FEMALE Hadwig (Hathui, Haduwich)

Wife of Otto der Erlauchten, duke of Sachsen (Saxony).

Date of birth: Unknown.
Place of birth:

Date of death: 24 December, probably 903.
Place of death: Unknown.
The necrology of Merseburg gives 24 December as her date of death ["VIIII. k. [Ian.] ... Hathuuui mat. Heinrici reg." Calend. Merseb., 127]. She was probably the countess Had(a)wih whose 903 death is given in the Necrological Annals of Fulda ["Hadwih com. / Ob. Hadawih" Ann. Necr. Fuld., MGH SS 13: 189].

Possible father: Heinrich, d. 28 August 886, marquis of Neustia.
Mother: Unknown.
Although Hadwig's father is not directly documented, the circumstantial evidence in favor of Heinrich is good. This is discussed below in the Commentary section, as is the conjectured origin of her mother.

Spouse: Otto "der Erlauchten", d. 912, duke of Sachsen (Saxony).
["Temporibus quondam regis Francorum Conradi, dux in tota Germania princeps extiterat nomine Otto, genere secundum seculi dignitatem nobilissimus, opibus pullens, et cunctos honore praecellens, quia virtutibus erat praeditus, cuius coniugium veneranda matrona Haduwich subierat moribus non dissimilis. His filiae procreantur et tres filii, ... horum unum nomine Heinricum maiori sustulit excellentia, ..." Vita Mahthildis reginae antiquior, c. 1, MGH SS 10; "... extitit in partibus Germaniae dux quidam Otto nomine, ... Cui Hathuwic, matrona venerabilis, coniugali copulabatur vinculo. Quibus duo gignebantur filii, ... Maior natu vocabatur Thancmarus, et alter Heinricus." Vita Mahthildis reginae, c. 1, MGH SS 4: 284; "Hic nobilissimo Ottonis et Hathui stemmate editus, ..." Thietmar, Chron., i, 2, MGH SS 3: 735]

See the page of Otto for details.

MALE Thankmar, d. bef. 30 November 912.

MALE Liudolf, d. bef. 30 November 912.

MALE Heinrich I, d. 2 July 936, king of Germany, 919-936;
m. (1) Hatheburg, (d. 21 June?),
daughter of Erwin, count of Merseburg.
m. (2) ca. 909,
Mathilde, d. 14 March 968, daughter of count Dietrich.

FEMALE Oda, living 30 December 952;
m. (1) 897, after Easter, Zwentibold, d. 13 August 900, king of Lorraine;
m. (2) 900, Gérard/Gerhard, d. 22 June 910, count.

Possible daughter:
See the page of Otto for details.

FEMALE Liutgard, d. 21 January 923, abbess of Gandersheim, 919-923.


Much has been written about the possible origin of Hadwig. The account given here owes much to the outline of Eduard Hlawitschka [Hlawitschka (2006), 39-52]. The parentage of Hadwig was a matter of speculation from a very early time. As the mother of a king (Heinrich I) who did not have any obvious genealogical claim to the throne, it was inevitable that someone would try to turn her into a Carolingian herself, or at least a descendant of that house. The earliest known attempt in that direction did not even give her the right name:

Falsely attributed name: Liutgard.
Falsely attributed father: Arnulf, d. 8 December 899, emperor.
This claim, repeated often in some of the older secondary sources, appears in one version of the universal chronicle of Ekkehard ["Cuonradus rex moriens coram principibus regni regem designat Heinricum, filium Ottonis Saxonum ducis ex Luitgarda filia Arnulfi imperatoris." Ekkehard, Chronicon universale, MGH SS 6: 175; Annales Magdeburgenses (following Ekkehard), s.a. 919, MGH SS 16: 142]. Even if we ignore the incorrect name of Heinrich's mother, the claim is chronologically impossible (Arnulf was born ca. 850). It is a transparent attempt to give the Saxon dynasty a Carolingian ancestry. A more recent variant of this error gives the correct name but keeps the false parentage [e.g., RFC2, 131 (line 172)].

