Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Richilde, countess of Hainaut - part 2a - relationships and onomastics

1,124 views
Skip to first unread message

Peter Stewart

unread,
Mar 13, 2023, 9:54:51 PM3/13/23
to
The name Richilde is not very helpful in narrowing down potential birth
families for the countess. This was most notably associated with Charles
the Bald's second wife, daughter of a count of Metz, whose only
descendants were through her own daughter married to a count named
Roger. Onomastics zealots may instinctively react "Bingo!" and mark
their cards accordingly, since Roger was the name of Richilde of
Hainaut's eldest son and she was said to have imperial blood. But of
course genealogy, like reality, does not work that way. The Roger
married to Empress Richilde's daughter was a count of Maine whose
dynasty cannot be shown to have used the name Roger ever again or that
of Richilde at all for certain, nor to have made any marriage
connections in the north-east of France. The name Richilde pops up
occasionally by the early 11th century in other families closer to
Hainaut, for example a countess of Blois whose descendants were counts
of Champagne (a younger son of one of them was successor to Richilde's
son Roger as bishop of Châlons-sur-Marne); and another Richilde of
unknown family origin married to Thierry I, duke of Upper Lorraine, with
descendants including counts of Bar, Mousson, Arlon etc, all
consanguineous with Herman of Hainaut via Hugo Capet through Thierry,
but no Rogers or Richildes on record before the time of interest.

The name Roger given to Richilde of Hainaut's eldest son, along with the
hereditary claim she and/or this son's father Herman had to
Valenciennes, led Henri Pirenne to speculate that she was a niece of
Arnulf of Cambrai, count of Valenciennes at the beginning of the 11th
century, who had a brother named Roger. The latter was proposed by
Pirenne as a possible father of Richilde, but since he was dead before
the end of June 983 he was certainly not the parent of a woman whose
last son was born ca 1055. However, Arnulf had another brother named
Reginar, and Platelle further suggested that this man may have been her
father - also perhaps the source of medieval confusion making her the
daughter of his namesake, her father-in-law Reginar V of Hainaut. This
scheme is hardly convincing from the chronology, since Arnulf's siblings
most probably belonged to the broad age-group of Richilde's grandparents
(unless she was born to a father in his 70s), but it has the advantage
of locating her family origin closer to the little else reported or
implied about her blood relatives. Pirenne pointed to the frequency of
the name Richilde in 11th-century charters from Hainaut as supporting
his conjecture (that he somewhat hopefully called a conclusion).

The name Roger occurs around the time of Richilde and her son in a few
families of comital rank in the region of the counts of Champagne that
included his bishopric of Châlons-sur-Marne: for example an archdeacon
in Reims and a count of Porcien in the Rethel family, and a count of
Bassigny and the Bolenois descended from Roger II of Laon who until the
early 940s had been count of Ostrevant, castellan of Douai and lay abbot
of Saint-Amand at Elnone, ruling territory close to Valenciennes (Michel
Bur suggested that the count of Porcien was also his descendant).
However, the remaining influence of the comital family of Laon would
more plausibly have directed relatives to the bishopric of Langres,
which incidentally became vacant in 1065 at almost exactly the same time
as Châlons-sur-Marne, than to the latter.

It is notable that Roger's two predecessors as bishop of
Châlons-sur-Marne were both also named Roger. It is possible that he was
given his name as a mark of candidacy for their office if he was lame
(as we are told by one source) from birth and closely related to one or
both of them. The choice of Châlons-sur-Marne would be a strange
coincidence if there was no family connection to these Rogers - if his
step-father Balduin of Flanders was behind a forced clerical career, as
represented in one source, then he could have had Roger placed just
about anywhere in France throughout the early 1060s since his own father
Balduin V was ruling the kingdom as 'procurator' in his capacity as
guardian of the young King Philippe I until 1066. The royal charter in
which Richilde's son Roger (III) first occurs as bishop is also the
first document in which Philippe explicitly announced his personal rule
after the end of Balduin V's regency. Bishop Roger II, immediate
predecessor of Richilde's son, had been one of the emissaries sent by
Philippe's father Henri I to Kiev seeking the marriage to his mother
Anna (aka Agnes) in 1049/50.

The name Agnes given to Richilde's daughter, presumably the otherwise
unnamed child of Herman said to have been consigned to a nunnery by her
step-father Balduin of Flanders, is not much more helpful. Roger was
probably born by 1036 and Agnes was apparently younger than him if she
was encloistered around 1051 and yet had the opportunity to leave and
possibly wish to marry by 1071, as mentioned before. The likelihood that
Agnes was Herman's daughter is indicated by her occurring in a charter
of her mother's son Arnulf of Flanders written after his father Balduin
VI's death on 17 July 1070. This was a donation for the souls of both
his father Balduin and his mother's prior husband Herman to Saint-Hubert
abbey in the Ardennes of allods in Huy (Ardenne) and Namur (Hesbaye),
which appear more likely to have come into his possession from
Richilde's own family or through her from the maternal inheritance of
Herman, and so either way concerning their daughter, than directly to
Arnulf through Balduin or by acquisition. In any case, King/Emperor
Heinrich III's wife from November 1043 was Agnes of Poitou and
Richilde's daughter may have been named in her honour if born ca 1044,
which would fit well enough with her not being a professed nun in the
early 1050s and still freely able to think of marrying in the early 1070s.

