If you carefully evaluate the evidences of Gundrada’s parentage that have been previously discussed in this ten year old thread; the two charters of Lewes Priory, Gundrada’s epitaph, Orderic Vitalis and the Liber Hyda, you’ll immediately see these evidences point in difference directions. As you pointed out you can’t assume one author knew more than another, both may have flawed conclusions. This is very true unless you analyze these to see what each proves; or fails to prove. This is how to go about resolving the conflicting evidence, which is the forth element of the Genealogical Proof Standard.
Earlier in this thread it was pointed out (correctly) that one of the charters had been altered by a later hand; inserting information to indicate she was a daughter of William the Conqueror. The second, the confirmation charter of Lewes Priory itself has been proved a later forgery. It included an allusion to Gundrada being the daughter of Matilda of Flanders. As they are both spurious neither can be relied upon as evidence of her parentage.
Regarding the inscription on Gundrada’s black marble grave slab, the first two lines (in Latin) read:
The key two words here being ‘stirps’ (of the line of ) and ‘ducum’. The latter word has been selectively translated as meaning dukes, however the word ‘ducum’ does not exclusively mean duke. A ‘duc’ is a military commander, a battle leader or general. While this certainly includes dukes and counts, it can also be applied to any line of commanders or leaders of note. The inscription includes the word ‘nobile’ which translates to noble so this points to any of a considerable number of noble commanders—dukes, counts, viscounts, generals, barons and advocates, among others. It would have to be a line of such, but there is nothing in this inscription or its wording to point to any particular line or family. Any assumption that it does is simply that, an assumption.
Next we have Orderic Vitalis, the English chronicler who was himself a Norman. In his ‘Historia Ecclesiastica’ he states clearly that Gundrada, the wife of William de Warenne was the sister of Gerbod the Fleming, Earl of Chester. This is corroborated by the Liber Hyda (the chronicle of Hyde Abbey). Both are near-contemporary sources.
So far Orderc and the Liber Hyda are the only solid evidence that survives any scrutiny. Her brother, Gerbod, the Earl of Chester has been shown by other sources to have been the son of another Gerbod, of Saint-Omer, the hereditary advocate of St. Bertin. E. Warlop shows this family in his ‘The Flemish nobility before 1300’. The ambiguity of the term ‘stirps ducum’ is that it applies as easily to a line of advocates as to a line of dukes or counts. An advocate of an abbey was typically a great baron tasked to represent and defend the abbey and to lead ecclesiastical knights and levies in battle. And the office was hereditary, passed down from father to son(s). In fact, according to an article by Elizabeth Van Houts, ‘Frederick, Brother-In-Law of William of Warenne,’ (published in Anglo-Saxon England, vol 28, 1999) on p. 218 she wrote: Charter evidence shows that the advocacy of St. Bertin stayed with both Warenne sons at least until the late 1090s and it links the Warenne family to land at Roquetoire near Saint-Omer until the late twelfth century.” Gundrada’s two sons, William II and Rainald, were also advocates of St. Bertin following in the footsteps of their grandfather and maternal uncles. This is further compelling evidence their mother was of this family.
There are also good indicators as to who her parents were not. If Gundrada had been a daughter of William the Conqueror, her marriage to William I de Warenne would have been consanguineous in the 3rd degree because he and the conqueror were second cousins. Likewise, if Gundrada had been the daughter of Matilda of Flanders, her son William II de Warenne could not have married Isabella de Vermandois because she and Matilda of Flanders were first cousins. William II and Isabella would have then been consanguious in the 2nd degree. In addition, Isabella of Vermandois had already come under papal scrutiny since an earlier proposed marriage had been barred due to issues of consanguinity, in this case of a more distant blood relationship (5th or 6th degree).
Given all this it should be clear that Gundrada was not and could not have been the daughter of either William the Conqueror or Matilda of Flanders. Significant evidence exists to show that she was a member of the noble Oosterzele-Scheldewindeke family, hereditary advocates of St. Bertin. If you have any questions or would like to follow up on any aspect of this, just ask. I’d be happy to help if I can.