You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

A Groom of One's Own

The medieval church and the question of gay marriage.

Same-Sex Unions in Premodern Europe

by John Boswell

Villard Books, 412 pp., $25

 I.

We find ourselves, all of us, in a historical crisis of gender. It has produced highly charged arguments over "Amendment 2" to the constitution of Colorado, and over the various legal actions that have stemmed from that controversial initiative. In Ontario, one of the larger provinces in my own country, it has produced acerbic debate and the defeat of a legislative bill that would have recognized same-sex unions as "marital" in nature, and would have granted them comparable rights and duties. No small

The relationship of historians and their work to this crisis is fraught and dangerous. The stakes are high. And so the appearance of a large book by a well-known historian from Yale University on what are, he says, historical precedents for homosexual marriages in Christian society and their official recognition by the Christian church, is bound to find a large readership and to stoke a vigorous debate. The publisher's announcement excitedly warns that the work is "bound to be as controversial as the publication of the Dead Sea Scrolls." For John Boswell claims to have discovered a series of medieval manuscripts that record Christian church ceremonials for creating and blessing "same-sex unions"--for what were, in effect, marriages between men.

Apart from a foray into the problem of abandoned infants in ancient and early-modern European society, Boswell is best known for his investigation of the problematic relations between male homosexuals and the Christian church. His Christianity, Social Tolerance and Homosexuality: Gay People in Western Europe from the Beginning of the Christian Era to the Fourteenth Century, which appeared in 1980, was a learned and groundbreaking investigation of a subject that the author rightly categorized as "taboo." More than twelve years in the researching and writing, his new book on same-sex unions is similarly intended to reshape our interpretations of the past and our practices in the present.

Boswell attempts to demonstrate that "gay marriage ceremonies" were an accepted part of the early Christian church, and that the rituals that formalized such marriages were only later deliberately and consciously effaced by the church. He laudably provides the reader with transcriptions of the documents in the original Greek, along with his own English translations of them. No less laudably, he guides the reader through interpretations of his material that differ from his own.

Since the material that Boswell has uncovered is unfamiliar and impressive and controversial, it is perhaps best to give the reader some sense of it--his own English version of the text of one of these ceremonies. What follows is from an eleventh-century Greek manuscript labeled Grottaferrata .B. ii, and I have inserted some of the significant original Greek words in transcription.

Office for Same-Sex Union [Akolouthia eis adelphopoiesin]

i.

The priest shall place the holy Gospel on the Gospel stand and they that are to be joined together place their right hands on it, holding lighted candles in their left hands. Then shall the priest cense them and say the following:

ii.
In peace we beseech Thee, O Lord.
For heavenly peace, we beseech Thee, O Lord.
For the peace of the entire world, we beseech Thee, O Lord.
For this holy place, we beseech Thee, O Lord.
That these thy servants, N. and N., be sanctified with thy spiritual benediction, we beseech Thee, O Lord.
That their love [agape] abide without offense or scandal all the days of their lives, we beseech Thee, O Lord.
That they be granted all things needed for salvation and godly enjoyment of life everlasting, we beseech Thee, O Lord.
That the Lord God grant unto them unashamed faithfulness [pistis] and sincere love [agape anhypokritos], we beseech Thee, O Lord....
Have mercy on us, O God.

"Lord, have mercy" shall be said three times. iii.

The priest shall say:

Forasmuch as Thou, O Lord and Ruler, art merciful and loving, who didst establish humankind after thine image and likeness, who didst deem it meet that thy holy apostles Philip and Bartholomew be united, bound one unto the other not by nature but by faith and the spirit. As Thou didst find thy holy martyrs Serge and Bacchus worthy to be united together [adelphoi genesthai], bless also these thy servants, N. and N., joined together not by the bond of nature but by faith and in the mode of the spirit [ou desmoumenous desmi physeis alla pisteis kai pneumatikos tropi], granting unto them peace [eirene] and love [agape] and oneness of mind. Cleanse from their hearts every stain and impurity and vouchsafe unto them to love one other [to agapan allelous] without hatred and without scandal all the days of their lives, with the aid of the Mother of God and all thy saints, forasmuch as all glory is thine.

iv.