Although this attempt to find a Carolingian ancestry for Heinrich is false, it is natural to ask what evidence there might be for such a descent. One thing that we can say with reasonable certainty is that Hadwig was not the daughter of a Carolingian king, for there is little chance that such a relationship, if true, would have gone unnoticed in the early sources. So, if Hadwig had a Carolingian connection, it would have to be of a more indirect nature. While there is no early source which unambiguously states that Hadwig was of Carolingian descent, there is one source which can be seen as possibly implying such a descent. In the life of Heinrich's aunt Hathumod, abbess of Gandersheim, the monk Agius states of Hathumod that "frater eius regum neptem in matrimonio habet" [Vita Hathumodae, c. 2, MGH SS 4: 167]. This is less than clear, because Agius does not say which brother of Hathumod married a neptis of kings. Some have thought that the passage refers to Hathumod's elder brother duke Bruno (d. 880) [e.g., the editor of Vita Hathumodae, MGH SS 4: 167 n. 2]. However, more recent opinion has tilted toward Bruno's younger brother duke Otto, due to the fact that Otto's son Heinrich I is called a nepos of the French king Charles the Simple by Thietmar of Merseburg ["Hic [Charles the Simple] Heinrici regis nostri, nepotis autem sui, ..." Thietmar, Chron., i, 13, MGH SS 3: 741].

One early attempt in this direction must be set aside:

Falsely attributed identification: Heilwig/Hélvide, d. after 894.
Falsely attributed father: Eberhard, d. 865×6, duke of Friuli.
Falsely attributed mother: Gisela, d. after August 874, daughter of Emperor Louis "the Pious".
Originally due to Eckhart in the eighteenth century, this theory was later revived by others, most notably Krüger [Krüger (1893)]. Krüger would identify the kings in Agius's statement "frater eius regum neptem in matrimonio habet" as Louis/Ludwig the German and Charles the Bald, and would interpret the word neptis to mean niece in the strict sense, thus making Hedwig/Hathui a daughter of a sibling of these two kings [Krüger (1893), 32]. Placing Otto's wife as a daughter of Eberhard of Friuli and his wife Gisela would satisfy this restriction, and thus Krüger identifies Hedwig/Hathui with Eberhard's similarly named daughter Heilwig. However, Hedwig and Heilwig are not the same name, and the identification has little to recommend it. Dümmler argued convincingly against it in the same year that Krüger's article appeared [Dümmler (1893)]. The identification must in any case be false if Hirsch's convincing outline of Heilwig's marriages is correct [see the page of
Heilwig for more details].

The most convincing attempt to conjecture a father for Hadwig involves a candidate who apparently has no Carolingian descent:

Possible father: Heinrich, d. 28 August 886, marquis in Neustia.

The most direct indication of a relationship between marquis Heinrich and Heinrich I is a passage from Widukind's history ["Nam cum bellum esset Cuonradi regis Cuonrado patri et Adelberto Heinrici ex sorore nepoti, primum interfectus est frater Adelberti; postea in ultionem fratris Cuonradus quoque occisus est ab Adelberto, ..." Widukind, i, 22 (pp. 26-7)]. Here, Heinricus is clearly the German king Heinrich I. The difficulty is how to interpret the words Adelberto Heinrici ex sorore nepoti. In the absence of other information, the most common translation of these words would be "Adalbert nephew of Heinrich through a sister", i.e., Adalbert, son of a sister of Heinrich. Indeed, possibly expanding on the above statement of Widukind, a number of later sources make Adalbert the son of a sister of Heinrich I named Baba [see the page of duke Otto of Saxony for more details].

The problem with this is that Adalbert appears in the records with his two (apparently younger) brothers in 898 in the context of a quarrel with bishop Rudolf (suggesting that they were then adults) [see the page of marquis Heinrich], and a son of a sister of Heinrich I would almost certainly not be an adult by 898. Adalbert was more likely in the generation preceding Heinrich I. Thus, since the word nepos can also have the wider meaning of "relative", it appears that in this case the words Adelberto Heinrici ex sorore nepoti should be translated as "Adalbert relative of Heinrich through a sister". Eckhardt saw nepos as a possible Latin translation of the old German neve, which in those days could also mean "uncle" [Hlawitshcka (1974), 142; Hlawitschka (2006), 1.2: 40; both cite Eckhardt (1963), 15, not seen by me]. Given that it was Adalbert who was in the earlier generation, the sister in question would probably be a sister of Adalbert rather than of Heinrich.

