Living Tradition
Editor: Msgr. John F. McCarthy, J.C.D., S.T.D.Distributed several times a year to interested members.
Associate Editor: Rev. Brian W. Harrison, O.S., M.A., S.T.D.  Not to be republished without permission.
Please address all correspondence    e-mail:
Living Tradition, Oblates of Wisdom, P.O. Box 13230, St. Louis, MO 63157, USA

No. 11 Roman Theological Forum | Article Index | Study Program May 1987

by John F. McCarthy


        Seeming contradictions loom large in the genealogies of Jesus in Matthew and Luke. Throughout the history of the Church these apparent errors have been an object of apologetic concern on the part of many believers, of curious and sometimes condescending interest on the part of others, and of triumphant delight on the part of some unbelievers. In modern times historical criticism has found more inconsistencies in these two genealogies than ever before, and yet by the goodness of God the pristine historicity of the Gospel accounts has also in our day begun to manifest itself with new vigor and clarity.

        In The Birth of the Messiah Raymond Brown advises his readers that "the Lucan genealogy is no less theological in purpose than Matthew's and no freer of historical difficulties." 1 He observes that "the Lucan list, while in some ways more plausible than Matthew's list, scarcely constitutes an exact record of Jesus' biological ancestry." He finds that "(w)hat one may say with surety of Luke's list is that, in part, it is artificially arranged in numerical patterns of seven and that it contains enough inaccuracies and confusions to suggest a popular provenance (rather than an archival provenance) among Greek-speaking Jews." His final conclusion is: "Luke adopted this list and adapted it for theological purposes by placing it between the baptism of Jesus and his temptations. This means that, while the two NT genealogies tell us how to evaluate Jesus, they tell us nothing certain about his grandparents or his great-grand-parents. The message about Jesus, son of Joseph, is not that factually he is also (grand)son of either Jacob (Matthew) or of Eli (Luke) but that theologically he is 'son of David, son of Abraham' (Matthew), and 'Son of God' (Luke)." 2

        Brown's obviously nuanced conclusion raises certain questions of historical method. Does a theological purpose need to conflict with an accurate historical presentation? Can the cautious historian be sure that the seeming inaccuracies and confusions in the genealogies of Matthew and Luke are really what they seem? Can the careful historian be certain that these two genealogies tell us nothing about the grandparents or great-grandparents of Jesus? How can non-historical genealogies convey how to evaluate Jesus in a way that serious historians can take seriously? What is the value of a theological message that is not based on historical facts? Questions such as these might well be kept in mind as we review the age-old problem of the two genealogies of Jesus, arising from the fact that St. Joseph appears as the son of Jacob in the genealogy of Matthew and as the son of Eli in the genealogy of Luke, and subsequently that the names in Matthew's list differ almost entirely from these in Luke's list going as far back as King David.


        To the problem of the historicity of the genealogies of Our Lord in Matthew and Luke, five solutions may be considered. The two classic solutions are the theory of levirate marriage and the theory of Marian genealogy.

        1. LEVIRATE MARRIAGE.   The Law of the Levirate states: "When brothers dwell together, and one of them dies without children, the wife of the deceased shall not marry another, but his brother shall take her and raise up seed for his brother. And the first son he shall have of her he shall call by his name, that his name be not abolished out of Israel" (Deut 25:5-6).

        According to an explanation going back in essence at least to Julius Africanus in the first half of the third century A.D. (who claimed to have heard it as handed down by the relatives of Jesus), Joseph's grandmother (Estha) bore Jacob to one husband (Matthan) and Eli to a second (Matthat). Joseph's mother married Eli, who died without children; then she married his uterine brother Jacob, who raised up Joseph as seed to Eli. Thus Joseph had Eli as his legal father and Jacob as his biological father. The genealogy of Matthew shows the biological ancestry of Jesus, and that of Luke the legal ancestry. 3

        This solution is excluded by Raymond Brown, who says: "The theory of a levirate marriage solves so little and has so many difficulties that it should be abandoned as a solution in the problem of the two genealogies, and even in the more restricted problem of Jesus' overabundance of grandfathers." 4

        The difficulties as seen by Brown are four in number:

        a) Matthan and Matthat are similar names. Thus one is faced with the "dubious coincidence" that the mother of Jacob and Eli married two men who had almost the same names. But similarity of first names is not unheard of in the case of successive husbands or in the case of brothers. Hence, no historian can exclude this explanation, handed down from early times, on the mere ground that the names of the two husbands are similar.

        b) "We are not certain how widely levirate marriage was practiced in Jesus' time." But it was probably practiced, as Brown himself admits. Therefore, no historian can exclude this explanation on the ground that it couldn't have happened. It would have taken only one instance to make it happen, and history abounds with unique happenings.

        c) If Joseph were the issue of a levirate marriage, it would be "very strange" to have a genealogical list going back through his biological father. It would not be strange at all, it seems to me, especially if Matthew had a particular purpose for doing so, such as tracing the more direct line of royal descent, which, in fact, he does. A prominent example of a known levirate marriage in the same list of names, Obed, has his ancestry traced back through Boaz, his biological father (Ruth 4). And in the case of Matthew and Luke both ancestries are preserved. Hence, the explanatian of a levirate marriage cannot be logically excluded on this ground.

        d) "Why does Matthew trace descent through David's son Solomon, while Luke traces it through David's son Nathan?" The answer to this is obvious: because two different lines of ancestry go back to David, one from Matthan through Solomon, the other from Matthat through Nathan. Hence, the levirate explanation cannot be excluded on the ground of this divergence.

