Be wifmannes beweddunge: Betrothals and Weddings in Anglo-Saxon England

Abstract

The Old English quasi-legal text Be wifmannes beweddunge (‘On the betrothal of a woman’) is a key source for understanding how marriages were contracted in late Anglo-Saxon England. This paper will use the nine clauses of Be wifmannes beweddunge as a window into a broader discussion of the Anglo-Saxon betrothal and wedding process. It will consider in turn the issue of licit and illicit unions, the economic and legal terms of the betrothal agreement, and the development of Christian wedding rites. It will argue that Be wifmannes beweddunge is fundamentally concerned with the legal, financial, physical and social protection of women within marriage. Moreover, it will argue that this text offers evidence for a gradual Christianisation of betrothal and wedding customs in late Anglo-Saxon England.

The Old English quasi-legal text Be wifmannes beweddunge (‘On the betrothal of a woman’) – preserved in two twelfth-century legal compendia, the Textus Roffensis (Rochester, Cathedral Library A.3.5, fols. 94v–95r), and Cambridge, Corpus Christi College 383 (fols. 57v–58v) – is a key source for understanding how marriages were contracted in late Anglo-Saxon England. ¹ Despite its importance, this text has rarely been studied; indeed, to my knowledge no single extended study has been devoted to it. ² This paper will use Be wifmannes beweddunge as a window into the Anglo-Saxon wedding process. What was the process from betrothal to wedding? Who, apart from the bride and groom, was party to the process? With what rituals was the event marked? In answering these questions, I will take my lead from the text of Be wifmannes beweddunge, which in nine brief paragraphs touches on a remarkable number of social, economic and theological issues. However, in order to understand these issues, it will be necessary to roam widely, drawing on a wide range of sources from across the period: other legal and quasi-legal texts, penitentials, charters, liturgical books, doctrinal and exegetical texts, as well the two surviving marriage agreements from the eleventh century. Anglo-Saxon England, especially the later part of the period, has sometimes been elided from scholarship on medieval marriage. ³ Be wifmannes beweddunge offers something quite unique to the study of marriage: an early, vernacular (quasi-)legal text solely dedicated to codifying betrothal and wedding customs. The text itself is fundamentally concerned with the legal, financial, physical and social protection of women within marriage, from the initiation of the betrothal process through to the wedding and beyond. Moreover, through a comparison of Be wifmannes beweddunge with other Anglo-Saxon texts, we can trace a gradual, though not always consistent, process of Christianisation in Anglo-Saxon betrothal and wedding customs. ⁴

BE WIFMANNES BEWEDDUNGE: DATE AND CONTEXT

Dorothy Whitelock notes that Be wifmannes beweddunge ‘belongs to a branch of law of which we should be glad to know more’; indeed, it is both generically and thematically a very unusual text in the corpus of Anglo-Saxon law. ⁵ This text probably dates to the early eleventh century or just possibly the late tenth. Liebermann dates it to c. 970 × 1030, while Whitelock is content to see it as ‘not early’, partially on the basis that the author uses an unusual Old Norse loanword, sammæle (‘united’, from ON sammæli, ‘agreement’) at c. 6. ⁶ This word first appears in the Old English corpus at the very end of the tenth century, in III Æthelred, a code of laws relating specifically to the Danelaw, issued in 997. It also appears in a Kentish land agreement written c. 990 × 1005 (S 1455) and in the Anglo-Saxon Chronicle (MS E) for 1018. While in III Æthelred and ASC E sammæle occurs with reference to a specifically Anglo-Scandinavian context, in the Kentish agreement (between Wulfric, abbot of Canterbury, and Eadred son of Lyfing, whose names show no immediate Scandinavian background) the word appears already to have been naturalised in Old English. Certainly, there is nothing else in Be wifmannes beweddunge in either vocabulary or context to suggest strong Scandinavian influence. On this basis, I would be tentatively inclined to narrow Liebermann’s date range slightly to perhaps 990 × 1030.

This places the text broadly within the same milieu as Archbishop Wulfstan of York (d. 1023). ⁷ The text does not appear to have been written by Wulfstan himself. There is, however, a textual link to Wulfstan’s own writings in the phrase ‘æfter Godes rihte ⁊ æfter woroldgerysnum’ in c. 1 of Be wifmannes beweddunge. ⁸ This complete phrase appears twice elsewhere in the OE corpus; these three places are also the only citations for the compound woroldgerysene. It appears in a very similar quasi-legal text known as Swerian, which is a step-by-step procedure for the proper swearing of oaths. ⁹ The phrase also appears in Wulfstan’s Institutes of Polity:

Riht is, ðæt ealle cristene men heora cristendom rihtlice healdan and ðam life libban, þe heom to gebyrað æfter Godes rihte and æfter woruldgerysenum. And heora wisan ealle be þam ðingan geornlice fadian, þe ða wisian, ðe hy wislice and wærlice wisian cunnon. ¹⁰