Further circumstantial evidence comes from several relatives of the German kings who are mentioned in various sources. Henri/Heinrich, bishop of Trèves (956-964), is called a propinquus of Otto the Great by Flodoard ["Episcopatus Trevirensis cuidam Hainrico, regis Othonis propinquo, datur." Flodoard, Annales, s.a. 956 (p. 143)]. He was a brother of Poppo (I), bishop of Würzburg (941-61) ["... Heinricus, eximia Francorum Suevorumque prosapia genitus, ... ad Herbipolim, quae a rusticis Wirciburg vocatur, veniret. ... quod frater eius, Poppo nomine, monarchiam illius episcopii tenebat, ..." Othlo, Vita S. Wolfkangi, c. 4, MGH SS 4: 528; "... Heinricus ab Ottone Magno Treverensem suscepit archiepiscopatum, ..." ibid., c. 7, p. 528], who was succeeded by his proximus Poppo (II) in 961 ["... Poppo Wirziburgensis episcopus ... XVI. Kal. Martii diem clausit extremum; cui proximus suus Poppo in episcopatu successit." Regino, Chronicon (continuation), s.a. 961 (p. 170)]. Furthermore, Poppo (II), bishop of Würzburg (961-83), is called a nepos by Otto II [Althoff (1984), 312, citing DD OII #132, #208]. Thus, we have the brothers Henri/Heinrich and Poppo (I) and their near relative Poppo (II) who were all relatives of the Saxon emperors Otto I and Otto II. There is no direct evidence about the ancestry of these three kinsmen, but the coincidence of the two names Heinrich and Poppo suggests that they were descended from one of the ninth century brothers Heinrich, marquis of Neustria, and Poppo, duke of Thuringia [both of whom had sons; see the page of Heinrich]. Another possible relative appears in an act of Heinrich I dated 18 October 927, where the king refers to a count Heinrich as propinquus noster ["... Heinricus divina favente clementa rex. ... quia nos interventu fidelis dilectique comitis ac propinqui nostri Henrici ..." MGH DD H I, 51 (#14)].

To these indications that king Heinrich I might have been closely related to the three "Babenberg" brothers, we can also add the onomastic indication of the name Heinrich, at that time rather rare. This name is not known to appear in the family of duke Otto, father of Heinrich I, and it may therefore be suggested with a high degree of plausibility that the name Heinrich appeared in Hadwig's family. This is made even more likely by the appearance mentioned above of men named Heinrich as relatives of the Saxon rulers. All of this evidence clearly points to marquis Heinrich of Neustria as the possible father of Hadwig. While this paternity for Hadwig remains unproven, no more convincing candidate for her father has emerged.

Conjectures regarding the mother of Hadwig

With Hadwigs father plausibly identified as someone with no likely Carolingian descent, it follows that if she did have any such ancestry, it probably came through her mother. A number of conjectures have been put forward about the possible Carolingian connections of Hadwig's mother.

Conjectured mother (improbable): NN, daughter of count Adelard, and sister of Adélaïde, second wife of Louis II le Bègue, king of France.
Since marquis Heinrich had a son named Adalhard/Adalard, Geldner conjectures that that name occurred in the family of the wife of Heinrich, and he places Heinrich's wife as a conjectured daughter of Adalard, the father-in-law of Louis II le Bègue, king of France [Geldner (1971), 12]. This Adalard has a conjectured Carolingian ancestry [see, e.g., Werner (1967), 429-441], which would in turn give the hypothesized ancestry to Hadwig. This conjecture is very thinly based and seems very unlikely.

Conjectured mother (unlikely): Engeltrude, b. say 837×840, d. after 2 April 870, daughter of Eberhard, duke of Friuli.
[Eckhardt (1963), 50ff., not seen by me, but outlined by Werner (1967), 452, n. 25, and by Hlawitschka (2006), 45] This conjecture is similar to the conjecture of Eckhart and Krüger, but with an added generation and a different daughter of Eberhard. In addition to making Hadwig a neptis regum, there is the onomastic argument that Heinrich had a son Adalhard, as did Eberhard. Eckhardt pointed out that the confraternity book of Reichenau contains an entry naming a Heimirichi and an Engildrud adjacent to one another [MGH Libri Confrat., 267 (#396, lines 15-16)]. This descent of the German king Heinrich I from Eberhard of Friuli has been accepted by some [e.g., Keats-Rohan (1997), 196-7, 201-2; Jackman (2000), 131-2], but others have objected. Metz pointed out that none of the names given to children of Eberhard appear among the children of Heinrich I of Germany. Also, as was pointed out by Werner and Hlawitschka, there is a good reason to doubt that Engeltrude was the wife of Heinrich "of Babenberg". On 2 April 870, Engeltrude's mother Gisela gave donations to Cysoing toward the burial of her and Engeltrude ["... ea ratione ut a die presenti idem locus ad quietem meam vel filie mee Ingeltrudis preparatus, ..." Cart. Cysoing, 8-9 (#4)]. It seems unlikely that Gisela would have been making arrangements for Engeltrude's future burial if she were then married to Heinrich "of Babenberg". It is more likely that Engeltrude was either unmarried or widowed at the time and living with Gisela [Werner (1967), 452, n. 5; Hlawitschka (1974), 163, n. 275]. Still, this is not a conclusive argument against the identification, and if Hadwig's mother was in fact a daughter of Eberhard, his eldest daughter Ergeltrude seems like the only chronologically feasible candidate, and even then the chronology would be very tight (see the remarks below under Judith). The connection seems unlikely.