Another Prayer for Same-Sex Union

O Lord Our God, who didst grant unto us all those things necessary for salvation and didst bid us to love one another and to forgive each other our failings, bless and consecrate, kind Lord and lover of good, these thy servants who love each other with a love of the spirit [tous pneumatike agape heautous agapesantas] and have come into this thy holy church to be blessed and consecrated. Grant unto them unashamed fidelity [pistis] and sincere love [agape anhypokritos], and as Thou didst vouchsafe unto thy holy disciples and apostles thy peace and love, bestow them also on these, O Christ our God, affording to them all those things needed for salvation and life eternal. For Thou art the light and the truth and thine is the glory.

v. Then shall they kiss the holy Gospel and the priest and one another, and conclude.

It is this ceremonial, and blessings like these, that Boswell claims to be part of a lost, or deliberately suppressed, tradition of church-legitimized same-sex marriages between men.

II.

Boswell's argument stands or falls on his interpretation of a series of documents relating to a singular ritual practiced in the Christian church during antiquity and the high middle ages, principally in the lands of the eastern Mediterranean. The bonds between men that are confirmed in these church rituals are cautiously (and a little coyly) labeled by him as "same-sex unions." For his arguments to have the force that he wishes them to have, however, the words "same-sex" and "union" must be construed to mean "male homosexual" and "marriage." If they signify other sorts of associations that happened to be same-sex in gender, or unions that were meant for purposes other than marriage or a permanent affective union, then his claims fail.

For this reason, the narrative chapters of his book are ancillary, in that they digress on other aspects of the general problems of marriage and family formation in a way that is designed to support Boswell's claims about the supposed same-sex marriage rituals. His larger investigation of the nature of "heterosexual" marriage and love, and their attendant vocabulary in the Greco-Roman world, is undertaken to demonstrate that his interpretation of the "same-sex union" rituals is the most probable one.

Given the centrality of Boswell's "new" evidence, therefore, it is best to begin by describing his documents and their import. These documents are liturgies for an ecclesiastical ritual called adelphopoiesis or, in simple English, the "creation of a brother." Whatever these texts are, they are not texts for marriage ceremonies. Boswell's translation of their titles (akolouthia eis adelphopoiesin and parallels) as "The Order of Celebrating the Union of Two Men" or "Office for Same-Sex Union" is inaccurate. In the original, the titles say no such thing. And this sort of tendentious translation of the documents is found, alas, throughout the book. Thus the Greek words that Boswell translates as "be united together" in the third section of the document quoted above are, in fact, rather ordinary words that mean "become brothers" (adelphoi genesthai); and when they are translated in this more straightforward manner, they impart a quite different sense to the reader.

Whatever effect these liturgical ceremonials were intended to achieve, it is clear that they used ecclesiastical formalities to make two men "brothers," and employed various rituals and symbolic claims to confirm this relationship within the confines of the church. All of Boswell's documents relate to practices rooted in the societies of Greece, the Balkans and the eastern Mediterranean between the twelfth and sixteenth centuries--though, as he rightly argues, they surely reflect practices that were current from periods dating back to the end of the Roman empire, and probably earlier. The original documents that he cites are therefore in Greek, the ecclesiastical lingua franca of the eastern Mediterranean. The only Western versions of them are translations made into Latin from the original Greek prayer and liturgical books--wherein, notably, it seems that the Latin translators did not understand the purpose of the originals very well.

The ecclesiastical rituals that bless adelphopoiesis, or the making of a brother, include prayers and invocations of Christian virtues, particularly agape, or the Christian concept of love. They note that conditions of peace, not conditions of hate or vituperation, should exist between the two parties. Appeal is also made to pairs of men in the Christian tradition who were thought to exemplify these virtues: Philip and Bartholomew, among the disciples of Christ, and Serge and Bacchus, among the martyrs of the early church. Other elements of the ceremonial include, most significantly, the shaking or "juncture" of right hands; the exchanging of tokens; the mutual bestowing of a ritualistic kiss; and the holding of a celebratory feast or banquet to mark the occasion.

Such agreements and rituals are "same-sex" in the sense that it is two men who are involved; and they are "unions" in the sense that the two men involved are co-joined as "brothers." But that is it. There is no indication in the texts themselves that these are marriages in any sense that the word would mean to readers now, nor in any sense that the word would have meant to persons then: the formation of a common household, the sharing of everything in a permanent co-residential unit, the formation of a family unit wherein the two partners were committed, ideally, to each other, with the intent to raise children, and so on.