Wulfstan’s own style is instantly recognisable here, with its rich alliterative and chiastic patterns – there is little doubt that this author and the author of Be wifmannes beweddunge are different. What are we to make of the recurring phrase then? It suggests either that the author of at least the first clause of Be wifmannes beweddunge was influenced by Wulfstan, or that Wulfstan himself discovered this phrase in the quasi-legal material and found it useful. ¹¹ Either way, this phrase gives us a clear textual link between Wulfstan and Be wifmannes beweddunge. Moreover, it is worth noting that the text is found in the legal compendia close by the Wulfstanian ‘Compilation on Status’, a group of six quasi-legal texts – Geþyncðo, Norðleoda laga, Mircna laga, Að, Hadbot, and Grið – each of which presents a Wulfstanian image of how English society functioned in the past, before the tumultuous later years of King Æthelred. ¹² However, we cannot push the Wulfstan link any further than this – nor do we need to, since Be wifmannes beweddunge is very much a standalone, self-contained text. It is safe to place the text in the same broad category as texts like Swerian and Hit becwæð, which ‘circulated widely … presumably because they were found useful, filling in some of the gaps left by royal legislation’. ¹³

LICIT AND ILLICIT UNIONS

Be wifmannes beweddunge begins with the phrase, ‘Gif man mædan oððe wif weddian wille …’. ¹⁴ The reasons for embarking upon marriage in the early medieval world were surely as many and varied as they are today. Patristic and early medieval theologians did not discuss the practical reasons for marriage so much as the spiritual purpose. Augustine of Hippo defined the three ‘goods’ of marriage as the bonum fidei, the bonum prolis and the bonum sacramenti – the good of faith, the good of offspring, and the good of ‘sacrament’. ¹⁵ That is, marriage is good because it allows a couple to express faithfulness to each other and to God; because it produces children; and because it was instituted by God as a spiritual mystery, an image of the union of Christ and the church. Augustine explicitly censured those who embarked upon marriage for no other reason than to satisfy sexual desire: ‘but someone ill uses that good [of marriage] if they use it in a bestial manner, so that their intention is in the pleasure of lust, not in the desire for offspring’. ¹⁶ Of course, this tacitly acknowledges that many people did in fact embark on marriage for just such a reason. Indeed, the apostle Paul recognised as much in I Corinthians, when he advised the unmarried that ‘if they cannot contain themselves they should marry; for it is better to marry than to burn’. ¹⁷ For Paul, whatever else marriage might signify, its practical purpose was to legitimise and sanctify sexual desire.

For Isidore of Seville, writing c. 600, a man looked for four things in a prospective wife: ‘virtue, family, beauty and wisdom’ – perhaps unsurprisingly he saw wisdom (sapientia) as the most important of these. ¹⁸ However, the surviving evidence – legal and ecclesiastical – for how Anglo-Saxons went about choosing marriage partners is almost entirely negative, defining limitations on choice rather than positive reasons. Be wifmannes beweddunge ends with an injunction ‘to take caution that someone knows that [the couple] are not too close in kinship’. ¹⁹ Affinity or consanguinity in marriage was an issue for Anglo-Saxon Christianity from its very inception. ²⁰ In 601, Augustine of Canterbury addressed a series of pressing pastoral questions to Pope Gregory the Great, whose responses are known as the Libellus responsionum (‘Book of Responses’). ²¹ To the question of whether two brothers might marry two unrelated sisters, Gregory simply responded, ‘it is allowed to happen by all means’. ²² In response to the broader question of to what degree Christians might marry, Gregory forbade marriages between first cousins, marriage between a step-son and step-mother, and marriage between a man and his brother’s wife. He did, however, allow for a degree of toleration for newly converted Christians ‘who, while they were yet in unbelief, are said to have been joined in this unlawful matrimony’. ²³ In subsequent centuries, there was a certain amount of confusion and disagreement on what degree of consanguinity was licit between prospective spouses, as can be seen especially in the correspondence of Boniface, as well as the Poenitentiale Theodori. ²⁴ The Legatine Council of 786 made no effort to rule decisively on the trickier questions of affinity, simply forbidding ‘all unrighteous and incestuous marriages, as much with a maidservant of God [that is, a nun], or with other forbidden persons, as with those closely related and related by blood, or with foreign women’. ²⁵ We might imagine that if such learned scholars as Boniface and the author of the Poenitentiale found it difficult to reconcile divergent modes of calculating affinity, then it must have been a struggle for any lesser clergyman, never mind a lay person. ²⁶ We can readily imagine why Be wifmannes beweddunge prescribes caution and careful working out, ‘lest someone afterwards put asunder what had previously been joined together wrongfully’. ²⁷

The opening clause of Be wifmannes beweddunge does hint at an alternative to marriage, which does not fit into the idealised situation envisioned by the rest of the text. In clause 1, the groom is told to make a pledge ‘that he desires her in such a way that he wishes to keep her according to God’s law, as a man must keep his wife’. ²⁸ This raises the question of whether there was a way to ‘keep’ a partner which was not æfter Godes rihte. This may have taken the form of concubinage (defined as keeping an additional partner (or several) in addition to a lawful spouse) or cohabitation (defined as a marriage-like state which is not legally recognised as such). ²⁹ Illicit unions are clearly prohibited in the laws of Wihtred of Kent (issued at Berghamstyde in 695), though the code does not define precisely what it means by unriht hæmed, which I have translated as ‘illicit union’: ³⁰

