The Brothers and Sisters of Jesus: An Epiphanian Response to John P.

Meier
Author(s): RICHARD BAUCKHAM
Source: The Catholic Biblical Quarterly, Vol. 56, No. 4 (October, 1994), pp. 686-700
Published by: Catholic Biblical Association
Stable URL: http://www.jstor.org/stable/43721789
Accessed: 07-07-2016 04:52 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

Catholic Biblical Association is collaborating with JSTOR to digitize, preserve and extend access to The
Catholic Biblical Quarterly

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 The Brothers and Sisters of Jesus:
An Epiphanian Response
to John P. Meier

RICHARD BAUCKHAM
University of St. Andrews
St. Andrews, Fife KY16 9AJ
Scotland

Both in the first volume of John P. Meier's book A Marginal Jew, 1 and
in his presidential address to the Catholic Biblical Association in 1991, pub-
lished in this journal,2 Meier has discussed once again the controversial issue
of the brothers and sisters of Jesus. His two discussions are identical, except
that in the latter he relates the purely historical conclusion he reaches in the
former to the doctrinal and ecumenical issue of "the hierarchy of truths." The
conclusion he reaches with regard to the historical evidence is, at the same
time, appropriately cautious as to the degree of certainty which is possible
in such a case and quite definite as to the most probable conclusion:

Needless to say, all of these arguments, even when taken together, cannot pro-
duce absolute certitude in a matter for which there is so little evidence. Never-
theless - if prescinding from faith and later church teaching - the historian or
exegete is asked to render a judgment on the NT and patristic texts we have
examined, viewed simply as historical sources, the most probable opinion is that
the brothers and sisters of Jesus were true siblings.3

In other words, of the three views which have been held since at least the
fourth century, he thinks the Helvidian view (that the brothers and sisters of

1 J. P. Meier, A Marginal Jew: Rethinking the Historical Jesus, 1 (AB Reference Library;
New York: Doubleday, 1991) 316-32.
2 J. P. Meier, "The Brothers and Sisters of Jesus in Ecumenical Perspective," CBQ 54
(1992) 1-28.
3 Meier, "Brothers and Sisters," 26 = Marginal Jew, 331.

686

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 687

Jesus were children of Joseph and Mary) more probable than the Epiphanian
view (that they were children of Joseph by a previous marriage) or the Hierony-
mian view (that they were not children of Joseph or Mary but were cousins
of Jesus).
My own discussion of this issue,4 written before I saw Meier's, was
published after his book A Marginal Jew. I reached a different conclusion:
that although the Hieronymian view is very improbable, the historical evi-
dence is not decisive in favor of either the Helvidian view or the Epiphanian
view. Meier's article was written after he had read my book; because in his
article he reproduces his earlier discussion unchanged, he does not engage
with my arguments in detail but only responds to them briefly in additional
or expanded footnotes. In the present article I shall discuss and reject his
reasons for thinking the Helvidian view more probable than the Epiphanian.
I shall also extend my discussion by adding a line of argument for the Epi-
phanian view which has not been suggested before. This additional argument
may tip the balance of probability slightly in favor of the Epiphanian view.
An unfortunate feature of Meier's discussion is that he tends to con-
struct the issue as a debate between the Helvidian view on the one hand and
the other two views on the other hand, and to present the arguments as
though the same arguments may serve to refute both the Epiphanian and
Hieronymian views and to support the Helvidian view against both. As a
result, the arguments are in fact directed primarily against the Hieronymian
view and give most attention to it. The impression given is that the Epipha-
nian view can be quite easily discounted. In my opinion, it is the Epiphanian
view which is the serious alternative to the Helvidian. By not taking the
Epiphanian view sufficiently seriously, Meier has missed the fact that it is not
vulnerable to the most cogent criticisms he makes of the Hieronymian view.

I. Clarifying the Issue

Before addressing Meier's arguments against the Epiphanian view it is
necessary to clarify the issue more carefully than he does. In the first place,
in the interests of isolating the strictly historical issue, it is important to
distinguish the question of the parentage of the brothers and sisters of Jesus
from the question of the perpetual virginity of Mary. Of course, the debate
about the fomer, at least since the fourth century, has been intimately con-
nected with the latter. The Helvidian view necessarily entails denial of the
perpetual virginity. The Epiphanian view has commonly been combined with

