An imaginary Beatrice of Rethel, purported sister of Balduin II of Jerusalem

[Peter Stewart]

The existence and marriage for this person in Genealogics

https://www.genealogics.org/getperson.php?personID=I00139824&tree=LEO

are drawn only from *The Rupenides, Hethumides and Lusignans: the
structure of the Armeno-Cilician dynasties* by W. H. Rüdt-Collenberg
(1963). The same authority has been accepted for the same details in
several academic works, most notably in Gérard Dédéyan's *Les Arméniens
entre Grecs, musulmans et croisés* (2003).

This is unfortunate, as Rüdt-Collenberg's book is not only extremely
ill-organised overall but in this particular instance a hot mess of
incomprehension, error, misprinting and bogus references.

There was most probably no such person as Balduin of Bourcq's alleged
sister named Beatrice, and his actual sister named Cecilia was most
probably never married to Leo I, lord of the (Cilician) Mountains.

The source relied on by Rüdt-Collenberg - and in turn by Dédéyan - is
Orderic Vitalis, who very clearly did not know what he was talking about
when he wrote that Leo was son of Thoros of the Mountains (absurdly
given the Norman name Turold) and uncle to the wife of Bohemond II of
Antioch ('Prefatus Leo [Armenicus] Turoldi de Montanis erat filius, et
uxoris Buamundi auunculus'). Bohemond's wife was indeed a daughter of
Balduin II of Jerusalem, but Leo was not the son of Thoros and Orderic
cannot in any case be seriously taken as using "avunculus" for an
uncle-by-marriage. Moreover, the name Beatrice for his purported wife
(Balduin's alleged sister) is imaginary.

William of Tyre named the siblings of Balduin II as brothers Manasses
and Gervais (wrongly stating that Balduin was the eldest of them) and
sisters Mathilda and Hodierna, whose husbands were the lords of Vitry
and Hierges respectively. In the same passage William disdained the
study of princely genealogies, and obviously did not trouble himself
over completeness since elsewhere he named Roger of Hauteville, regent
of Antioch (killed 1119), as a brother-in-law of Balduin. We know from
her own charter dated 1126, for a donation with the consent of her late
husband's cousin Bohemond II of Antioch, that Roger's widow was named
Cecilia, that she was lady of Tarsus and that she was Balduin II's
sister ('Ego Cecilia domina Tharsensis, scilicet soror regis
Ierosolimorum ... concessu d[omi]ni Boamundi Antiocheni principis').

The possession of Tarsus by Cecilia in 1126 had been part of an old
controversy about whether it belonged to Antioch or to the Armenian
Cilician lordship - leaving aside this charter of Cecilia the question
might appear more open than it really is, but if Leo I of the Mountains
had been the lord of Tarsus (that is on the Cilician plain, not in the
mountain region ruled by his elder brother Thoros in 1126) he would
surely have been mentioned in his wife's charter given with the consent
of another suzerain. The obvious conclusion, that Tarsus was her dower
as widow of a ruler of Antioch, was persuasively argued by Valeriy
Stepanenko in several articles in *Vizantiyskiy vremennik* in the 1980s
and 90s.

There is no good reason to conflate the real Cecilia with the fictitious
Beatrice. Rüdt-Collenberg made her into the mother of at least two of
Leo's sons, consequently making the subsequent Armenian royal lineage in
Cilicia into her descendants, but he did not even understand the short
and simple text of Orderic quoted above, which he misrepresented in
quotation marks as indicating '"Only Leo, and _not_ his brother Thoros
uncle of Baudouin's II [sic] daughter Alice"'.

He went on to claim wrongly that 'Hethum and Vahran ... say that [Leo's
sons] Mleh, Stephané (and Constantine?) went to their _maternal_ uncle
or their _maternal_ relatives in Edessa. Jocelin's I [sic] mother {line
missing} that Mleh, Stephané and probably Constantine were the offspring
of the "Rethel" Lady ...'. In fact, Varham of Edessa did write that Mleh
and Stephané in 1137 went to their maternal uncle (քեռի) the count of
Edessa, but Hethum the Historian did not say anything about this. The
count of Edessa in 1137 was Joscelin II, whose mother was a sister of
Leo I of the Mountains so that he was a first cousin, not maternal
uncle, to the latter's sons.

I would guess that the missing line in Rüdt-Collenberg's sloppy passage
(incidentally, he lavished praise on the book's printers in his
acknowledgements) may have been something like: 'Jocelin I's mother {was
a sister of Hugue de Rethel's wife, suggesting} that Mleh, Stephané and
probably Constantine were the offspring of the "Rethel" Lady ...' which
would barely help even if he had been right about Hethum supposedly
describing 'maternal relatives'.

In the following annotation Rüdt-Collenberg asserted of Balduin I's
Armenian wife Arda that 'Jean de Vitry [sic, he meant Jacques de Vitry]
confounds her with [Balduin II's Armenian wife] Morfia'. I am unable to
find any mention of either lady in Jacques de Vitry's *Historia
orientalis*, though possibly he mentioned and conflated them in one of
his letters that I have not checked. However, this does suggest a way
that Orderic might have come by his misunderstanding that Balduin II's
daughter was a niece of Leo I - he wrongly thought that Leo's father was
Thoros of the Mountains, and perhaps he had the latter confused with
Thoros the governor of Edessa who at one point adopted Balduin I as his
son. If Orderic had confused two sets of namesakes, he may have meant
that Leo was an adoptive uncle of a daughter of the wrong Balduin. In
any event, Dédéyan's explanation in favour of Rüdt-Collenberg's
proposal, based on his analysis of geo-politics, is unconvincing.

Peter Stewart