Richilde herself donated to Saint-Hubert in 1071 an estate at Chevigny,
less than 15kms south of the abbey, that was specifically said to come
from her patrimony but possibly may have come to her instead from the
property of Herman's mother. This lady was the daughter of Herman of
Verdun, margrave of Ename, from the dynasty of Ardenne. His agnatic
first cousin Gozelo (incidentally a son of his paternal uncle named
Reginar, and brother of a bishop of Laon) was count probably at Bastogne
around 32 kms north-east of Chevigny. Gozelo, who was living in the late
1020s, is known to have had a daughter who died as a recluse at
Saint-Hubert, and he was himself buried there. He had no other known
offspring, but of course that does not absolutely preclude the
possibility. However, the names Richilde and Roger cannot be associated
with him, and if he - as a first cousin of Herman of Hainaut's maternal
grandfather - had been closely related to Richilde it is not plain to
see why her marriage to her first husband would have been allowed.

I have run out of puff for the present, so will have to split this part
of my comments into two.

Peter Stewart

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

Peter Stewart

unread,
Mar 14, 2023, 12:26:54 AM3/14/23
to
On 14-Mar-23 12:54 PM, Peter Stewart wrote:

<snip>

> Richilde herself donated to Saint-Hubert in 1071 an estate at Chevigny,
> less than 15kms south of the abbey, that was specifically said to come
> from her patrimony but possibly may have come to her instead from the
> property of Herman's mother. This lady was the daughter of Herman of
> Verdun, margrave of Ename, from the dynasty of Ardenne. His agnatic
> first cousin Gozelo (incidentally a son of his paternal uncle named
> Reginar, and brother of a bishop of Laon) was count probably at Bastogne
> around 32 kms north-east of Chevigny. Gozelo, who was living in the late
> 1020s, is known to have had a daughter who died as a recluse at
> Saint-Hubert, and he was himself buried there. He had no other known
> offspring, but of course that does not absolutely preclude the
> possibility. However, the names Richilde and Roger cannot be associated
> with him, and if he - as a first cousin of Herman of Hainaut's maternal
> grandfather - had been closely related to Richilde it is not plain to
> see why her marriage to her first husband would have been allowed.

My attention was lagging at this stage - in fact it is precluded that
Gozelo had any other surviving offspring apart from the daughter who
became a recluse at Saint-Hubert: she was specifically called his only
daughter ("unica") and her entire patrimony was given by Emperor
Heinrich III to her father's second cousin Frederic of Luxemburg, duke
of Lower Lorraine (died 1065), who frequently visited her at Saint-Hubert.

Some blood relationship between Richilde and Gozelo more distant than
(grand)father/(grand)daughter is possible - for instance, he had a
brother named Bardo who had at least one son living in 985, but nothing
more is known about descendants. At any rate Chevigny evidently did not
come to Richilde by inheritance from the comital family whose territory
she had to pass through to get there on her way back to Mons from Rome
ca 1084, because Arnulf of Chiny chased and tried to capture her when
she passed through his land without permission, forcing her to take
refuge in Saint-Hubert abbey. Gislebert of Mons attributed this incident
to her daughter-in-law Ida of Louvain, but no-one says there was any
family connection between Arnulf and his quarry.

lancast...@gmail.com

unread,
Mar 14, 2023, 3:57:46 PM3/14/23
to
Peter apart from the so-called House of Ardenne, with all its Gozelons, the places you mention seem to indicate a centre of gravity with the princebishopric of Liège, and perhaps the pre-eminent family there in this period were the lords of Montaigu.

lancast...@gmail.com

unread,
Mar 14, 2023, 4:15:13 PM3/14/23
to
On Tuesday, March 14, 2023 at 8:57:46 PM UTC+1, lancast...@gmail.com wrote:
> Peter apart from the so-called House of Ardenne, with all its Gozelons, the places you mention seem to indicate a centre of gravity with the princebishopric of Liège, and perhaps the pre-eminent family there in this period were the lords of Montaigu.

Another question Peter. Do you see Arnulf of Cambrai as a different person to Arnulf of Valenciennes?

Peter Stewart

unread,
Mar 14, 2023, 6:01:04 PM3/14/23
to
On 15-Mar-23 6:57 AM, lancast...@gmail.com wrote:
> Peter apart from the so-called House of Ardenne, with all its Gozelons, the places you mention seem to indicate a centre of gravity with the princebishopric of Liège, and perhaps the pre-eminent family there in this period were the lords of Montaigu.

Richilde does not appear to have had very strong sway in the diocese of
Liège - she tried to get Bishop Wazo to arrest her first husband and
hand him over to the emperor, but he ignored her request.

When I can get round to continuing part 2 of the thread there will be
some evidence for Richilde's origin west of Liège, in the diocese of
Cambrai, and not necessarily from a pre-eminent family there.

Peter Stewart

unread,
Mar 14, 2023, 6:07:39 PM3/14/23
to
It seems more likely to me that this was one person - whether he was
connected to the earlier namesake counts who had contested with Roger of
Laon for control of Ostrevant, perhaps introducing his name into their
family through a reconciliatory marriage, is unknown. But there has to
be some limit to how many different counts named Arnulf a region could
produce in a century.

Peter Stewart

unread,
Mar 15, 2023, 5:16:36 AM3/15/23
to
On 15-Mar-23 9:01 AM, Peter Stewart wrote:
> On 15-Mar-23 6:57 AM, lancast...@gmail.com wrote:
>> Peter apart from the so-called House of Ardenne, with all its
>> Gozelons, the places you mention seem to indicate a centre of gravity
>> with the princebishopric of Liège, and perhaps the pre-eminent family
>> there in this period were the lords of Montaigu.
>
> Richilde does not appear to have had very strong sway in the diocese of
> Liège - she tried to get Bishop Wazo to arrest her first husband and
> hand him over to the emperor, but he ignored her request.

This noisome piece of Wikirot has been brought to my attention: "His
[Herman of Hainaut's] wife, who preferred an alliance with the emperor,
attempted to get Herman to imprison Wazo, Bishop of Liège, but he
refused", see https://en.wikipedia.org/wiki/Herman,_Count_of_Hainaut.
How on earth this could thwart Herman in his alliance against Heinrich
III, or help the emperor in line with Richilde's preference, is not
explained.