        The conclusion would seem to be that Brown does not present cogent grounds for abandoning the possibility of a solution through levirate marriage, since four flimsy reasons do not add up to one good reason. Difficulties and improbabilities do, indeed, lead one to be wary of accepting this solution as proven historical fact, but the historian needs real historical evidence in order to exclude it as a possible historical fact. The historian can recognize beneath the problem of the two genealogies a unique historical background created by the Law of the Levirate, whereby levirate marriage was not merely permitted, but was legally imposed with a force that is clearly expressed in Deut 25:5-10. With that background in mind, he may question, he may doubt, but he may not exclude, pending actual historical evidence to the contrary.

        In a monumental piece of research, published in French in 1982, 5 Jacques Masson reviews the ancient argument of a levirate marriage of Joseph's mother. To deepen his analysis of the last few generations in the two genealogies, he first presents a study of all the preceding generations. The wealth of material that he brings to bear on the question and the painstaking care with which he has sifted and arranged the data make his book a classic work on the subject. Masson does not prove that there was in this case a levirate marriage, but he so increases the area of discussion and so challenges the reader to continue sifting the data that his work cannot be ignored in any serious treatment of the question. Not only does this book make clear the complexity of the question, it also brings the reader closer up to the history behind the Gospel text and behind the discussion of the text that has gone on now for nineteen centuries.

        The possibility of a levirate marriage as presented by Julius Africanus was questioned by Jacques-Paul Migne, 6 Urban Holzmeister, 7 and others on the ground that the Law of the Levirate did not apply to uterine brothers, seeing that neither did a uterine brother carry the same male seed nor would he keep the heredity within the same family. Masson sees general validity in this objection, but he finds high probability in a levirate marriage of Joseph's mother with a relative of her deceased husband who had a common ancestor with him. Such a common ancestor, according to the two genealogies, would obviously be David, but more proximately he finds Salathiel and even more proximately Eliud/Esli.

        Masson argues as follows. Jechonias, having no son, adopted Salathiel, husband of his daughter and son of Neri, who was descended from David through Nathan. Salathiel thus became the legal son and successor of Jechonias. By the Law of the Levirate, the name of Salathiel'a biological father disappears from the genealogy and the name of Jechonias appears. Salathiel becomes a common ancestor of both Jacob and Eli.

        More proximately, both Jacob and Eli are descendants of Eliud/Esli. Achim of Matthew's genealogy died without children, and Naggai of Luke's genealogy begot Eliud/Esli as Achim's legal son. Matthew's Eleazar was the eldest son of Eliud/Esli. Naum was a younger son. Thus a levirate adoption by Jechonias and later levirate marriages raising up seed to Achim and to Jacob, the legal. father of Joseph, solve the contradiction of the two fathers of Joseph and the disparate lines of descent from David to Joseph. 8

        Masson's careful argumentation leading up to these conclusions should not be lightly dismissed. 9 The data that he presents can indeed be interpreted in different ways, but there is a substratum of truth that should not be ignored. The question, however, broadens at this point from an elementary base into issues related to the second solution, that of a possible genealogy of Mary, which Masson takes up in lesser depth.

        2. MARIAN GENEALOGY.   The theory of a Marian genealogy in its simplest form is based on a reading of Matthew and Luke such that Matthew presents the ancestry of Joseph and Luke presents the ancestry of Mary. The theory is based on the fact of the Virginal Conception, which both Matthew and Luke clearly present, and on the reasoning that, because of this fact, the real biological descent of Jesus is only through his mother. If Matthew gives the legal descent of Jesus, and if Luke gives the real descent, then Luke gives the genealogy of Mary.

        Raymond Brown excludes this theory completely: "What influences this suggestion is the centrality of Joseph in Matthew's infancy narrative, as compared with the spotlighting of Mary in Luke's. Even at first glance, however, this solution cannot be taken seriously: a genealogy traced through the mother is not normal. in Judaism, and Luke makes it clear that he is tracing Jesus' descent through Joseph. Moreover, Luke's genealogy traces Davidic descent and despite later Christian speculation, we really do not know that Mary was a Davidid." 10

        Brown's reasons for excluding this theory are questionable. What, indeed, may be called 'normal' Jewish genealogical practice in the case of a virginal conception, which Luke clearly presents? There is simply no precedent for this. 11 Again, since Luke says "as was supposed," it is too much to affirm that Luke "makes it clear" that he is tracing the descent of Jesus through Joseph, except in some possibly secondary or superficial sense. Again, "we do not know" (that Mary was a Davidid) cannot be used as an argument for exclusion, since ignorance of the facts has very limited probative value.

        René Laurentin, in The Truth of Christmas, 12 maintains that "the hypothesis of a Davidic filiation with regard to Mary is foreign to the two evangelists." Laurentin affirms: "Nothing is lost in Mary's not being biologically the daughter of David. The rigor with which the evangelists have avoided this easy solution gives a new indication of their exactitude. They do not invent in order to appease current expectations, as those who came after them did. On the contrary, they accepted the paradoxes which caused the difficulty. This honesty led them to a great theological profundity."

        Great theological profundity there is in the accounts of the two Evangelists, but here the question is one of historical truth, and it is a mistake to suppose that an actual biological descent of Mary from David would somehow lessen the theological depth. Basically, where the history fails, the theology fails, unless the theology is clearly independent of history. As far as some aspects of the theology are concerned, nothing would be lost if Mary's biological father were an Egyptian. That could better explain why the Holy Family fled into Egypt. But such tampering with the historical facts would cause other aspects of the theology to suffer.

        Jesus was born "from the seed (ek spermatos) of David according to the flesh" (Rom 1:3). "Remember this: the Lord Jesus Christ rose from the dead from the seed (ek spermatos) of David according to my gospel" (2 Tim 2:8). Laurentin avers that in these passages the Greek word sperma (usually translated as 'seed') does not mean 'sperm' in the restricted biological sense that this word has in French (and in English), but rather means 'posterity,' 'lineage,' or 'descent' in a very broad sense. Thus, he says, according to the Law of the Levirate, a man was to marry his brother's widow to raise up 'seed,' that is, posterity, for his brother, not for himself. "The meaning is juridical and not biological."