4 R. Bauckham, Jude and the Relatives of Jesus in the Early Church (Edinburgh:
T. & T. Clark, 1990) 19-36.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 688 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

belief in the perpetual virginity, but although the Epiphanian view (like the
Hieronymian view) of the brothers and sisters of Jesus makes the perpetual
virginity a possibility, it does not entail it. My own historical argument for
the Epiphanian view of the brothers and sisters of Jesus should not be taken
to be an argument for the perpetual virginity of Mary, for which I think there
is no good historical evidence.5 In the absence of such evidence, I find it more
natural to assume that, if the Epiphanian view is correct, Mary and Joseph
had normal marital relations after the birth of Jesus but produced no children.
This is, after all, not an unusual phenomenon.
Failure to distinguish clearly between the question of the parents of Jesus'
brothers and sisters and the question of the perpetual virginity of Mary leads
Meier to treat evidence from Irenaeus (Adv. haer. 3.21.10; 3.22.4) which may
indicate that Irenaeus did not believe in the perpetual virginity as though it
were evidence that Irenaeus held the Helvidian view of the brothers and sisters
of Jesus.6 But it is one thing to assume that Joseph and Mary had sexual
relations, another to think that Jesus' brothers and sisters were offspring of
Joseph and Mary. The close connection between these two points, which was
forged by later controversy about the virginity of Mary, should not be read
back into Irenaeus without further argument.
Second, the relationship between the Epiphanian view and the virginal
conception is more problematic, but it is even more important to clarify. The
Epiphanian view, as it has been traditionally held, presupposes the virginal
conception. In this case, it means that the brothers and sisters of Jesus were
not blood relatives of Jesus at all. But it is possible to combine the Epiphanian
view with denial of the historicity of the virginal conception. In this case, it
means that the brothers and sisters of Jesus were his half brothers and half
sisters, sharing with Jesus a common father (Joseph) but not a common
mother. In this case, they were as closely related to Jesus as they were
according to those who combine the Helvidian view with belief in the virginal
conception as a historical fact. There is an inconsistency in Meier's argument
at this point. His own discussion of the virginal conception claims that the
historical evidence is inconclusive,7 yet he equates the Epiphanian view
with the view that the brothers and sisters of Jesus were "stepbrothers" and
"stepsisters" (i.e., that they had no biological parent in common with

5 This statement may serve to show that 1 have no doctrinal investment in the issue of the
brothers and sisters of Jesus. I neither believe in the perpetual virginity nor have any particular
hostility to it. I do not even think the question at issue between Meier and myself has any
historical importance, since it does not seem to have any significant bearing on other historical
issues. But it raises interesting questions of historical method.
6 Meier, "Brothers and Sisters," 25-26 = Marginal Jew, 330-31.
7 Meier, Marginal Jew, 220-22.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 689

Jesus). On his own strictly historical premises, his argument should have
allowed for the possibility that they were half brothers and half sisters of
Jesus, sharing a common father.
The point is relevant to exegesis. Exegetes who do not regard the virginal
conception as historical fact almost always think it was unknown to Paul and
Mark and often think it was unknown to John or not accepted by him. This
means that, supposing these authors knew that Jesus' brothers and sisters
were children of Joseph by a previous marriage, they would have considered
them half brothers and half sisters, not, as Meier seems to presume, step-
brothers and stepsisters. Even Matthew and Luke, who do not represent the
virginal conception as a matter of common knowledge to people in Jesus'
lifetime, would have supposed, on the Epiphanian view, that Jesus' contempo-
raries thought his brothers and sisters were his half brothers and half sisters,
just as they mistakenly thought Joseph his biological father. Furthermore,
even exegetes who accept the historicity of the virginal conception do not
usually think it was common knowledge during Jesus' lifetime, and they by
no means always think it was known to Paul and Mark. Therefore, whether
or not the virginal conception is considered historical, the Epiphanian view
means that in many of the texts the brothers and sisters of Jesus may be
regarded as his half brothers and half sisters.
Finally, however, we need to ask precisely what relationship between
Jesus and his brothers and sisters is implied by the Epiphanian view if the
virginal conception is presupposed. This would be what Matthew and Luke,
at least, took to be the relationship if they held the Epiphanian view. Meier
is quite correct to insist on the distinction between this relationship and that
of half siblings,8 but his use of the terms "stepbrother" and "stepsister" for
this relationship is potentially very misleading. Although I myself used these
terms in the same way in my book, I now regret doing so.
Relationships between stepfathers and stepchildren and relationships
between stepsiblings have real social meaning and legal implications only in
a society where the children of parent A by a first marriage and the children
of parent B by a first marriage become the joint family of A and B when A
and B marry. It is doubtful that this normally happened in Jewish society.9
In a society where patriarchal lineage is what counts, the children of a woman
by her first husband would not become part of her second husband's family

8 Meier, "Brothers and Sisters," 6 n. 9 = Marginal Jew, 354 n. 1 1.

9 This may account for the fact that, so far as I have been able to tell, there is no instance
of stepsibling relationships in prerabbinic Jewish literature. Lev 18: 1 1 probably refers to a man's
stepsister (his father's wife's daughter), but the LXX, like many modern scholars, understands
it to refer to a man's half sister (his own father's daughter).

