Hostname: page-component-74656f9884-c52kr Total loading time: 0 Render date: 2026-11-06T18:10:20.421Z Has data issue: false hasContentIssue false
  • English
  • Français

The Accommodations of Joseph and Mary in Bethlehem: Κατάλυμα in Luke 2.7*

Published online by Cambridge University Press:  28 May 2010

Stephen C. Carlson
Stephen C. Carlson
Affiliation:
Duke University, 209 Gray Bldg., Durham, NC 27708, USA. email: stephen.carlson@duke.edu
Rights & Permissions [Opens in a new window]

Abstract

The identity of the κατάλυμα in Luke 2.7 has been debated among Western scholars for over five hundred years. Proposals have ranged from an inn to a guest room. This article argues that the term κατάλυμα has a generic sense of ‘place to stay’ and that the final clause of Luke 2.7 should be rendered ‘because they had no space in their place to stay’. Moreover, three clues in the context—Joseph's compliance with the census order, the betrothal of Mary, and the manger—suggest that the accommodations presupposed by Luke are a marital chamber too small for giving birth.

Keywords

Lukan infancy accountLuke 2.7κατάλυμαmarital chamberbetrothalcensus

Information

Type
Articles
Information
New Testament Studies , Volume 56 , Issue 3 , July 2010 , pp. 326 - 342
Copyright
Copyright © Cambridge University Press 2010

1. Introduction

The inn and the stable have long been familiar elements of Christmas pageants and nativity crèches, but the Lukan infancy account is hardly explicit about the accommodations for Joseph and Mary in Bethlehem. Nevertheless, traditional interpretations of the story can be tenacious even in the light of scholarship showing that they are unsupported by the text. In 1584, the great and outspoken philologist at the University of Salamanca, Francisco Sánchez de las Brozas, known as ‘El Brocense’, discovered this the hard way when his students reported him to the Spanish Inquisition.Footnote 1 He had criticized the depictions of the nativity in church paintings, and one of his criticisms was that Jesus was not born in the stable nor were his parents rejected by an innkeeper as commonly thought, but that Mary gave birth in a private home belonging to friends or relatives.Footnote 2 Summoned before the Inquisitors later in September, El Brocense defended his positions in writing, and, as a result, the files of the Spanish Inquisition contain one of the earliest historical-critical exegeses of the Lukan birth narrative.Footnote 3

El Brocense's defense of his position is still cogent five centuries later, and many of his arguments enjoy support from contemporary scholars.Footnote 4 For example, he recognized that neither the Greek term κατάλυμα in Luke 2.7 nor even its Vulgate rendering diversorium necessarily means an ‘inn’ as evident from the use of the same term in Luke 22.11 referring to an upper room.Footnote 5 Moreover, there would have been no need for an inn, El Brocense argued, because Joseph had to return to his own town according to the decree, so he must have had family—if not his own house—in Bethlehem where he could stay.Footnote 6 El Brocense also denied that there would have been a throng of census registrants descending upon Bethlehem because subjects did not need to register on a specific day.Footnote 7 As for the placement of the baby in a manger, El Brocense pointed out that this was hardly unusual because farmhouses often kept animals in the same part of the house where the people slept.Footnote 8 Unfortunately, these arguments did not sway the Inquisitors, and El Brocense was reprimanded.Footnote 9

Five hundred years after El Brocense argued that the Lukan account did not refer to an ‘inn’,Footnote 10 this idea that Joseph and Mary were turned away from the inn retains its hold upon scholars of Luke's infancy narrative.Footnote 11 Part of the reason for this tenacity is that the ‘inn’ continues to be a staple among leading translations of Luke. For example, the New Revised Standard Version renders the description of the accommodations of Joseph and Mary in Luke 2.6–7 as follows:

⁶ Ἐγένετο δὲ ἐν τῷ εἶναι αὐτοὺς ἐκεῖ ἐπλήσθησαν αἱ ἡμέραι τοῦ τεκεῖν αὐτήν, ⁷ καὶ ἔτεκεν τὸν υἱὸν αὐτῆς τὸν πρωτότοκον, καὶ ἐσπαργάνωσεν αὐτὸν καὶ ἀνέκλινεν αὐτὸν ἐν φάτνῃ, διότι οὐκ ἦν αὐτοῖς τόπος ἐν τῷ καταλύματι.