        Laurentin's exclusion of the biological meaning of the word sperma (seed) in these passages is not logical, because it is not an either/or situation. The extended legal or social meaning has an original and fundamental biological meaning as well. The reason especially why a brother was ordered in Deuteronomy to "raise up seed" is because brothers were carrying the same biological male seed as did their father. St. Paul says "according to the flesh," and Laurentin excludes any biological meaning; St. Paul says "rose from the dead," and Laurentin excludes any biological meaning from the seed of David. It is a question not of whether the word sperma has other meanings as well, but rather of how one can justifiably exclude the biological meaning in a context such as this.

        Again, in Acts 2:30, Peter teaches that the Prophet David spoke of Jesus as "the fruit of his loins," with reference to the Resurrection of Jesus and to the fact that "his flesh did not see corruption." According to Laurentin, this and other texts "in no way focus on the biological connections, nor on the virginal modality." But in some way they certainly do. St. Peter is explicitly speaking about the flesh of Jesus, and St. Paul is speaking about the origin of Jesus "according to the flesh."

        It is a mistake to imagine that in a given passage there must be only one meaning of a word. The fact that a word has wider implications does not exclude the particular meaning. Laurentin rules out the explicit reference to bodily processes in these passages and assumes an exclusively transferred sense of the word 'seed.' There is no way that Peter or Paul could get a biologically focused message through the barrier of this kind of interpretation, which impedes rather than produces knowledge.

        St. Augustine and other Church Fathers saw a biological meaning in these passages. Augustine reasoned that Mary had to be of the family of David, because Jesus was born "from the seed of David," and no male seed was involved in Jesus' conception. 13 Laurentin observes that any such deduction "was foreign to the Apostle Paul," who "confined himself to quoting an ancient profession of faith." Paul was a sharp thinker. To assume (without evidence) that he mindlessly repeated an ancient expression without logical reflection on his own part does not do full justice either to the mind of the Apostle or to the rules of literary interpretation.

        Laurentin "will not discuss the hypothesis of Annus [sic] of Viterbo (c. 1490), who made the Lucan genealogy apply to Mary," because "it is not maintained by anyone today." Having disposed of the theory with no discussion and no argument, Laurentin goes on to affirm: "Mt 1 and Lk 1 are silent regarding Mary's lineage, and this seems to be because they were not able to call her a descendant of David." But, according to this undiscussed theory, Luke is not silent; he need not have spoken about Mary's lineage in chapter one, if he was going to present it fully in chapter three. And reasons for the silence of Matthew can also be suggested, if, indeed, Matthew was as silent as Laurentin thinks.

        That Mary was a direct physical descendant of King David is a teaching of the Fathers of the Church. 14 Whether this fact is illustrated in the genealogies of Luke and Matthew is a more particular question, but the idea that Mary's descent can be traced directly from David has been held since ancient times. A.J. Maas testifies as follows:

        "Tradition tells us that Mary too was a descendant of David. According to Num 36:6-12, an only daughter had to marry within her own family so as to secure the right of inheritance. After St. Justin (Adv. Tryph. 100) and St. Ignatius (Eph. l8), the Fathers generally agree in maintaining Mary's Davidic descent, whether they knew this from an oral tradition or inferred it from Scripture, e.g., Rom 1:3; 2 Tim 2:8. St. John Damascene (De Fide Orthodoxa IV:14) states that Mary's great-grandfather, Barpanther, was Heli's cousin; and her father, Joachim, was a cousin of Joseph, Heli's levirate son. ... At any rate, tradition presents the Blessed Virgin as descending from David through Nathan." 15

        St. Ignatius, second successor of St. Peter to the See of Antioch, martyred in the Colosseum of Rome about 110 A.D., declared: "I offer my life's breath for the sake of the Cross, which is a stumbling block to unbelievers, but to us is salvation and eternal life. What has become of the philosopher? What of the controversialist? What of the vaunting of so-called intellectuals? The fact is, our God Jesus Christ was indeed conceived by Mary according to God's dispensation of the seed of David, but also of the Holy Spirit." 16

        St. Justin, martyred in Rome about 165 A.D., in his Dialogue with Trypho the Jew, asserts time and again that Jesus was born of the Virgin Mary, who was physically descended from King David. In one place (No. 100) he is more specific: "Therefore (Christ) revealed to us all that we have perceived by his grace out of the Scriptures, so that we know Him to be the first-begotten of God, and to be before all creatures; likewise to be the Son of the patriarchs, since He became flesh from the Virgin descended from them, and submitted to become a man without comeliness, dishonoured, and subject to suffering. ... He said that He was the Son of man, either because He had been born from the Virgin, who was, as I said, descended from the stock of David and Jacob and Isaac and Abraham: or because Abraham was the father both of Himself and of those whom I have enumerated, from whom Mary derives her descent. For we know that the fathers of women are the fathers likewise of those children whom their daughters bear." 17

        St. Irenaeus (ca. 140-202 A.D.), in his work Against Heretics, speaks of Jesus, "Who was of a virgin who was of the stock of David." Irenaeus, after reasoning that Jesus could not properly have inherited the promises through Joseph, because in the genealogy of Joseph (Mt 1:2-16) his ancestor Jechonias had been prophetically deprived of the Messianic inheritance, turns to the genealogy of Luke and speaks of the Virgin Mary. 18

        The Greek philosopher Celsus, a great opponent of Christianity, in his so-titled True Discourse, written about 178 A.D., accuses Christians of insolence for having drawn up a genealogy of Jesus that traces his descent through the kings of the Jews all the way back to the first man and declares: "And so the wife of a carpenter would not have been unaware that she arose from such great ancestry!" Thus, the pagan Celsus in the second century saw the genealogy of Luke, and possibly that of Matthew, as pertaining to the Virgin Mary, and he attacked Christians for believing that this was the descent of Mary. Origen, writing Against Celsus in about 246 A.D., does not deny that Christians believe this about Mary. Rather he says: "So what? Say that she was not unaware. What does that do to us? Say that she was unaware. Does it follow from this that she was not descended from the first man or that her lineage does not pertain to the kings of the Jews? Does Celsus think that poor people are always born from poor people and kings from kings? It seems useless, therefore, to tarry longer over these things, since it is clear even in our age that some who are poorer than Mary have been born of rich and illustrious parents, and potentates and kings born of the least noble." 19 The point to be drawn from this exchange is that in the second and third centuries Christians were attributing to Mary a genealogical descent identical in at least some elements with the genealogies of Matthew and Luke.