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 690 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

on her remarriage after her first husband's death;10 they would continue to
belong to her first husband's wider family. The only steprelationship having
social reality would be that between stepmother and stepchildren. Thus, if
Joseph had been Jesus' stepfather in the ordinary sense of the word - i.e., if
Jesus had been Mary's son by a previous marriage - he would probably not
have been regarded as Jesus' father in any sense. But Matthew and Luke do
not in fact portray Joseph as related to Jesus in the way that a stepfather would
be. Since Jesus had no biological father, the relationship is strictly unique, but
it is much more like an adoptive relationship than like a steprelationship.
Since Jesus was conceived and born when Mary and Joseph were either
betrothed or married, and since Mary was not accused (let alone convicted)
of adultery by Joseph or by anyone else, Joseph in effect accepted Jesus as
his own son. Legally, and in public opinion, he was Jesus' father (Matt 13:55;
Luke 2:41,43,48; 3:23; 4:22), as he would not have been had he been Jesus'
stepfather. Therefore, if Joseph had children by a previous marriage, Jesus'
relation to them would have much more reality than the relation of a step-
sibling would. He would be their adoptive brother. As we shall see, this point
is of great importance to Meier's linguistic arguments, to which we now turn.

II. The Linguistic Argument

In the NT and other early Christian literature (as well as Josephus, Ant.
20.9.1 §200) the word used to describe the relationship of Jesus' brothers to
him is invariably áôetapóç, and on the few occasions on which his sisters are
mentioned a5etapf| is used. The question is, therefore, whether the Epipha-
nian view is consistent with this usage, as the Helvidian view unquestionably
is. Much of Meier's linguistic argument is directed against the claim that the
Hieronymian view is consistent with this usage, but he also firmly denies that
the Epiphanian view is. His argument is simply that in the NT àSsX(póç and
á5etapf| cannot mean "stepbrother" and "stepsister."
Examining the use of áSsXtpóç in the NT, and prescinding from the
disputed case of the brothers of Jesus, Meier claims that when the word is
used literally of family relationships, not figuratively or metaphorically (e.g.,
of fellow Christians or fellow countrymen), it always means "full brother,"
except in one case (Matt 14:3 || Mark 6:17 || Luke 3:19), where it means "half
brother." This he considers sufficient evidence to exclude the possibility that
it can mean anything else in the case of the brothers of Jesus:

10 This probably explains why the OT never mentions children by the first marriages of
famous widows who remarried (Ruth, Abigail, Bathsheba).

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 691

In the NT adelphos, when not used merely figuratively or metaphorically but

rather to designate some sort of physical or legal relationship, means only full
or half-brother, and nothing else. Outside our disputed case, it never means
step-brother (the solution of Epiphanius), cousin (the solution of Jerome), or
nephew. When one considers that adelphos (in either the literal or metaphorical
sense) is used a total of 343 times in the NT, the consistency of this "literal" usage
is amazing. To ignore the strikingly consistent usage of the NT in this regard . . .
and to appeal instead to the usage of koine Greek in various Jewish and pagan
texts cannot help but look like special pleading."

This is in more than one respect an odd argument. We may notice at once
that the total number of occurrences of àSeXcpóç in the NT (in both literal
and metaphorical senses) is quite irrelevant to the consistency of the literal
usage. Meier omits to tell us that, of the 343 occurrences of àSsXxpóç, 268 are
in a metaphorical sense, and only 75 refer to literal family relationships (14
of these to the brothers of Jesus). But this is a relatively minor flaw in the
argument.
What is quite extraordinary is the assertion that the general NT usage
of a word exclusively determines its meaning in particular instances in the
NT, excluding meanings which are attested in literature outside the NT.12
This is to treat the NT as though it were written in some kind of linguistic
ghetto with a range of linguistic usage all its own, whereas, of course, it is
obvious that NT writers were free to exploit whatever range of meaning a
word had in their linguistic environment. When Paul wrote of the brothers
of the Lord (Gal 1:19; 1 Cor 9:5 - the only two occurrences of àSstapóç in a
literal sense in Paul's writings), his meaning was not determined by the use
of àSs tapóç in other NT writings, none of which were known to him. Of
course, it is true that the early church developed its own semitechnical use of
vocabulary for its own religious practices and ideas. The use of àSstapóç to
mean "fellow Christian" is an instance of this. But in the ordinary, literal use
of àôstapóç Christian writers participated in the common linguistic milieu of
their Greek-speaking contemporaries. If à5eX(póç in that linguistic milieu
could be used of family relationships other than full brother and half brother,
then it could also be so used by NT writers. The fact that they generally used
it in its most common sense of full brother cannot exclude the possibility of
their using it in less common senses on particular occasions. Moreover, in
relation to the meaning "stepbrother" which Meier thinks is excluded by the
consistency of NT usage, this consistency would be "amazing" only if there
were occasions on which NT writers referred to stepbrothers in other ways.