⁶ While they were there, the time came for her to deliver her child.⁷ And she gave birth to her firstborn son and wrapped him in bands of cloth, and laid him in a manger, because there was no place for them in the inn. (nrsv)

This familiar translation rests on a series of questionable exegetical decisions. First, it renders καταλύματι as ‘inn’, even though many scholars recognize that this translation is unsatisfactory and favor something more general such as ‘guest room’.Footnote 12 Second, its rendering of the definite article τῷ before καταλύματι as ‘the’ reinforces the overly specific translation of the noun—not only does the v. 7 refer to an ‘inn’, but to ‘the inn’, as if the inn was well known or the only one in town.Footnote 13 Finally, this translation construes the dative αὐτοῖς, not as a dative of possession with the verb ἦν in accordance with leading grammars of the NT,Footnote 14 but with τόπος as a dative of advantage: no ‘place for them’ in the inn.Footnote 15 The English thus portends that Joseph and Mary never had a room in the inn at all, leading readers to reflect upon the cruel inhospitality of the innkeeper who rejected them in their time of need so that Jesus had to be born in a dirty stable.

Other translations are hardly better at rendering the Greek and some are even worse. The NIV, for example, follows the AV in translating the final clause of v. 7 as ‘because there was no room for them in the inn’. This rendition of τόπος with the English ‘room’, which can refer to a chamber as well as a space, strengthens the mistranslation of κατάλυμα as ‘inn’.Footnote 16 The updated NIV, called Today's New International Version (TNIV), rewrites the clause as ‘because there was no guest room available for them’, merging τόπος and κατάλυμα into a single noun phrase, ‘guest room’. By this device, the TNIV manages to avoid the familiar ‘inn’ when translating κατάλυμα, but, by preserving the interpretation of αὐτοῖς as a dative of advantage, it maintains the same theological reading of the Lukan infancy account undergirding that of the traditional ‘inn’—the entry of Jesus into the world was accompanied by human inhospitality and rejection.

A common exegetical pitfall plaguing the interpretation of κατάλυμα in Luke 2.7 is that interpreters begin by being too specific as to its meaning. One example is Raymond Brown.Footnote 17 His analysis of the meaning of κατάλυμα does not begin with ascertaining its sense but with surveying various of its proposed referents, which he lists as: ‘A private home’, ‘A room in an unidentified place’, and ‘The inn, or more specifically, the well-known traveler's inn at or near Bethlehem’.Footnote 18 The first two possibilities are presented with an English indefinite article while the third has the definite article, so it should not be surprising that Brown rejects the first two options due to the Greek article in the phrase ἐν τῷ καταλύματι: ‘The definite article before katalyma remains an obstacle to translating it “a room” ’.Footnote 19 While the ‘inn’ option has Brown's sympathies,Footnote 20 he is troubled by Luke's use of a more specific term for inn in the parable of the Good Samaritan at 10.34, πανδοχεῖον.Footnote 21 Ultimately, Brown gives up and decides that is impossible to ascertain what Luke meant, opting for a more ambiguous term ‘lodgings’.Footnote 22 Had Brown begun with κατάλυμα's sense instead of its possible referents,Footnote 23 on the other hand, it would have been possible to settle on the scope of meaning for κατάλυμα and its appropriate English translation, and only then to determine what kind of accommodation κατάλυμα refers to in the particular context of Luke 2.6–7.

2. The Sense of Κατάλυμα

To ascertain the sense of κατάλυμα it is important to consider its derivation, the word's usage in Hellenistic Greek texts, how it was used in the Septuagint (lxx) to render the Hebrew, its NT usage, and ancient translations of it into Syriac, Coptic, and Latin at Luke 2.7. All of this evidence bears out the conclusion that κατάλυμα was a generic term with the sense of ‘place to stay’ or ‘accommodations’. Due to its broad sense, the term has referred to various kinds of accommodations, whose particular identification can only be determined by examining the specific context in which the term was used.