Although it is difficult to state precisely what these ritualized relationships were, most historians who have studied them are fairly certain that they deal with a species of "ritualized kinship" that is covered by the term "brotherhood." (This type of "brotherhood" is similar to the ritualized agreements struck between members of the Mafia or other "men of honor" in our own society.) That explains why the texts on adelphopoiesis in the prayerbooks are embedded within sections dealing with other kinship-forming rituals, such as marriage and adoption. Giovanni Tomassia in the 1880s and Paul Koschaker in the 1930s, whose works Boswell knows and cites, had already reached this conclusion.

This likely interpretation is made more likely by an extensive modern study of which Boswell appears to be unaware. In 1987 Gabriel Herman, a professor of history at the Hebrew University in Jerusalem, published Ritualized Friendship and the Greek City. In that book, and in several papers and articles on the subject published in leading journals of history and literature, Herman has analyzed the phenomenon of fictive "brotherhood" and "friendship" in the context of the world of the Greek city-state, and also in the cultures of the ancient Near East and in the regions that would later become parts of Slavonic Europe. In Herman's studies one finds all the phenomena regarded as indicative of "same-sex marriage" by Boswell: the ritual of the handshake, the exchange of tokens and right hands (dexiai), the declarations of love and friendship and of "no hostility or animosity" between the two parties, the exchange of a ritualistic kiss and the celebration of a common feast or banquet at the time of the formation of the compact.

Such ceremonials created ritualized friends who often spoke of each other as "brothers" and forged a close bond of brotherhood between themselves. They were "made brothers" rather than "brothers by nature." Hence the terminology, in Boswell's documents, of adelphopoiesis, or the ritual connected with "the making of a brother," and the phrases in his liturgical documents that specify that the two men "are not joined by the bond of nature, but rather by means of faith/trust and spirit," or similar words. This is why the documents contain references to the right of "protective asylum" (asylon anepereastos) and "safe conduct" (asphaleia) as divine attributes.

The kinds of words used to express the new relationship of "brothers" (words that are also found in Boswell's ecclesiastical rituals) were employed precisely because the men often entered into these relationships not out of love, but out of fear and suspicion. Hence the effusive emphasis on safety and trust. These relationships form as close a parallel in social institutions and practices as one could wish to have as background to the church ceremonials described in the texts cited by Boswell. Although such rituals did create fictive kinship links between the parties to them, these links were never mistaken or confused with the union of marriage. They were not undertaken primarily for erotic or affective reasons, for household formation, nor, even theoretically, for the procreation of children and the continuation of household lines.

There is only one segment of one document in Boswell's book that contains part of a liturgical service designed for a marriage ceremony: the fifth and sixth sections of the Grottaferrata manuscript of the eleventh century. Its words, which do refer to a wedding (gamos) and to the ceremonial use of crowns (stephanoi) in the ritual "crowning" of the bride and groom, give Boswell grounds to expatiate on the significance of these terms and the ceremonials involved. But there is no mention here of a same-sex union. From even a cursory reading of all of the documents, it is apparent that the original text of the "making of a brother" ceremony terminates at the end of section four of the manuscript in question. What Boswell prints as section five and section six of this document, as if they were a seamless continuation of the ritual of adelphopoiesis, belong in fact to an entirely different and separate document, which was indeed connected with a ceremony of marriage. The questionable joining of the two documents as if they were one enables Boswell to appeal throughout his main text to the totality of the documents as if they are all variant types of a marriage ceremonial, which they are not.

III.

The rest of Boswell's book does nothing other than provide an interpretive basis for the erroneous claims made about these documents, and there is little in it that is new or significant. His first chapter on "the vocabulary of love and marriage" rightly emphasizes the ambiguity of the terms used to describe love and marriage--he discusses, for example, the strong contextualization of what "to love" or "to kiss" might mean. But tendentious arguments constantly slip into the presentation. Two examples will have to suffice.

To justify the reading of "love" in his sources as having erotic content, Boswell has to demonstrate that the noun agape, "love," and the verb agapan, "to love," could be used to describe any sort of love relationship, including a wholly erotic one. This is clearly not the case. Any thorough study of the term agape will justify the traditional view that this was an unusual Christian coinage when applied to love. It signaled, as a novel virtue, an ability to accept and to embrace one's entire condition as a part of a millennially transformed world. As Boswell himself notes, the vocabulary of pre-Christian physical and sexual love--for example, the verb eran, meaning to love erotically--is almost wholly absent from the New Testament. Although the Christian term agape did occasionally enter the discourse of other late Greek writers as an alternative manner of expressing physical or erotic love, such usages are extremely rare. But that is hardly the point. What remains indisputable is the significance of the word in ecclesiastical, theological and liturgical writings--in the specific genres of Boswell's "same-sex union" documents.

