancestral traditions
and drew on the same or very similar material social and ideological resources’.
Chambers goes so far as to assert that during the second half of the eighteenth-
century British planters received perhaps twice as many Igbo as any other African
ethno-historical group, and perhaps four times as many as came from the Gold
Coast. In Jamaica and elsewhere, he detects evidence of the inﬂuence of this Igbo
migration in the customs, diet, language and cultural forms of enslaved peoples,
from jonkonu and obeah to the use of Igbo words such as okro for okra. Other histor-
ians have agreed, suggesting that in the late eighteenth century far more slaves left
Bonny when compared with Old Calabar or Elem Kalabari (New Calabar) in the
Bight of Biafra, and that the Igbo were the dominant ethnic group, far outnumbering
the Ibibio (or Mocos) who were less popular in Jamaica. Diptee goes so far as to
388 Simon P. Newman et al.
contend that the ‘Bight of Biafra was the most signiﬁcant area of provenance for the
enslaved in Jamaica’, accounting for 44 per cent of the enslaved.
Such numbers and the oft-stated preferences of planters and merchants obscure as
much as they reveal. The Jamaican planter Simon Taylor was typical when he sought
out ‘150 or 200 young Ebo’, reiterating several months later that they should be ‘Eboes
[from Bonny] and not Mocos [from Calabar or Elem Kalabari]’. It seems likely that the
massive increase in slave shipments from Bonny, and the corresponding decline in
shipments from the other ports in the Bight of Baiafra, had relatively little to do
with planter and merchant demands, and were in fact the result of changing situations
in the African hinterland, so that the same peoples were in fact funnelled to one port
rather than another. As the Aro’s trade with Europeans increased, they ventured farther
into the Biafran hinterland, establishing new settlement and enhancing trading routes,
in the process shifting an older trade away from Old Calabar and Elem Kalabari and
transferring it to the more centrally located and more easily accessible port of Bonny.
Perhaps the often stated preference of planters for Gold Coast slaves, and their
regular statements that slaves from the Bight of Benin, and to a lesser extent the
Bight of Biafra were less desirable, combine to suggest the possibility that a signiﬁcant
proportion of the slaves from the Bights of Biafra and Benin were immediately sold on
to Spanish, French or other European colonists. A striking weakness of the Transatlan-
tic Slave Trade database is that the disposal of slaves following their arrival in the New
World is all but invisible. Between 1789 and 1808, approximately 14 per cent of all
slaves brought to Jamaica may have been re-transported off the island to other desti-
nations: Jamaica was a vital source for the restocking of slave populations in South
Carolina and Georgia in the quarter-century following the end of the American
War for Independence. If the large majority of the slaves transported off Jamaica
were of one particular ethnicity, which is very difﬁcult to discern with any accuracy,
the effect on the ethnic proﬁle of new slaves on Jamaica would have been signiﬁcant.
Both David Eltis and John Thornton have argued far more slaves arriving in Jamaica
from the Bight of Biafra were sold on to Spanish America: six slaves from other African
regions were sold to the Spanish for every slave from the Gold Coast who was sold on.
Perhaps as many as 80 per cent of slaves on late-seventeenth- and early-eighteenth-
century Jamaican plantations may well have been drawn from the relatively culturally
homogenous areas of the Gold Coast and the Slave Coast.
Thus, historians face a
series of connected problems. While the Voyages database reveals with startling accu-
racy how many Africans disembarked from speciﬁc ports on the West African coast
bound for Jamaica, we know far less about the actual regional identities of these
people. Furthermore, although it is clear that many did not remain in Jamaica and
were sold off the island, we have little hard evidence to indicate what proportion of
speciﬁc African ethnic groups were sold or retained, making our understanding of
the ethnic composition of Jamaica’s enslaved population all the more imprecise.
Trevor Burnard’s analysis of the ways in which Africans arriving in seventeenth-
century Jamaica were sold, bought by merchants and planters, and sent on to planta-
tions and slave pens also casts doubt on the persistence of ethnic identity amongst
slaves on plantations. Patterns changed over time, and employing the Transatlantic
Slavery & Abolition 389
Slave Trade database Burnard demonstrates that during the ﬁrst quarter-century of
English settlement of Jamaica, 60 per cent of African slaves bound for Jamaica
embarked not from the Gold Coast but from ports in the Bight of Biafra. In the
late seventeenth and early eighteenth centuries far fewer slaves arrived in Jamaica
from the Bight of Biafra, and during the 1680s and 1690s between three-quarters
and four-ﬁfths of Jamaica’s slaves had left from the Bight of Benin and West-
Central Africa. Then, during the ﬁrst quarter of the eighteenth century, the Gold
Coast rose to prominence, accounting for 45.7 per cent of the slaves leaving Africa
for Jamaica. In short, no one African region dominated the shipment of slaves to
Jamaica, whatever the stated preferences of Jamaican planters may have been: ‘The
overall impression is of rapidly changing provenance zones’.