        St. Hilary of Poitiers (ca. 315-367 A.D.) testifies in the fourth century: "Many are of the opinion that the genealogy which Matthew lists is to be ascribed to Joseph and the genealogy listed by Luke is to be ascribed to Mary, in that, since the man is called the head of the woman. her generation is also named for the man. But this does not fit the rule or the question treated above [by Hilary], namely where the character of the genealogies is demonstrated and most truthfully solved." 20 Hence, according to Hilary, by his time "many" had maintained that Luke's genealogy presents the ancestry of the Blessed Virgin Mary.

        St. Epiphanius (315-402 A.D.) notes that Luke, in his genealogy, brings the line of descent back through Abraham and Noah all the way to Adam "in order that he who first of all (men) had been formed was sought unto salvation by Him who had been begotten from his (Adam's) substance, namely, from the Blessed Virgin. ... But finally he goes beyond Adam and says 'who was of God.' And thus it is quite clear that He was the Son of God, who came from the seed of Adam, endowed with flesh by an unbroken succession." 21

        St. John Damascene (ca. 645-750 A.D.) sees the ancestry of Mary represented in Luke's genealogy: Mary was descended from David through his son Nathan. Mary's father Joachim was the son of Barpanther and the grandson of Panther. Panther was the brother of Melchi and the son of Levi in Luke's genealogy. Therefore, all of the names in Luke's genealogy from Melchi to Adam are direct ancestors of Mary. 22

        Cornelius a Lapide, in his linear commentary, interprets Lk 3:23 as saying that Jesus was only "supposed to be" the son of Joseph, but was really the son of Eli. 23 He notes that an extrabiblical tradition identifies the parents of the Virgin Mary as Joachim and Anne. 24 How then could the father of Mary have been "Eli"? A Lapide explains that Joseph (Lk 3:23) is called the son, that is, the son-in-law, of Eli, and Eli by apocope is short for Eliachim, which is an alternative name for Joachim. To be sure, we read in 4 Kg (2 Kg) 23:34 that "Pharao Nechao made Eliakim, the son of Josias, king in place of Josias his father, and he changed his name to Joakim" (cf. 2 Chron 36:4). Again, in Judith 4 the high priest is called Joakim in the Septuagint and Eliachim in the Vulgate and in some other manuscripts; in fact, the same man is called by the Vulgate "Eliachim" in Judith 4 and "Joachim" in Judith 15:9. The etymological basis for the interchangeability is that the syllables 'Jo' (Jehova) and 'El' (Elohim) are both names of God and that both names mean "God makes firm" or "God raises up."

        Francis Xavier Patrizzi, in an elaborate treatise on the genealogies of Matthew and Luke published in 1853, 25 claims that Annius of Viterbo was the first writer who attempted to show that Eli of Lk 3:23 was the biological father of the Virgin Mary and the biological grandfather of Jesus. 26 Subsequently, many Catholic and Protestant exegetes adopted the theory, including Cornelius a Lapide, as stated above. In his treatise, Patrizzi severely criticizes this hypothesis, not for seeing the ancestry of Mary within Luke's genealogy, which Patrizzi himself upholds along with many before him beginning in ancient times, but solely and simply as understanding Eli to be the Father of the Virgin Mary. Patrizzi claims that this is an unacceptable reading of the text.

        a) Patrizzi claims that there is no example in Sacred Scripture to illustrate the shortening of Eliachim (or Eliakim) to Eli; in fact, he points out, the 'E' in 'Eli' is actually a different vowel in the Greek from the 'E' in 'Eliachim,' and, in the Hebrew 'Eli' comes from a different root altogether.

        b) Again, he says, there was no other way that a man could be called the 'son' of another man who was not his biological father except by way of levirate marriage. Joseph could not have been the adopted 'son' of Eli. The Fathers of the Church saw both genealogies as those of Joseph, and never did one writer before Annius of Viterbo seek to solve the problem by attributing Luke's genealogy to the Virgin Mary.

        c) Again, he says, Luke would not have made a parenthesis like this: "Jesus ... (as was thought, the son of Joseph), who was of Eli." Would not the sacred writer rather have said: "Jesus ... who (although he was thought the son of Joseph) was of Heli"?

        d) Again, he says, the name of Mary would have had to be mentioned by Luke. If Matthew broke the Jewish custom of mentioning Joseph, who did not beget Jesus, how much the more would not Luke have broken the Jewish custom by mentioning Mary? And yet there was no such fixed custom among the sacred writers. Matthew, a Jew writing to Jews, mentions several women in his genealogy. Why would Luke, a non-Jew writing to non-Jews, have been more careful not to mention the woman who brought forth Jesus by virginal conception? And, in fact, the Jews did trace genealogies through women (Judith, Sarvia, Abigail, Rebecca, and the daughter of Sesan).

        e) Again, grammatically, the Greeks did not write genealogies in this manner, and no early translation of the Gospels brings out a reading that links Jesus to Eli.