Another misleading lexical interpretation is Boswell's treatment of the word adelphos or "brother." He believes that the word "brother" designates a homosexual lover. "The term `brother,'" he writes, "was widely understood in the Roman world to denote a permanent partner in a homosexual relationship." This is a possible meaning in some genres (in, say, Greco-Roman romantic novels); but context must be given priority in interpretation, since it is context that determines the meaning that was normally to be understood. One cannot abuse a mere possibility as sufficient grounds for asserting that a word, in this instance the word "brother," must have had precisely this significance in a series of Christian liturgical texts.

Boswell's tendency to misconstrue evidence extends beyond simple matters of definition, however, to the very social institutions that are central to his analysis. So, in his treatment of "heterosexual matrimony" in the Greco-Roman world, he is careless not only with proofs and concepts, but also with modern vocabulary. To label matrimony in Roman society as "heterosexual," as Boswell consistently labels it, is misleading, since it suggests that there was some form of "homosexual" matrimony recognized by the same social order. The plain fact is that there never was such a formally acknowledged alternative. Boswell describes marriages in Roman society as things that were "primarily a property arrangement" or "business deals"; he claims that "divorce was very common"; and, perhaps most astoundingly, he asserts that "children were not integral to the ideal of marriage." A brief consideration of some of the best documented upper-class families of the period, such as the family of Cicero, will readily demonstrate that even on this highly property-conscious level of society, marriage and marital relationships can hardly be reduced to the crude scheme presented by Boswell.

As for the purpose of producing children, what else can one add when the tabulae nuptiales, or marital agreements, which were part of the contractual arrangements between the married couple, specifically stated, in conformity with the norms espoused by "pagan" philosophers and Christian bishops and theologians, that marriage was primarily intended for the procreation of children, or liberorum procreandorum causa? In all these matters, Boswell creates a portrait of "heterosexual" marriages that is a caricature, and not at all a fair reflection of the day-to-day realities of such unions. Although he frequently cites Susan Treggiari's Roman Marriage, the most recent comprehensive analysis of matrimony in the Roman world, Boswell ignores Treggiari's basic findings: that love and affection, a type of (admittedly asymmetrical) sexual fidelity between the partners, the procreation of children, the sharing of resources, the hope of a lifelong union only to be ended by death and other such ideals were central to the Roman concept and the Roman practice of marriage.

It must be admitted that the problems and readings of family history faced by Boswell are of a very high order of difficulty. The best historians who have tried to cope with definitions of marriage and the family in the past have often found themselves in an obstinate morass, for past practices were no less diverse and nebulous than those of the present, and the gap between ideological presentations of family life and the realities of family life was no less difficult to gauge. One can no more speak of the model Roman family than one can speak of the model American family today. The potential variation was great. Then, as now, the model of what a family was relied upon tacit negative definitions or assumptions about what it was not.

Still, the definitions that we do have, the pre-Christian Roman ones, legal and literary, and the later Christian theological and canonical ones, are unanimous in regarding marriage or matrimonium invariably as a union of opposite sexes: male and female. The core Roman legal definition of marriage is explicit on the matter. It quite consciously and deliberately does not refer to distinctions of social status (husband and wife as maritus and uxor), or even to those of gender (man and woman as vir and femina), but instead it rather carefully refers to sexual differences, and defines marriage as the joining of male and female, coniunctio maris et feminae. (I refer to Digest, 23.2.11, and also 1.1.3, where marriage is explicitly associated with the procreation of children.)

This sexual definition was explicit and constant, while the romantic, loving or affective elements were variable. Hence different expectations of such a union are expressed; but no possibility is ever considered other than one constituted of these basic elements. Similarly, what a family was varied in terms of space, time, life cycle, region, culture and social class. It began with postulates of marriage and family formation by means of children, but it then diverged from these central assumptions and legal definitions. A single man, if he was a legally autonomous person, together with his household of personal slaves, would constitute a familia. There were numerous deviations, then, from the core definition, numerous possibilities; but there were always limits, negative frontiers were always encountered and beyond them there never existed any conception that the union of two males could possibly constitute a family--even in cases of the frereche, the consortium or sharing of property between two unmarried brothers.