11 Meier, "Brothers and Sisters," 6 n. 9 = Marginal Jew, 328.

12 J. Murphy-O'Connor, review of Meier, Marginal Jew ; RB 99 (1992) 782, makes the
same criticism in milder terms.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 692 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

Since it is unusual to refer to stepbrothers at all, there is nothing amazing

about the NT writers' failure to use àSsXyóc, in this sense outside the dis-
puted case.
To realize how extraordinary Meier's argument is, we need only to notice
the effect it would have if it were applied consistently to NT vocabulary.
There are many Greek words which in the NT usually have one meaning but
occasionally have another. Meier's principle would mean that this rare
meaning, however well attested outside the NT, should not be allowed within
the NT because the general usage of the NT (its amazing consistency) would
exclude it. For example, TpárceÇa occurs 15 times in the NT, normally with
the meaning "table." Just once, in Luke 19:23, it means "bank," according to
all translators and exegetes, but Meier's principle would have to disallow this.
To take another example, xéXoç occurs 41 times in the NT, always with the
meaning "end," except on three occasions when it means "tax" (Matt 17:25;
Rom 13:7). This ratio of common meanings to rare ones (38 to 3) is much
more unfavorable to the rare meaning than is the ratio of àôetapóç in
undisputed instances of the meanings "full brother" and "half brother" to
àSstapóç in the disputed instances referring to the brothers of Jesus (61 to
14). If in the latter case the general usage must determine the meaning in
the remaining 14 instances, then how much more must the 38 instances of
TéXoç in the sense of "end" require the same meaning in the remaining
three instances.
It is probably unnecessary to labor the point that Meier's argument
contradicts what nearly all translators and exegetes assume: that the range
of use from which the meaning of a word in the NT must be chosen is the
range of use in the language, not the range of use in the NT. Meier in fact
confuses his argument about the use of à8 eXxpóç in the NT with a quite
different argument about the relevance of context to meaning: "no amount
of parallels from outside the NT can tell us a priori what the NT texts mean;
only a detailed exegesis of the NT texts in their own context can tell us that"
(his italics).13 If this means that the meaning of, say, Mark 6:3 depends on
its context in Mark's Gospel, the point has validity, but it cannot exclude the
relevance of evidence extraneous to the literary context, whether from else-
where in the NT or from outside the NT. When Meier himself argues that
áSstapóç in Mark 6: 1 7 must mean "half brother" rather than "full brother," 14
there is nothing in the context of this verse in Mark, or indeed in the rest of
the NT, which enables him to know this. He has to assume that information
he has from Josephus was common knowledge to Mark and his readers. The

13 Meier, "Brothers and Sisters," 21 n. 36 = Marginal Jew, 360 n. 35.

14 Meier, "Brothers and Sisters," 20 = Marginal Jew, 327-28.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 693

case is parallel to that of the brothers of Jesus. Defenders of the Epiphanian

view do not claim that the information that the brothers of Jesus were sons
of Joseph by a previous marriage can be deduced from the immediate context
of references to them in the NT. It is sufficient to claim that their actual
relationship to Jesus can be deduced from other evidence.
Some more specific comments are now required on Meier's claim that
the Epiphanian view is disproved because there is no clear instance of the
meaning "stepbrother" for àSetapóç in the NT.
1 . We have seen that the Epiphanian view actually requires áSsXcpóç to
mean "half brother" in biblical statements which do not presuppose the
virginal conception (either because the virginal conception is not presup-
posed by the writer himself, or because he does not imagine it to be pre-
supposed by the speakers in his narrative). There are undoubtedly some texts
in this category (e.g., Mark 6:3), and there may be many. If Matthew and
Luke are considered the only NT writers who accept the virginal conception,
then it would not be difficult to argue that their use of áSs^cpóç for the
brothers of Jesus does not really take the virginal conception into account
but follows the already well-established usage of their traditions. In this case,
the Epiphanian view would be consistent with the meaning "half brother,"
which Meier allows is a NT sense of áSetapóç, in all NT references to the
brothers of Jesus.
2. Meier assumes that the Epiphanian view requires the meaning "step-
brother." Although he tells us what Paul should have said instead of àSstapóç
if he had meant "cousin" (ávsvj/ióç),15 he does not tell us what Paul should
have said if he meant "stepbrother." Perhaps Meier is aware that Greek has
no word specifically for "stepbrother." It is possible that Paul could have used
KTļ8eoxfļ(;, which means "relative by marriage" and is used of fathers-in-law,
stepfathers, and brothers-in-law, and so could presumably be used of
stepbrothers. But to specify the relationship of stepbrother precisely, only a
cumbersome expression could have served.
3. Meier might have come to different conclusions had he extended his
study of àôstapóç to other words for family relationships. When Luke calls
Joseph and Mary the "parents" of Jesus (2:41,43: oi yoveîç aulou) and has
Mary refer to Joseph as Jesus' "father" (2:48: ó 7taxf|p cou), Meier must
presume that these terms designate Joseph Jesus' stepfather, yet the use of
these terms for literal family relationships elsewhere in the NT provides no
example of a meaning other than biological parenthood. The point is that the
Epiphanian view postulates Jesus' standing in the same kind of relationship
to his brothers and sisters as he did to Joseph. If Luke can call Joseph Jesus'

