THE HERBERT FAMILY PEDIGREE
By Darrell Wolcott
In the mid-1400's, two brothers assumed the surname Herbert;
these were Sir William, Earl of Pembroke and Sir Richard of Coldbrook, sons of Sir William ap Thomas and his wife Gwladys
Ddu, daughter of Sir Dafydd Gam. This was probably done at the time Sir William was named Earl of Pembroke by King Edward
IV. Apparently these brothers knew their grandfather, Thomas, was the son of Gwilym ap Jenkin ap Adam but further back
they were unsure.
We are told King Edward IV commissioned one Hywel
Lloyd and others to find the true pedigree of the Earl of Pembroke. The result of that project traced Jenkin ap Adam
back through a succession of men, ending with a "Lord Herbert, Duke of Cornwall son of Godwin"
 said to have lived prior to the 1066 Norman conquest. Based on those now extant, this early attempt at a family
pedigree probably looked like this:
Sir William==Gwladys Ddu
Since the first Herbert shown in this
chart was subsequently identified as a companion of William the Conqueror, early genealogists scoffed at the notion he
could have been the son of a Saxon and Godwin was dropped from the pedigree. Soon, amended pedigrees were being cast
making the second-named Herbert a base son of King Henry I. That failed to convince others who said the Henry in
the pedigree was the "Henry the Treasurer" found in the 1086 Domesday Book, but we suspect all the early generations were
men called Herbert. Yet others claimed an early marriage to Emma de Blois, half-sister to King Stephen and a daughter
of Adelia, a base daughter of William the Conqueror.
Other early marriage matches inserted
in the various versions of the pedigree included Nest ferch Rhys ap Tewdwr, Julia daughter of Robert Corbet and Lucy
daughter of Milo fitz Walter. It was almost certainly not Julia but Sibyl Corbet that married a Herbert. A
son of the Corbet marriage would fit chronologically with Lucy and the liason with Lucy produced Peter. A public record
calls "Sibilla" the grandmother of "Petr fili Herbti". This section of the early Norman family probably looks like this:
1035 Herbert, Count of Vermandois
1085 Nest vz Rhys ==/==Herbert
II===Emma de Blois 1080
l 1070 l
1105 Sybil Corbet====Herbert III St William,
archbishop ob 1154
Herbert IV====Lucy dau of Milo fitzWalter 1135
fitz Herbert==Alice dau Robert FitzRoger 1175
Without spending a lot of time
dissecting the charted family, the chronological time line is possible. If any of those men sired a child by the infamous
Nest ferch Rhys ap Tewdwr it would have to be the one called Herbert II although those pedigrees who include
Nest match her with the father of the Herbert who married Lucy. (See Appendix below)
Reginald fitz Peter (son of Peter in
the above chart) is said to have married Joan, daughter of William Fortibus (1190-1241) and their son, Peter fitz Reginald,
died in 1323 leaving a son Herbert fitz Peter who was 48 years old at the inquisition taken in 1323....thus born in 1275.
It is the marriage match claimed for Peter fitz Reginald which, we believe, was faked in order to connect this family to the
later Earl of Pembroke. The lady in question is Alis ferch Bleddyn Broadspear, heiress of the manors of Llanllowel
and Beachley. We shall return to this lady momentarily, but first we should present a chart showing the remainder of
the pedigree together with estimated birthdates which will expose the fraud:
1275 Herbert fitz Peter (birthdate on record)
Gwilym obit 1377
1360 Thomas obit 1438
Sir William ap Thomas obit 1446
We date the Adam in the
above chart by his marriage, which all versions of the pedigree agree was Cristyn ferch Gwarin Ddu. This man was
descended from Ynyr, king of Upper Gwent, as follows:
1095 Ynyr Fychan
1155 Sir Gwarin
1230 Gwarin Ddu
1265 Cristyn========Adam 1255
While the medieval genealogists
had made a number of emendations to the basic pedigree, some seeking to paper over the chronological problem by attaching
Adam to Reginald instead of to his grandson, all insisted the family at the bottom of the chart descended from the one at
the top. But in 1876, George T. Clark debunked the pedigree as "a forgery and a clumsy one, and its statements not to
be reconciled with independent dates and records". Rather than be content that the ancestry of Adam clearly was
not that claimed in the pedigree, we sought to learn not only his correct ancestry but why the earlier genealogists had connected
him to the old Norman family.
