Ẹrẹ̀ Ìbejì: The Twin Figures
The carved figures made when a Yoruba twin dies, what the family does with them, the genuinely exceptional twinning rate that produced them, and how tens of thousands of them ended up in Western collections.
When one of a pair of Yoruba twins dies, the family may commission a small carved wooden figure, an ẹrẹ̀ ìbejì, and thereafter treat it as they would have treated the child: washing it, oiling it, dressing it, feeding it, carrying it. The figure is not a portrait and it is not a toy. It is a place for the dead twin to be, so that the twin remains inside the family and so that the surviving twin is not left half of a pair. Because the Yoruba have the highest twinning rate ever recorded in a human population, and because the practice was widespread and continued into the twentieth century, an enormous number of these figures exist, and a very large proportion of them are now outside Nigeria.
Why there are so many twins
The Yoruba twinning rate is not folklore, it is one of the better-documented facts in human population genetics. The rate concerned is dizygotic twinning, that is, fraternal twins arising from two separate ova, which is the type that varies between populations; monozygotic twinning is roughly constant everywhere at about four per thousand.
Ìgbó-Òra in Oyo State, which markets itself as the twin capital of the world, has a recorded dizygotic twinning rate of about 45 per 1,000 live births . The global range for dizygotic twinning is generally 8 to 16 per thousand . The Yoruba rate is therefore roughly three to five times the world average, and the work establishing it was done by Patrick Nylander from the late 1960s .
Two explanations are on the table and they are not exclusive.
The first is genetic. Elevated dizygotic twinning runs in families and populations, is inherited maternally, and is mediated by follicle-stimulating hormone levels producing multiple ovulation. This is the mainstream explanation for population-level differences.
The second is dietary and is the one that gets repeated in popular coverage. Nylander proposed that phytoestrogens, particularly in yam species of the genus Dioscorea, might stimulate the ovaries and produce multiple ovulation . This hypothesis is decades old, is intuitively attractive because Yoruba diet is yam-heavy, and has not been demonstrated. It should be reported as a hypothesis and not as a finding, which is not how it is usually reported.
A further detail from Nylander's work is more securely established and less often mentioned: the peak twinning rate among Yoruba mothers falls at ages 30 to 34, whereas in other populations the peak is in older maternal age brackets .
There is one genuine methodological caution. Modern rates measured in a specific town are not automatically valid for the historical Yoruba population as a whole, and Ìgbó-Òra is exceptional even within Yorubaland, which is why it is famous. The safe statement is that Yoruba dizygotic twinning is very high by world standards and Ìgbó-Òra is the highest recorded locality.
The cultural point that follows is straightforward. A society in which twins are several times more common than elsewhere, and in which infant mortality was historically high, will have a large number of families with one surviving twin and one dead twin. The ẹrẹ̀ ìbejì is an institutional response to a demographic fact.
Táíwò and Kẹ́hìndé
Yoruba twins receive fixed names rather than chosen ones. The first born is Táíwò, from a contraction usually explained as tọ́ ayé wò, "taste the world" or "come and see the world," sometimes given in the fuller form Táíyéwò or Táyélolú. The second born is Kẹ́hìndé, "the one who comes last," or Ọmọ́kẹ́hìndé.
The tradition inverts the obvious reading of who is senior. Kẹ́hìndé, born second, is regarded as the elder, on the account that Kẹ́hìndé sent Táíwò out first as a scout to test whether the world was fit to be entered, and followed only when word came back . Táíwò is therefore the junior, the one who was sent, and Kẹ́hìndé the senior who did the sending. A third child born after twins is named Ìdòwú, and the child after that Àlàbá.
The oríkì of twins
Twins have their own body of oríkì, sung by mothers and by the ìyá ìbejì, and the opening formula is one of the best known lines of Yoruba oral poetry.
Ẹ̀jìrẹ́ ará ìṣokún, Ọmọ ẹdún tíí ṣeré orí igi.
Ẹ̀jìrẹ́: the two-who-are-one, the paired ones (from èjì, two). ará ìṣokún: person of / native of Ìṣokún. Ọmọ: child, offspring. ẹdún: the colobus monkey. tíí: who habitually. ṣeré: plays, sports, makes play. orí igi: on top of the tree, at the head of the tree.
Ẹjirẹ, native of Isokun, Child of the monkey that plays in the treetops.
Translator: Ìpilẹ̀ṣẹ̀ editorial rendering, unreviewed
The rendering here is mine from the widely circulated Yoruba text and is marked as such; the confidence on this specific translation is medium and the text is given because the formula is so standard, not because a single authoritative published transcription was located.
