Dear Leo
It is now established that WIlliam de Longespee's mother was Ida de Tosny - the SGM archive will show the trail to this conclusion.
For years it was known that William called his mother Ida "countess" and it was assumed that she became the wife of Roger Bigod, earl of Norfolk (the Magna Carta surety) as this was the only woman of the right name and rank at the appropriate time.
Some eyars ago Ray Phair posted proof to the newsgroup that this was true - Ida's Bigod son was called brother to William de Longespee and vice versa.
Then a new book on the Bigod earls came out in 2005, citing an account from the Hundred Rolls where a jury in the following century recorded that Earl Roger's wife Ida was surnamed "de Toeny".
The question still remains, was she a daughter of Roger II of Tosny & Ida of Hainaut, or their granddaughter (i.e. a daughter of Raoul V of Tosny & Margaret de Beaumont). Both are chronologically possible - I strongly incline to the former, but this can't be definite on the available evidence.
I agree that the chronology would support Ida being either (A) a
somewhat late-born daughter of Roger de Tosny and Ida of Hainaut
[married before 9 August 1138], or (B) an older daughter of the
marriage of their son Ralph to Margaret de Beaumont.
I am strongly inclined to placing her as a daughter of the latter
couple, for two reasons:
1) Onomastics (admittedly weak), on the basis of Roger le Bigod
and Ida de Tony having a daughter Margaret (m. William de Hastings of
Aston Flamville, co. Leics.). All the names of the other issue of the
marriage have an ascertainable Bigod or Tony source via either
marriage - with the exception of Mary [source unknown, but possibly
religious only] - but Margaret would appear to be the namesake of
Margaret de Beaumont, who I think was her maternal grandmother.
2) Chronology also points in this direction. Roger le Bigod and
Ida de Tony had 8 known children, which would have likely occurred
over a period of say 12 to 16 years. * One might argue for a period
of 6 to 8 years as possible, but this seems extreme, assuming possible
multiple births, no stillbirths, and relatively rapid consecutive
pregnancies *. Roger and Ida are held to have been married in late
1181, so we might expect the first birth to have occurred in later
1182. If Ida is then conjecutured to have stopped giving birth ca.
1194-1198, she was probably born no earlier than say 1150-1155. She
is usually held to have been born ca. 1155-1160 given the estimated
birth of William Longespee ca. 1175. Ralph de Tony and Margaret de
Beaumont were married before 1155, according to what I recall from
Paul Reed and Ray Phair.
The foregoing is still not conclusive, but I think provides a
greater likelihood (not certainty) for Ida being a daughter of Ralph
de Tony and Margaret de Beaumont.
Cheers,
John
Ida was a royal ward in 1181
Roger was very much alive and kicking, Ralph however was long dead.
Will Johnson