Furthermore, Burnard’s examination of Royal African Company records of the sale
of slaves to Jamaican planters suggests that whatever their stated preferences for slaves
from particular regions, almost all planters bought whatever slaves they could, from
whichever regions of Africa they came, and that consequently their plantations were
stocked with slaves from a bewildering array of African locations. While planters
may have preferred Gold Coast slaves to others, it was likely quite difﬁcult to translate
that preference into reality. Although there was relatively little difference in the cost of
healthy adult slaves from different regions in the late seventeenth century, that did not
mean that supplies of men and women from particular regions were always available.
Inventories of slaves compiled between 1723 and 1735 include indications of the
African ethnicity of 320 enslaved men, women and children: nearly two-thirds of
these appear to have come from the Gold Coast and the Slave Coast, with the latter
region dominating, while signiﬁcant minorities came from West-Central Africa and
the Bight of Biafra. High white mortality in late-seventeenth- and early-eighteenth-
century Jamaica meant that plantations and slaves changed hands frequently, with
the result that estates and their slaves were frequently sold, all of which meant
further disruption to the far from stable nascent Afro-Caribbean communities.
Burnard is suggesting that the ‘randomization’ theories of a generation ago may
come back into vogue, and he has suggested that no one ethnicity dominated
within the enslaved Afro-Caribbean society of Jamaica between the late seventeenth
and early nineteenth centuries.
One problem obscured by our growing awareness of the numerical contours of the
Transatlantic Slave Trade is our far less complete understanding of family and social
formation amongst the enslaved in colonies such as Jamaica. If, for example, larger,
older and more established plantations tended to buy relatively fewer new slaves
over the course of the eighteenth century, with the enslaved reproducing themselves,
then perhaps it is to the African migrants of the later seventeenth and early eighteenth
century that we must look for cultural formation and continuities over time. If the
slaves imported during the later eighteenth century went in larger proportions to
newer and smaller plantations, were they less likely to enjoy a signiﬁcant formative
role in the create of an Afro-Caribbean society and culture? Michael Craton has
demonstrated a proportional decrease in the number of Africans in Jamaica over
the course of the eighteenth century, from 89.4 per cent in 1712, to 76.4 per cent in
390 Simon P. Newman et al.
1752, to 64.4 per cent in 1772. By 1807, according to Barry Higman, only 45 per cent of
Jamaican slaves had been born in Africa.
Edward Long, Lieutenant Governor of Jamaica and a keen observer of life and
society on the eighteenth-century island, was convinced that the Gold Coast Coro-
mantee dominated the language of slaves: their’s was, he observed ‘more copious
and regular than any other of the Negro dialects’. As long as a century later, William
Gardner echoed Long, asserting that ‘the inﬂuences of the Coromantyns seems to
have modiﬁed, if not entirely obliterated whatever was introduced by other tribes’.
Long considered Gold Coast Africans to be ‘well made’, with ‘features very different
from the rest of the African Negroes’, and he considered that even their dancing and
music helped to keep alive their particularly martial spirit. The Coromantee ‘build
their houses distinct from the rest’, he went on, and they ‘do not mix at all with the
other slaves’. Scholars of linguistics have argued that Akan-speakers were ‘centers of
linguistic and cultural conservatism for a large part of the island’, and there is evidence
that Africans from other areas learned elements of the Akan language once in Jamaica.
Mervyn Alleyne has argued that Twi-Asante, generally known today as Akan, almost
completely dominated the language of Jamaica’s Maroons: other African languages
‘progressively and rapidly decayed until they disappeared completely. Twi-Asante,
on the other hand, changed little... [and] ﬂourished among Maroons during the
period of slavery’. However, Robin Law has argued that immigration from the Gold
Coast and European use of Gold Coast workers introduced their language to the
Slave Coast, and it seems likely that the European presence in West Africa helped
increase awareness and understanding of Akan languages all along the Guinea coast.
Thus, at least some Twi-Asante may well have familiar to slaves from many different
regions, and with Gold Coast guardians working as intermediaries between crew
and enslaved cargo on the Middle Passage, it is likely that the Gold Coast language
was being employed by some non-Gold Coast slaves long before they arrived in
Jamaica. The point here is that evidence of Twi-Asante cannot be taken as conclusive
evidence of the numerical superiority of Gold Coast Africans in Jamaica. More recent
work has suggested that creolization likely involved deployment of different African
languages, as well as English, and it has proved extremely difﬁcult to point to ethnic
dominance as demonstrated by language.
It is quite likely that African day-names were carried from the Gold Coast and
employed in Jamaica: thus, a male born on a Sunday might be named Quashie,
while a female born on Friday might be named Phibba. Evidence suggests that the
use of these day-names was relatively common amongst the Maroons. On slave plan-
tations, however, over times the names were corrupted, with masters choosing and
imposing names at random, and combined names such as ‘Juba’s Quasheba’ develop-
ing. Trevor Burnard’s examination of Jamaican plantation records from the later eight-
eenth century indicates that African names, including Gold Coast day-names, were
often imposed by white masters and overseers, who regarded them as generic
African names, and that consequently these names inevitably lost their particular
meanings and cultural resonance.