As in most historical societies, formal definitions, elastic practices and assumed boundaries regarding family and marriage existed simultaneously in Rome. Some of them stood in manifest contradiction. One of the dominant models of the Roman familia in the legal codes was that of a multigenerational male descent lineage under the domination of a very powerful and senior household head, the "father of the family" or paterfamilias. But the tendency of most Roman families that were not in the propertied elite, and that formed the majority of the population in the cities and towns of the empire, was to the continual re-forming of small "nuclear" or elementary families. The power elite, needless to say, lived in a rather different sexual and familial world, and it is the world that overwhelmingly dominates our surviving literary records. It would be a mistake to take the legally prescriptive texts of the wealthy and the powerful as offering realistic definitions of family and marriage for most persons in this society. While recognizing some of the complexity of Roman marriage and family patterns, Boswell leaves the matter there. He does not advance to an explicit recognition of the fact that, all the haziness notwithstanding, there were well-known and firmly set boundaries to the conception and the conduct of marriage in the Roman world.

IV.

To sustain his argument, Boswell must constantly tear words, sentences and larger statements out of the social and literary contexts in which they were embedded. Of this, I will give only a single example, but it is a particularly characteristic one. The essay titled Toxaris, which was composed by the second-century Greek writer Lucian, is invoked several times by Boswell to demonstrate the intense homoerotic content of unions formed between two men, and to sustain one of his arguments that such "friendships" were in fact male marriages. The problem is that a few lines or paragraphs wrenched from the context of the whole work impart a distorted picture to the reader of what is actually happening.

Lucian's essay is a fictitious dialogue between a Greek called Mnesippos and a Scythian named Toxaris on the subject of "friendship" (philia) between men. It is not just any sort of friendship, however, but precisely the sort of "ritualized" or fictive friendship between two powerful and potentially hostile men that has been cogently analyzed by Herman. Whatever homoerotic feelings and sentiments might have been part of the affective content of such a relationship, and there is no denying that in some cases homosexual men may have found emotional satisfaction in such socially sanctioned "friendship," this does not add up to an understanding of the social institution itself, and it certainly does not make it a "same-sex union" in the sense of being a marriage. It is this kind of decontextualization that permits Boswell to string together isolated pieces of evidence that lead him in the wrong direction.

The text of Lucian's Toxaris is important, since Boswell uses it as a critical proof for one of "three types of formal unions," of which "more detail is known." This union, he argues at length, was a type of same-sex union of two men, explicitly stating that it is a strict analogue to "heterosexual" marriage. The positive evidence that Boswell adduces in support of same-sex marriage practices is worth examining in detail. As the Scythian Toxaris makes clear in his summation of Mnesippos's examples of "friendship" between men in Greek societies of the time, such friendships permit these men to achieve heroic deeds. But note what such accomplishments are: marrying an ugly wife without a dowry, giving money to the amount of two talents to the daughter of a friend on her marriage, and the sharing of a period of temporary imprisonment.

When the Scythian Toxaris gives examples of male philia in his own society, he begins by noting that the Greeks "have no great occasions at all" on which to display their philia because they live in conditions of "profound peace," or eirene batheia. The Scythians, by contrast, are constantly involved in war and fighting. The words of Toxaris that follow are worth quoting not only because Boswell repeatedly avers to them, but also because they provide us with one of the closest analogues to the relationship referred to in his ecclesiastical documents:

First of all, I wish to tell you how we make our friends. Not through being drinking buddies, like you people do, not because a man has been in army cadet training with us [synephebos] and not because he's our neighbor. No. When we see a brave man capable of achieving great things, it's then that we all become eager to get close to him. And just as you behave when you are trying to get married, we think it right to behave in this same way in forming friends--wooing them at great length and with much effort, and doing everything to ensure that we do not lose their friendship and are rejected. Then when the decision has been made to accept someone as a friend [philos], an agreement is made and a great oath is sworn that we will live together and, if necessary, even die for each other. We actually do this. For when we have cut our fingers, let the blood drip into a cup, dipped our sword points into it and then, together, both at the same time have raised it to our lips and have drunk, there is nothing that happens afterwards that can possibly break the link between us. We are allowed to enter into three such alliances at most, since we think that a man who has [too] many friends [polyphilos] is like immoral and adulterous women, and we consider that the strength of his friendship is no longer the same when it has been split between [too] many loyalties [pollai eunoiai].

