Contemporary chronicle evidence and the testimony of Edward I's wardrobe
accounts confirm that Joan's first child by Gilbert, another Gilbert, was
born in May 1291. The dates of birth of her last three children, all
daughters, are not as specifically recorded and must be recovered from the
inquisitions post mortem taken following the death at Bannockburn in 1314
of their brother Gilbert, the last Clare earl of Hertford and Gloucester.
Only two of the sisters' births can be determined in this way, however.
Elizabeth, the youngest, was born in November 1295, only weeks before her
father's death.
But it appears to have been the eldest of the the three sisters, Eleanor,
who was born in October 1292 and not Margaret, the second of the three.
All contemporary record sources agree that Eleanor was the eldest, and as
the human reproductive cycle is fixed within certain limits there is, even
biologically speaking, barely enough time between May 1291, when young Gilbert
was born, and October 1292 when it is alleged Margaret was born, for yet
another child to have been born. If we factor in the prevailing custom of
a woman's confinement for a period of up to 60 days (two months) between the
birth of a child and her churching or purification, it becomes even less likely
that Joan "of Acre" could have borne Gilbert in May 1291, then Eleanor and
then Margaret in October 1292.
Frances Underhill's biography of the youngest sister Elizabeth (to be published
in November by St Martin's Press in New York) date Eleanor's birth to October
1292, Margaret's c. 1295 and Elizabeth's in November 1295.
John Parsons
John Carmi Parsons wrote:
Does the evidence preclude that Eleanor and Margaret might have been twins, with
Margaret the younger? Or was it customary for the chronicles to identify twins?
Thanks for addressing this.
Henry Sutliff
John P.