Slavery & Abolition 391
What white planters interpreted as the martial prowess and ﬁerce pride of slaves
from the Gold Coast region may have made them strong workers, but it also made
them into potential rebels, and it is no coincidence that Jamaicans labelled a series
of slave uprisings as the Coromantee Rebellions. During the late seventeenth and
early eighteenth centuries, a great many Gold Coast slaves poured into Jamaica, but
the ensuing high incidence of slave rebellions may well have contributed to a dramatic
drop in Jamaican imports of Gold Coast slaves between 1725 and 1745. The bloody
campaigns against Maroon communities during the 1720s and 1730s brought
matters to a head, and Jamaicans took as few as 3 per cent of the enslaved Africans
who left the Gold Coast in the 1730s: the result was that Gold Coast slaves were, at
that point, accounting for as little as 1 per cent of Jamaica’s imports.
One of the Gold Coast slaves who arrived in mid-eighteenth-century Jamaica was
Apongo, an elite ‘prince’ in one of the tributary states of the Dahomey kingdom.
He met Governor John Cope at Cape Coast Castle, and was entertained as a dignitary
by Britain’s leading ofﬁcial in Africa. Years later, following his capture, enslavement
and passage to Jamaica, Apongo once again met Cope, who had become a planter:
the former governor remembered the former leader, and treated him well. Apongo
was one of a great many Gold Coast Africans who had experience of the territorial
wars that characterized the region during the late seventeenth and the eighteenth cen-
turies. Gold Coast slaves may well have been valued by planters for their strength and
resilience, but they were feared too for the experiences that African military service had
given them. Apongo was a good example of this: also known as Wager, he became an
important leader in Tacky’s Rebellion of 1760, the greatest slave uprising in Jamaican
Both contemporaries and subsequent historians have regarded African ethnicity as
an essential element of Jamaica’s Maroon communities, the most signiﬁcant societies
of their kind in the British Americas. Although a relatively small island, some 140 miles
long and 45 miles wide, Jamaica has a mountainous and heavily forested terrain that
made it idea for the establishment of independent Maroon communities by runaway
slaves. Mountains, hills, hidden valleys or ‘poljes’ and ‘cockpits’ provided plenty of
cover, and even as British colonists spread over the island, large areas remained undis-
turbed. On the eastern, Windward side the forested slopes of the Blue Ridge Moun-
tains, with slopes of as much as 60-degrees, provided a base for one set of Maroon
communities, while in the islands interior the nearly 500 square miles of the
Cockpit Country were all but impenetrable, and a safe haven for a second group of
Leeward Maroon communities. As sugar plantation agriculture increased in the
years after 1670, English planters brought in increasingly large numbers of African
slaves, many of them Akan- and Adangme-speaking Coromantees from the Gold
Coast. The prevalence of Gold Coast names such as Cudjoe, Accompong, Quaco,
Cuffee and Quao attributed to Maroon leaders suggests the signiﬁcance of the Coro-
mantee nation amongst Jamaica’s Maroons. According to one of the most detailed
studies of the Jamaican Maroons, the Leeward Maroons were dominated by Coroman-
tee leaders, who established an Asante-style autocratic polity. Non-Coromantee run-
aways were inducted into the Leeward Maroons only after a rigorous probationary
392 Simon P. Newman et al.
periods, and on occasion bitter and violent conﬂict erupted with rival ethnic Maroon
bands, as in the armed conﬂict with ‘Madagascar’ slaves from Don’s plantation, who
had established themselves in the hills of Westmoreland Parish in the 1720s. Appar-
ently English became the day-to-day language amongst the Leeward Maroons,
although ‘Coromantee’ remained highly signiﬁcant amongst them. Perhaps an
amalgam of different Gold Coast languages and dialects, this syncretic form was
used in a variety of rituals by captains and obeahmen, providing passwords and
code words that protected the community. Perhaps building on their African past as
free or enslaved soldiers in the series of imperial wars that raged through the late-
seventeenth- and early-eighteenth-century Gold Coast, the men of the Leeward
Maroons were warriors, protecting their villages and hunting and ﬁshing in the
forests. Women enjoyed few rights, and were responsible for farming and cooking:
it seems likely that for many years men outnumbered women, and the Maroon com-
munities depended upon runaways to maintain their numbers.
One researcher in the 1970s found a far greater number of ‘African lexical survivals’
amongst the descendants of the Windward Maroons than was true for Jamaican
language as a whole. These Maroon communities were for many years regarded as sep-
arate and distinct from Jamaican society as a whole, and their African words and cul-
tural practices were often seen in derogatory terms by the larger society. A good many
of the surviving words appear to have had Akan origins, and while the meanings of
some were readily discernible, others continued to have secret meanings and signiﬁ-
cance, well over a century after the end of slavery.
Recent research employing the Transatlantic Slave Trade database indicates that
enslaved women and children arrived in Jamaica in more signiﬁcant proportions
during the second half of the eighteenth century and the early years of the nineteenth
century: if this coincided with a narrowing of the ethnic composition of those slaves
who arrived in and remained on the island, then this was likely to have been a signiﬁ-
cant period for the formation of unprecedented numbers of families, and of the trans-
mission of African cultural forms between generations. While historians have generally
accepted that throughout most of the colonial era planters preferred to purchase
healthy young adult male slaves, Audra Abbe Diptee has suggested that from the
mid-eighteenth century on Jamaican planters began requesting and purchasing
more women and children. Between 1751 and 1775, women and children accounted
for 60 per cent of all slaves shipped to Jamaica, dropping to just under half during
the period between 1776 and 1800. The rise of abolitionism in Britain, the disruptions
occasioned by the American War for Independence, and then by the wars of the French
Revolution and the Haitian Revolution combined to give planters a strong incentive to
create a self-sustaining slave population, a move with huge implications for the devel-
opment of Afro-Caribbean society and culture.