The opposite, of course, is true. The source for the episode, rather
than the Wikireversal of history, is Anselm of Liège's 'Gesta
episcoporum Leodiensium', written in the 1050s, stating (in a chapter
headed "How he [the bishop] was incited by the countess of Mons") that
Richilde sent a messenger enjoining him (Bishop Wazo of Liège) to come
with an armed force at a place and time where and when he might seize
her husband for handing over to the emperor ("mandans illi per nuntium,
ut cum armatis veniat, locum et tempus, ubi et quando maritum capiat, ut
imperatori tradat, denuntiat").

lancast...@gmail.com

unread,
Mar 15, 2023, 8:03:11 AM3/15/23
to
On Tuesday, March 14, 2023 at 11:07:39 PM UTC+1, Peter Stewart wrote:

> > Another question Peter. Do you see Arnulf of Cambrai as a different person to Arnulf of Valenciennes?
> It seems more likely to me that this was one person - whether he was
> connected to the earlier namesake counts who had contested with Roger of
> Laon for control of Ostrevant, perhaps introducing his name into their
> family through a reconciliatory marriage, is unknown. But there has to
> be some limit to how many different counts named Arnulf a region could
> produce in a century.

I agree. Arnulf of Valenciennes must be of interest here anyway because he appears in Flemish records in Gent while he clearly also held lands in both Brabant and what is now Limburg (see his mother's grants to Sint-Truiden). His widow was also holding Hanret. Also west of the imperial boundary he and his mother clearly had a presence in "Caribant" near Lille. There is also evidence that he held the castle of Visé near Liège which later seems to have been a castle associated with the Antwerp march. I think most of my notes on this can be found in my Loon article. Before the Verdun family took over he seems to have been the partner of their ancestor Godfrey the captive when it came to representing the empire on the frontier with Flanders. (His ancestry apart from his mother is on the other hand not something I've ever been able to find much evidence for apart from the various normal speculations.)

In any case he is the type of relative who might explain the later implied claims of Richild, although direct descent seems doubtful unless via a daughter. His son Adalbert predeceased him. As you know, it is typically presumed that the castellans of Valenciennes after him were also relatives.

Peter Stewart

unread,
Mar 15, 2023, 6:32:46 PM3/15/23
to
I think it likely that Hugo, Isaac and Emissa "the countess", incumbents
of the castellany of Valenciennes from the mid-11th to the mid-12th
century, may represent the kindred that Richilde and Herman bought off
for possession of Arnulf's countship/margraviate.

Platelle's notion that Richilde was perhaps the daughter of Arnulf's
brother Roger or the niece of both men is flawed chronologically yet may
point in the right direction. Descent from Arnulf seems to me less
plausible than a collateral link.

The purpose of setting out the possibility mentioned upthread of
Richilde's having inherited property she donated in Ardenne and the
Hesbaye is just to record indicators we can find, not to advocate for
her birth family's placement in that region east from Mons as I suspect
she originated from west of there - allods at Somal (in the county of
Huy) and Taviers (in Namur) may well have fallen into her possession
through the maternal inheritance of her first husband.

lancast...@gmail.com

unread,
Mar 16, 2023, 11:31:51 AM3/16/23
to
Arnulf's brother Roger must have died in the early 980s, which is presumably why you find him an unlikely parent. As explained in my Loon article I think Arnulf had brothers named Geveard and Herman. Another close relative was apparently Bishop Balderic II of Liège, and therefore presumably also the future counts of Loon.

Peter Stewart

unread,
Mar 16, 2023, 5:44:13 PM3/16/23
to
Not just an unlikely parent but an impossible one - Richilde had two
sons to her second husband Balduin of Flanders, whom she married in
1051, so that her father cannot have been a man recorded as dead by 29
June 983 - unless that record is false, in which case we have no
reliable evidence for the existence of Roger in the first place.

Richilde's birth cannot have been earlier than ca 1010 to allow for the
birth of her youngest son ca 1055, or later than ca 1020 if her eldest
son was born by 1036 as proposed upthread. Her first husband was born
after 1015, and it is likely enough that she was too.

> As explained in my Loon article I think Arnulf had brothers named Geveard and Herman. Another close relative was apparently Bishop Balderic II of Liège, and therefore presumably also the future counts of Loon.

The record of Arnulf of Valenciennes having a brother named Roger says
that on 29 June 983 Arnulf donated to Saint-Pierre abbey in Ghent for
his own soul and that of his deceased brother Roger. Another record from
Saint-Pierre abbey represents a Roger with brothers named Arnulf, Odo
and Rainer as making a donation on 2 October 983, by when Roger the
brother of Arnulf of Valenciennes reportedly had been dead for more than
3 months, and a pseudo-original charter forged ca 1035 dated 29 June 960
- presumably drawing on the transaction recorded with the same date in
983 - represents a Roger as donating to Saint-Pierre abbey maning five
of his brothers as subscribers, Odo, Hugo, Arnulf, Rainer and Robert.
Evidently the forger had not read your Loon article.

lancast...@gmail.com

unread,
Mar 17, 2023, 4:10:13 PM3/17/23
to
Yes Peter the question about the Arnulf-Roger brothers revolves around the question of the dating on the charters. Koch believed the dates to be falsified as per the various citations. I don't feel qualified to comment on that, but in any case many of them must have been from about the right time. I tend to think that there might be two sets of Arnulf-Roger brothers because (1) the titles were important, and (2) because of the evidence I laid out for Arnulf of Valenciennes having a whole different set of brothers. (A conclusion I derive mainly from the analysis of Bas Aarts.)