        Patrizzi's arguments batter the idea of a Marian link to Eli, but they do not utterly destroy it.

        a) Vowels and consonants often get changed in the popular shortening of names. In English, John (Johann) becomes Jack, Charles becomes Chuck, William becomes Will or Bill, James becomes Jim, Robert becomes Bob. Therefore, Eliachim could possibly have become Eli.

        b) A man could in Jewish custom be called the 'son' of another man also by adoption (see below).

        c) The word order of Lk 3:23 is somewhat unusual by any reading. Most translations rearrange or add words, such as "the son" or "who was" in continuing the genealogy. Patrizzi himself, in discussing the alleged parenthesis, does not present the word order correctly, which in the Greek is as follows: "... Jesus ... being son (as was thought of Joseph) of Eli, of Matthat...." By the correct order of the words (which the Vulgate gives) the parenthetical phrase is awkward but not quite as strange as Patrizzi claims.

        d) Luke was not prohibited by custom from mentioning the name of Mary, but he would not have been obliged to mention her if he had a reason for not doing so.

        e) Nor did the Greeks write genealogies exactly in the manner that Luke writes this one. A feature in the Greek text not to be ignored is the fact that Joseph is set off from Eli and the following names by not having the definite article before his name. Is this difference suggestive of a parenthesis? St. Jerome may have left this possibility open in his translation of the passage, even though he did not follow the Marian theory himself.

        What seems to follow from Patrizzi's analysis of the Marian theory is that the genealogy of Luke belongs to Joseph in the plain and simple reading of the text, but that there could be a connection with Mary in a deeper historical reading of the text, especially if an historical reason could be suggested to explain the actual manner of expression used by Luke.

        Can one correctly surmise the following from the accounts of Matthew and Luke? When Joseph went down to Bethlehem with Mary his espoused wife, who was with child, to register for the census (Lk 2:1-5), he could well have registered his ancestry more or less according to the genealogy reported in Matthew 1:2-16a. In view of the attempt shortly afterwards by Herod to murder Jesus as a possible claimant to the throne, this genealogy became a dangerous credential. Is it out of question to suppose that, in the process of returning from exile in Egypt, Joseph decided to adopt and adapt the genealogy of Mary as a new credential for himself, thus transferring it from her real ancestry to his supposed ancestry? It would have been necessary to make Mary's ancestry look like that of Joseph. If they were cousins, the two lists might have joined anyhow a certain way back. Joseph might have substituted for the name of Mary's father, Joachim, the equivalent name Eliachim, and shortened it to Eli. This would not have been a falsification, but it would have been enough of a change to throw suspicious outsiders off the track. And it is not inconceivable that one or two other immediate ancestors of Mary (Matthat and Levi) might have been dropped from this legal credential or listed according to their substitute names.

        There is no actual evidence that such an adaptation was made by Joseph, but the real and established threat to the life of Jesus as a descendant of the Davidic monarchs might have constituted a motive for doing this over and above the concealing to outsiders of the fact of the Virginal Conception. If the adaptation was made by Joseph, then Luke is presenting the adapted and merely supposed genealogy of Joseph, and beneath the adapted genealogy, concealed for historical reasons, is the real genealogy of Mary.

        3. LEGAL ADOPTION.   Urban Holzmeister transforms a Lapide's theory into a theory of adoption. The reading of Lk 3:23 as directly presenting a biological bond of Jesus with Eli, to the exclusion of Joseph, he finds to be unconvincing and in violence to the text. But, he says, if Mary was an only child, as we have every reason to believe, it would have been entirely in keeping with Old Testament law and custom for her father to adopt her husband and transfer to him all of his rights and possessions. If this happened, then the genealogy of Luke could well be materially the genealogy of Mary, but formally (and gramatically) the genealogy of Joseph, who had inherited Mary's ancestry from her father by way of adoption. Hence, all of the names in Luke's genealogy beginning with Eli are ancestors of Mary, but she is not named. He finds a precedent for this kind of adoption in 1 Chron 2:34. Sesan had no sons, so he gave his daughter (unnamed) in marriage to his Egyptian servant Jeraa, and she brought forth to him (Sesan) a son named Ethei. Thus Ethei was the son of Sesan through his unnamed daughter and his adopted (and named) Egyptian son-in-law.

        The theory of "special adoption" proposed by Holzmeister and others replaces the theory of "generic adoption" held for a time by St. Augustine of Hippo which contemplated the (childhood) adoption of Joseph by Eli. St. Augustine withdrew this theory in favor of the theory of levirate marriage after he had seen and studied the solution of Julius Africanus. 27

        Jacques Masson rejects Holzmeister's theory of the adoption of St. Joseph by the father of the Blessed Virgin Mary on the ground that Holzmeister resorted to this to save the historicity of the text, since he was convinced that the Law of the Levirate could not have applied to uterine brothers. Masson resolves Holzmeister's problem with the provision of Jewish law that the two successive husbands of Joseph's mother need not have been uterine brothers but rather could have been close relatives descended from a common male ancestor. Nevertheless, Masson does not really exclude the legal adoption of Joseph as a possibility. In fact, Masson readily admits the possibility in general of the adoption of a son-in-law in the Jewish law and customs of the times, and he includes an instance of it (Salathiel) in his own explanation. 28

        Holzmeister's idea that Joseph was adopted by Eli, the father of Mary, is untenable in Masson's estimation, because, according to St. John Damascene, Joachim (son of Barpanther, son of Panther, son of Levi) was Mary's father. But Patrizzi, studying the same testimony of John Damascene, concluded that Joseph was Mary's uncle; 29 and Masson does not refute Patrizzi's reasoning. The data can be interpreted differently, and Eli may even be Joachim.

        If Mary was an only child, Masson cannot logically exclude that her father might have arranged her marriage to her cousin Joseph and then adopted him in keeping with Jewish law and custom. But that would make the genealogy in Luke the real ancestry of the Blessed Virgin Mary, as Holzmeister maintains. The two theories tend to merge in the sense that even for Masson the names in Luke's genealogy from Levi, father of Matthat, all the way back to Adam are also the ancestors of Mary. Only Eli and Matthat are exclusive ancestors of Joseph in the theory of Masson.