Conjectured mother (extremely improbable): Judith, d. after 863×4, daughter of Eberhard, duke of Friuli.
[Decker-Hauff (1955), 292-309] The logic behind this conjecture is similar to the previous one, but in this case picking the apparently youngest daughter of Eberhard and Gisela makes the chronological problem even more severe: Heinrich I of Germany was born about 876 [he was aged about 60 at his death in 936, Widukind, i, 41, MGH SS 3: 435-6]. Since Heinrich had two elder brothers who died young, his parents are unlikely to have been married after 873, and they were probably married a few years earlier. Allowing 30 years for two 15-year generations, we see that it is very unlikely that Heinrich's maternal grandmother was born later than 843, with a significantly earlier date being much more probable. Gisela could not have been any older than 24 in 843, and she was probably younger. Since Judith appears to have been the youngest of at least nine children of Gisela, it is very improbable that she was born by 843. It is true that in his outline of the family of Gisela, Decker-Hauff provides estimated dates which would give Gisela seven children by 840, but he does this by placing her birth late in 819 (the earliest possible) and by giving her one child in each of the years 834 through 840. However, such extremes are far from probable, and Decker-Hauff's attempt to reshuffle the order of Eberhard's children is not convincing.

Falsely attributed mother: Bava/Baba, daughter of Berengario, count of Spoleto.
Kimpen got the name Baba/Bava from several sources which make Heinrich's son Adalbert a son of Baba, sister of king Heinrich I [
MGH SS 10: 137; MGH SS 6: 174; MGH SS 6: 28; MGH SS 6: 590; see the page of marquis Heinrich under his falsely attributed wife Baba for further details]. However, Kimpen rejected the claimed parents of "Baba" and instead conjectured that Hadwig's mother was a daughter of count Berengario of Spoleto by Helletrude, daughter of the emperor Lothair I [Hlawitschka (2006), 1.2: 46-7, outlining arguments in Kimpen, "Abstammung Konrads I. u. Heinrichs I.", 742-759, and Kimpen, "Genealogie d. bayr. Herzoge", 62, the latter two not seen by me]. There is no good reason to accept either the name or the conjectured parentage.

Conjectured mother (possible): NN, granddaughter of count Egbert and St. Ida.
The basis of this conjecture is a passage in the ninth century Translatio S. Pusinnae, which states that Haduwy (Hathuwig), abbess of Herford, was related to king Charles [the Bald] in the third and fourth degrees ["... ad regimen Herifordensis monasterii venerabilis Haduini ... Erat autem ei aditus facilis ad ipsum, sive consanguinitatis gratia, cum ei [king Charles] tertio quartoque cognatione gradu iungeretur, ..." Translatio S. Pusinnæ, c. 3, Wilmans (1867), 1: 542; see also MGH SS 2: 682]. This gives us two women, Hadwig and Haduwy, with the same name or variants of the same name, both of whom have indications of a close relationship to the Carolingians. Thus, it was suggested by Metz and following him, by Hlawitschka, that the two women were related and that their Carolingian relationship had the same source [Metz (1964), 276ff.; Metz (1971), 143; Hlawitschka (1974), 146ff.; Hlawitschka (2006), 47-52]. However, they differed in the exact descent conjectured. Metz suggested that Hadwig's mother was a sister of Haduwy [Metz (1964), 281; Metz (1971), 143]. Hlawitschka, who had also conjectured that Haudwy's father was an uncle of Liudolf of Saxony, needed to avoid making Hadwig a close relative of her husband, and thus made Hadwig's mother a daughter of count Cobbo (uncle of Haduwy) or one of his siblings [Hlawitschka (1974), 162; Hlawitschka (2006), 1.2: 51]. See the page of count
Egbert for more details about the relationships of these individuals. These scenarios seem more plausible than the other attempts at finding the origin of Hadwig's mother, but they fall far short of proof. Metz's variation making Hadwig a niece of Haduwy is a more direct fit for the onomastic evidence of those two names. As for the supposed Carolingian descent, Metz suggested that the Carolingian ancestry of Heinrich I came through an otherwise unknown daughter of either Charlemagne or Louis the Pious [Metz (1964), 286]. Hlawitschka thought that the Carolingian ancestry came through St. Ida, whom he placed as a daughter of king Carloman, brother of Charlemagne [Hlawitschka (1974), 146-165; Hlawitschka (2006), 1.2: 44-52; see the page of St. Ida for more details].