Burnard and Morgan have argued that ‘Heterogeneity of ethnic origins is a con-
spicuous feature of African migration to Jamaica’, and it has become increasingly
clear that a more nuanced analysis of African ethnic identity is necessary, which
takes into account the shifting dynamics of ethnic identity of enslaved arrivals over
time, and the tremendously complex and constantly changing dynamics of
Slavery & Abolition 393
interacting African, European and creole cultures and identities. Plantations on
Jamaica usually included Africans from a variety of different regions, many of
them comparatively recent arrivals, and it is unlikely that the enslaved in Jamaica
were able to fashion distinct and coherent ethnically-based communities. However,
this does not mean that African ethnicity had no signiﬁcance in the development
of life, culture and society amongst the island’s enslaved population. With stagger-
ingly high mortality rates, and the island drawing more Africans than any other
portion of the British Atlantic World, Africans were numerically dominant and
African culture was ubiquitous in Jamaica throughout the period before 1807.
African identity and ethnicity were vitally important, and we must continue to
develop as complete an understanding as is possible of the ethnic identities of
enslaved Jamaicans, and of the ethnicity of enslaved women who built upon West
African traditions and the harsh realities of plantation life in raising children and
transmitting beliefs, culture and language to successive generations of Jamaican
slaves. The DNA research moves beyond the data available through the Voyages data-
base, which has provided evidence that more enslaved Africans left the Bight of Biafra
for Jamaica than from any other region. We have demonstrated that the mtDNA pro-
ﬁles of present-day Jamaicans reveals a rather different picture. Whatever their port
of disembarkation, the maternal lineages and ancestry of present-day Jamaicans
suggests that more of their enslaved maternal ancestors were drawn from the Gold
Coast region than from any other part of Africa, including the Bight of Biafra. The
DNA evidence suggests the existence and persistence of a core population of Gold
Coast Africans on Jamaica, and that women drawn from this population gave
birth to children and most likely passed on to them a variety of social and cultural
beliefs and practices: their descendants are on the island to this day. This analysis
ﬁts well with much of the seventeenth- and eighteenth-century literary record, in
which planters’ preferences for Gold Coast slaves appear frequently. Names from
the Gold Coast, the purported attributes of Gold Coast Africans, and even the
ethnic character of ‘Cormantee’ slave rebellions all suggest the signiﬁcance of this
region in the making not just of the enslaved population, but also the creation of
a distinct Afro-Jamaican society and culture. DNA analysis reveals the Gold Coast
ethnicity of many of the enslaved women who bore and raised children on
Jamaica, but it can only hint at their role in the perpetuation of Gold Coast language,
belief and culture in the increasingly creolized population.
 Olaudah Equiano, The Interesting Narrative and Other Writings, ed. Vincent Carretta
(New York: Penguin, 2003), 34, 171, 172. In his autobiography, Equiano claimed that he was
born in western Africa in about 1745, and then enslaved and sent to the West Indies at
about the age of 11. Vincent Cartetta has found evidence to suggest that Equiano may in
fact have been born in South Carolina, and that his African identity may have been a composite
account of an entire people, constructed from published sources and the personal recollections
of others. Whether or not Equiano had direct experience of West Africa, his account of its
people and culture is nonetheless both detailed and, in a manner of speaking, authentic. See
394 Simon P. Newman et al.
Carretta, Equiano, The African: Biography of a Self-Made Man (Athens, GA: University of
Georgia Press, 2005), xiv–xv.
 Ira Berlin, Many Centuries Gone: The First Two Centuries of Slavery in North America (Cam-
bridge, MA: Belknap Press, 1198), 39; David Eltis and David Richardson, ‘The “Numbers
Games” and Routes to Slavery’, in Routes to Slavery: Direction, Ethnicity and Mortality in the
Atlantic Slave Trade, ed. David Eltis and David Richardson (London: Frank Cass, 1997), 1.
 Trevor Burnard, ‘E Pluribus Plures: African Ethnicities in Seventeenth and Eighteenth Century
Jamaica’, Jamaican Historical Review, 21 (2001): 8; Michal Craton, Testing the Chains: Resistance
to Slavery in the British West Indies (Ithaca: Cornell University Press, 1982), 99; Amy Marie
Johnson, ‘Expectations of Slavery, African Captives, White Planters, And Slave Rebelliousness
in Early Colonial Jamaica’ (PhD diss., Duke University, 2007), 36.
 Voyages: The Transatlantic Slave Trade Database, David Eltis, Martin Lambert et al., http://www.
slavevoyages.org/ (accessed April 13, 2011). Initially published as a CD-ROM by Cambridge
University Press, and now mounted as a freely accessible web resource by the WEB. DuBois
Institute at Harvard University, the Voyages database includes at least partial information
about as many as 95 per cent of the slave ships that left British ports, as well as less complete
data for the slave ships of Dutch, French and other nations. All told some 35,000 voyages are
included in the database, with information about the point of departure, the ethnicity, and the
point of sale of hundreds of thousands of African slaves. Estimates drawn from Voyages. Actual
numbers, and known percentages of females were drawn from the database on 13 August 2011.