15 Meier, "Brothers and Sisters," 18 = Marginal Jew, 326.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 694 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

parent or father without implying blood relationship, then it is arbitrary

to insist that reference to Jesus' brothers and sisters must imply blood
relationship.
4. We have seen, in section 1 above, that "stepfather" is not really the
appropriate term for Matthew's and Luke's view of Joseph's relationship to
Jesus. Since they understood Joseph to be Jesus' father in social reality and
in law, i.e., in every respect other than biological paternity, the idea of adop-
tion comes much closer to what they envisage. On the Epiphanian view,
Jesus' status as Joseph's adopted son makes him the brother of Joseph's
natural children. The linguistic evidence shows that for the relationships
created by adoption the ordinary Greek words for the corresponding natural
relationships could be used.16 This can be shown even from the NT, where
Moses, understood to have been adopted by Pharaoh's daughter, is called
her son (Acts 7:21; Heb 1 1:24: uióç).17 That àSeXçóç can be used for the
relationship between natural and adopted children is shown by Paul's
metaphorical use of the term to designate the relation between Jesus and
Christians (Rom 8:29) in a context which makes it clear that Christians are
adopted children of God (Rom 8:15, cf. 8:16-17).

III. The Redaction-Critical Argument

While Meier rightly recognizes that the language of Matt 1:25 cannot be
pressed to imply that Joseph and Mary had sexual relations - let alone that
they had children - after the birth of Jesus, he argues that Matthew's redac-
tion of Mark in Matt 12:46 (|| Mark 3:31) and Matt 13:55 (|| Mark 6:3) shows
that Matthew thought the brothers and sisters of Jesus were children of
Joseph and Mary.18 However, these arguments are examples of oversubtle
and overambitious use of redaction criticism.
That Mary, Joseph, Jesus, and his brothers and sisters formed a family
unit is certainly presupposed by Matt 12:46-50 (|| Mark 3:31-35; cf. Luke 8:19-
21). Meier rightly observes that the Hieronymian view would deprive this
pericope of much of its force. But the Epiphanian view, which makes Mary
the stepmother of Jesus' brothers and sisters and makes Jesus their adoptive
brother, is quite consistent with this pericope. What matters is simply that

16 Alternatively, 0p87nóç (foster son), 0p87cxf| (foster daughter), and xpocpeúç (foster fa-
ther) could be used, but they do not exclude the use of the terms for natural family relationships.
This can be seen from a Jewish inscription from Rome, commemorating an adopted daughter:
she is called xpeÇirxfi (sic, i.e., 0pe7ixf|, foster daughter), but her parents are rcaxpòç Kai nT|xpóç:
CIJ 21, reprinted in H. J. Leon, The Jews of Ancient Rome (Morris Loeb Series; Philadelphia:
Jewish Publication Society of America, 1960) 267.
17 Compare also John 19:16-17.
18 Meier, "Brothers and Sisters," 8-15 = Marginal Jew, 320-24.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 695

Jesus and his relatives compose a nuclear family. Precisely how that family
is constituted, by blood relationship or by marital or legal ties, is beside the
point. It is overinterpretation of the differences between Matthew and Mark
to see Matthew's phrase r' |ir|TTļp Kaì oi áSetapoi aoxou (Matt 12:46, in place
of fļ jifjTTļp aÔToC Kai ot àSetapoì aòioC in Mark 3:31) as a deliberate attempt
to reinforce "the impression that the mother and brothers naturally belong
together as blood relations."19 It is impossible to be sure that such a change
is deliberate at all, or that, if it is deliberate, it represents more than a stylistic
preference. The regular association of Jesus' mother and brothers in the
gospel traditions by no means requires that they were blood relatives; it
requires only that they constituted a family unit.
In the case of Matt 13:55, Meier makes much of the fact that the first
question refers to Jesus' father, while the second groups together his mother
and brothers. "Matthew," he claims, "is at pains to separate the legal-but-
not-biological father of Jesus from Jesus' real, biological mother. Faced with
this great divide that he himself creates, Matthew chooses to place Jesus'
brothers with his biological mother, not his legal father."20 Again, this is
overinterpretation. Matthew is representing what the people of Nazareth
(whom he would not expect to know of the virginal conception) would say.
In the cultural context, it is natural that they should give Jesus' paternity
priority, with a significance of its own. Other members of the family follow
in order of importance. Moreover, the division of Matt 13:55 into two
questions may well be no more than the result of Matthew's redactional
substitution of xou tektovoç for Mark's t£ktcov. His first question copies
Mark's, but the change in the description of Jesus makes a second question,
concerning Mary and the brothers, grammatically necessary.