By the way Geveard/Gebhard was a fairly unusual name in this region so it is maybe worth pointing out that there is such a fellow mentioned by Alpertus of Metz. He ended up crossing the Lambert(Leuven)/Gerhard(Metz)/Balderic(Upladium) team successfully, but getting his comeuppance. And of course all our sources hate that team, but seem fascinated with them. It really is a fascinating group. I can't help imagining something like the old cowboy movies when In read Alpertus and Thietmar. Arnulf of Valenciennes was apparently on the "government" side when younger, with a silver star on his chest, but that probably does not tell us much about who he was related to.

Another family we can only see shadows of is that of Godizo the son of Richizo whose wife Gebhard married. There are many hints that the family feuds we know from this period, such as "Reginars" versus "Verduns" are only a pale reflection of a much more complex multi-family feud which may have put Game of thrones to shame (unless you need dragons for a good story).

Peter Stewart

unread,
Mar 17, 2023, 7:13:59 PM3/17/23
to
This speculation is based partly on a late-15th century version of a
confirmation by Otto I dated 24 January 966, reciting a list of
benefactions to Nivelles, in which several donors are accorded the title
'count' but Bertha the mother of Arnulf, Herman and Gerard or Gebhard
(all three brothers untitled) is not called countess ("Bertha cum filiis
suis Harnulfo, Hermanno, Girardo [in the MGH edition]/Giuardo [in the
'Oorkondenboek van Noord-Brabant' edition]"). Titles, as you say, were
important, not least to the imperial chancery. Nonethelss counts and
their wives sometimes did occur without stating their rank, even in
their own charters - but in imperial diplomatic, not so much

Another document in question is a confirmation by Thierry of Alsace,
count of Flanders, dated 1146 in which a donation by Coucnt Arnulf to
Sint-Truiden was witnessed by no less than six other counts, all titled
so, along with nameless others ("Huic traditioni facte ab Arnulfo comite
interfuerunt Eremfridus comes Hermannus comes Raynerus comes Rodulfus
comes Geueardus comes Rogerus comes et alii multi"). Arnulf's family
must have been rarely if not uniquely successful if he and two of his
brothers were all counts simultaneously, and yet oddly unfraternal
enough for his brothers to be separated among counts who were not
Arnulf's siblings on such a red-letter occasion with seven counts present.

In other words, I can't agree with the analysis of Aarts.

lancast...@gmail.com

unread,
Mar 18, 2023, 4:17:47 AM3/18/23
to
Thanks for those remarks Peter. Just to be clear, the way I understand it the witnesses you mention are normally understood to have been witnesses of the original 10th century grant, and not from the time of the much later confirmation. I don't think you are questioning that, but just to be sure.

Concerning your main point I have nothing conclusive to offer, but I think in the 10th century we do find families with many "counts" at the same time, such as for example the so-called Ardennes family and Arnulf was apparently of a similar rank.(He and Godfrey the captive were mentioned together several times and generally understood to have been assigned by the empire to be marcher lords confronting Flanders. This is also indicated by his connection to Visé.) As discussed elswhere it is difficult to know exactly how the term "count" was delimited in this period, but it does not seem to have had much to do with what we would now call counties. I think it was on the one hand a status term, but on the other hand the families of "consular" status, as they sometimes called it, were clearly being positioned all over the place into various advocacies, castles, and other jurisdictions complementing their clerical cousins and siblings who were at the peak in terms of territorial control, but weren't supposed to be killing people. (In this period eldest sons sometimes even appear to have been sent to the church, which medieval families supposedly never did.) This means that a whole set of brothers, often younger brothers, could end up being assigned with different "comital" offices at the same time, because all those offices and jurisdictions had not yet settled back down to being heritable (as always inevitably seems to happen throughout history). The great experiment of trying to run the empire via imperially selected bishops was of course particularly important in eastern Belgium.

Peter Stewart

unread,
Mar 18, 2023, 4:30:59 AM3/18/23
to
Of course they are supposed to be the original 10th century witnesses -
kindly refrain from patronising me or SGM readers generally with such
fatuous considerations.

> Concerning your main point I have nothing conclusive to offer, but I think in the 10th century we do find families with many "counts" at the same time, such as for example the so-called Ardennes family and Arnulf was apparently of a similar rank.(He and Godfrey the captive were mentioned together several times and generally understood to have been assigned by the empire to be marcher lords confronting Flanders. This is also indicated by his connection to Visé.) As discussed elswhere it is difficult to know exactly how the term "count" was delimited in this period, but it does not seem to have had much to do with what we would now call counties. I think it was on the one hand a status term, but on the other hand the families of "consular" status, as they sometimes called it, were clearly being positioned all over the place into various advocacies, castles, and other jurisdictions complementing their clerical cousins and siblings who were at the peak in terms of territorial control, but weren't supposed to be killing people. (In this period eldest sons sometimes even appear to have been sent to the church, which medieval families supposedly never did.) This means that a whole set of brothers, often younger brothers, could end up being assigned with different "comital" offices at the same time, because all those offices and jurisdictions had not yet settled back down to being heritable (as always inevitably seems to happen throughout history). The great experiment of trying to run the empire via imperially selected bishops was of course particularly important in eastern Belgium.

Produce examples of sets of three brothers all counts at the same time
and yet witnessing the senior brother's act with no distinction from
four other non-sibling counts among whom they are intermingled, and then
I will take your point seriously.

lancast...@gmail.com

unread,
Mar 18, 2023, 7:11:18 AM3/18/23
to
Peter if I understand correctly your point is not that there are no other examples of multiple brothers being counts, but that they would always appear next to each other in witness lists, and they would generally be noted as brothers, especially if the document involved the family inheritance. I am not sure how strict we can be about that, but you are certainly making a reasonable point. On the other hand we do not know who the other counts in this list are, and whether they also had a claim on the inheritance. (Roger may be another brother for example, and I guess it is likely that all or most of the group were all close relatives. That is how 20th century historians all seem to have interpreted it, but there has been a lot of speculation about them as you know. FWIW I think this Eremfried may well have been a count with local jurisdiction, because one with this name appears in the area a few decades later. Similarly Rodulf may well be the count from the Ardennes who that later record reports to have held Velm in the county of Eremfried.)