Unrihthæmde mæn to rihtum life mid synna hreowe tofon oþþe of ciricean genaman [sc. gemanan] ascadene sien. Æltheodige mæn, gif hio hiora hæmed rihtan nyllað, of lande mid hiora æhtum ⁊ mid synnum gewiten; swæse mæn in leodum ciriclicæs gemanan ungestrodyne þoligen. Gif ðæs geweorþe gesiþcundne mannan ofer þis gemot, þæt he unriht hæmed genime ofer cyngæs bebod ⁊ biscopes ⁊ boca dom, se þæt gebete his dryhtne C scll’ an ald reht. Gif hit ceorlisce man sie, gebete L scll’; ⁊ gehwæder þæt hæmed mid hreowe forlæte. Gif preost læfe unriht hæmed … sio he stille his þegnungæ oþ biscopes dom. ³¹

Margaret Clunies Ross suggests that unriht hæmed must refer specifically to concubinage in this instance. ³² However, she also suggests that concubinage per se was only ever prevalent at the highest levels of Anglo-Saxon society. ³³ The specific reference to ‘illicit’ unions among ceorls suggests that the practice proscribed in the laws was rather more widespread. A ceorl was surely more likely to cohabit with a partner in a de facto relationship – or to carry on an adulterous relationship outside of marriage – than he was to keep a concubine in addition to a lawful wife. ³⁴ Clunies Ross also refers to II Cnut, c. 54.1 (produced in quite a different cultural context) to justify her reading of unriht hæmed, even though that code uses the more specific term cifes to refer to a concubine taken in addition to a lawful wife (rihtwif). ³⁵ It seems more reasonable to allow for a certain greyness in the nature of marital and quasi-marital relationships, especially in the earliest years of Anglo-Saxon Christianity. There were rihte and unrihte hæmedu and, while some of the latter might be mapped onto our word ‘concubinage’, others may have looked more like ‘cohabitation’. However, Be wifmannes beweddunge has a clear idea of what constituted rihtlif, and does not spend any time considering unriht alternatives – it neither legislates against nor makes allowances for concubinage. ³⁶ This, of course, does not mean that concubinage and cohabitation did not happen, but that there was significant (and perhaps increasing) social and religious pressure to follow a legally and ecclesiastically recognised procedure of marriage.

THE BETROTHAL AGREEMENT

After one had found a suitable partner – who was not too closely related – and decided to take them into rihtlif instead of unriht hæmed, how did one go about arranging a marriage? The first clause of Be wifmannes beweddunge establishes the fundamental principle of consent, which courses through the text: although the process of betrothal is initiated by a man, it may only proceed if ‘it is pleasing to her and to her “friends”’. ³⁷ The use of the word freond (pl. frynd in this text) is ambiguous. ³⁸ The word had a much wider semantic range in Old English than ‘friend’ does today; certainly, in some other legal texts it seems to refer to ‘relatives’, but even in these contexts it may be better to think of it as something closer to ‘advisor’ or ‘trusted one’. ³⁹ The ambiguity of freond leaves open the possibility that an unmarried woman may not have been reliant on her biological parents (who, of course, may not have been living), but on other relatives or even on non-familial connections. Clause 1 includes a single reference to another party on the bride’s side, the foresprecan (sg. forespreca). In its origins, this word evidently has the sense of ‘spokesman’ or ‘advocate’. It probably has something of this sense here: the foresprecan’s role is not merely to consent but formally to receive the man’s wedd (pledge). ⁴⁰ The Latin translation of Be wifmannes beweddunge in the Quadripartitus translates foresprecan as paranymphi, which specifically refers to nuptial sponsors (including what we would now call groomsmen and bridesmaids). ⁴¹ This may suggest that the foresprecan also had a liturgical or ritual role in the marriage process, though I can find little other evidence for the role of paranymphi in Anglo-Saxon marriages. A third word is used later in the text: magas (sg. mæg), which does have the specific sense of ‘kinsmen’. Only after the entire betrothal agreement has been finalised does it fall to the prospective bride’s kinsmen ‘to pledge their kinswoman as a wife and in lawful matrimony to him that desired her’. ⁴² While it is highly likely that the roles of frynd, magas and foresprecan would have overlapped in practice, the text sees these as distinct roles: the frynd are the woman’s closest counsellors who help initiate the process; the foresprecan take a specific legal role as ‘sponsors’; the woman’s family as a whole gives her into marriage. Finally, it is worth noting that the man is also expected to lean heavily on his frynd, even if their role is slightly different. Thrice in the text (c. 1, c. 3, c. 4) we find the formula ⁊ aborgian frynd þæt (‘And his “friends” will stand surety for that’).