The derivation of κατάλυμα suggests that the word has a broad, generic meaning. This noun is a verbal substantive formed by adding the −μα, −ματος suffix to the stem of the verb καταλῦσαι, ‘to untie’ or ‘to unloose’, which, when used for unpacking one's belongings, means ‘to lodge’.Footnote 24 This suffix denotes the result of the verb's action,Footnote 25 so one would expect κατάλυμα to mean ‘the result of unpacking’ or ‘the result of lodging’, yet by its earliest attestation in the third century bce, it has already acquired the sense of ‘the place of lodging’. As a result, the usage of κατάλυμα, not just its derivation, is important for determining the scope of its meaning.

The earliest attested uses in the Hellenistic era also substantiate a generic sense for κατάλυμα. The term appears twice in Polybius's History. At 2.36.1 Hasdrubal, the Carthaginian general and Hannibal's brother-in-law who had governed Spain for eight years, was murdered at night in his own quarters (ἐν τοῖς ἑαυτοῦ καταλύμασι). The sense of the word is hardly specific; it can perhaps refer to the rooms where the general slept in his palace. The other instance occurs at Hist. 32.12.2 where Polybius states that the Roman ambassadors sent to Illyria were not properly furnished with room and board (μήτε κατάλυμα δοθῆναι σφίσι μήτε παροχήν) by their Dalmatian hosts. In the Letter of Aristeas 181, the term is used when the king orders that the finest quarters (καταλύματα…τὰ κάλλιστα) near the citadel be assigned to the seventy translators of the Hebrew scriptures. As another example, Diodorus Siculus used the term to describe the reward of public accommodations (δημόσιον…κατάλυμα) that the Romans gave to Timasitheus for rescuing their ambassadors from pirates (Bibliotheca historica 14.93.5). Furthermore, κατάλυμα also occurs several times among the Greek papyri recovered in Egypt. According to a detailed study of housing terms in the papyri by Geneviève Husson, κατάλυμα and its cognates καταλυμάτιον and κατάλυσις were employed for the quarters supplied for kings, generals, soldiers, artisans, and pilgrims.Footnote 26 She concludes that all these usages have a common, underlying sense: ‘the dwelling where one stays or remains for some time’.Footnote 27

The diversity of referents for κατάλυμα in the lxx corroborates a broad sense for the word.Footnote 28 Indeed, κατάλυμα translates seven different Hebrew words. In Exod 4.24, it renders מלון in reference to a place where Moses and his family spent the night on his way into Egypt (cf. also Jer 14.18). Κατάλυμα twice translates נוה, a ‘shepherd's abode’, at Exod 15.13 and Jer 33.12 lxx (40.12) in a metaphor for God's dwelling in the promised land. Its reference to a nomad's dwelling is evident at 2 Kgdms 7.6, 1 Chron 17.5, and Sir 14.25, where it is used in parallelism with σκηνή (‘tent’). In Jer 25.38 lxx (32.38), it is used in the simile that ‘he has left his lair like a lion’ (עזב ככפיר סכו). In Ezek 23.21 it was added by the translator as a general term for Oholibah's dwelling place during Israel's sojourn in Egypt. Three times in the lxx (but corresponding only once to a particular Hebrew word) it refers to the accommodations of both room and board given to priests at a temple (1 Kgdms 1.19; 9.22 לשׁכתה; 1 Chron 28.13). Finally, the author of 1 Macc 3.45 bewails the fact that Jerusalem has become a place for Gentiles to stay (κατάλυμα τοῖς ἔθνεσιν).