 Here, we are taking the known percentages of females assessed from the surviving records of
voyages which are available through Voyages, and applying them to the estimates of the total
numbers of enslaved Africans proposed by the database compilers, in order to estimate the
total number of women from each of these three key regions. One of the best studies of the
purchase of enslaved Africans in Jamaica, including consideration of their ethnic identity, is
Trevor Burnard and Kenneth Morgan, ‘The Dynamics of the Slave Market and Slave Purchasing
Patterns in Jamaica, 1655– 1788’, William and Mary Quarterly, 3d. ser., 58 (2001): 205– 28.
 Daryll Forde, ‘Kinship and Marriage Among the Ashanti’, in African Systems of Kinship and
Marriage, ed. A.R. Radcliffe-Brown and Daryll Forde (London: for International African Insti-
tute by OUP, 1956), 261 – 4. See also Barbara Bush, Slave Women in Caribbean Society, 1650–
1838 (Kingston: Heinemann Publishers Ltd., 1990), 83– 93.
 Similar studies have been undertaken for other populations of African descendants in the
Americas. See Antonio Salas et al., ‘The African Diaspora: mitochondrial DNA and the Atlantic
Slave Trade’, The American Journal of Human Genetics, 74, no. 3 (2004), 454– 65; Nicolas
Brucato et al., ‘The Imprint of the Slave Trade in an African American Population: Mitochon-
drial DNA, Y Chromosome and HTLV-1 Analysis in the Noir Maroon of French Guiana’, BMC
Evolutionary Biology, 10 (2010), 314; Klara Stefﬂova et al., ‘Dissecting the Within-Africa ances-
try of populations of African Decent in the Americas’, PloS One 6, no. 1 (2011), e14495.
 The scope and detail of the mtDNA evolutionary tree is well illustrated by the rendering of the
tree on the website http://www.phylotree.org/tree/main.htm. The African branch of the tree is
represented by ‘L’, and can be seen at http://www.phylotree.org/tree/subtree_L.htm.
 mtDNA is twice as sensitive to genetic drift as nuclear DNA because it is inherited from the
mother alone. Only those women who reproduce pass their mtDNA on, and it will pass on
to future generations through daughters. Peter A. Underhill & Toomas Kivisild, ‘Use of Y
Chromosome and Mitochondrial DNA in Tracing Human Migration’, Annual Review of Gen-
etics, 41 (2007), 539–64.
 Salas et al., ‘The African Diaspora’; Isabel Mendizabal et al., ‘Genetic Origin, Admixture, and
Asymmetry in Maternal and Paternal Human Lineages in Cuba’, BMC Evolutionary Biology,
8 (2008), 213.
 David C. McLean, Jr., et al., ‘Mitochondrial DNA (mtDNA) Halotypes Reveal Maternal Popu-
lation Genetic Afﬁnities of Sea Island Gullah-Speaking African Americans’, American Journal of
Slavery & Abolition 395
Physical Anthropology 127 (2005), 427–38; J. Benn Torres et al., ‘Mitochondial and Y chromo-
some diversity in the English-speaking Caribbean’, Annals of Human Genetics, 71 (2007),
782–90; Salas et al., ‘The African Diaspora’, 454– 65.
 Precise information about town, parish and county of birth is known for 344 (86 per cent) of
the volunteers. The percentages of these 344 who were born in each parish are as follows, with
that parish’s percentage of the Jamaican population in 2010 in brackets (as revealed in the 2010
census): Kingston and St Andrew, 36 per cent (25 per cent); St Thomas, 3 per cent (3 per cent);
Portland, 1 per cent (3 per cent); St Mary, 3.5 per cent (4 per cent); St Ann, 4 per cent (6 per
cent); Trelawny, 4 per cent (3 per cent); St James, 4 per cent (7 per cent); Hanover, 1 per cent (3
per cent); Westmoreland, 4 per cent (5 per cent); St Elizabeth, 13.5 per cent (6 per cent); Man-
chester, 8.5 per cent (7 per cent); Clarendon, 7 per cent (9 per cent); St Catherine, 10.5 per cent
(19 per cent). For Jamaican parish populations as revealed by the 2010 census, see ‘Population
by Parish 2010’, Demographic Statistics, Statistical Institute of Jamaica, http://statinja.gov.jm/
populationbyparish.aspx (accessed August 19, 2011).
 For a technical description of the DNA extraction procedure, see Appendix 1. For information
about the statistical methodology, see Appendix 2.