Although the reference to frynd does qualify or limit a bride’s own legal agency, the text still stresses the importance of her consent. At c. 3, this is restated, here without reference to any third party: the groom is to proclaim what he intends to bestow upon her ‘providing she accepts his will’. ⁴³ It is interesting to compare this to the late seventh- or early eighth-century Poenitentiale Theodori:

Puellam disponsatam non licet parentibus dare alteri viro nisi illa omnino resistat; tamen ad monasterium licet ire, si voluerit. Illa autem desponsata, si non vult habitare cum eo viro cui est desponsata, reddatur ei pecunia, quam pro ipsa dedit, et tertia pars addatur; si autem ille noluerit, perdat pecuniam quam pro illa dedit. Puella autem XVI. annorum sui corporis potestatem habet. Puer usque ad XV. annos sit in potestate patris sui; tunc ipsum potest monachum facere, puella vero XVI. vel XVII. annorum, quae ante in potestate parentum erat. Post hanc aetatem patri filiam suam contra ejus voluntatem non licet in matrimonium dare. ⁴⁴

There is a degree of confusion over the exact age at which a woman ‘has power over her own body’ – different manuscripts of the Poenitentiale give fourteen, thirteen or seventeen. ⁴⁵ A later Old English translation of these canons, the so-called Scrift boc (written in the tenth or possibly the ninth century) gives fourteen as the age and seems to alter the sense of the final provision:

Feowertynewintre mæden mot agan hire lichaman geweald … Fæmne oð ðæt heo sy þreottynewintre oððe feowertynewintre sy heo in hire eldrena mihtum. ⁴⁶ Æfter ylde hire hlaford ⁴⁷ hi mot gifan mid hire willan. ⁴⁸

Where the Poenitentiale Theodori allows the possibility of a father giving his daughter in marriage against her will before the age of majority, the Scrift boc stresses that any marriage should be made with the woman’s consent; at the same time, it gives the woman’s father or ‘lord’ the task of arranging marriage even after that age. Taken together with Be wifmannes beweddunge, the penitential evidence may point to a gradual strengthening of the principle of consent over the course of the Anglo-Saxon period, although we would need more evidence to say this with confidence. At the same time, it would appear that women rarely, if ever, had the right to arrange marriages entirely off their own bat.

Be wifmannes beweddunge makes several references to the financial aspect of the betrothal agreement. The text makes three discrete provisions: c. 2 states that the groom must provide fosterlean (‘payment for bringing up’); c. 3 has the groom publicly state what he will ‘bestow’ (geunnan) on the bride (not on her family); while c. 4 states that the woman will inherit half of the man’s property if he dies without issue and the entirety if they have a child. Fosterlean is a somewhat obscure term. It appears elsewhere in the Old English corpus only in the Menologium, a poetic course through the liturgical year probably dating from the second half of the tenth century, where it refers to the heavenly reward given to Mary for raising Jesus. ⁴⁹ Given that there is no other evidence for the term in Anglo-Saxon legal discourse, perhaps we should view fosterlean not as a specific legal/financial term but more as a kind of justification for payment of a bride-price to the bride’s family; it is a way of acknowledging the family’s role in raising the woman to adulthood and of recompensing them for the ‘loss’ of their daughter. ⁵⁰ Note also the requirement in c.3 to ‘make known to whom her fosterlean belongs’ – another acknowledgement that the woman might not be in the care of her biological parents, but that fosterlean must be paid regardless. ⁵¹ While this is paid to the woman’s family or guardians, the gift bestowed in c. 3 is clearly to be given to the woman herself. Further evidence for the nature of any payments made must be sought in the ‘secular’ laws. The laws do not make any provisions for the size of any such payments, which was entirely left to the agreement of the parties. The earliest laws do, however, make provisions for two distinct payments, one to the family and one to the woman. The laws of Æthelberht of Kent (c. 602–603?) ⁵² and Ine of Wessex (695) both refer to the practice of ‘buying a wife’, which seem to refer to the bride-price, a payment made to the family rather than to the woman herself. ⁵³ Æthelberht, c. 81 makes a reference to a separate payment, the morgengifu (‘morning gift’), which was to be made to the bride herself after the wedding. ⁵⁴ Although the term morgengifu does not reappear in a legal code until II Cnut, c. 73a, some 400 years later, it was evidently part of Anglo-Saxon custom throughout that period. ⁵⁵ It appears several times in wills and charters; for instance, a letter to King Edward the Elder (r. 899–924) concerning a dispute over land at Fonthill, Wilts., states that a man named Helmstan held land ‘just as Ætheldryth had given it for Osulf to own for a fair price, and she said to Osulf that she owned it with the right to sell it to him because it was her morgengifu when she first came [to marry] Athulf’. ⁵⁶ The prevalence of this practice is attested in place-name evidence. A number of place-names derive directly from the word morgengifu, especially though by no means exclusively in Essex and Sussex. ⁵⁷ Carole Hough has further argued that the place-name evidence suggests that the practice of morgengifu was not limited to the upper classes; small paddocks with names like le Moryivegarston in Surrey may have been the gift of ‘small farmers and freemen’. ⁵⁸ It would appear that the financial provisions involved in betrothal agreements were remarkably stable, at least in theory, over the course of the Anglo-Saxon period. It is also worth noting that Be wifmannes beweddunge exclusively talks about financial provisions being made to the bride and her family – there is no mention of a dowry, as in a payment made by the bride’s family to that of the groom.