IV. The Second-Century Evidence

In my book I argued that, whereas most references to the brothers and
sisters of Jesus in the Christian literature of the first and second centuries are
consistent with either the Epiphanian or the Helvidian view, only the Epipha-
nian view is unambiguously attested before Tertullian.21 It is found around

19 Meier, "Brothers and Sisters," 13 = Marginal Jew, 323.

20 Ibid.
Bauckham, Jude and the Relatives, 24-32. Meier ("Brothers and Sisters," 22-24 = Mar-
ginal Jew, 329-30) seems to think Hegesippus adopts the Helvidian view. Although I agree with
Meier that Hegesippus cannot have taken the Hieronymian view, on the choice between the
Helvidian and the Epiphanian views Hegesippus' statements are as ambiguous as those of the
NT : see Bauckham, Jude and the Relatives, 30-3 1 .

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 696 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

the middle of the second century in three works of Syrian Christian prove-
nance: the Protevangelium of James 9:2; 17:1-2; 18:1, 22 the Gospel of Peter
(according to Origen In Matt. 10. 17), 23 and the Infancy Gospel of Thomas
16: 1-2. 24 I suggested that the tradition common to these three works might
preserve an accurate historical memory of the relationship of Jesus to his
brothers and sisters, though I did not consider this proven.
Meier evidently considers the evidence of these works not even worth
considering, because of their date and their nature,25 but it is not good
historical method to dismiss evidence simply because it occurs in relatively
late sources rich in imagination. Especially in the field of ancient history, for
which many original sources are lost, information from earlier sources which
are no longer extant is often preserved in later sources. In the case of gospel
traditions, not only early written gospels now lost or preserved only in
fragments but also the oral gospel tradition survived into the early second
century. The use of late sources certainly requires caution, but it should not
be disallowed out of hand. My suggestion is not that these three works
themselves should be trusted for historical information, but that the common
tradition which evidently predates them may be of historical value.
The Protevangelium of James and the Infancy Gospel of Thomas are
certainly works of imagination, not of historiography, but the idea that Jesus'
brothers and sisters were Joseph's children by a previous marriage is taken
entirely for granted in these works as something the readers already know to
be the case. It is the only piece of nonbiblical information common to these

22 Text in E. de Strycker, La forme la plus ancienne du Protévangile de Jacques (Subsidia

Hagiographica 33; Brussels: Société des Bollandistes, 1961); A. de Santos Otero, Los evangelios
apócrifos (BAC 148; 6th ed.; Madrid: Editorial católica, 1988) 130-70. Meier ("Brothers and
Sisters," 5 n. 8) is incorrect in denying that the Protevangelium of James is clear about the status
of the brothers and sisters of Jesus. He is confusing the perpetual virginity, which is only implied,
with the Epiphanian view of the brothers and sisters of Jesus, which is clear both in Joseph's
explicit statement, "I already have sons" (9:2), and in the appearance of his sons in the narrative
of the birth of Jesus (17:1-2; 18:1). As I have argued {Jude and the Relatives, 39-41), Salome
also (20:1-3) is probably one of Joseph's daughters. Cf. also what James the brother of Jesus,
the supposed author of the work, says in 25:1.
23 Origen's text is given in H. B. Swete, The Akhmîm Fragment of the Apocryphal Gospel
of St. Peter {London: Macmillan, 1893) x n. 1; L. Vaganay, L'Evangile de Pierre (EBib; 2d ed.;
Paris: Lecoffre, 1930) 8.
24 In 16:1, this work implies that Joseph's son James is older than Jesus. For the text, see
de Santos Otero, Evangelios apócrifos, 279-97 (Greek); W. Wright, Contributions to the Apoc-
ryphal Literature of the New Testament (London: Williams & Norgate, 1865) 6-1 1 (Syriac). The
textual history of the Infancy Gospel of Thomas is complex and obscure (see S. Gero, "The
Infancy Gospel of Thomas: A Study of the Textual and Literary Problems," NovT 13 [1971]
46-80), but the story of Jesus and his brother James occurs in all versions.
25 See his brief dismissal of my discussion: Meier, "Brothers and Sisters," 6 n. 11.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 697