The NT usage of κατάλυμα apart from Luke 2.7 coheres with its having a broad meaning. At both Luke 22.11 and its parallel at Mark 14.14, Jesus instructs his disciples to ask a man carrying a jar in Jerusalem about accommodations for eating the Passover: ποῦ ἐστιν τὸ κατάλυμα.Footnote 29 Translations usually render this instance of κατάλυμα rather specifically as ‘guest room’, but the generality of κατάλυμα is evident from the further specification in both Luke and Mark that the place to stay is a ‘large, furnished upper room’ (ἀνάγαιον μέγα ἐστρωμένον). We know that κατάλυμα refers to a ‘guest room’ in this context, not because the sense of the word is so specific, rather because the context makes its reference specific. Moreover, when Luke wanted to be specific about an inn, as in the parable of the Good Samaritan, the author used a precise term, πανδοχεῖον (Luke 10.34).Footnote 30

The earliest translations of Luke 2.7 into Syriac, Coptic, and Latin bear witness to a generic meaning for κατάλυμα.Footnote 31 Among the Syriac translations, the Peshitta for instance renders ἐν τῷ καταλύματι broadly with the clause ܐܰܝܟ݁ܳܐ ܕ݁ܰܫܪܶܝܢ ܗ̱ܘܰܘ, ‘where they were dwelling’. As another example, the ultra-literal Harklensis translation follows its Greek Vorlage in the use of a prepositional phrase: ܒܒܝܬ ܡܫܪܝܐ, ‘in the house of dwelling’. As for the Old Syriac, the translator of the Sinaitic may have felt the generic κατάλυμα to be superfluous and omitted it, rendering the final clause of v. 7 simply as ܡܛܠ ܕܠܝܬ ܗܘܬ ܠܗܘܢ ܕܘܟܬܐ, ‘because they had no place’.Footnote 32 Like the Syriac translations, the Coptic translations also render κατάλυμα with a generic phrase, in this case, as ‘place of dwelling’: maNqoile (Sahidic) and maNouox (Bohairic). At first blush, the Vulgate and most of the Old Latin translations of ἐν τῷ καταλύματι as in diversorio may seem to support a stricter understanding for the phrase as ‘in the inn’, but in Classical times, the noun diversorium (or devorsorium) still had a wider sense of a ‘lodging-place’.Footnote 33 As El Brocense had pointed out, Cicero wrote in a letter to Gallus: ‘Nor more willingly would I buy a place to stay (diversorium) at Tarricina, so I would not ever be bothersome to my host’ (Book 7).Footnote 34 In this context, the meaning of ‘inn’ for diversorium is difficult to sustain. Another Old Latin translation, found in Codex Palatinus (e), renders κατάλυμα with the word stabulum, which meant a ‘standing-place, abode, habitation, dwelling’ as well as a ‘stable’ or ‘cheap hostel’.Footnote 35 Accordingly, ancient translators of the Greek text agree that κατάλυμα is a generic term.

This wide-ranging survey of the evidence for the sense of κατάλυμα demonstrates that this term had a wide meaning of a ‘place to stay’, ‘lodging’, or ‘accommodations’. Because translations should preserve—to the extent feasible—a scope of meaning commensurate to that of the source text, a faithful translation of κατάλυμα in Luke 2.7 ought to be as general as its Greek original. Thus, translations specifying that Joseph and Mary's place to stay in Bethlehem was an ‘inn’ or even a ‘guest room’ ought to be avoided because they are too specific. A translation faithful to the sense of κατάλυμα should be satisfied with merely stating that it was a ‘place to stay’ or ‘accommodations’.Footnote 36

3. The Meaning of διότι οὐκ ἦν αὐτοῖς τόπος ἐν τῷ καταλύματι in Luke 2.7

Any reconsideration of the lodging of Joseph and Mary in Bethlehem, moreover, cannot stop at ascertaining the meaning of κατάλυμα according to Luke. The exegesis of the other words in the immediate context—especially the definite article τῷ before the noun and the dative pronoun αὐτοῖς—have depended on the understanding that κατάλυμα meant ‘inn’. Because meaning is so often dependent on context, it is wrong to assume that their meanings would hold when one of the terms in the context, κατάλυμα, has been clarified. Thus, the reconsideration of Joseph and Mary's accommodations in Bethlehem must also entail a revisiting of the exegesis of κατάλυμα's context.