So, the chief concern of Be wifmannes beweddunge seems to be the legal, financial, physical and social protection of women within marriage. ⁵⁹ This is evident from the repeated demands for consent and agreement, strengthened by multiple ‘pledges’ and ‘sureties’; from the multiple levels of financial provision for the bride; and from the multiple levels of ‘friendship’ and kinship which provide witnesses, sponsors and counsellors for both parties. This principle of protection is most clearly laid out in clause 7:

Gif hy man ðonne ut of lande lædan wille on oðres þegnes land, ðonne bið hire ræd, ðæt frynd ða forword habban, ðæt hire man nan woh to ne do, ⁊ gif heo gylt gewyrce, ðæt hy moton beon bote nyhst, gif heo næfð, of hwam heo bete. ⁶⁰

This passage acknowledges the potential for abuse when a woman marries out of her own community, away from her established networks of frynd and magas. ⁶¹ The clause ‘ðæt hire man nan woh to ne do’ seems to be directed not at the prospective husband but at his family and community. Both this and the succeeding clause about compensation may indeed be envisioning a situation where a woman is left on her own in another community, because her husband has either died or been called away on business or campaign. If she incurs a fine – the usual Anglo-Saxon means of punishment – and is unable to pay, she is not to be left to her own devices; her old kinship group continues to have the right to intervene on her behalf. ⁶²

We might infer from clause 7 of Be wifmannes beweddunge not only that it was considered normal enough for women to marry outside their immediate communities – this much can be discovered from surviving wills and marriage agreements, at least for higher-status women – but that this situation left the new bride open to abuse, from which she needed legal protection. ⁶³ The Wife’s Lament, a poem preserved in the late tenth-century Exeter Book, gives us a sense of a wife’s emotional response to being sundered from her community, preyed upon by her husband’s family, and separated from her husband:

Ða ic me feran gewat folgað secan,

wineleas wræcca, for minre weaþearfe.

Ongunnon þæt þæs monnes magas hycgan

þurh dyrne geþoht, þæt hy todælden unc,

þæt wit gewidost in woruldrice

lifdon laðlicost, ond mec longade.

Het mec hlaford min her [h]eard niman,

ahte ic leofra lyt on þissum londstede,

holdra freonda. ⁶⁴

There is much in The Wife’s Lament that defies explanation and the poem has attracted a vast range of interpretations as a result. ⁶⁵ Many of the central questions – the dating of the poem, whether it was authored by a man or a woman, whether refers to a lost legendary story, even whether the central character is alive or dead – cannot be answered with any certainty. ⁶⁶ However, recent scholarship has at least tended to accept that the poem is about a woman who is separated first from her own family and then from her husband. ⁶⁷ The use of the impersonal mon in The Wife’s Lament (‘heht mec mon wunian | on wuda bearwe’) echoes, coincidentally but revealingly, the use of the impersonal in Be wifmannes beweddunge: ‘hy man ðonne ut of lande lædan wille … ðæt hire man nan woh to ne do’. ⁶⁸ In both texts, it is not specified who is doing or causing the abuse – it may be the husband’s family, the husband’s lord or the husband himself. For the wife and for the community she left behind, it scarcely mattered.

Two Eleventh-Century Marriage Agreements

Some evidence of how an agreement of the sort envisioned by Be wifmannes beweddunge might be formulated in practice can be found in a pair of marriage agreements which survive from the 1010s (making them roughly contemporaneous with the quasi-legal text). These are worth quoting in full (using Whitelock’s translations, with my alterations marked by square brackets). Firstly, a marriage from Worcestershire, dated by Lapidge to 1014 × 1023, concerning the marriage of Archbishop Wulfstan’s sister to one Wulfric:

Her swutelað on ðysum gewrite ymbe ða forwerda ðe Wulfric ⁊ se arcebisceop geworhtan ða he begeat ðæs arceb. swuster him to wife · ꝥ is ðæt he behet hyre ꝥ land æt Ealretune ⁊ æt Ribbedforda hire dæg · ⁊ he behet hire ꝥ land æt Cnihtewican · ꝥ he wolde hit hire begytan ðreora manna dæg æt ðam hirede on Wincelcumbe · ⁊ sealde hyre ꝥ land æt Eanulfintune to gyfene ⁊ to syllene ðam ðe hire leofest wære on dæge ⁊ æfter dæge · ðær hire leofest wære · ⁊ behet hire. L. mances goldes · ⁊ XXX. manna · ⁊ XXX. horsa. Nu wæs ðyses to gewitnesse Wulfstan arceb · ⁊ Leofwine ealdorman · ⁊ Aeþelstan bisc · ⁊ Aelfword abb · ⁊ Brihteh munuc · ⁊ manig god man toeacan heom · ægðer ge gehadode ge leawede · ꝥ ðas forewerda ðus geworhte wæran. Nu syndon to ðysum forwordan twa gewrita · oþer mid ðam arceb · on Wigereceastre · ⁊ oþer mid Aeþelstane bisc · on Herforda. ⁶⁹

Secondly, a Kentish marriage agreement, dated to 1018 × 1020, concerning the marriage of one Godwine and the daughter of a Brihtric:

Her swutelaþ on þysan gewrite þa foreward þe Godwine worhte wið Byrhtric þa he his dohter awogode, ꝥ is ærest ꝥ he gæf hire anes pundes gewihta goldes wið þonne þe heo his spæce underfenge, ⁊ he geuþe hire þæs landes æt Stræte mid eallan þon þe þærto herð, ⁊ on Burwaramersce oðor healf hund æcera, ⁊ þærto þrittig oxna, ⁊ twentig cuna, ⁊ tyn hors ⁊ tyn ðeowmen. Ðis wæs gespecen æt Cincgestune before Cnute cingce on Lyfinges arcebisceopes gewitnesse, ⁊ on þæs hiredes æt Cristescircan, ⁊ Ælfmeres abbodes, ⁊ þæs hiredes æt S. Augustine, ⁊ Æþelwines sciregerefan, ⁊ Siredes ealdan, ⁊ Godwines Wulfeages sunu, ⁊ Ælfsige cild, ⁊ Eadmer æt Burham, ⁊ Godwine Wulfstanes sunu, ⁊ Kar ðæs cincges cniht, ⁊ þa man ꝥ mædan fette æt Byrhtlingan, þa eode þyses ealles on borh Ælfgar Syredes sunu, ⁊ Frerþ preost on Folcestane, ⁊ of Doferan Leofwine preost, ⁊ Wulfsige preost, ⁊ Eadræd Eadelmes sunu, ⁊ Leofwine Wærelmes sunu, ⁊ Cenwold rust, ⁊ Leofwine Godwines sunu æt Hortune, ⁊ Leofwine se reade, ⁊ Godwine Eadgeofe sunu, ⁊ Leofsunu his broðer: ⁊ swa hwæðer heora læng libbe fo to eallan ætan ge on ðam lande þe ic heom gæf ge o ælcon þingan. Ðyssa þinga is gecnæwe ælc dohtig man on Kænt, ⁊ on Suþsexan on ðegenan ⁊ on ceorlan, ⁊ þyssa gewrita synd ðreo · an is æt Cristescyrcan, oðer æt S. Augustine, ꝥ þridde hæfð Byrhtric self. ⁷⁰

These documents both present a very similar procedure for arranging a marriage. In both cases the agreement is made between the prospective husband and the bride’s guardian (whose role may correspond to that of the forespreca in Be wifmannes beweddunge). These are both close relatives – a father in the Kentish document, a brother in the Worcestershire document. The woman’s consent is more clearly affirmed in the Kentish document (with the offer of land being made only on the condition that ‘heo his spæce underfenge’). ⁷¹ In both agreements, as in Be wifmannes beweddunge, the husband bestows very significant gifts on the wife. However, the different payments are not as clearly demarcated as they are in Be wifmannes beweddunge or in other texts. All of these gifts are understood to be granted to the bride herself – there is no evidence of fostorlean or bride-price in these texts. There does seem to be a difference in the Worcestershire document between land given for the duration of the bride’s lifetime (hire dæg), and land given both for her lifetime and after her lifetime (on dæge ⁊ æfter dæge) – the latter is more likely to represent the morgengifu, which the wife would be free to dispose of as she wished either during her lifetime or in her will (this presumably being the force of the doublet (to gyfene ⁊ to syllene).

Both documents provide long lists of witnesses, corresponding to the frynd of Be wifmannes beweddunge. Be wifmannes beweddunge makes separate provisions for the groom to provide guarantors of the bride’s safety, and for the bride’s magas to pledge their kinswoman to the groom; there seem to be different roles for different witnesses or guarantors at different times. However, in the Kentish and Worcestershire agreements, it is difficult to untangle these witnesses – to figure out which ones speak for the bride and which for the groom. In the Worcestershire agreement, all the witnesses are given in one long list. This might suggest that they are all considered the frynd of the bride, given that Wulfstan is included among them. However, the final provision – that two copies of the agreement be made, one for Worcester and one for Hereford – may suggest that the Bishop of Hereford was acting as freond for Wulfric. In the Kentish agreement, there is first a list of witnesses to the marriage, including the King and the Archbishop. There is then a separate list of those who, in a striking echo of Be wifmannes beweddunge, ‘acted as surety’ (eode … on borh) when the bride was ‘fetched’ from Sussex to Kent. Again, it is unclear whether these guarantors belonged to the bride’s party or the groom’s. However, given that the match was made across two shires, it is perhaps the most natural reading of the text to suggest that the first list of witnesses represents Godwine’s guarantors/frynd (the Kentish party) and the second list represents the bride’s frynd (the Sussex party). As in the Worcestershire agreement, the Church seems to have acted almost as a broker for the marriage – high-status clergy acted as witnesses, the agreement itself was (presumably) drafted by a cleric, and copies of the agreement were kept in church archives. ⁷²

While it is not always possible to match the details of these documents to the specific provisions of Be wifmannes beweddunge, the surviving agreements do largely follow the pattern set down in the quasi-legal text. There is evidence of multiple levels of frynd/magas acting as guarantors for both parties; there is also evidence of multiple levels of financial provision, granted directly to the bride for her to dispose of as she wished. Of course, these were both very high-status marriages between prominent and well-connected people. At the same time, however, the two texts seem to be entirely independent of each other. They come from far-removed parts of the country, and they are associated with different dioceses. Apart from the opening her swutalað clause (common in wills from the same period), there are few textual links between the texts – notably the Worcestershire document’s use of dæg (in the sense ‘lifetime’) is not mirrored in the Kentish document. The Worcestershire document would have been drafted by a Worcester scribe, while the Kentish document is strongly associated with Christ Church, Canterbury. ⁷³ The fact that they are remarkably similar to each other, and that they both largely accord with the picture in Be wifmannes beweddunge is evidence that there was a somewhat standardised way of forming a betrothal agreement in late Anglo-Saxon England.