This is a distorting "colonialist" fiction, a Greek writer imagining what the barbarians do, and therefore a literary interpretation of the lived realities of Scythian life. Still, there is enough in the account that is reliable to make a basic assumption in it clear: that the fictional Scythian speaker is not referring to anything like same-sex male marriages, but rather to an intense social bond that was formed between men of power for the purpose of coping with the lack of formal institutions in their society and the violent behavior that pervaded it. He likens the actions undertaken by Scythian men in forming this bond to the sorts of rituals used by the civilized and peaceful Greeks in the pursuit of the women whom they wish to make their wives. But the comparison is clearly and forcefully signaled as a metaphor, as a social simile. Scythian "friendship" is clearly not marriage, since most of the men are described as having wives and children, and Toxaris's third story of such a "friendship" involves three "friends" assisting one of their own to achieve a marriage--to help the man in the successful pursuit of the woman whom he loves and wishes to wed.

Toxaris specifically states that the purposes of forming such male alliances are personal protection and violence: raiding; the conduct of wars and vengeance; the protection of possessions of herds, pasture lands and wagons; the defense of community. He provides illustrations of how such ritual friendship or philia allows men to marshal the resources of kin and friends to assemble whole armies, and to use the ascribed power of blood rituals, compacts, promises and exchanges of tokens to control resources vital to their survival. The details of this philia given by Toxaris sometimes involve three men simultaneously, though it is clear that such men also have families, wives and children and other species of property that these personal alliances were meant to protect and to subserve. These "friendships" are specifically stated as giving men power to effect aims that they would not be able to achieve on their own. But here is Boswell's recapitulation of these philia relationships between Scythian men:

They [artificial kinship relations] were often, for example, symbolic ratifications of peace, pledges of cooperation between warriors or peoples, or means of forming ties between families or tribes.... But there is absolutely no suggestion of any of these functions in Toxaris's description of the Scythian practice ... there is no indication that any other family or tribal members are involved; and the bond is not presented as political or strategic, or as having any broader social significance than personal emotional attachment. Its significance is manifestly and unmistakably personal and affectional.

Apart from Boswell's hyperbolic denials of direct statements in the Toxaris, there is little I can say except that he and I must be reading different texts.

Boswell then juxtaposes three texts from Roman legal codes and claims that they demonstrate his "third type of formal same-sex union," which involved the legal practice of "collateral adoption": one man adopted another as his brother and hence as a marital partner. These matters are legal and complex, and it is best to take them in turn. The first text (Digest 38.2.59) says nothing at all about adoption. It merely states the conditions under which a person can nominate someone as an heir to his or her property. The jurist Paul then adds that if someone is not a biological brother but you treat him as one, you may call the person by the name "brother" when you designate him as heir, and the testament will still be valid. That is all. There is nothing said about adopting the person. The second passage (Digest, 38.8.3) does indeed mention "adoptive brothers," but from the context it seems clear that the jurist is treating nothing other than the ordinary problem of siblings, one of whom might happen to be adopted, and the consequences that would follow from this fact for inheritance.

So far, therefore, there is no evidence whatever for adult males adopting other adult men as their brothers. The singular reference to such a possibility is contained in a late third-century legal decision issued by the Roman emperors (Codex Justinianus, 6.24.7). It is a solitary case referring to an eastern Mediterranean context that was not understood by the emperors themselves. In any event, they denied the legal validity of the attempt at fraternal adoption. That is it. Boswell then claims that scholars have deliberately refused to consider the institution of adult men adopting other men as their "brothers" as a formal type of homosexual union, "because it could not be honestly considered in the moral and intellectual climate of Europe or the United States in the last two centuries." "The only convincing explanation for collateral adoption," he continues, "would seem to be its peculiar personal and emotional value, about which scholars writing in the last century have shown so slight a curiosity as to border on aversion." This verges on paranoia. Except for one brief imperial legal decision, there is no evidence for the supposed institutional practice of adult male adoption as a form of same-sex marriage. The uniqueness and the lack of context of that one legal ruling, and not any malign conspiracy of silence, are more than sufficient to account for the lack of scholarly curiosity.