Three things go missing in English. Ẹ̀jìrẹ́ is not "twins", which is ìbejì; it is a praise-name built on èjì, two, and carries the sense of a matched or paired excellence, so calling a child Ẹ̀jìrẹ́ is closer to "you paired one" than to a statement of fact. Ìṣokún is a place name and the line asserts that all twins originate there, which is an origin claim rather than biography, and no twin is expected to have any actual connection to Ìṣokún. Ẹdún is a specific animal, the colobus monkey, and the association of twins with monkeys is a substantive part of the tradition: monkeys, which commonly bear multiple young and move in troops, are linked to twins across a body of Yoruba narrative, and in some accounts twins are said to descend from or be kin to them. Rendering ẹdún as a generic "monkey" loses the species and loses the point. The full oríkì runs much longer and continues through further epithets.
Elisha Renne's chapter "The Ambiguous Ordinariness of Yoruba Twins," in the collection Twins in African and Diaspora Cultures, is the place to go for how ambivalent this material actually is . Twins are treated as bringers of wealth and blessing and simultaneously as dangerous and demanding, and the same practices that honour them are also practices of appeasement.
The figure itself
When it is made. If a twin dies, whether in infancy or later, the family consults a diviner, and if the divination calls for it a carver is commissioned to make a figure . Where both twins die, two figures are made. The figure belongs to the mother, and passes on her death to the surviving twin or to another family member.
What it looks like. Typically 20 to 30 centimetres high, a standing figure on a round base, arms at the sides, hands often against the thighs, an elaborate coiffure, and often carved with the family's facial marks . That last detail is important: the figure carries the lineage identification of the child it stands for, which makes it specific to a family even though it is not a portrait.
It is carved as an adult. The figure has adult proportions, adult genitals, and in the female case adult breasts, even when the child it represents died at a few days old. This confuses first-time viewers and it is not a mistake. It follows the aesthetic criterion of ọ̀dọ́, the requirement that a figure be shown at the prime of life, and it follows the logic of the object: the dead twin continues to exist and continues to grow, so the figure shows the person the twin now is, not the infant who died .
What is done with it. This is where the object differs most from anything in a Western category. The figure is washed, rubbed with oil, powdered with camwood, dressed in cloth, given beads and cowries, fed, and carried on the mother's back in a wrapper as a living child would be carried . It is placed with the household's other important things and it is addressed. The care is not commemorative in the sense of remembering; it is continuing, and one commonly stated reason for it is that treating the dead twin well protects the surviving twin, whose fate is understood to be bound to the other's .
The physical consequence is the diagnostic feature of the genre. Ìbejì that were used for decades have worn faces, softened features, sometimes the nose and mouth almost gone, and a thick dark accretion of camwood, indigo and oil. Collectors call this patina and value it. In the Yoruba account it is not patina, it is the record of the object's whole life, which is what ìwà means applied to a thing . A crisp, sharp, unworn ìbejì is either recent, or was never used, or was made for sale.
When both a mother and a figure become unavailable. In the later twentieth century families sometimes substituted a mass-produced plastic doll, or a photograph, for a carved figure. Marilyn Houlberg documented these substitutions in her Nigerian fieldwork . The practice adapted; it did not simply stop.
The collection history
There are a very large number of ẹrẹ̀ ìbejì in Western hands, more than of any other single Yoruba object type, and the reason is a combination of the twinning rate, the small size and portability of the figures, and a twentieth-century market that specifically sought them.
Fausto Polo, who compiled a photographic reference of ìbejì by carver, put the number of identifiable ìbejì carvers active between about 1830 and 1960 at three to four hundred . Gert Stoll published the first reference book on the subject in 1980 . Both of these are collector-market publications rather than academic ones, and their attributions of figures to carvers rest on stylistic grouping, so they should be used as typologies rather than as documented artist attributions of the kind established for Olówè.
The scholarly documentation is thinner and more valuable. Marilyn Houlberg photographed among the Yoruba from 1961 to around 2007, and her collection at the Smithsonian's Eliot Elisofon Photographic Archives runs to 8,515 colour slides, of which roughly 854 are her ìbejì series, dated from about 1969 to about 1990 . Her field notes are separately archived . That is the closest thing to a systematic photographic record of the figures in use rather than on plinths.
Tim Chappel's 1974 article in Africa set the historical framing, arguing that the twin cult should be understood as having a history and a development rather than as a timeless custom .
The ethical shape of the collection history is worth stating plainly and without exaggeration. Ìbejì were not looted in a military action, as the Benin bronzes were. They were bought, very often from the families that owned them, in a market where the price offered was substantial by local standards and where missionary and later Pentecostal pressure was actively defining the figures as idols to be given up. A sale under those conditions is a real sale and it is also not a free transaction in any strong sense. What was sold in each case was an object that a specific mother had washed and fed for years and that stood for a specific dead child. The museum label "Twin figure, Yoruba peoples, Nigeria, 20th century" is the entire remaining record of that.