Althoff (1984) = Gerd Althoff, Adels- und Königsfamilien im Spiegel ihrer Memorialüberlieferung (Munich, 1984).

Calend. Merseb. = Ludwig Hesse, "Calendrium Merseburgense", Zeitschrift für Archivkunde, Diplomatik und Geschichte 1 (1834): 101-150.

Cart. Cysoing = Ignace de Coussemaker, Cartulaire de l'abbaye de Cysoing et de ses dépendances (Lille, 1883).

Decker-Hauff (1955) = Hansmartin Decker-Hauff, "Die Ottonen und Schwaben", Zeitschrift für Württemburgische Landesgeschichte 14 (1955), 233-371.

Dümmler (1887-8) = Ernst Dümmler, Geschichte des Ostfränkischen Reiches, 3 vols. (2nd. ed., Leipzig, 1887-8).

Eckhardt (1963) = K. A. Eckhardt, Genealogische Funde zur allgemeinen Geschichte (1963). [I have not seen this work.]

Flodoard, Annales = Ph. Lauer, ed., Les Annales de Flodoard (Paris, 1905).

Geldner (1971) = Ferdinand Geldner, Neue Beiträge zur Geschichte der "Alten Babenberger" (Meisenbach, 1971).

Hlawitschka (1974) = Eduard Hlawitschka, "Zur Herkunft der Liudolfinger und zu einigen Corveyer Geschichtsquellen", Rheinische Vierteljahrsblätter 38 (1974): 92-165.

Hlawitschka (2006) = Eduard Hlawitschka, Die Ahnen de hochmittelalterlichen deutschen Könige, Kaiser und ihrer Gemahlinnen. Ein kommentiertes Tafelwerk. Band I: 911-1137, 2 vols. (MGH Hilfsmittel, 25, Hannover, 2006).

Jackman (2000) = Donald C. Jackman, "Cousins of the German Carolingians", in Keats-Rohan & Settipani, eds., Onomastique et Parenté dans l'Occident médiéval (Oxford, 2000), 117-139.

Keats-Rohan (1997) = K. S. B. Keats-Rohan, "Poppa of Bayeux and her Family", The American Genealogist 72 (1997): 187-204. Also available in French as "Poppa 'de Bayeux' et sa famille", in Keats-Rohan & Settipani, eds., Onomastique et Parenté dans l'Occident médiéval (Oxford, 2000), 140-153.

Metz (1964) = Wolfgang Metz, "Die Abstammung König Heinrichs I." Historisches Jahrbuch 84 (1964): 271-287.

Metz (1971) = Wolfgang Metz, "Heinrich 'mit dem goldenen Wagen' ", Blätter für deutsche Landesgescichte 107 (1971): 136-161.

MGH Libri Confrat. = Paul Piper, ed., Monumenta Germaniae Historica, Libri Confraternitatum Sancti Galli Augiensis Fabariensis (Berlin, 1884).

MGH SS = Monumenta Germaniae Historica, Scriptores series.

Regino, Chronicon = Friedrich Kurze, ed., Reginonis abbatis Prumiensis Chronicon cum continuatione Treverensi (MGH SRG, Hannover, 1890).

RFC2 = Roderick Stuart, Royalty for Commoners (2nd ed., Baltimore, 1992).

RHF = Recueil des historiens des Gaules et de la France.

Werner (1967) = Karl Ferdinand Werner, "Die Nachkommen Karls des Großen bis um das Jahr 1000 (1.-8. Generation)" (Exkurs II: Königin Adelheid), Karl der Große 4 (1967): 429-441.

Widukind = Georg Waitz & Karl Andreas Kehr, eds., Widukindi monachi Corbeiensis Rerum Gestarum Saxonicarum libri tres (4th ed., MGH SRG 55, Hannover & Leipzig, 1904).

Wilmans (1867) = Roger Wilmans, Die Kaiserurkunden der Provinz Westfalen 777-1313 (Erster Band: Die Urkunden des Karolingischen Zeitalters 777-900) (Münster, 1867).

Compiled by Stewart Baldwin

First uploaded 3 April 2011.

Return to Henry Project home page

Go to Henry Project index page

Go to Henry II ancestor table