works, the works themselves show no signs of a literary relationship, and so

the information can reasonably be considered a tradition which predates both
works. These works are, therefore, evidence of a well-established tradition in
(probably early) second-century Syrian Christianity that Jesus' brothers and
sisters were children of Joseph by a previous marriage. Accordingly, the
historical value of this tradition should not be prejudiced by the unhistorical
character of the Protevangelium of James and the Infancy Gospel of Thomas
but should be judged on its own merits. Since the tradition was also incor-
porated into the Gospel of Peter ; it may well belong to the oral gospel tradition
of the church of Antioch, to which the Gospel of Peter was probably indebted.
That this gospel, independently of other extant gospels, preserves some old
traditions as well as later legendary developments is entirely possible. The
possibility that the Epiphanian view of the brothers and sisters of Jesus was
one such old tradition does not rest on uncritical credulity towards late and
legendary sources but on critical distinguishing of what might be good
tradition from what clearly is not.
The Protevangelium of James also contains an account of the miraculous
birth of Jesus which preserves Mary's virginity (the virginitas in partu).
Although the perpetual virginity of Mary is not explicitly stated, the Epipha-
nian view of the brothers and sisters of Jesus combines with the virginitas in
partu to imply it. The Protevangelium of James did not invent the idea of the
miraculous birth of Jesus any more than it did the Epiphanian view of the
brothers and sisters of Jesus. The miraculous birth is implied or recounted
in other Syrian Christian works of the early second century (Ignatius
Eph. 19:1; Odes Sol. 19:8-9; Ascension of Isaiah 11:9-14), 26 but only the
Protevangelium of James combines the two themes.
It may be that a development of the theme of Mary's virginity, involving
both the virginitas in partu and the perpetual virginity, occurred in early
Syrian Christianity, and that the Epiphanian view of the brothers and sisters
of Jesus was a result of this concern with the virginity of Mary. In that case,
the second-century evidence for the Epiphanian view would have no historical
value. But it is equally possible that the brothers and sisters of Jesus were
remembered not to have been children of Mary, and that precisely this
tradition made possible the development of the idea of Mary's perpetual
virginity. In that case, the Epiphanian view could rest on good historical
tradition, and the fact that it was later used apologetically to defend the
perpetual virginity should not be allowed to prejudice our judgment of
its value.

26 On these texts, see J. A. de Aldama, María en la patrística de los siglos I y II (BAC 300;
Madrid: Editorial católica, 1970) 189-211.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 698 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

The second-century evidence, therefore, leaves the question open, but

it should be stressed again that nothing in first- or second-century Chris-
tian literature contradicts this tradition of the Epiphanian view in early
second-century Syria. Other literature can be shown not to take the Hierony-
mian view, but it cannot be shown to take the Helvidian rather than the
Epiphanian view.

V. "Son of Mary" (Mark 6:3)

In sections 2 and 3 we have seen that NT references to the brothers
and sisters of Jesus do not, as Meier claims, support the Helvidian view
rather than the Epiphanian. In section 4, 1 have argued that second-century
evidence positively supporting the Epiphanian view should not be dismissed
too quickly, but this evidence seems too problematic to tip the balance of
probability in favor of the Epiphanian view. In the present section, I wish to
suggest a new interpretation of Mark 6:3 which would constitute good early
evidence for the Epiphanian view, if it should prove more plausible than
other interpretations.
The fact that in Mark's account of Jesus' visit to Nazareth, the people
of Nazareth call him "son of Mary" (Mark 6:3) rather than "son of Joseph"
(cf. -Matt 13:55; Luke 4:22; John 6:42) has never been satisfactorily ex-
plained.27 Suppositions that it was Jewish custom to use a metronymic to refer
either (a) to the son of a widow,28 or ( b ) to an illegitimate son have been shown
to have no convincing support from Jewish parallels.29 In a recent article Ilan
has shown that there is some evidence from Josephus and rabbinic tradition
that (c) "a man would be called after his mother when she possessed superior
lineage."30 She suggests that Mark, unlike Matthew and Luke, knew nothing
of Joseph's Davidic lineage, and could, therefore, consider Mary the more
important parent.31 In fact, it is probable that Mark believed in Jesus' Davidic
descent (Mark 10:48; 11:10), which was the widespread belief of the early
church (Acts 2:30; 13:23; 15:16; Rom 1:3; 2 Tim 2:8; Heb 7: 14; Rev 5:5; 22:16;
Did. 10:6), and he would have needed good reason not to assume that this

27 On the text-critical problem in Mark 6:3, see J. Blinzler, Die Brüder und Schwestern
Jesu (SBS 21; Stuttgart: Katholisches Bibelwerk, 1967) 28-30, who argues for the originality of
the reading ó iéktcov ó uiòç Mapíaç. For a different view, see H. K. McArthur, " 'Son of Mary1,"
NovT 15 (1973) 46-52.
28 T. Ilan, "'Man Born of Woman . . .* (Job 14:1): The Phenomenon of Men Bearing
Metronymes at the Time of Jesus," NovT 34 (1992) 23 n. 3, supplies two examples of sons of
widows known by the names of their fathers.
29 McArthur, "'Son of Mary'," 44-46, 52-53, followed by Meier, Marginal Jew, 226.
30 Ilan, "'Man Born of Woman'," 43.
31 Ibid., 45.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 BROTHERS AND SISTER OF JESUS 699