OTOH, the proposal that we can equate the two Berthas and their two eldest sons named Arnulf does not totally rely on any of the witnesses in Sint-Truiden. In the end though, the extra information the other Bertha could bring to the discussion is not very much. It is already very significant that we can connect Arnulf to acts made in Gent, places in Artois, and important imperial offices in Lotharingia. This shows that he must have had a very interesting family background (which we can now only guess at). Unfortunately the parentage of his mother Bertha proposed by Vanderkindere has to be rejected.

Peter Stewart

unread,
Mar 18, 2023, 7:33:54 PM3/18/23
to
The Sint-Truiden chronicle is not highly reliable, and it identifies
this Arnulf as count of Flanders rather than of Valenciennes, but it
states that his donation to the abbey was made at the request of his
mother Bertha while she was on her deathbed after visiting Aachen and
that she died on 16 July 967 (at odds with other information).

Even supposing that her illness had brought together seven counts at
Sint-Truiden, of whom at least three were her own sons, how then to
explain why Emperor Otto I in the year before had not deigned to call
this widowed great lady countess or any of her sons count?

In 967 Arnulf was evidently not yet the father of his only recorded son
Adalbert, who occurs three decades later, so that however many brothers
he may have had were his presumptive heirs with an equal interest in
carrying out their mother's dying wishes. The idea that in these
circumstances they would be named merely as present at his ceremonial
handover along with four other counts not stated to have any family
connection is a pretzel-stretch of credulity that Aarts may have
accomplished, but I won't try to emulate.

Just because some aspects of comital titulation may appear hazy to you
does not mean that these were equally murky to medieval observers or
that they might have scattered the title count around like confetti to
mark their way through the fog.


> OTOH, the proposal that we can equate the two Berthas and their two eldest sons named Arnulf does not totally rely on any of the witnesses in Sint-Truiden. In the end though, the extra information the other Bertha could bring to the discussion is not very much. It is already very significant that we can connect Arnulf to acts made in Gent, places in Artois, and important imperial offices in Lotharingia. This shows that he must have had a very interesting family background (which we can now only guess at). Unfortunately the parentage of his mother Bertha proposed by Vanderkindere has to be rejected.

The proposal that we can equate Bertha the mother of Arnulf of
Valenciennes with the donor to Nivelles in Otto I's confirmation is
practically baseless without tacking onto it the tendentious proposal of
Aarts. If you think otherwise, why not elucidate this rather than just
tossing it into a post without detail?

The names Bertha and Arnulf were far too common to conclude that every
mother/son pair must be the same people, and as pointed out before if
the second son named by Otto I was Girard as the MGH editor read it
rather than the peculiar form Givard then Aarts is short of one corner
for his triangulation that was implausible anyway.

If you want to have a discussion about Vanderkindere's unacceptable
proposal of Bertha's parentage, it would be courteous to SGM readers to
start a new thread and specify this in more detail than just to say it
"has to be rejected" - not everyone here has ready access to, or hangs
on, everything he published.

lancast...@gmail.com

unread,
Mar 19, 2023, 5:47:06 AM3/19/23
to
Hi Peter, Although I don't really I agree with how strongly you describe the doubts, I'm happy to file the Nivelles proposal of Bas Aarts under "uncertain". (I think he would too.) Perhaps I will indeed write a quick explanation about Bertha, as proposed. However I am interested to check a few points.

1. Can you explain what you are referring to with these words? "Even supposing that her illness had brought together seven counts at Sint-Truiden, of whom at least three were her own sons, how then to explain why Emperor Otto I in the year before had not deigned to call this widowed great lady countess or any of her sons count?" Is there a specific record from the year before that you have in mind?

2. You write: "The Sint-Truiden chronicle is not highly reliable, and it identifies this Arnulf as count of Flanders rather than of Valenciennes". The chronicle is really at least 4 different works. IMOH there has been a problem of historians calling it unreliable for the bits that they don't like, but then relying on other parts of it. So we need to look at the details. Granted, this record is in a 14th century part, but it concerns an important grant which was still being commemorated and the remarks make it clear that the abbey still had documentation. When it comes to grants, the chronicle is full of details and quite careful. I think the most important concern about those parts is that the abbey was of course trying to promote its own side, but I don't see that as a big issue for us in this case? Perhaps most importantly I think the error you mention is not an error. Other evidence confirms that Arnulf and Bertha were apparently "Flemish" in important ways. (On this point Bas Aarts originally seemed to accept something like the same reasoning which you are following. He was even more cautious and even doubted that the 967 Count Arnulf was Arnulf of Valenciennes. I think this was because he was looking at the argumentation of Vanderkindere and Dhondt as presented by Baerten, which is confusing on this point.)

In fact, the reason that we also know about this grant from a later confirmation by the count of Flanders is because the grant involved lands in Provin, which is near Lille, ie "Flanders" (or more correctly Artois). This was no one off. Count Arnulf (and his wife and son, who you claim is only known from much later) appear in Gent records in the 980s, giving grants of lands in exactly the same pagus of Caribant. So how can we claim that Arnulf and his mother are known not to be Flemish? I think we can't. By the way there is one medieval record giving a hint about Bertha's ancestry and everyone seems to ignore it. The same 14th century continuation of the Sint-Truiden chronicle gives the actual lines of praise which were still being sung for their benefactor, and they seem to suggest that she had royal ancestry. This could indicate many things, but given the period, and her regional affiliations, the first thing which comes to mind is Carolingian connections such as claimed by the counts of Flanders, or the Ardennes family. Indeed there were still "real" Carolingians in France, and they were apparently holding southern parts of Flanders in the mid 10th century.



lancast...@gmail.com

unread,
Mar 19, 2023, 6:39:37 AM3/19/23
to
A third question. "The Sint-Truiden chronicle [...] states that [...] that she died on 16 July 967 (at odds with other information)." Which other information is it at odds with?