 Sidney W. Mintz and Richard Price, The Birth of African-American Culture: An Anthropological
Perspective (Boston: Beacon Press, 1992); Edward Braithwaite, The Development of Creole
Society in Jamaica, 1770–1820 (Oxford: Oxford University Press, 1971); John W. Thornton,
Africa and Africans in the Making of the Atlantic World, 1400– 1800 (Cambridge: Cambridge
University Press, 1992). For further examples of scholars following Thornton’s lead in exploring
ethnic clustering in slave arrivals, and cultural retention and survival in slave communities see
Gwendolyn Midlo Hall, Africans in Colonial Louisiana: The Development of Afro-Creole Culture
in the Eighteenth Century (Baton Rouge: Louisiana State University Press, 1992), and James
Sweet, Recreating Africa: Culture, Kinship, and Religion in the Afro-Portuguese World, 1441 –
1770 (Chapel Hill: University of North Carolina Press, 2003). For an excellent discussion of
this historiography, see Trevor Burnard, ‘The Atlantic Slave Trade and African Ethnicities in
Seventeenth-Century Jamaica’, in Liverpool and Transatlantic Slavery, ed. David Richardson,
Suzanne Schwarz and Anthony J. Tibbles (Liverpool: Liverpool University Press, 2007),
 Thomas Phillips, ‘A Journal of a Voyage Made in the Hannibal of London, Ann. 1693, 1694,
from England to Cape Monseradoe in Africa: And thence along the Coast of Guinney to
Whidaw, the island of St. Thomas. And so forward to Barbadoes’, in A Collection of Voyages
and Travels, Some Now Printed from Original Manuscripts, Others Now Published in English,
ed. Awnsham Churchill and John Churchill (London: 1732), VI, 214; Edwyn Stede and
Stephen Gascoigne to Royal African Company, Barbados, May 30, 1681, Abstract of Letters,
October 22, 1678 to May 30, 1681, Records of the Company of Royal Adventurers of
England Trading With Africa and Its Successors, National Archives, T70/15, 63; Peter
Thomson, ed., ‘Henry Drax’s Instructions on the Management of a Seventeenth Century
Barbadian Sugar Plantation’, William and Mary Quarterly, 3d. ser., 66 (2009): 585.
 Francis Grant to James Rogers & Co., December 30, 1778, Miscellaneous Accounts, Papers and
Correspondence of James Rogers, Merchant of Bristol, National Archives, C107/7, Box 2;
Simon Taylor to Robert Taylor, December 29, 1803, quoted in Audra Abbe Diptee, ‘Atlantic
Connections: An African Cohort in the Making of a Slave Society, Jamaica, 1775–1807’
(PhD. diss., University of Toronto, 2006), 60–1; Return Relation to the General State of the
Trade in Africa (1777), Records, National Archives, T70/177, 4. Johnson, ‘Expectations of
Slavery’, 34; Robert P. Stewart, ‘Akan Ethnicity on Jamaica’, The Maryland Historian,28
 David Eltis and David Richardson, ‘West Africa and the Transatlantic Slave Trade: New
Evidence of Long-Run Trends’, in Routes to Slavery, 20; Johnson, ‘Expectations of Slavery’,
143; Stephanie E. Smallwood, Saltwater Slavery: A Middle Passage from Africa to American
396 Simon P. Newman et al.
Diaspora (Cambridge, MA: Harvard University Press, 2007), 9; Diptee, ‘Atlantic Connections’
(PhD diss., University of Toronto, 2006), 1; Paul E. Lovejoy, Transformations in Slavery: A
History of Slavery in Africa (Cambridge: Cambridge University Press, 1983), 8.
 Eltis and Richardson, ‘The Numbers Game’, 6; Eltis and Richardson, ‘West Africa and the
Transatlantic Slave Trade’, 16.
 John Thornton, ‘The Coromantees: An African Cultural Group in Colonial North America and
the Caribbean’, The Journal of Caribbean History, 32 (1998), 161– 2; D.B. Chambers, ‘Ethnicity
in the Diaspora: The Slave Trade and the Creation of African “Nations” in the Americas’, Slavery
& ABolition, 22 (2001), 25–39.
 The Royal Gazette of Jamaica (Kingston), April 28– May 5, 1781, January 5–12, 1793, quoted in
Diptee, ‘Atlantic Connections’, 131, 62; Edward Long, The History of Jamaica: Or, General
Survey of the Antient and Modern State of that Island (London: for T. Lowndes, 1774), II, 472.
 Eltis and Richardson, ‘West Africa and the Transatlantic Slave Trade’, 20 – 1; Douglas
B. Chambers, ‘“My own nation”: Igbo Exiles in the Diaspora’, in Routes to Slavery, 73, 73–5,
84–5; Diptee, ‘Atlantic Connections’, 66–7, 136; Ugo Nwokeji, ‘The Atlantic Slave Trade and
Population Density: A Historical Demography of the Biafran Hinterland’, Canadian Journal
of African Studies, 34 (2000), 616–55. Diptee claims that between 1775 and 1800, Bonny
exported approximately 68,000 slaves to Jamaica, compared to 16,500 from Old Calabar and
8000 from Elem Kalabari: Diptee, ‘Atlantic Connections’, 66.
 Simon Taylor to Robert Taylor, March 26, 1804, and Simon Taylor to Robert Taylor, December
29, 1803, quoted in Diptee, ‘Atlantic Connections’, 62 – 3; Ugo Nwokeji, ‘The Biafran Frontier:
Trade, Slaves, and Aro Society, c.1750–1905’ (PhD diss., University of Toronto, 1999), 38–66.
See also Diptee, ‘Atlantic Connections’, 67–8.
