Since this vexed subject came up in a recent thread, it may be worth
starting a new one focused on the problem.
According to Ademar of Chabannes, Vulgrin when already ageing was sent
by his close relative Charles the Bald to be count in Angoulême and
Périgueux after the death (in June 866) of Emeno, count of Angoulême,
resulting from local strife. Vulgrin by Ademar's account took with him
two sons, Hilduin and William, and once there claimed Agen owing to his
unnamed wife being the sister of William of Toulouse. He ruled the three
cities for 17 years ("Carolus ... Vulgrimnum propinquum suum ... direxit
in Aquitaniam, et prefecit eum comitem Egolisme simul et Petragorice ...
eratque jam senex quando eum Carolus Calvus fecit comitem supradictarum
urbium. Venerunt cum eo a Francia duo filii, Alduinus et Willelmus.
Aginnum quoque urbem habebat, quam assumens vindicavit propter sororem
Willelmi Tolosani, quam in matrimonium acceperat. Tenuit principatum in
his tribus civitatibus per XVII annos.")
The details here can't all be correct, since the circumstances do not
entirely fit with information stated or implied in sources closer to the
time and place, and generally more reliable than Ademar anyway.
For starters, it's scarcely credible that Charles would have sent a
brother-in-law of his deceased enemy William of Toulouse to govern in a
distant and troubled region. William was very probably count of Bordeaux
from 845 to 848, and in that capacity would have controlled Agen, but he
led opposition to Charles in Aquitaine favouring the king's nephew
Pippin II against him. William was almost certainly captured by Normans
in 848 and dispossessed, after which he briefly usurped power in
Ampurias and Barcelona before being killed in 849 or 850. His father
Bernard of Septimania had been executed in 844 and was hardly the late
father-in-law of a relative whom Charles would highly trust, since apart
from political differences he had been accused of adultery with the
king's mother empress Judith.
Also Vulgrin's wife was the mother of grown sons who were sent to
support her ageing husband in 866, whereas any sister of William of
Toulouse could not have been born before 842 and (since she was not
mentioned in his mother Dhuoda's manual written in the early 840s)
probably not until a few years later if such a sibling ever existed at
all. In that case, the onomastic link from her purported brother to
Vulgrin's son William is broken since he would necessarily have been son
of a prior wife and his inheritance of Agen along with Périgueux (while
Angoulême went to his brother Hilduin) becomes another problem.
Agen in 866 was presumably controlled by the count of Bordeaux at that
time, and as far as we can tell this was Arnald, duke of Gascony, son of
count Immo of Périgueux. The sole source for Arnald is an account of the
translation of relics of St Fausta in 864 from Vic-Fezensac to Brivezac,
where the journey ended on 15 June as recorded in the mid-12th century
martyrology of Solignac abbey here (folio 51v):
https://gallica.bnf.fr/ark:/12148/btv1b10032632p/f53 ("XVII Kl [julii]
... Briuaciaco translacio sanctę Faustę uirginis").
Since Arnald sent the expedition to collect St Fausta's relics after
succeeding his maternal uncle Sancho Sanchez Mitarra as ruler of the
Gascon duchy he must have been duke before 15 June 864, and since one of
the men he sent was his 'nepos' Godefrid he was himself evidently
already the uncle (if not the grandfather, but in any event the senior
relative) of a grown man, i.e. most probably not born in 840 as asserted
without evidence by Christian Settipani in 2004.
The next ruler of Gascony on record, perhaps Arnald's successor but
maybe his rival or sharer in partitioned rule over the Bordeaux-linked
duchy and an ethnic Gascon principality respectively, was evidently his
first cousin Sancho Sanchez, who was perhaps the man called 'king' as if
he ruled all Gascons in 867 by his father-in-law Galindo Aznárez of
Aragón ("ego Galindo Asinari comes deprecor Sancium regem generum
meum"). This identification is debatable, but at any rate Sancho Sanchez
ruled Gascony including Bordeaux not long after 864 - Arnald reportedly
met an untimely death between the translation of St Fausta in June 864
and the writing of the account apparently not long afterwards from which
we know of him. According to this, he wished to become a monk at
Solignac himself but was prevented by unexpected death ("spondens
voueret se in eodem pro Christi amore comam capitis sui depositurum, et
monasticis semetipsum subdendum disciplinis. Postea vero id implesset,
nisi inopinata morte praeuentus fuisset"). One of these men, either
Arnald or his successor in Bordeaux, must have ceded Agen to Vulgrin on
demand in 866. Given the precarious condition of the region under Viking
incursions this concession is not very surprising.