Once κατάλυμα in Luke 2.7 is understood broadly as a ‘place to stay’, the role of the definite article before it becomes much clearer in the phrase ἐν τῷ καταλύματι. The article is anaphoric, pointing back to the accommodations of Joseph and Mary presupposed in v. 6: Ἐγένετο δὲ ἐν τῷ εἶναι αὐτοὺς ἐκεῖ ἐπλήσθησαν αἱ ἡμέραι τοῦ τεκεῖν αὐτήν, ‘It happened that, while they were there, the days for her to give birth were fulfilled’. Joseph and Mary had to stay somewhere until her full term became due, and the κατάλυμα in v. 7 points to the place where they were staying in Bethlehem. As a result, it is not necessary to consider the specialized usage of the definite article that their κατάλυμα was somehow unique (as if the only inn in town) or that the κατάλυμα must have been sufficiently well known.Footnote 37 The problem facing Joseph and Mary in the story was not that they were denied a particular or well-known place to stay when they first arrived, but that their place to stay was not such that it could accommodate the birth and neonatal care of the baby Jesus.

The usage of the Greek article is somewhat different from the definite article in English, and translations should reflect that difference. Greek tends to tolerate the definite article more often than English when the anaphora is indirect and the noun was merely implied in the context, as is the case here. Joseph and Mary's being in Bethlehem implies that they must have had some place to stay. English, on the other hand, has different strategies for dealing with indirect anaphora. For example, when the context indicates that the object referred to by the noun is possessed by or belongs to a person in the context, English often employs a possessive pronoun for Greek's definite article.Footnote 38 For instance, the statement of Luke 5.2 in the story of the great catch of fish, οἱ δὲ ἀλιεῖς…ἔπλυνον τὰ δίκτυα, is translated into English as ‘the fishermen…were washing their nets’ (nrsv). The definite article τά before δίκτυα ‘nets’ implicitly refers to the nets that fishermen use, that is, ‘their nets’. Similarly the article before κατάλυμα in Luke 2.7 refers back to the accommodations that Joseph and Mary had at the beginning of v. 6, so it is appropriate to render the Greek definite article with an English possessive pronoun ‘their’. Thus, the preposition phrase at the end of 2.7 ought to be rendered as ‘in their accommodations’ or ‘in their place to stay’.

Some commentators have disputed the interpretation of Luke 2.7 as referring to the place where Joseph and Mary were already staying, arguing that the dative pronoun αὐτοῖς refers to the parents, not the newborn.Footnote 39 Had Luke wanted to imply that there was no place for the baby Jesus, the argument goes, he would have used the singular αὐτῷ instead or he would at least have specified ‘no other place’.Footnote 40 The flaw in this argument is not the premise that αὐτοῖς refers to the parents (which it reasonably does), but the assumption that αὐτοῖς is a dative of advantage construed with τόπος as ‘a place for them’. What makes this assumption dubious is that the use of a personal dative with the linking verb εἶναι (here, as the imperfect ἦν) usually signifies a dative of possession, not a dative of advantage. Further supporting this conclusion is the word order with αὐτοῖς following the verb ἦν, not the noun τόπος. Leading Greek grammars of the NT even cite this very verse as an example of the dative of possession.Footnote 41 Accordingly, οὐκ ἦν αὐτοῖς τόπος should not be translated as ‘there was not a place for them’ but rather as ‘they did not have a place’.

Putting these exegetical conclusions together, the entire clause should be rendered as ‘because they did not have space in their accommodations’ or ‘because they did not have room in their place to stay’.Footnote 42 This clause means that Jesus had to be born and laid in a manger because the place where Joseph and Mary were staying did not have space for him. Luke's point is not so much any inhospitality extended to Joseph and Mary but rather that their place to stay was too small to accommodate even a newborn.

4. The Referent for Κατάλυμα in Luke 2.7

Understanding the sense of κατάλυμα in Luke 2.7 does not end the analysis. It may still be possible to deduce with some specificity what kind of accommodations the narrative presupposes for Joseph and Mary. Notwithstanding the generality of the term κατάλυμα, the context surrounding Luke 2.7 provides three clues as to its nature: Joseph's compliance with the census order in vv. 3–4, the betrothal of Mary in v. 5, and the manger in v. 7.