 Diptee, ‘Atlantic Connections’, 29; David Eltis, The Rise of African Slavery in the Americas (Cam-
bridge: Cambridge University Press, 2000), 245– 50; Thornton, Africa and Africans, 189 – 90;
Burnard, ‘The Atlantic Slave Trade’, 143 – 6. See also Greg O’Malley, ‘Final Passages: The
British Inter-Colonial Slave Trade, 1619 – 1807’ (PhD diss., Johns Hopkins University, 2007).
 Burnard, ‘The Atlantic Slave Trade’, 142.
 Ibid., 146– 54. See also, Burnard and Morgan, ‘Dynamics of the Slave Market’, 208, 211 – 9.
 Michael Craton, ‘Jamaican Slave Mortality: Fresh Light from Worthy Park, Longville and the
Tharp Estates’, Journal of Caribbean History, 3 (1971), 5, 24; Barry Higman, Slave Population
and Economy in Jamaica, 1807–1834 (Cambridge: Cambridge University Press, 1976), 76.
 Long, The History of Jamaica, II, 474, 475; W.J. Gardner, A History of Jamaica: From Its Discov-
ery by Christopher Columbus to the Year 1872 (London: E. Stock, 1873), 184; Long, History of
Jamaica, II, 474. See also Michael Mullin, Africa in America: Slave Acculturation ad Resistance
in the American South and the British Caribbean, 1736 – 1831 (Champaign, IL: University of Illi-
nois Press, 1992), 270–1; Mervyn Alleyne, ‘Language in Jamaica’, in The Language, Ethnicity
and Race Readers, ed. Roxy Harris and Ben Rapton (London: Routledge, 2003), 56– 7; R.B.
LaPage and David De Camp, Jamaican Creole (London: Macmillan, 1960), 77; Robin Law,
The Slave Coast of West Africa, 1550 –1750: The Impact of the Atlantic Slave Trade on an
African Society (Oxford: Clarendon Press, 1991), 22–4; Stephanie E. Smallwood, ‘African Guar-
dians, European Slave Ships, and the Changing Dynamics of Power in the Early Modern Atlan-
tic’, William and Mary Quarterly, 3rd ser., 64 (2007), 679 – 716. See also Mullin, Africa in
America, 270– 71.
 David De Camp, ‘African Day-Names in Jamaica’, Language, 43 (1967), 139 – 43; Trevor
Burnard, ‘Slave Naming Patterns: Onamastics and the Taxonomy of Race in Eighteenth
Century Jamaica’, Journal of Interdisciplinary History, 31 (2001), 325–46.
 Stewart, ‘Akan Ethnicity’, 85 – 6.
 Diptee, ‘Atlantic Connections’, 1–4.
 Barbara Kopytoff, ‘The Development of Jamaican Maroon Ethnicity’, Caribbean Quarterly,22
(1976), 40; Kopytoff, ‘The Maroons of Jamaica: An Ethnohistorical Study of Incomplete
Slavery & Abolition 397
Polities, 1655– 1905’ (PhD diss., University of Pennsylvania, 1973), 71; Craton, Testing the
Chains, 67, 77–8.
 David Dalby, ‘Ashanti Survivals in the Language and Traditions of the Windward Maroons of
Jamaica’, African Language Studies 12 (1971), 31–51.
 Kristin Mann, ‘Shifting Paradigms in the Study of the African Diaspora and of Atlantic History
and Culture’, Slavery & Abolition 21 (2001), 3–21; Diptee, ‘Atlantic Connections’, 33, 38–40.
 Burnard and Morgan, ‘Dynamics of the Slave Market’, 208; Burnard, ‘E Pluribus Plures’, 10;
Philip D. Morgan, ‘The Cultural Implications of the Atlantic Slave Trade: African Regional
Origins, American Destinations and New World Developments’, in Routes to Slavery: Direction,
Ethnicity and Mortality in the Atlantic World, 1400– 1680, ed. David Eltis and David Richardson
(London: Frank Cass, 1997), 122–45.
 The buccal swabs are rigid ﬁbrous swabs attached to small rods. Either the subject or the
researcher scrubs the inside of the cheek rigorously for a minute or two to collect buccal
cells. The small rod is ﬁtted with a quick release system for depositing the swab into a centrifuge
tube containing cell lysis solution, a liquid used to break open the cell walls. These were pro-
vided by Medical Packaging Corporation, Camarillo, CA.
 Using a series of chemical washes and specially designed ﬁlters, human genetic material was
isolated while oral bacteria, salivary enzymes and other general debris were removed (Qiagen
Ltd., Crawley, UK).
 For details, see Robert Scott et al., ‘Mitochondrial DNA Lineages of Elite Ethiopian Athletes’,
American Journal of Human Genetics, 140, no. 3 (2005), 3, 497– 503.
 Richard M. Andrews et al., ‘Reanalysis and Revision of the Cambridge Reference Sequence for
Human Mitochondrial DNA’, Nature Genetics, 23, no. 2 (1999), 147.
 The sequencer was produced by FMC Bio-Products, Rockland, Maine, USA.
 Chromas Lite is produced by Technelysium, Queensland, Australia. Thomas Hall, ‘Bioedit: A
User-Friendly Biological Sequence Alignment Editor and Analysis Program for Windows 95/
98/NT’, Nucleic Acids Symposium Series, 41 (1999), 95 – 8.
