Etiennette of Longwy - NOT!!!!
Todd A. Farmerie
Prosopographical Research - Prosopographica et Genealogica,
vol.3, Onomastioque et Parente dans l'Occident medieval, K. S. B.
Keats-Rohan and C. Settipani, eds. (2000). I have not had a
chance to read beyond the first couple of articles, but even the
first one will be of interest to many in this group. Szabolcs de
Vajay here contributes a reevaluation of his elegant work on
Etiennette/Estefania/Stephanie, wife of William, Count of
Burgundy.
In a 1960 article appearing in Annales de Bourgogne, he evaluated
the surviving sources, and focussed particularly on two
contemporary ones that appeared to illuminate the question. One
of these indicated that Pope Callixte (son of Etiennette) was of
mixed parentage, one parent being Burgundian, and the other from
Lorraine. The second source mentioned a Countess Ermesende of
Longwy - taken to be the mother of Clemence, Countess of
Luxembourg, and wife of William VII of Aquitaine. Vajay was led
to conclude that these two women, apparently both of Lorraine but
with Mediteranean names, represented a single family group. At
about this time Longwy was held by an Adelbert, identified with
the Duke of Lorraine, who was then tapped as father. At the same
time, the names Ermesende and Etiennette appeared in the family
of Bernard Roger, Count of Foix (another sibling was Raimond, a
name which Etiennette of Burgundy would give a son), so the
mother of the Lorraine Ermesende and Etiennette was hypothesized
as another sibling in this family, provisionally named Clemence
(the name of daughters of both Ermesende and Etiennette). Thus
the question has stood for four decades (with only a few
historians dissenting). (Thus it has likewise appeared in
secondary sources which failed to indicate the hypothetical
nature of the solution, from which it has entered innumerable
personal and public databases.)
In his new analysis, entitled "Parlons encore d'Etiennette" (pp.
2-6), Vajay goes through his previous proof, and step by step he
destroys it.
The first source, describing the parentage of Callixte, is now
known not to refer to the son of Etiennette, but to her grandson
of the same name, who was son of Louis of Montbeliard and
Ermesende of Burgundy - parentage that matches his described
mixed origins. The other source can now be shown to refer not to
the mother of Clemence Countess of Luxembourg, but to her
daughter Ermesende who married Adelbert de Dabo - the Adelbert of
Longwy mistakenly associated with the Duke of Lorraine. Taken
together, there is no longer any support for a connection with
Lorraine (and no reason to think that Duke Adelbert married or
had any issue at all). More critically, it becomes clear that
with these two pieces of evidence removed from relevance, there
is no contemporary source that bears dierctly on the question.
One recent author, A. Beau, approached the question by returning
to the work of Pere Anselme, who wrote that Etiennette was
daughter of Raimond II of Barcelona and his wife Sancha of
Navarre. With minor modifications, this possibility is
attractive. The man Anselme appears to have had in mind was
Count Berenger I Raimond II, who married Sancha Sanchez of
Castile. This Berenger is now thought to have had a sister
Etiennette (for whom he might have named a daughter), while in
his fourth wife (who Beau would make mother of our Etiennette) we
have Guisla of Lluca, for whom Etiennette would have named her
daughter Gisela, and this solution would also explain the use of
Raimond.
Vajay is, however, not entirely enthusiastic about this
solution. First of all, Beau's reasoning that Anselme's account
was likely based on an accurate tradition/source is belied by the
fact that Beau then gives her a father Berenger and mother Guisla
(with Berenger's first wife, Sancha, being from Castile), while
Anselme had named them Raimond and Sancha of Navarre. Likewise,
this solution fails to explain the appearance of the name
Clemence (unknown among the Barcelona Counts, although Guisla's
ancestry is not fully characterized), while finally, there is the
problem of the relationship between Henry, Count of Portugal and
Raimond, Count of Galicia (Etiennette's son).
The word "congermanus" is used to describe the relationship
between Henry and Raymond. This word is usually used to refer to
what we would now call second cousins or even more distant
relations, the children (or descendants) of first cousins. By
Vajay's old solution, this would be the case, both being
grandchildren's-grandchildren of Count Roger of Carcassonne, with
Henry's mother postulated as daughter of Berenger and Guisla.
Moving Raimond's mother to the same family renders the two Counts
first cousins, which is too close. (Vajay would now instead
place Henry's mother as sister of Berenger, but the problem
remains.) This can not be simply dismissed by suggesting that
"congermanus" is being used somewhat atypically here to refer to
first cousins, since Raimond's sister Sybil married Henry's
brother Eudes and a marriage of first cousins (? once removed)
certainly would have been considered too close. In light of
this, Vajay tends to dismiss this solution as well.
He concludes by stating that the fate of his earlier theory
should serve as a warning about the uncertain nature of such
arguments based more on onomastics and hypothesis than direct
proof. This is perhaps a good warning with which to start this
book, as the articles that follow derive from a group of
researchers that uses just such methods to reconstruct
relationships among family groups.
What are we left with? - not much. Cut loose of the Lorraine
connection, we are now free to throw our net a little wider for
potential candidate families. This seach must now be based on
the two remaining useful bits of information - a kinship (but not
too close) to the Counts of Barcelona (there seems no
disagreement that the mother of Henry of Portugal derived from
this clan, as both of Henry's brothers, Hughes Borell and Eudes
Borell, were given the byname of Count Raimond Borell of
Barcelona, father of Berenger Raimond); and the use of some of
the names (i.e. Raimond, Clemence, Gisela) which first arise
among the Counts of Burgundy in the generation following the
marriage of Count William to the mysterious Etiennette. What we
appear to lack is any documentary evidence with which to compare
our hypothesis, in order to evaluate its likelyhood.
On a more practical level, we are left with a discredited theory
that has been so widely spread that, in spite of the author's own
disproof, it will continue to infest the uncritical literature
(and on-line sources) for decades (even centuries, perhaps) to
come. If anyone doubts this, just do a Google search for
"Berenger Sancha Sanchez", and see how many references marry her
to "Raimond Berenger" and not "Berenger Raimond". (Some of these
derive from an editorial error which appeared in Vajay's first
Etiennette manuscript, and was copied from there into Ancestral
Roots [and reprinted in later editions in spite of the error
being brought to the compiler's attention], and from there into
Royalty for Commoners. This, mind you, was not even a theory -
just an editorial error.) It is truly disheartening.
taf
Roger LeBlanc
able to follow the entire evaluation, but in light of the recent
discussion, I wonder if someone would clarify whether Etiennette is
yet known to be the sister of Ermesinde/Ermengarde, wife of William
VII of Poitou (ca 1023- ca 1058).
If yes, then how is that relationship as sisters known?
If not, then does the Longwy ancestry still hold up for Ermengarde?
Roger LeBlanc
Todd A. Farmerie
>
> I missed many of the postings on this thread, and perhaps was not
> able to follow the entire evaluation, but in light of the recent
> discussion, I wonder if someone would clarify whether Etiennette is
> yet known to be the sister of Ermesinde/Ermengarde, wife of William
> VII of Poitou (ca 1023- ca 1058).
> If yes, then how is that relationship as sisters known?
The evidence for sisterhood is directly linked to the proposed,
and now discredited Longwy ancestry.
> If not, then does the Longwy ancestry still hold up for Ermengarde?
The source thought to provide Longwy ancestry for Ermesinde can
now be shown to have refered to an entirely different woman, and
thus her Longwy connection fails, independent of whether or not
they were sisters.
What we know: They both lived at about the same time, they both
had southern names, and they both named a daughter Clemence.
taf