The first clue is that Joseph's residence in the Lukan account is located in Bethlehem. Chapter 2 begins with a decree from Caesar Augustus for all the empire to get registered (v. 1), and the text continues with a note that ‘everyone was going to get registered, each to his own town’ (ἕκαστος εἰς τὴν ἑαυτοῦ πόλιν, v. 3). Then, ‘Joseph too goes up (ἀνέβη δὲ καὶ Ἰωσὴφ, v. 4) from Galilee out of a city Nazareth into Judea into David's city, which is called Bethlehem, (because he was of the house and lineage of David) to get registered (ἀπογράψασθαι, v. 5)’. Joseph's compliance with the edict to get registered in his own town by going up to Bethlehem establishes that his own town, according to Luke 2.3, is Bethlehem. This conclusion accords with our understanding that Roman censuses registered people by their residence and by where they own their fields.Footnote 43

It has been contended, however, that Bethlehem was not actually Joseph's residence but merely his ancestral town. Under this position, Joseph's actual home town would have been Nazareth, and the census would then have to require Joseph to go, not to his own town, but to his ancestral town.Footnote 44 To be sure, the text does note that Joseph was a descendent of David (v. 4b), making Bethlehem also his ancestral town, but, as Raymond Brown acknowledges: ‘In Roman censuses there is no clear evidence of a practice of going to an ancestral city to be enrolled; the oft-cited examples from Egypt are not the same as what Luke describes’.Footnote 45 Given the lack of evidence that an ancestral census had ever been practiced in the Roman Empire,Footnote 46 the position that Luke must have intended an ancestral census is an extraordinary claim that requires extraordinary exegetical evidence. The comment that Bethlehem was Joseph's ancestral town, without more, is insufficient to countermand the plain reading of vv. 3–4. Nevertheless, Brown opposes the position that Bethlehem was Joseph's town for the following reasons in his exegesis of v. 4b:

Still another suggestion is that he was returning to his home in Bethlehem (‘his own city’ of vs. 3) after having gone to Nazareth to claim Mary as his bride who lived there. These suggestions run against the reference to Nazareth as ‘their own city’ in 2.39 and against the indication in 2.7 that Joseph had no place to stay in Bethlehem…Footnote 47

Of these two additional reasons why Luke must have been referring to an extraordinary census by ancestry, the second one depends on a misunderstanding of κατάλυμα in v. 7 as explained above. That verse does not say that Joseph had no place to stay in Bethlehem. To the contrary, it states that Joseph did have a place to stay (the κατάλυμα after all), though it was inadequate to accommodate the newborn Jesus.

As for Luke 2.39, Brown's rendering of εἰς πόλιν ἑαυτῶν as ‘their own city’ may well be an example of exegetical inertia. ‘Their own city’ is the wording of the AV, and it had been perfectly fine for the inferior text it translated: εἰς τὴν πόλιν ἑαυτῶν with the article before πόλιν.Footnote 48 Yet earlier and superior manuscripts lack the article, as does the Nestle-Aland critical text, and Brown's argument does not take the absence of that article into full account. Without the article, the phrase εἰς πόλιν ἑαυτῶν is better rendered as ‘into a city of their own’. As A. T. Robertson had observed with respect to Luke 13.19 and 19.13, the sense of the possessive when the governing noun is anarthrous is ‘not quite the same’ as when the governing noun takes the article.Footnote 49 More particularly, the anarthrous constructions in Luke 13.19 (εἰς κῆπον ἑαυτοῦ, ‘a garden of his own’) and 19.13 (δέκα δούλους ἑαυτοῦ, ‘ten slaves of his own’) do not limit the men in the respective parables to having only one garden or to possessing just ten slaves. By the same token, Luke 2.39 does not restrict Nazareth as the only town of Joseph and Mary; indeed, the narrative had already identified two such towns: Nazareth as Mary's town (Luke 1.26, 56) and Bethlehem as Joseph's town (2.3–4). This summary statement, therefore, does not establish that Nazareth was ‘their own’ town earlier in the narrative when Jesus was born—only that it was so by the time they went back.Footnote 50