Boswell's analysis of the Christian church and matrimony is designed to downplay the former's concerns with family formation and marriage, and thus to reduce the latter to a more secular, contractual relationship that would be more easily transferable to "same-sex unions." He ends his discussion with an extensive analysis of the late antique and early medieval hagiographical accounts of the Roman soldier-martyrs Serge and Bacchus. The data adduced are peculiar selections that favor his own interpretation. Thus, in Boswell's words, the public parading of miscreants as a punishment either by itself, or as a prelude to execution, "does recall the penalty for homosexual acts described by Procopius, Malalas and Theophanes." Those authors alluded to the same ritualistic punishment for pederastic relations. But the fact is that this punishment was in no way especially limited to persons committing "homosexual acts." Public exposure, and the humiliation achieved by parading criminals through public streets, was habitually enforced on many types of outcasts in the Roman empire, including Christians. The use of the punishment might hint at an element of homoerotic relations between Serge and Bacchus, but not necessarily, since the entire martyrological account has them executed because they were Christians and for no other reason.

Boswell's analysis of "same-sex unions" is on firmer ground when he creates a typology that extends from males who were sexually exploited by other men because they were dominated or owned by them to free and volitional relationships between male lovers. But the leap to the formation of permanent relations that were in fact marital--"the fourth type of homosexual relationship known in the ancient world consisted of formal unions"--is not underwritten by any solid or persuasive data. Almost all his examples are ones repeated from his first book--for example, the vituperative attacks by the imperial biographer Suetonius and the historian Tacitus on the extravagant behavior of Nero, including his "marriage" to one Sporus. But neither Suetonius nor Tacitus regards this "union" as a genuine marriage, and the entire effect of their explicit condemnation of Nero's behavior is founded on the assumption that their readers share the same view.

Language, context and assumed meaning clearly show that the other Roman authors cited by Boswell, including the satirist Martial, were never remotely referring to a separate and parallel sphere of male "same-sex marriages," but rather to what they perceived as a perversion of the only known, and acceptable, type of marriage. That is what made their satire work. It was designed, after all, as an aggressive attack on the character of the disparaged person. Hence it is astounding to see Boswell reach the conclusion, on the basis of a single passage from Juvenal's Satires, that homosexual marriages were "absolutely commonplace." If that were so, Juvenal's lines would lack the wit and the mordant punch that were obviously intended by their author. Such lapses are, alas, too frequent in Boswell's attempt to provide a literary and historical basis for his argument. Boswell translates a passage from the Greek historian Xenophon as "the man and the boy live together like married people," only to admit in a footnote that "like married people is not literally expressed." In fact, the words are not in the original at all; they are gratuitously provided by Boswell.

V.

The practices and the rituals performed by the men in Boswell's documents, and also the emotional and erotic connections that are so richly described by him, may seem unusual or frightening to us, given our codes of civility, morality and masculinity. There is a nice irony here: the ancient and medieval world about which Boswell writes was not riven by the same anxieties and repressions that mark our own. In that world, public and affective bonds between men were typical, even banal. But this is not the same thing as the legitimization, or the sacralization, of homosexuality. The "new" documents that Boswell has unearthed are nothing more than a few additional texts that shed more light on a primitive and basic power linkage between men in the ancient Mediterranean, and the rituals attendant on its formation.

By the time Boswell's ecclesiastical documents celebrated or blessed this type of personal arrangement, it had been brought at least partially under the aegis of the Christian church. As the structures of the law and the civil institutions of the state became more dominant, particularly in Western Europe, the church wished to divest itself of a ceremonial that was intended to substantiate a type of personal power that, in synchrony with the state, it now excoriated. Ritualized friendship naturally survived much longer in parts of the eastern Mediterranean, and especially in the mountainous regions of the Balkans, where more primitive forms of personal power have tended to subsist. There might well have been homoerotic elements to some of these "brotherhood" relationships, and a rather alien Greek ritual may have been misunderstood by some of its Latin translators; both of those possibilities deserve more attention than they have received by historians. But same-sex marriages forged with the approval of the Christian church, and with its rituals? No. Such a reading is very misleading.

The data of the past may not be all that happy for the liberationist movements of our time. Why else would those movements come into being? But what the sources record is, for better or for worse, what the sources record. A good part of what they record, certainly, is made up of systematic and successful repressions; but tinkering with the moral balance of the past is a disservice to the study of history and to the reform of society. The past is dead. We cannot change it. What we can change is the future; but the way to a better future requires an unsentimental and accurate understanding of what happened in the past, and why. A more civil and humane modernity will not be achieved by tendentious misreadings of antiquity.