THE WEDDING RITE

The second-to-last provision of Be wifmannes beweddunge concerns the wedding itself: ‘At the wedding by right there ought to be a priest, who should unite them with God’s blessing on their union in all prosperity’. ⁷⁴ Clearly the author of this text envisions some sort of Christian ceremony as the final step in organising a marriage ‘æfter Godes rihte’ – or it may be preferable to see a two-stage process here, a formal betrothal agreement ([be-]weddung), followed at some later date by a marriage ceremony (gyft). ⁷⁵ Clause 8 provides little guidance for such a ceremony, apart from the properness of a priestly blessing. However, this in itself is quite significant and implies much about the Christianisation of marriage in the early Middle Ages. ⁷⁶ A Christian marriage service or nuptial blessing was by no means taken for granted in the practice of the early church. Although conduct within marriage was much discussed in the New Testament, the wedding itself seems to have been a matter for secular custom. ⁷⁷ Even though Augustine had referred to marriage as a sacramentum, the idea of marriage as one of the ‘official’ sacraments of the church did not develop until the late Middle Ages. ⁷⁸ Nevertheless, the idea that marriages ought ideally to be blessed by a priest developed by the fifth century at the latest, at least in Gaul. ⁷⁹ The Poenitentiale Theodori stated that ‘for a first marriage a priest ought to conduct a mass and bless both [the bride and groom]’. ⁸⁰ In the Eastern Church, the Emperor Leo VI (r. 866–912) ruled that a priestly benediction was necessary for a marriage to be considered legitimate. ⁸¹ However, in the West this blessing was never made a condition for a valid marriage until the Reformation. Marriage was evidently not seen in the first order of a priest’s duties. The eighth canon of the Council of Clofesho in 747 admonished priests to ‘unceasingly remind themselves’ (indesinenter reminiscant) of the purpose of their ordination: to officiate at the altar; to ‘look after’ (conservare) the house of prayer; to spend time faithfully in reading, prayer, masses and psalmody; and to assist their abbots and abbesses in teaching and admonishing the people. ⁸² The blessing of marriages was evidently a secondary task of the priesthood, part of the priest’s general duties of looking after the lay flock rather than a distinct ministry.

How did the Christian ritual of marriage develop over the course of the early medieval period? The Poentitentiale Theodori had prescribed both a mass and a blessing, specifically for a first wedding (in primo conjugio), implying that remarriages, though licit, were not to be blessed in the same way. ⁸³ Evidence from North Africa shows that a nuptial mass of some sort featured in Latin liturgy from at least the end of the third century; it is difficult, however, to trace its development or prevalence in the first eight centuries of western Christianity. ⁸⁴ In the absence of any firm evidence, there is little we can say about nuptial masses in early Anglo-Saxon England. It seems probable, however, that most early nuptial blessings were primarily or exclusively domestic affairs, conducted in the home, perhaps with blessings composed or modified ‘on the spot’ by the priest. For more isolated communities, it is possible that any blessing might have occurred some time after the wedding – this is perhaps what the laws of Wihtred envisioned when it called on people to ‘legitimise’ their relationships (hiora hæmed rihtan). ⁸⁵ From the eighth century onwards, liturgical books become increasingly common in England and on the continent, meaning that we can begin to trace both aspects of the nuptial blessing – the mass and the domestic ritual. The earliest complete nuptial masses appear in the Gelasian Sacramentary (early-mid-eighth century) and the Gregorian Sacramentary (late eighth century). ⁸⁶ The latter was obtained by Charlemagne directly from Pope Hadrian I, in the hope of establishing a uniform text of the missal; however, elements of both the Gelasian and Gregorian texts ultimately formed the basis of the Roman Missal. ⁸⁷

The earliest English liturgical books to include nuptial blessings are the Leofric Missal (Oxford, Bodleian Library 579) and the Durham Collectar, also known as the Durham Ritual (Durham, Cathedral Library A.IV.19), both dating to the late ninth or early tenth century. ⁸⁸ Both largely follow the Gregorian text of the mass; however, the Durham Collectar contains the text of the epistle and Gospel readings, as well as a set of blessings which evidently derive from an earlier ‘domestic’ ritual. ⁸⁹ This allows us to see a complete nuptial blessing from ninth-/tenth-century England. ⁹⁰ The Gospel reading is taken from Matt. XXII.1–14, the parable of the wedding feast. The epistle is rather more direct, a warning from Paul (1 Cor. VI.15–20) not to unite oneself with a prostitute (adheret meretrici) but to ‘flee sexual immorality’ (fugite fornicationem). The Gregorian benediction during the mass is, interestingly, directed specifically towards the bride, as seems to have been the case with most of the earliest nuptial liturgies. ⁹¹ It calls on God’s blessing ‘over this your maidservant’ (super hanc famulam tuam), and calls for God to help her become (among other things) an imitatrix of the wives of the patriarchs: to be as loving as Rachel, as wise as Rebecca, and as faithful and long-living as Sarah. ⁹² The text of the mass is then followed by the series of blessings, which by their reference to hoc thalamum and lectum istum were evidently meant to be said in the home of the newly-wed couple. This sequence, which may well draw upon a much earlier and long-lasting tradition of domestic blessings, recurs exactly (with two additions) in the mid-tenth-century Egbert Pontifical (Paris, BnF, Lat. 10575) – given the way the Durham version breaks off mid-sentence during the seventh prayer, I am confident that the version in Egbert represents the full version of the sequence. Given that there is a degree of repetition in the prayers, it is possible that the priest or bishop was meant to select only some of the prayers on any given occasion:

1. 1) Aaronic blessing (adapted from Num. VI.24–27)

2. 2) Blessing of the marriage covenant (conuentio)

3. 3) Blessing of the marriage bed/chamber (thalamus)

4. 4) Trinitarian blessing

5. 5) Blessing of the couple (ad[h]ulescentuli)

6. 6) Blessing of the ring ⁹³

7. 7) Blessing of the bed

8. 8) General blessing

9. 9) Blessing of the generations (adapted from Tobit IX.10–11)

Elements of this ‘Durham-Egbert sequence’ recur in a number of eleventh-century prayer-books, including the Benedictional of Archbishop Robert (Rouen, Bibliothèque municipale, 368) and the Red Book of Darley (Cambridge, Corpus Christi College, 422). ⁹⁴ Later Anglo-Saxon prayer-books contain evidence of other nuptial blessings, entirely separate from the tradition of Durham/Egbert, which may represent local traditions or ad hoc prayers. For instance, the eleventh-century Canterbury Benedictional (London, British Library, Harley 2892), alone of all western liturgies, preserves the idea of ‘crowning’ the newlyweds, which had been common practice in Greece, Egypt and Syria/Palestine. ⁹⁵ We do not know whether the original author of this text drew inspiration from eastern texts, whether it was suggested by Pope Nicholas’ responses to the Bulgarians, or whether (just possibly) he preserved an idea which had been brought to England in the seventh century by Theodore of Tarsus. ⁹⁶ Although there is more to be said about the earliest marriage liturgies, the overall impression is of a loose and often idiosyncratic tradition.

The picture of how Christian wedding ceremonies were conducted in England becomes slightly clearer at the very end of the period. By the turn of the twelfth century, liturgical books from England, Normandy and Brittany had largely coalesced around an ordo which began at the door of the church, where the bride and groom acknowledged their consent, confirmed the dowry and rehearsed their vows; they would then move inside the church for a nuptial mass, before repairing to the marital home for the blessing of the bridal chamber. ⁹⁷ Overall, as Stevenson notes, the earlier domestic rituals represented by Durham/Egbert recede into the background, as the ‘church wedding’ becomes increasingly settled. ⁹⁸ Of course, it remains unclear to what extent church weddings, or even weddings blessed by the priest at home, had become the norm by the end of the Anglo-Saxon period – just because Be wifmannes beweddunge sees it as part of the ‘normal procedure’ and there is liturgical evidence that such rituals existed does not necessarily say anything about how common it was. There were doubtless marriages which were blessed post facto if at all, marriages which were conducted clandestinely, not to mention well-attested practices of cohabitation and concubinage. At the same time, we should not assume that the procedure in Be wifmannes beweddunge was not common or normal, just because we can find evidence for alternatives.

CONCLUSION

Perhaps the best summary of the text comes from the text itself. Be wifmannes beweddunge describes the procedure for arranging a marriage ‘æfter Godes rihte ⁊ æfter woroldgerysnum’. ⁹⁹ The author of this text offers an idealised picture of how these two demands – God’s law and secular customs – might be satisfied. The history of marriage in Anglo-Saxon England suggests that the balance between divine and earthly ideals and realities was constantly shifting. Indeed, the nature of what exactly godes riht and woroldgerysnu demanded underwent enormous changes over the course of the Anglo-Saxon period, as marriage ceased to be a primarily domestic arrangement and gradually ‘became a part of the religious life of Christendom’. ¹⁰⁰ At the same time, godes riht and woroldgerysnu should not be seen as opposing forces, with the former representing the inexorable reach of the church, the latter representing the survival of ‘pagan’ customs. ¹⁰¹ This model may be somewhat relevant to the very earliest days of Anglo-Saxon Christianity, although even at that time there was a degree of alignment between worldly and Christian expectations of marriage. In later periods, there was rarely a sharp distinction between religious and secular demands; even if they did not always work in perfect harmony, they tended to be mutually reinforcing. Undoubtedly there must have been a great many breaches of both godes riht and woroldgerysnu; the recorded instances of cohabitation and concubinage suggest as much, as do laws concerning rape and adultery. Yet it is surely unnecessarily cynical to suggest that the idealised procedures represented in Be wifmannes beweddunge and the liturgical books bore no relationship to reality at all. The Church gradually defined what it meant to live in rihtlif (‘lawful marriage’) and made this an attainable ideal for all lay people.

Amid these changes and occasional tensions, there stood a couple, making a commitment to share their lives, to be gathered into a ‘union’ (gesomnung). ¹⁰² Marriage was one of the defining events in the transitory lives of millions of women and men in early medieval England. They were not alone, of course: their families, friends, lords, priests and bishops all potentially had a say in the matter. Collectively, these people shared in an experience of marriage; an experience which was shaped and to some extent defined by godes riht and woroldgerysnu, but which had the potential to confound both. ¹⁰³