descent was through Joseph. Even if Mark did not think Jesus was descended
from David, we are still at a loss to know why he should have thought Mary
the genealogically more distinguished parent. Another possibility, ( d ) that
Mark refers to the virginal conception, is unlikely. None of the gospels
represent the virginal conception as a matter of public knowledge, and so,
even supposing Mark knew the tradition of the virginal conception, it is
unlikely that he meant to refer to the virginal conception when he put the
designation "son of Mary" on the lips of the people of Nazareth.
Finally, McArthur and Meier, having rightly rejected other proposed
solutions, except (c), which they did not know, propose ( e ) that the phrase is
not a formal designation but an "informal description" occasioned by the
context.32 Wishing to point out that Jesus was an ordinary member of their
own community, the Nazarenes naturally think of him as the son of his still
living parent, who is presumably there in the synagogue as they speak.33 This
is the best explanation that has so far been offered, but it is not quite
convincing. If Jesus were normally known as "the son of Joseph," as Jewish
custom seems to require, one would expect the Nazarenes to make their point
by first calling him this, and then going on to refer to those relatives who were
there with them: his mother, brothers, and sisters.
It is surprising that another possible explanation seems not to have been
suggested: that in Nazareth Jesus would have been known as "the son of
Mary" because this distinguished him from the children of Joseph by his first
wife. This usage can easily be paralleled from the OT. Women only occa-
sionally occur in biblical genealogies, but in a large majority of the cases
where they do, their function in the genealogy is that of distinguishing a man's
sons by one wife from his sons by another wife (e.g., Gen 4:19-22; 22:20-24;
36:10-14; 46:10; Exod 6:15; 1 Chr 2:2-4,18-19,21,24,25-26,46,48-49; 3:1-9).
This concern to distinguish the sons of different mothers means that sons of
men who had children by more than one wife can be designated by their
metronymic, instead of the usual patronymic. Thus Hur, who was the son of
Caleb by one of his wives (1 Chr 2:18-19) is known as "Hur the firstborn of
Ephrath" (1 Chr 2:50; 4:4). Similarly, David's son Adonijah is known as
"Adonijah the son of Haggith" (1 Kgs 1:5,1 1; 2:13; cf. 2 Sam 3:4; 1 Chr 3:2),
distinguished from sons of David by other wives (2 Sam 3:2-5; 1 Chr 3:1-9).
In rabbinic literature, not only Adonijah but also other sons of David are
referred to by their metronymics ( b . B. Bat . 109b; b. Ketub. 62b). Similarly,
the sons of Jacob by his two wives and two concubines are sometimes
designated by their metronymics (Philo Fug. 73; Joseph and Aseneth 22:1 1;

32 McArthur, " 'Son of Mary'," 54.

33 Meier, Marginal Jew, 226-27.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms
 700 THE CATHOLIC BIBLICAL QUARTERLY I 56, 1994

26:6; 27:6-7; 28:1,9). It is possible that this is also the explanation of the most
famous use of metronymics in the OT: the brothers Joab, Asahel, and Abishai
are always called "the son(s) of Zeruiah," their mother (1 Sam 26:6 and
twenty-three times in the OT; also Josephus Ant. 7.1.3,6 §§1 1,45; 7.3.2 §65;
7.9.1 §265; b. Qidd. 49b; b. Sank. 95a; only Josephus Ant. 7.1.3 §1 1, perhaps
following 1 Chr 4:14, names their father). This usage cannot be intended just
to highlight their relationship to David, who was Zeruiah 's half brother,
since Amasa, son of Zeruiah 's sister Abigail, is known by his patronymic
(2 Sam 17:25; 1 Kgs 2:5,32; cf. 1 Chr 2:16-17). But it is possible that Zeruiah 's
husband had sons by another wife and that her sons are distinguished by
reference to her.34 (A less likely possibility is that the sons of Zeruiah had the
same mother but different fathers and that they used the metronymic to
indicate their relationship to each other.)
It is easy to suppose that, whereas outside Nazareth Jesus would have
to be identified as "the son of Joseph," in Nazareth, where the family was
known, the children of Joseph's two wives would be distinguished by their
metronymics. Jesus would be called "the son of Mary" precisely because
James, Joses, Judas, and Simon were not sons of Mary. This understanding
of Mark 6:3 does not, of course, depend on the improbable assumption that
Mark preserves an accurate historical report of what the people of Nazareth
said. It simply assumes that Mark attempts to portray with verisimilitude
what they would have said, just as he represents them as calling Jesus "the
carpenter" because this is what people in his home village, though not else-
where, would be likely to call him.
In conclusion, Meier is right that, since "the data are sparse and
ambiguous," any conclusions about the relationship of Jesus to his brothers
and sisters must be "quite limited and tentative."35 1 hope to have shown that
much of the data is more ambiguous than he allows, while there is also some
suggestive evidence which he neglects. I should be content to have demon-
strated at least that the issue between the Epiphanian and Helvidian views
must remain more open than Meier concluded it should.

34 See D. N. Freedman in Meier, Marginal Jew, 250 n. 101.

35 Meier, "Brothers and Sisters," 7 = Marginal Jew, 319.

This content downloaded from 131.172.36.29 on Thu, 07 Jul 2016 04:52:08 UTC
All use subject to http://about.jstor.org/terms