Solignac where Arnald wished to retire is about 10 kms south of Limoges,
making it difficult to be sure where he was count apart from Bordeaux.
His father is said to have been Immo, count of Périgueux, and whoever
this may have been he was almost certainly not the man identified as his
father by Settipani, Emeno of Angoulême who was killed in 866 before
Vulgrin's arrival. First, there is no sound evidence that this
Emeno/Immo was ever count in Périgueux, and secondly the chronology is
overstretched for him to have been father of Arnald who was duke of
Gascony by 864. Emeno of Angoulême had sons named Ademar (also count of
Angoulême, who married Vulgrin's granddaughter Sancia) and Adalelm. An
Arnald, evidently a third son, occurs in a charter dated June 923 as
brother of count Ademar ("Signum Arnaldi, fratri Ademaro comite" - the
editor glossed this as a copyist's error made in the 12th/13th century,
substituting the name Ademar for his successor in Angoulême, Guillem I
Taillefer, but since 'Ademaro' is not at all similar to 'Willelmo' this
is implausible). A man who was duke by the summer of 864 and who had met
an untimely death before he could enter a monastery not long after was
plainly not still living without a title in 923. He was also surely not
alive until 884, as asserted without citing evidence by Settipani in
2004, following Édouard de Saint Phalle in 2000 making him son of Emeno
of Angoulême by a first wife, born long before Ademar and Adalelm from a
purported second marriage.
There are a number of occurrences of counts named Immo/Emeno in the
region of Aquitaine from the late-8th century through the 9th. In 1969
Otto Gerhard Oexle wrily remarked that historians had tried to identify
these with as few individuals as possible, and the subsequent Saint
Phalle/Settipani proposal falls into this unwarranted pattern.
From an onomastics point of view the obvious place to look for the
count Immo of Périgueux whose son Arnald briefly became duke of Gascony
is a family in the region using also the name Godefrid that belonged to
duke Arnald's 'nepos'. In the comital family of Quercy (Cahors) both
Immo/Emeno and Godefrid are recorded, though not Arnald. An Immo was
appointed count in the pagus, not necessarily in the county, of Périgord
by Pippin the Short after he had campaigned through Aquitaine in 767.
This Immo evidently lived until 823, as he had kept a hostage taken in
767 at Turenne in servitude until the man was liberated and given back
his property by Louis I in November 823. Turenne is not in the county of
Périgord, and in any case we know that in 778 Charlemagne made Widbod
count of Périgord with other counts newly appointed in Bourges,
Poitiers, Auvergne, Velay, Toulouse, Bordeaux, Albi and Limoges at the
same time. No new count was named in Cahors, perhaps because Immo
remained in office there. He may have been the father, uncle or
grandfather of Rodulf who occurs as count from November 823, just when
Immo had evidently disappeared allowing for his unjustly-retained
hostage to seek freedom and recompense from the emperor after more than
50 years in servitude. Rodulf had sons named Godefrid and Immo/Emeno -
the latter, or an unknown namesake relative, possibly became count of
Périgueux and father of Arnald.
As for Ademar of Chabannes and 'William of Toulouse', he may have simply
guessed at the name on learning that Vulgrin's brother-in-law had been
the count controlling Agen in or before 866. If the correct name was
Arnald rather than William, Vulgrin's wife would have been the daughter
of Immo, count of Périgueux. However, this is merely conjecture:
Ferdinand Lot reasonably suggested and long maintained that Vulgrin's
wife was probably a Gascon lady who brought the name Sancia into his
family - this does not occur in the comital line of Quercy, although the
absence of a female name in the extant record is hardly strong evidence
that it was never used.
Peter Stewart