Peter Stewart

unread,
Mar 19, 2023, 6:28:01 PM3/19/23
to
The confirmation by Otto I dated 24 January 966, as discussed in the
text of this thread copied above.

>
> 2. You write: "The Sint-Truiden chronicle is not highly reliable, and it identifies this Arnulf as count of Flanders rather than of Valenciennes". The chronicle is really at least 4 different works. IMOH there has been a problem of historians calling it unreliable for the bits that they don't like, but then relying on other parts of it. So we need to look at the details. Granted, this record is in a 14th century part, but it concerns an important grant which was still being commemorated and the remarks make it clear that the abbey still had documentation. When it comes to grants, the chronicle is full of details and quite careful. I think the most important concern about those parts is that the abbey was of course trying to promote its own side, but I don't see that as a big issue for us in this case? Perhaps most importantly I think the error you mention is not an error. Other evidence confirms that Arnulf and Bertha were apparently "Flemish" in important ways. (On this point Bas Aarts originally seemed to accept something like the same reasoning which you are following. He was even more cautious and even doubted that the 967 Count Arnulf was Arnulf of Valenciennes. I think this was because he was looking at the argumentation of Vanderkindere and Dhondt as presented by Baerten, which is confusing on this point.)

Being Flemish in a general sense, or even being descended from the
counts of Flanders specifically, is not the same as being "comitissa
Flandrie" as the Sint-Truiden continuator called Bertha. This means
straightforwardly countess "of Flanders", not "in the Flanders area".

> In fact, the reason that we also know about this grant from a later confirmation by the count of Flanders is because the grant involved lands in Provin, which is near Lille, ie "Flanders" (or more correctly Artois). This was no one off. Count Arnulf (and his wife and son, who you claim is only known from much later) appear in Gent records in the 980s, giving grants of lands in exactly the same pagus of Caribant. So how can we claim that Arnulf and his mother are known not to be Flemish? I think we can't. By the way there is one medieval record giving a hint about Bertha's ancestry and everyone seems to ignore it. The same 14th century continuation of the Sint-Truiden chronicle gives the actual lines of praise which were still being sung for their benefactor, and they seem to suggest that she had royal ancestry. This could indicate many things, but given the period, and her regional affiliations, the first thing which comes to mind is Carolingian connections such as claimed by the counts of Flanders, or the Ardennes family. Indeed there were still "real" Carolingians in France, and they were apparently holding southern parts of Flanders in the mid 10th century.

The epitaph given for Bertha by the continuator was written after her
remains had been moved to a different tomb under abbot Adalard II, who
died in December 1082. The third line of this says "Stemma prefulsit ei
regalis progeniei", i.e. she rejoiced in a royal pedigree. Given her
name, general location and the choice of "royal" as opposed to recasting
the line to describe her ancestry as "imperial" fitting the metre,
suggests that by the late 14th century she may have been remembered as a
Carolingian descendant through the illegitimate Vermandois lineage,
traced from a king of Italy, rather than legitimately from any emperor
including Charlemagne himself. But if so that is just a vestigial
implication, not particular evidence.

I hope Hans Vogels is not upset by my posing as too knowledgeable.

Peter Stewart

unread,
Mar 19, 2023, 6:36:02 PM3/19/23
to
On Sunday, 19 March 2023 at 9:39:37 pm UTC+11, lancast...@gmail.com wrote:

<snip>

> A third question. "The Sint-Truiden chronicle [...] states that [...] that she died on 16 July 967 (at odds with other information)." Which other information is it at odds with?

The obituary of Saint-Lambert de Liège places the death of countess Bertha on 30 October, and Ruffini-Romzani and others place her death after 967 (though I can't at present recall or check why so).

There you go Hans, something I don't know.

Peter Stewart

lancast...@gmail.com

unread,
Mar 19, 2023, 6:36:59 PM3/19/23
to
Peter, your write "This means straightforwardly countess "of Flanders", not "in the Flanders area"." Could it not mean "a countess of Flanders"?

lancast...@gmail.com

unread,
Mar 19, 2023, 6:40:28 PM3/19/23
to
On Sunday, March 19, 2023 at 11:36:02 PM UTC+1, Peter Stewart wrote:
> On Sunday, 19 March 2023 at 9:39:37 pm UTC+11, lancast...@gmail.com wrote:
>
> <snip>
> > A third question. "The Sint-Truiden chronicle [...] states that [...] that she died on 16 July 967 (at odds with other information)." Which other information is it at odds with?
> The obituary of Saint-Lambert de Liège places the death of countess Bertha on 30 October, and Ruffini-Romzani and others place her death after 967 (though I can't at present recall or check why so).

Thanks for that reference. There are standard Vanderkindere/Baerten reasons for saying the date is wrong, and indeed Bas Aarts also accepted those in the older articles I mentioned. However most of them they are based on various assumptions about the witness list. However it does not ring a bell with me that Ruffini-Ronzani made any remarks on this.

Peter Stewart

unread,
Mar 19, 2023, 6:46:10 PM3/19/23
to
Not in remarks, but in a table - without references, hence my lack of
further information without checking that I can't do just now. See here:
https://journals.openedition.org/trajectoires/2272.