 Mannis van Oven and Manfred Kayser, ‘Updated Omprehensive Phylogenetic Tree of Global
Human Mitochondrial DNA Variation’, Human Mutation, 30, 2 (2009), E386 – 94. Antonio
Salas et al., ‘Charting the Ancestry of African Americans’, American Journal of Human Genetics,
77, no. 4 (2005), 676–80.
 The exact test of population differentiation is an accurate and unbiased method to calculate any
difference between the distribution of groups by exploring the probability with a Markov chain,
a method introduced in Michael Raymond and Francois Rousset, ‘An Exact Test of Population
Differentiation’, Evolution, 49 (1995), 6, 1280– 3.
 The deﬁnition stated in David P. Rowell, ‘Teleconnections between the Tropical Paciﬁc and the
Sahel’, Quarterly Journal of the Royal Meteorological Society, 127 (2001), 575, 1683–706.
 Salas et al., ‘The African Diaspora’, 456.
 Mendizabal et al., ‘Genetic origin, admixture, and asymmetry’.
Appendix 1: Methodology
Buccal swabs were collected from each individual and stored in cell lysis solution.
The total DNA was then extracted using the Qiagen buccal spin protocol.
hypervariable segment (HVS-1) of the mitochondrial genome was then ampliﬁed
using a polymerase chain reaction. A reverse primer was used to generate all
A forward primer was also used to read the sequence prior to the poly-
cytosine tract between base pairs 16,184– 16,193 according to the revised mtDNA
reference sequence (rCRS).
Sequencing product was then separated by 5 per cent
398 Simon P. Newman et al.
denaturing Long Ranger gel and detected using an Applied Biosystems 377 DNA
Chromatograms were later read and converted to text using Chromas
Lite, then manually aligned to the rCRS using Bioedit.
After sequencing, basepair
differences from the rCRS were recorded, controlling for the length polymorphism
found between base pairs 16,184– 16,193, and haplogrouped according to the compre-
hensive full mtDNA genome phylogenetic tree Phylotree at a comparable resolution to
that in the literature.
It then became possible to compare the DNA of the Jamaican
subjects with a variety of parental West African populations, and thereby assess the
African origins of this sample of the Jamaican population.
Appendix 2: Statistical
An exact test of population differentiation was performed on haplogroup proﬁle fre-
quencies between each African coast to investigate the discreet nature of each group in
light of any effect population migration or the inter-African slave trade may have
played on the clustering of populations into distinct geographic regions.
logroups not found in sub-Saharan Africa were excluded from the Jamaican sample
for the admixture analyses. Additional analyses were preformed excluding more mar-
ginal groups including the pygmy groups of equatorial Africa and groups from the
Sahel, a region geographically deﬁned here as 11.25 – 18.758N, 16.8758W – 35.6258E.
Admixture coefﬁcients were estimated using two separate methods. Firstly, a
Markov chain Monte Carlo posterior sampling method assuming a multinomial dis-
tribution for the mtDNA haplogroup proﬁles, a method introduced in detail by Salas
et al. (see note 7):
To quantify the magnitude of the impact of each African region on cluster frequen-
cies in the Americas, we ﬁtted the following model. The number of mtDNAs in each
cluster in the sample from a region of the Americas (n
:1≤i≤C, the number of clus-
ters) was assumed to be a draw from a multinomial distribution with parameters
i=1ni, the sample size in the American region, and pi=R
(1≤i≤C), where R is the number of source regions in Africa, f
is the frequency
of the ith cluster in the jth source region (assumed to be known), and the a
the admixture coefﬁcients. This model describes samples from an urn with C differ-
ent kinds of ball, where the urn has been created by mixing together R other urns in
proportions given by the admixture coefﬁcients. We chose to analyze this model in a
Bayesian framework, which meant that we had to explore the distribution of the
admixture coefﬁcients, given the data. The prior distribution of the admixture coef-
ﬁcients was taken to be uninformative – namely, uniform on
posterior distribution of a
was explored with the Metropolis-Hastings algorithm,
using a simple proposal, and was summarized by the posterior mean of each a
and its root-mean-square deviation about the mean.
Secondly, a similar approach was used to calculate the admixture coefﬁcients at a
much ﬁner resolution by comparing haplotypes between Jamaica and coastal regions
of Africa. The nature of the design makes it susceptible to the number of haplotypes
in each group, so a bootstrap resampling procedure was run to assess the conﬁdence
of the model ﬁt. This method is best explained in Mendizabal et al. (see note 10):
Slavery & Abolition 399
we calculated the probability of origin of each subcontinental region by a Bayesian
approach. The probability of origin of each of the subcontinental region was com-
puted as p0s=1
where, n is the number of Cuban sequences with matching
(≥1) in the whole continental dataset; k
, the number of times the sequence i is
found in the Cuban sample; p
, the frequency of the sequence i in the subcontinental
region dataset; and p
, the frequency of the sequence i in whole continental dataset.
In order to provide conﬁdence intervals for each of the estimations for the
subcontinental regions, we also computed the standard deviation as
400 Simon P. Newman et al.