The sample,141 mouths, Only six islands,The Only Ancestry Survey of Living Maldivians,
There is still only one published ancestry survey of living Maldivians: Jeroen Pijpe, Alex de Voogt, Mannis van Oven, Peter Henneman, Kristiaan van der Gaag, Manfred Kayser, and Peter de Knijff, “Indian Ocean Crossroads: Human Genetic Origin and Population Structure in the Maldives,” American Journal of Physical Anthropology 151 (2013): 58–67 (doi:10.1002/ajpa.22256). The mouth swabs were taken in 2003. The article is on Wiley and, as a free copy, on PubMed Central. Everything later said about “Maldivian DNA” in origin debates is, at bottom, a reading of this paper,or a complaint that it is too small.
Who did it, and with what tools
The work sat in Dutch forensic and human-genetics labs: Leiden University Medical Center and Erasmus MC in Rotterdam, with de Voogt then at the American Museum of Natural History. Ethics followed Leiden rules. The Maldivian government approved collection. Acknowledgments thank staff of the National Centre for Linguistic and Historical Research in Malé and named local helpers. The paper does not list a named external grant.
They did not sequence whole genomes and did not run a modern SNP-chip ancestry painting. They used three layers:
mtDNA , the mother’s line: control-region haplotypes assigned to haplogroups
Y chromosome , SNPs for haplogroup plus STRs for finer male haplotypes
autosomal STRs , a small set of forensic-style repeats on the ordinary chromosomes
That design can say which regional pools mothers and fathers resemble. It cannot estimate “eight percent East African autosomal ancestry” in a Feridhoo family. Mitochondrial DNA and the Y are passports. Mixed ancestry after many marriages lives on the other twenty-two pairs of chromosomes.
The sample: 141 mouths, six islands They collected 141 unrelated mouth swabs at six medical clinics, including Malé Hospital. The design was deliberate, not a random street draw. They wanted people whose both parents came from the same island, to catch older island structure rather than recent mixing in the capital. Of the 141, 119 were tied to six named settlements spanning almost the full length of the chain:
Island | Atoll | Dialect group | n (approx.) |
Huvarafushi | Haa Alifu | Northern | 20 |
Kandholhudhoo | Raa | Northern | 20 |
Feridhoo | Ari (Alifu) | Northern | 12 |
Nilandhoo | Faafu | Northern | 20 |
Fuvahmulah | Gnaviyani | Southern | 24 |
Addu Gan | Addu | Southern | 23 |
Feridhoo, the island most often named in African oral history,is the weakest cell. Several island collections were male-heavy, so maternal history is thinner than the headline number. Clinic sampling also favors people who could reach a clinic in 2003.
Maternal resultsThey found 63 mtDNA haplotypes in 29 haplogroups, mostly in the great South Asian clades M, R, and U. A few common groups,M6, M30, M36, M39, M41, M66, R5, U1, U2,together covered about two-thirds of the people. Two haplotypes inside macrohaplogroup R had not been reported before.
West Asian maternal types that would shout “Arabia”,JT, HV, R0,and African L were almost absent. Eastern types typical of Island Southeast Asia,B, R9, M7, E,were almost absent. One person carried M71, a type known at low frequency in Southeast Asia: an exception, not a wave.Using LEADMIX on haplogroup frequencies against three reference blocks (West, Central/South Asia, East), they estimated that 99.2 percent (95 percent CI 90.6–99.5) of the observed mtDNA could come from the Central block: India, Pakistan, Sri Lanka. West and East maternal contributions were statistically negligible (about 0.4 and 0.3 percent). That is the number behind every claim that “Maldivian mothers are South Asian.
Paternal results
They found 66 Y-STR haplotypes in 10 haplogroups. The common ones were the South Asian set: H1, J2, L, R1a1a, R2. Other groups appeared at low frequency, including C, G, Q, and a few odd lineages that might be local. Four Q1a chromosomes sat in the network.
Y diversity was low on every island and in the country as a whole,lower than many island peoples, in the range of bottleneck or founder populations. AMOVA put about 7.4 percent of Y-STR variance among islands: men had not moved freely between atolls. That is the basis for “independent male founder effects on each atoll.”
On the same LEADMIX table, the Y chromosome looks less one-sided than mtDNA: about 74.6 percent Central and about 25.3 percent West, with the East near zero. The authors refused to treat that quarter as a clean Arabian immigration. The West-looking slice is driven mainly by J2-M172, a haplogroup frequent in West Asia and in South Asia. Combined uniparental frequencies still put the Central block first (about 71 percent) and the West second (about 29 percent)
The clearest non-South Asian paternal trace was two men on Gnaviyani with haplogroup A(M91) and identical STR haplotypes matching types in Ethiopia and Uganda. That is a real East African paternal crumb. It is two men.
Feridhoo’s twelve samples did not carry the classic African mtDNA or Y tags the island’s reputation would predict. The paper says so explicitly. That is not proof of no African blood. It is proof that twelve uniparental tests are the wrong tool.
What they concluded-and what they excluded
Qualitative and quantitative haplogroup comparison said Maldivian genomes, on these markers, do not differ much from India,Pakistan,Sri Lanka. Eastern settler haplogroups that would support a major Austronesian stopping colony are missing. The authors therefore rejected the idea that the Maldives were a peopling bridge from Southeast Asia to Madagascar. Atolls were more likely watering stops than settler homelands. Female gene flow from West Asia was dismissed. Male West Asian flow was left as possible but inseparable from South Asia.
Inside the country they found subtle structure that does not line up neatly with miles or with the north/south dialect split. Reduced female movement between atolls fit older reports of matrilocality. Reduced male diversity fit separate island founders.
Their last sentence is careful: the data agree with the usual South Asian origin story, and they suggest multiple independent immigrations and unequal movement of women and men.
Why this survey still cannot close the historical file
Ibn Battuta already saw Africans in the 1340s. H. C. P. Bell mapped Buddhist hawitta. Duckworth, reading Gardiner’s measurements, guessed a third wave from Arabia or Africa. Clarence Maloney argued a Tamil floor under Dhivehi. Thor Heyerdahl wanted sun-worshippers. Pijpe 2013 can bless Bell and Maloney on the majority. It cannot weigh a few percent of East African autosomes after twenty generations of marriage.
There is still no published whole-genome painting of the Maldives and no well-known ancient DNA from those hawitta. Medical programs typed thalassemia and HLA. They did not replace this paper
So the 2013 survey remains what it always was: the only origin study, strong on South Asia, honest about two African Y chromosomes and one eastern mtDNA, silent where the slave ledgers and boduberu drums are loudest. Until someone runs hundreds of genomes from Feridhoo, Kudafari, old Malé, and the south, the argument will keep orbiting these 141 swabsnot because they are complete, but because they are almost alone
Comments