Peter Stewart

unread,
Mar 19, 2023, 6:53:21 PM3/19/23
to
In theory it could, but then the same continuator later in the passage
referred to her son Arnulf as "Arnulfo illustri viro, filio suo,
Flandrensi comiti". The author apparently did not know that the
countship ought to be Valenciennes, though that is not surprising as
earlier documentation available at Sint-Truiden would most probably
would not have specified where they had lived. Designating counts by
territory was unusual in the 10th century, and even more so for their
widowed countesses.

Peter Stewart

unread,
Mar 19, 2023, 11:39:59 PM3/19/23
to
On 19-Mar-23 8:47 PM, lancast...@gmail.com wrote:

<snip>

> Count Arnulf (and his wife and son, who you claim is only known from much later) appear in Gent records in the 980s, giving grants of lands in exactly the same pagus of Caribant.

Can you please post citations for these 980s Gent records that include
Arnulf with his wife and son?

There is a single charter of Arnulf for Saint-Pierre abbey dated 29 June
983, already discussed in this thread, donating an inherited estate in
the pagus of Caribant for the soul of his deceased brother Roger - but
no wife or son of Arnulf are mentioned in it.

Apart from that I can only find two charters of Arnulf with his wife
Lietgard and son Adalbert, dated 1 January 994 and 30 September 998
respectively, which is why I wrote that Adalbert does not occur until
three decades after the 960s. He certainly does not appear among the
crowd of counts who were drawn like moths to the dying flame of his
grandmother's sickbed at Sint-Truiden in 967.

lancast...@gmail.com

unread,
Mar 20, 2023, 1:56:43 AM3/20/23
to
Those are indeed the charters. You were therefore probably correct that Adalbert does not appear until roughly 3 decades later. The only point I'd add is that Koch believed the dates to be wrong. Quibbling, he appears between 2 and 3 decades after 967.

mike davis

unread,
Mar 20, 2023, 9:26:28 PM3/20/23
to
On Tuesday, March 14, 2023 at 1:54:51 AM UTC, Peter Stewart wrote:
> The name Richilde is not very helpful in narrowing down potential birth
> families for the countess. This was most notably associated with Charles
> the Bald's second wife, daughter of a count of Metz, whose only
> descendants were through her own daughter married to a count named
> Roger. Onomastics zealots may instinctively react "Bingo!" and mark
> their cards accordingly, since Roger was the name of Richilde of
> Hainaut's eldest son and she was said to have imperial blood. But of
> course genealogy, like reality, does not work that way. The Roger
> married to Empress Richilde's daughter was a count of Maine whose
> dynasty cannot be shown to have used the name Roger ever again or that
> of Richilde at all for certain, nor to have made any marriage
> connections in the north-east of France. The name Richilde pops up
> occasionally by the early 11th century in other families closer to
> Hainaut, for example a countess of Blois whose descendants were counts
> of Champagne (a younger son of one of them was successor to Richilde's
> son Roger as bishop of Châlons-sur-Marne); and another Richilde of
> unknown family origin married to Thierry I, duke of Upper Lorraine, with
> descendants including counts of Bar, Mousson, Arlon etc, all
> consanguineous with Herman of Hainaut via Hugo Capet through Thierry,
> but no Rogers or Richildes on record before the time of interest.
>
> The name Roger given to Richilde of Hainaut's eldest son, along with the
> hereditary claim she and/or this son's father Herman had to
> Valenciennes, led Henri Pirenne to speculate that she was a niece of
> Arnulf of Cambrai, count of Valenciennes at the beginning of the 11th
> century, who had a brother named Roger. The latter was proposed by
> Pirenne as a possible father of Richilde, but since he was dead before
> the end of June 983 he was certainly not the parent of a woman whose
> last son was born ca 1055. However, Arnulf had another brother named
> Reginar, and Platelle further suggested that this man may have been her
> father - also perhaps the source of medieval confusion making her the
> daughter of his namesake, her father-in-law Reginar V of Hainaut. This
> scheme is hardly convincing from the chronology, since Arnulf's siblings
> most probably belonged to the broad age-group of Richilde's grandparents
> (unless she was born to a father in his 70s), but it has the advantage
> of locating her family origin closer to the little else reported or
> implied about her blood relatives. Pirenne pointed to the frequency of
> the name Richilde in 11th-century charters from Hainaut as supporting
> his conjecture (that he somewhat hopefully called a conclusion).
>

I read that Gilles D'Orval writing c1250, said that her father was Reginar
son of Reginar. I dont have a ref for this but it seems he was cited
by Van Droogenbroeck work which I havnt seen either. I cant believe that
GD meant by this reginar V son of Reginar IV, or else she would have married
her own brother, clearly ridiculous. But what if this indicates that her father
was Reginar son of Reginar the brother of Arnulf of Valenciennes?

Something like this:

1 Reginar [983? bro of Arnulf of Valenciennes d1011/12]
2 Reginar son of Reginar
3 Richilde [c1020-1086] dau of Reginar son Reginar

I dont think the chronology is such a stretch.