Our journey began with two pedigrees
which said the father of Adam was actually named Cynhaethwy and a third pedigree which said Peter fitz Herbert had
a brother named Cynhaethwy, and that man had a son named Adam. Down to this point in the pedigree, all men
bore common English names and married English/Norman ladies but suddenly we encounter a purely Welsh name. Further research
led us to the family of Adam Gwent. Among his sons were Adam Fychan and Cynhaethwy; while his pedigree did not chart
the descendants of Cynhaethwy, it does identify a son of Adam Fychan who is styled "Lord of Beachley".
Return now to a lady mentioned earlier:
Alis ferch Bleddyn Broadspear. Most of the citations which mention Bleddyn identify him as Lord of Llanllowel and Beachley.
The fake Herbert pedigrees show Llanllowell descending to Sir Thomas ap Adam, older brother of Jenkin ap Adam.
They omit any mention of Beachley, a fact "corrected" by Joseph Morris in 1858. He simply takes Sir John ap
Adam Fychan (the man who did inheirit Beachley) and makes him the brother of Jenkin ap Adam and assigns Llanllowel to him.
Acting on the theory that both the families we
find holding those two manors c. 1300 had a common ancestor, and that ancestor was a son of Alis ferch Bleddyn,
we decided to see where it might lead us if we were to identify the Cynhaethwy claimed to be a brother of Peter fitz
Herbert (and father of Adam) with the man of that name cited as a son of Adam Gwent. Our result was this chart:
Caradog of Penrhos Bleddyn Broadspear
1220 Adam Fychan
1255 Sir John
Lord of Beachley l
Beverstone Sir Thomas 1285 1290 Jenkin
Lord of Llanllowel Lord of Wern Ddu
Iorwerth ap Caradog in this chart
was steward to Hywel ap Iorwerth ap Owain Wan of Caerleon, ruler of Lower Gwent c. 1175. Adam Gwent was steward to Morgan
ap Hywel c. 1210 Our construction is chronologically stable now that we have disconnected Jenkin ap Adam from the
Norman fitz-Herbert family and placed Alis ferch Bleddyn in a marriage that produced the descendants that actually inherited
her lands. Bleddyn's ancestry is nowhere given, but it is a common Welsh name and the form "Alis ferch Bleddyn"
used in the pedigrees should identify him as a Welshman. Our guess is he was related to the ruling family in Caerleon,
in whose realm Llanllowel is located and in whose service we find Iorwerth ap Caradog the father of Adam Gwent.
We can only guess as to
why the 15th century Earl of Pembroke did not want his ancestor listed as Adam Gwent, a nobleman of purely Welsh ancestry.
We even suspect Adam's ancestors represent a junior cadet of one of the families who ruled Gwent in the tenth and eleventh
centuries. But Sir William Herbert, as he styled himself, was an ardent Yorkist and must have preferred to claim
ancestry from an illustrous Norman family, not a "lowly" Welshman.
His genealogists thus did
a cut and paste job to accomplish his desire, ignoring chronology as they went. But their principal error was in bringing
Alis ferch Bleddyn into the pedigree to account for a Thomas ap Adam holding her manor of Llanllowel while failing to include
the family holding Beachley. While our construction cannot be guaranteed correct, it follows a stable timeline and places
both of Alis' manors in the hands of Adam Gwent whose descendants are known to have inherited them.
 Llyfr Baglan, pp 13
 LB 51
 LB 51, 79
 John B. Burke "Roll of the Battle Abbey", 1848 (1998 reprint, pp 55)
where the companion of William I is called Herbert, Count of Vermandois.