        Here, again, the idea of levirate marriage appears in a more sublime way. Mary conceived without male seed by the intervention of the Holy Spirit. Thus was raised up a descendant to Eli (Joachim), to David, and to Abraham, Isaac, and Jacob according to the promises. Joseph appears as adopted son-in-law of Eli (Joachim) and as servant of the Holy Spirit, inheriting the promises in a spiritual way.

        4. CONSANGUINITY.   St. Jerome affirms that Mary was a close relative of Joseph, and, therefore, a member of the tribe of Judah and of the family of David. This testimony is reinforced by St. Ambrose, Tertullian, St. John Chrysostom, and St. John Damascene. The Fathers of the Church almost unanimously defend the Davidic descent of Mary. 30

        Cornelius a Lapide expounds an elaboration of the Marian theory whereby both genealogies present the ancestry of Mary. Still commenting on Lk 3:23, he affirms that, while Eli (of Luke's list) was the father of Mary, Matthan (of Matthew's list) was the biological grandfather of both Joseph and Mary, because Jacob was the brother of Anne, the mother of Mary. By a Lapide's theory, Matthew gives the ancestry of Mary through her mother Anne, but Anne's brother Jacob is mentioned in the list instead of Anne, just as Joseph is mentioned instead of Mary. Thus the customary male genealogical tenor is observed and the maternal ancestry of Mary is retained. In this way, concludes a Lapide, both genealogies express the real (biological) ancestry of Jesus.

        In defense of the idea that Mary and Joseph could have been cousins, a Lapide notes that according to Num 36:6-10 women who are heiresses of their parents are ordered to marry, not only within the same tribe, but also within the same kindred and closely-related family, lest the inheritance pass to outsiders. Mary seemingly was an only child.

        A Lapide maintains that, by their respective genealogies, Matthew and Luke show that Jesus was son and heir of David by a double title, by descent from Solomon, who reigned after David, and by descent from Nathan, who was next to Solomon in order to the throne. He cites Ambrose, Jerome, Theodoretus, Jeremiah, Bernard, and Suarez as holding that the Blessed Virgin Mary was a descendant of David through Solomon and therefore, he says, through the genealogy of her mother, as presented by Matthew.

        Patrizzi maintains that both genealogies are those of Joseph by levirate marriage; yet they both reflect the ancestry of Mary and the biological descent of Jesus from King David because of the blood-relationship between Mary and Joseph. St. John Damascene, Andrew of Crete, Hugo Grotius, and others had presented explanations to show this which Patrizzi finds to be inexact. He prefers the explanation of Possini, Zaccaria, and others according to which Joseph was actually the paternal uncle of Mary and the brother of her father Joachim. He finds this explanation to be in conformity with the data on Mary's family provided by St. Epiphanius and also with the testimony of Julius Africanus, who says that Joseph was the third son of Jacob, and with some very early Christian records. Thus Joachim, Cleophas, and Joseph were the three sons of the last Jacob in Matthew's genealogy. Marriage of an uncle with his niece would seem to have been forbidden by Jewish law, but Patrizzi maintains that some exceptions were made, especially by way of the Law of the Levirate, since Joachim and Joseph were brothers. He cites also the case of Aristobulus, King of the Jews, whose uncle Absalom became his son-in-law. 31

        Fillion in the early twentieth century agreed with a Lapide that St. Anne, the mother of the Virgin Mary, was the sister of Jacob and the aunt of St. Joseph. Masson, after further genealogical studies, concludes that Mary and Joseph were second cousins on her father Joachim's side, inasmuch as Levi in Luke's genealogy was the great-grandfather of Joseph and the great-great grandfather of Mary. But Masson also agrees with a Lapide that St. Anne was the sister of Jacob and the aunt of St. Joseph. Therefore, according to Masson, Mary and Joseph were first cousins on her mother's side. 32

        5. HISTORICAL RESERVATION.   A fifth possible solution to the problem of the two genealogies stems from the laws of historiography as related to the text of Matthew and Luke.

        a) Mt 1:1-17. Matthew presents his genealogy with the words, "The book of the generation (biblos geneseos) of Jesus Christ, son of David, son of Abraham" (Mt 1:1). The Greek word biblos means a written tablet (as does the corresponding Hebrew word seper). Therefore, the "book of the generation" referred to by Matthew seems to be a written genealogical record to which he is referring - a document that he is quoting. This is therefore an explicit citation of a document.

        According to the laws of historiography, an historian has diminished personal responsibility for the historical accuracy of an account which he explicitly quotes, and he has no responsibility if he dissociates himself from the veracity of its contents. Now Matthew in chapter one sets up an explicit contrast between the record of the generation of Jesus that he is quoting (Mt 1:2-16a) (which, he says, does not really link up with Jesus) and the way in which the generation of Jesus really took place (Mt 1:18-25). Therefore, it is possible that Matthew did not intend to guarantee the historical accuracy of the genealogy or that he even implies that it is inaccurate.

        b) Lk 3:23-38. Luke begins his genealogical presentation with the words, "as was supposed." Now, it is clear that Luke is saying that Jesus was only supposed to be the son of Joseph, whereas in biological fact he was not. But it is possible that the phrase "as was supposed" applies as well to other links in the genealogy, or even to the whole genealogy. Thus, Luke may not be guaranteeing the historical accuracy of the genealogy which he presents; in fact, he may be implying that the genealogy is not historically accurate.

        From these indications regarding Matthew and Luke, it is possible to retain the complete historical accuracy of the two inspired writers according to their express intention, and at the same time to consider possible or probable inaccuracies and confusions in the lists of names themselves. Analysis of the texts of Matthew and Luke according to the principles of true historiography does not, on the one hand, exclude the possibility that one or both of these sacred writers, in presenting a genealogical list, intended to present doubtful or erroneous names to contrast with what they were really affirming about the origin of Jesus. On the other hand, textual analysis does not show unambiguously that either Evangelist did in fact intend not to guarantee his list or that either list is in fact to some degree historically inaccurate.