I havnt looked at how to square this with the reference in Flandria
generosa that she was the neice/neptis of Pope Leo or he was
her uncle until i've read your part2b.


snip
>
> The name Agnes given to Richilde's daughter, presumably the otherwise
> unnamed child of Herman said to have been consigned to a nunnery by her
> step-father Balduin of Flanders, is not much more helpful. Roger was
> probably born by 1036 and Agnes was apparently younger than him if she
> was encloistered around 1051 and yet had the opportunity to leave and
> possibly wish to marry by 1071, as mentioned before. The likelihood that
> Agnes was Herman's daughter is indicated by her occurring in a charter
> of her mother's son Arnulf of Flanders written after his father Balduin
> VI's death on 17 July 1070. This was a donation for the souls of both
> his father Balduin and his mother's prior husband Herman to Saint-Hubert
> abbey in the Ardennes of allods in Huy (Ardenne) and Namur (Hesbaye),
> which appear more likely to have come into his possession from
> Richilde's own family or through her from the maternal inheritance of
> Herman, and so either way concerning their daughter, than directly to
> Arnulf through Balduin or by acquisition. In any case, King/Emperor
> Heinrich III's wife from November 1043 was Agnes of Poitou and
> Richilde's daughter may have been named in her honour if born ca 1044,
> which would fit well enough with her not being a professed nun in the
> early 1050s and still freely able to think of marrying in the early 1070s.
>

French wiki has Roger and the nun Gertrude from her first marriage
and Alix, Arnulf III Baldwin II and Agnes from her 2nd. It says the
2nd marriage was annuled and they were excommunicated, but
obtained special dispensation from Leo IX. Theres no ref, but
Van Droogenbroeck work [which youve already trodden on] is
cited at the end. You have already cited the Flandria Generosa
as saying they excommunicated by their bishop, but does the
other info also come from this source or is it just assumed?

snip

> Richilde herself donated to Saint-Hubert in 1071 an estate at Chevigny,
> less than 15kms south of the abbey, that was specifically said to come
> from her patrimony but possibly may have come to her instead from the
> property of Herman's mother. This lady was the daughter of Herman of
> Verdun, margrave of Ename, from the dynasty of Ardenne. His agnatic

In many places on french sites, Richildes mother is called Matilda of
Verdun! I see that Pope Leo is occasionally 'given' a sister called
Matilda who was the wife of Richwin count of Scarponne 1019-43.

Mike

Hans Vogels

unread,
Mar 21, 2023, 2:33:10 AM3/21/23
to
Op dinsdag 21 maart 2023 om 02:26:28 UTC+1 schreef mike davis:
[knip]
> I read that Gilles D'Orval writing c1250, said that her father was Reginar
> son of Reginar. I dont have a ref for this but it seems he was cited
> by Van Droogenbroeck work which I havnt seen either.

The publications of Frans Van Droogenbroeck can be downloaded from Academia.
https://trismegistos.academia.edu/FransVanDroogenbroeck

>[knip]
> Mike

With regards,
Hans Vogels

Peter Stewart

unread,
Apr 6, 2023, 7:01:28 PM4/6/23
to
The passage in question is here (pp. 79-80):
https://www.dmgh.de/mgh_ss_25/index.htm#page/79/mode/2up.
It says that Countess Richilde of Hainaut was the daughter of Count
Reginar son of Count Reginar Long-neck, and that she was the sole
heiress of the county of Hainaut ("Fuit autem ista Richildis comitissa
Haynonie filia Raineri comitis, filii comitis Raineri Longi colli. Hanc
autem, quia unica erat heres comitatus Haynonie ..."). Obviously Gilles
d'Orval did not understand that Herman was the sole heir as son of
Reginar the son of Reginar Long-neck, so he was not in any way
suggesting an incestuous marriage but just making a wrong assumption
from Richilde's holding Hainaut in her own right after Herman's death
when she married Balduin of Flanders.

A decade or so earlier in the 13th century the same mistake had been
made by Aubry of Troisfontaines. Today we see that journalists with far
better resources for gathering and checking information make such
mistakes and misinterpretations, but some readers have an irrational
difficulty in accepting that medieval writers ever made plain and
complete blunders. Trying to minimise the error in this case - as in all
others - is a pointless exercise that only raises new difficulties. Why
for starters would an eldest son with a father named Herman and two
grandfathers both named Reginar be given the dynastically unprecedented
name Roger?
This has been covered upthread - there is no medieval evidence known to
me that Richilde had a daughter named Gertrude and the recorded daughter
Agnes (not Alix) may have been from her first marriage, while the
characterisation of Leo IX's reported off-hand or at any rate (under
canon law) peculiar response when hearing of Richilde's second marriage
as a "special dispensation" is inventive to say the least. This is
recounted in 'Flandria generosa' here (p. 320, lines 30-35, misnaming
the bishop who had allegedly excommunicated the couple):
https://www.dmgh.de/mgh_ss_9/index.htm#page/320/mode/1up.

> snip
>
>> Richilde herself donated to Saint-Hubert in 1071 an estate at Chevigny,
>> less than 15kms south of the abbey, that was specifically said to come
>> from her patrimony but possibly may have come to her instead from the
>> property of Herman's mother. This lady was the daughter of Herman of
>> Verdun, margrave of Ename, from the dynasty of Ardenne. His agnatic
>
> In many places on french sites, Richildes mother is called Matilda of
> Verdun! I see that Pope Leo is occasionally 'given' a sister called
> Matilda who was the wife of Richwin count of Scarponne 1019-43.

The attribution of Matilda of Verdun as mother (rather than correctly
mother-in-law) of Richilde comes from clinging onto a remnant of the
mistake by Gilles d'Orval and Aubry of Troisfontaines as above. The
putative sister of Leo IX who married a count in the Charpeigne (of
Scarpone or Montbéliard/Mömpelgard) was named Hildegard not Mathilde,
and this alleged relationship was one of several refuted by Frank Legl
as discussed upthread. The supposed connection as either 'neptis' or
'soror' of Richilde to Leo should not be taken in isolation from the
other legendary sisters and nieces ascribed to the pope as a result of
imaginative interpretation of his visits to them, which may be have been
due to political support for the emperor rather than selective family
reunions.

Peter Stewart

Peter Stewart

unread,
Apr 6, 2023, 7:27:53 PM4/6/23
to
Apologies for my lapse of memory - Frank Legl deduced that Hildegard the
wife of a count in the Charpeigne was probably a sister of Leo (though
this is not directly evidenced), and she was not among the alleged
relatives visited by the pope.
0 new messages