 Dwnn i, 293
 Cited as a landholder in Hampshire; since his entry immediately follows
land held by a "Herbert the Chamberlain", some would say they were father and son. That this conjecture is untrue can
be seen from the dates; if both held land in capite from King William I in 1086, both must date from before 1066
and if related at all, they are from the same generation. But nothing compels belief they were related.
 John of Hexham, Historical Works (ed by Thomas Arnold), 1882-85, vol
ii, pp 305-307
 LB 79, 215, 269; Dwnn i, 196
 LB 79; Dwnn i, 196, 292, 293
 LB 79, 215, 269, 323; Dwnn i, 196, 292, 293
 Archaeologia Cambrensis, 1858, pp 21
 Born c. 1085, Nest spent her childhood as a hostage at the court of William
Rufus and Henry I before marrying Gerald of Windsor. She is known to have had out-of-wedlock children with several important
men at court, including King Henry I. In 1109, she was abducted from her husband's bed by Owain ap Cadwgan of Powys "who was
moved by passion and love for the woman". Her only complaint to Owain was "if you want me for yourself, at least send
my husband's children back to him".
 Dwnn i, 196 may contain the identification of Peter's actual wife, Alice
daughter of John St. John, although it matches her with the earlier Peter. Burke's Dormant and Extinct Peerages
says Peter had a sister named Alice who married John St John, Lord of Basing born c. 1235. If that man had a base daughter born
c. 1255 before his marriage, she would fit chronologically with Peter.
 Limbus Patrum Morganiae et Glamorganiae, pp 250-251
 Dwnn ii, 42, 43
 LB 323
 Joseph Bradney "A History of Monmouthshire", vol 3, part 2, pp 218
 LB 215, 269, 323; Dwnn i, 293
 Archaeologia Cambrensis, 1858, pp 30
 see note 17
 Caradog ap Rhiwallon appears in several Llandaff Charters; in a charter
which scholars date to c. 1045, he described himself as "decomitibus" or "comes" to King Meurig (ap Hywel ap Owain ap Morgan
Hen). Text of the Book of Llan Dav, pp 261
 Sir John ab Adam was a baronet called to Parliament and to military service
many times between 1297 and 1307.
 Sir John had Beverstone from his mother and Beachley from his father.
 Jenkin was the younger son of Adam and received his mother's manor.
She was a daughter of Gwarin Ddu, whose pedigree can be found in LB 257 and Bradney's History of Monmouthshire, vol 1, part
2b, pp 360
 See note 17
 Llanllowell was located in the parish of Llantrisant Fawr in the hundred
of Usk, just north of Caerleon.
 Rhiwallon, the father of Caradog, could well have been a younger brother
of King Meurig ap Hywel. Not only does the chronology fit, but the nephew of a king could be expected to hold a
key role at his court...perhaps even his penteulu
The tradition that Nest ferch Rhys ap Tewdwr was
the mother of one of the early Norman Herberts (but not married to the father) is widely believed. The Dictionary of
Welsh Biography, 1959, page 683 cites a number of illegitimate sons she bore by liasons with not only King Henry I, but by
other highly-placed Norman men. That one of these was the Herbert ancestor who was married to Emma de Blois is
suggested by the fact that none
of the Herbert pedigrees found in Llyfr Baglan or in Dwnn's Visitations of Wales cite Emma de Blois as mother to any of the
early Herberts, but several do say Nest ferch Rhys ap Tewdwr was. One can never be sure whether the "marriage matches"
shown in medieval pedigrees were meant to indicate (a) that the lady was married to the man to whose name she is connected,
or (b) that she was the mother of the next following person in the pedigree, whether or not married to the father.
Our best guess is that Emma was
excluded from the pedigrees because she was not the mother of any cited son; not all the offspring of each generation
was included. According to the 12th century chronicler John of Hexham, Emma and Herbert of Winchester were the parents
of William, archbishop of York. On the other hand, Nest ferch Rhys ap Tewdwr may have been named because she
was the mother of a Herbert, not because she was actually married to his father.