        Pending more conclusive evidence to the contrary on the part of exegetes, all of the following appear to be historically viable possibilities regarding the relationship of the two genealogies of Jesus in Matthew and Luke:

        a) Both Matthew and Luke give true historical genealogies of Joseph (by levirate marriage or by legal adoption);

        b) Matthew gives the true historical genealogy of Joseph and Luke gives the true historical genealogy of Mary (by bracketing Joseph in Luke's genealogy);

        c) Matthew gives the true historical genealogy of Joseph and Luke gives the true historical genealogy of both Joseph and Mary (by Levirate marriage or legal adoption together with consanguinity);

        d) Both Matthew and Luke give the true historical genealogies of both Joseph and Mary (by levirate marriage or legal adoption together with consanguinity);

        e) Matthew gives the true historical genealogy of Joseph and Luke gives a non-historical genealogy of Joseph (by historical reservation);

        f) Luke gives the true historical genealogy of Joseph and Matthew gives a non-historical genealogy of Joseph (by historical. reservation);

        g) Both Matthew and Luke give non-historical genealogies of Joseph (by historical reservation).

        In the various solutions proposed to the problems of the two genealogies of Our Lord, there seems to be a certain underlying tendency towards convergence. Both the theory of levirate marriage and the theory of special adoption have tended to include the factor of consanguinity between Mary and Joseph. Joseph, the son of a levirate marriage, could later have been adopted by the father (Joachim) of the Blessed Virgin Mary, his spouse.

        The Marian reading of Luke's genealogy is weak in the plain reading of the text, but it converges with the levirate and the adoption theories after two or three generations, because of the factor of consanguinity. Therefore, there is a deeper Marian meaning beneath the genealogy of Luke and possibly also of Matthew. The first four theories are all saying ultimately that the genealogies of both Matthew and Luke are presenting the ancestry of Mary.

        What comes forth from a consideration of all of the theories is the split-level meaning of the genealogies, even in their literal sense. The fact that the genealogies are of Joseph does not mean that they are not also of Mary. The solution by historical reservation is important in the study of the purpose of the genealogies. It gives a certain freedom to the sacred writers in the way in which they have selected the names and arranged the lists. But the mass of historical evidence prevents the interpreter from depending too much on an a priori exclusion of historical accuracy. As the purposes of the genealogies are examined, the five solutions should present themselves in a fuller light.


1. Raymond E. Brown, The Birth of the Messiah (Doubleday, 1977), p. 90.

2. Ibid., pp. 93-94.

3. Julius Africanus, Letter to Aristides, in Migne, Patrologia Graeca, vol. 10, col. 64. Cf. Eusebius of Caesarea, in Migne, PG, vol. 20, col. 93; PC, vol. 22, col. 901. Julius Africanus has Eli as the son of Melchi, rather than of Matthat. See the Catholic Encyclopedia, (1913 edition) p. 411.

4. Brown, The Birth of the Messiah, p. 504.

5. Jacques Masson, Jésus, Fils de David dans les Généalogies de Saint Mathieu et de Saint Luc (Paris: Téqui, 1982). This is the publication of a doctoral thesis presented at the University of St Thomas in Rome in 1979. René Laurentin was one of the first to recognize and proclaim the importance of this work. The massive and painstaking research that Masson here presents on the two genealogies of Jesus, carried out according to the methods of modern genealogical inquiry, so expands the data and is so carefully put together that it should be considered essential for any further development of the discussion. It should be published in English translation at the earliest opportunity.

6. Note by J.-P. Migne in PG, vol. 20, cols. 98-100.

7. U. Holzmeister, "Genealogia S. Lucae," in Verbum Domini (1943), [pp. 9-18], pp. 11-12.

8. Masson, op. cit., (p. 456), sees a probability that Eli was the biological father of Joseph, and Jacob was the legal father. For this he relies on his genealogical analysis and especially on the general assumption that Matthew shows the legal descent of Joseph in relation to the monarchy.

9. Raymond Brown ("Gospel Infancy Narrative Research from 1976 to l986: Part I" in the Catholic Biblical Quarterly, July, 1986, p. 479) finds that Masson's research displays a "passion for demonstrating scriptural accuracy" and is "largely futile." Brown wonders "what theological difference it would make if the lists are not reconcilable," or "what historical difference so long as Jesus is the (legal) son of Joseph who was a Davidid (on which both genealogies agree)." But Jacques Masson's work will not be futile, if the results of his research are properly taken up and developed. A passion for demonstrating Scriptural accuracy is a better attribute than a passion for demonstrating questionable Scriptural inaccuracies. Theological interest of the Evangelists does not eliminate the need of an historical base, or at least the need of an historical context, relating to what they write.

10. Brown, The Birth of the Messiah, p. 89.

11. Apart from the singularity of the Virginal Conception, there are some precedents for the tracing of the genealogy of a woman. Thus, in Judith 8:1 the genealogy of Judith is given. Compare this with Lk 2:36: "Anna the prophetess, daughter of Phanuel of the tribe of Aser." See also note 27 below (second sentence).

12. R. Laurentin, The Truth of Christmas beyond the Myths (St. Bede's Publications: Petersham, Massachusetts, 1986), pp. 342-345 (French original: Les Evangiles de 1'Enfance du Christ, pp. 403-404).

13. St. Augustine, Contra Faustum, Patrologia Latina 42, cols. 471-472.

l4. "There is little doubt that the Savior's Davidic descent was part of the primitive kerygma, and it was eventually incorporated into the written gospel. ... (On the supposition that Luke's genealogy is that of Mary) we should read Lk 3:23 as follows: 'And Jesus Himself, when He began His work, was about thirty years of age, being - as was supposed - the son of Joseph [but in reality the grand-] son of Heli....' ... The real difficulty with the Marian hypothesis, as with that of Africanus, is that neither explains how Salathiel and Zorobabel appear as descendants of David in Luke's genealogy. It is useless to invoke the solution of a levirate marriage again, for we know the names of all the sons of Jechonia and none of them is called Neri (cf. I Chr 3:17-18)" (J. Edgar Bruns, "Genealogy of Jesus," in the New Catholic Encyclopedia, vol. 6, pp. 319-321). Masson and others have presented solutions to this "real difficulty." See above.

15. A. J. Maas, "Genealogy of Christ," in the Catholic Encyclopedia (19l3 ed.), vol. 6, p. 411.

16. Ignatius Martyr, Letter to the Ephesians, ch. 18, in PG 5, col. 660. Trans.: James A. Kleist, ed., Ancient Christian Writers, vol. 1, p. 67.

17. Justin Martyr, Dialogue with Trypho the Jew, in PG, vol. 6, col. 709. Cf. Alexander Roberts-James Donaldson eds., The Ante-Nicene Fathers (Grand Rapids: Eerdmans, 1979), p. 249.

18. Irenaeus, Adversus Haereses, in PG, vol. 7, cols. 951-958.

19. Origen, Contra Celsum, in PG, vol. ll, col. 852.

20. Hilary of Poitiers, in Angelo Mai ed., Novae Patrum Bibliothecae (Rome: Propaganda Fide, 1852), pp. 477-478. The editor notes that the text in which Hilary claims to have refuted the Marian theory is not extant.

21. Epiphanius, Adversus Haereses, bk. 2, Tome 1, Heresy 51, No. 11; in PG, vol. 41, col. 908.

22. John Damascene, De Fide Orthodoxa, in PG, vol. 15, cols. 1155-1157.

23. Cornelius a Lapide (Cornelius van den Steen: 1567-l637 A.D.), Great Commentary, commenting on Lk 3:23. Among those who have held or at least have been open to the possibility that Luke gives the genealogy of Mary, a Lapide lists Augustine of Hippo, Denis the Carthusian, Cajetan, Peter Canisius, Melchior Cano, Dominic Soto, Francis Suarez, and others. Holzmeister, op. cit., pp. 10 and 14, adds the names of six ancient writers (Justin Martyr, Celsus, Origen, Irenaeus, Epiphanius, and John Damascene) and eleven twentieth-century writers up to that time (1943) who held the theory: p. Vogt, J.M. Heer, V. Hartl, J. Pfättish, E. Mangenot, p. Pous, R. Riezler, Th. Innitzer, Simon-Prado, J. Geslin, and E. Ruffini. Jacques-H. Vosté, in De Conceptione Virginali Jesu Christi (Rome: Collegio Angelico, 1933), p. 100, lists nine late nineteenth and twentieth-century writers who hold the theory, including B. Weiss, A. Capecelatro, P.H. Didon, and E. Le Camus.

24. Joachim and Anne are presented as the parents of the Blessed virgin Mary in the Protoevangelion of James (second century A.D.).

25. Francis Xavier Patrizzi, De Evangeliis (Freiburg in Breisgau, 1853), vol. 2, pp. 82-105. Regarding Annius of Viterbo: pp. 84-91.

26. Annius of Viterbo (Giovanni Nanni: born in Viterbo in 1432 - died in Rome in 1502) wrote Antiquitatum variarum volumina XVII, a collection in seventeen volumes of fragments of ancient writings, all of which are apocryphal according to the Grande Dizionario Encyclopedico Utet (1967), art. "Annio." It is not known whether Annius was himself a falsifier or accepted these documents in good faith. In any case, he does not make a very reputable "founder" of the theory of Marian genealogy, and Patrizzi tears his historical references to shreds. But did Annius originate this theory? According to the indication given in footnote 23 above and elsewhere in this article, he did not. Annius presents his theory about the genealogy of Mary in his commentary on the Breviarium de Temporibus of Philo the Jew.

27. Holzmeister, op. cit., pp. l5-18. Holzmeister maintains that the theory of general adoption is very weak, but the special adoption of one's son-in-law by a man without sons is highly tenable. Holzmeister points out indications of such special adoption also in Esd 7:63 and in Num 32:41 (taken together with 1 Chron 2:21 and Gen 50:22). Augustine discusses the problem of the two genealogies in nine different places. See Holzmeister, in Zeitschrift fur Katholische Theologie 47 (1923), pp. 205-209. Augustine presents the theory of general adoption in PL, vol. 34, cols. 1072-1073; he withdraws the theory in his Retractationes PL, vol. 32, cols. 632-633.

28. Masson, op. cit., pp. 437-439, 347.

29. Patrizzi, op. cit., p. 98.

30. Jerome (345-420 A.D.), Super Matthaeum, commentary on Mt 1:18: (in PL, vol. 26, col. 24): "Since Joseph is not the father of Our Lord and Savior, how does a genealogy coming down to Joseph pertain in any way to Our Lord? To this we reply, first, that it is not the custom of the Scriptures that the category of women be woven into genealogies; secondly, that Joseph and Mary were from one tribe: whence by law he was obliged to receive her as a near relative, and they were registered together in the census in Bethlehem, as having sprung from one tribe." So also, for testimonies of Irenaeus, Ambrose, and Tertullian regarding the consanguinity of Mary and Joseph, see H. Didon, Jesus Christ (Paris, 1891), p. 416.

31. Patrizzi, op. cit., pp. 99-103. See Josephus, Jewish Antiquities, bk. 14, ch. 4.

32. Masson, op. cit., pp. 494 and 498. Cf. C. Fillion, Vita di Gesù Cristo (11th ed. Turin, 1940), pp. 105 ff.

Go to: Roman Theological Forum | Living Tradition Index | Previous Issue | Next Issue