Thus, the clue that Luke considered Bethlehem to be Joseph's own town for census purposes tells us that the κατάλυμα presupposed in the narrative is unlikely to have been a commercial inn. In accordance with contemporary norms of hospitality, Luke's audience would have expected Joseph's relatives in his own town to have provided a place to stay for him and Mary if he had no house of his own.Footnote 51 Indeed, the temporary and tiny nature of his accommodations bolsters the supposition that Joseph's place to stay was not a house of his very own. But this is not the only indication in the narrative that Bethlehem was Joseph's home, there is another—often overlooked—hint: Mary was still betrothed to Joseph in Luke 2.5.

The second clue about the nature of Joseph and Mary's accommodations in Bethlehem is the detail that Joseph went up to Bethlehem ‘with Mary, his betrothed’ (2.5, σὺν Μαριὰμ τῇ ἐμνηστευμένῃ αὐτῷ). According to Luke, Mary was still betrothed on the way to Bethlehem, but by the time she gave birth to Jesus in v. 7, she was cohabitating with Joseph. According to Jewish practices in antiquity, marriages were initiated by a betrothal (אירוסין) and finalized by a ‘home-taking’ (נישואין) in which the bride is taken to her husband's house.Footnote 52 Both events were celebrated by a public feast, the former at the bride's house and the latter at the groom's house. Accordingly, in the logic of the narrative, the point that Mary was still betrothed upon her arrival in Bethlehem (v. 5) but later cohabited with him there (v. 7) means that Bethlehem was the site of their wedding, when Joseph concluded the betrothal period by taking her into his home.

To be sure, a Bethlehem location for the wedding of Joseph and Mary in the Lukan infancy account is hardly considered in the literature.Footnote 53 Some critics locate the wedding back in Nazareth, contending that a betrothed woman would not travel with her groom.Footnote 54 This contention, however, runs into two difficulties. First, it conflicts with 1 Macc 9.37–39, which shows that grooms were indeed known to travel along with their betrothed as part of a larger wedding party.Footnote 55 Second, it requires dubious interpretations of the term ἐμνηστευμένῃ, which ordinarily refers to a pre-marital state of betrothal.Footnote 56 For example, some commentators propose that the detail of Mary's being betrothed was merely to emphasize Mary's virginityFootnote 57 or at least hint that Joseph was not the real father.Footnote 58 Raymond Brown supposes the term is more inconsequential, arguing that Luke ‘simply reused the term “betrothed” previously employed in 1.27, without any detailed reflection on the steps in the matrimonial procedure’.Footnote 59 Joseph Fitzmyer even warns us to ‘avoid overliteral readings of this description’.Footnote 60 These proposals, however, operate as devices for disregarding the force of the term ‘betrothed’—an expedient that should be taken only if it is otherwise virtually impossible to make sense of the text with that meaning. But, as seen above for Luke 2.3–4, the text does make sense with ‘betrothal’, because Bethlehem as Joseph's own town is an ideal location for him to get married. As a result, the clue that Joseph and Mary finalized their betrothal in Bethlehem means their accommodations would have to be appropriate for newlyweds. In fact, the narrative presupposes accommodations appropriate for a long-term stay, because Joseph and Mary then spend at least forty more days there for her purification after the birth (Luke 2.22; cf. Lev 12.2–8). This clue thus makes it unlikely that Luke was thinking of the newlyweds' accommodations as an inn designed for an overnight stay.Footnote 61

The third clue is that Luke 2.7 states that Mary ‘gave birth to her firstborn son, swaddled him, and laid him down in a manger, because they did not have room in their place to stay’. Childbirth was the riskiest moment in the entire pregnancy during antiquity, potentially lethal for both the mother and child.Footnote 62 Whenever possible, women about to give birth relied on the help of relatives, friends, and midwives in and around town.Footnote 63 Yet Mary's accommodations did not have room for giving birth, so the birth had to occur elsewhere, in a place that included a manger. This detail does not mean, as it would to Western Europeans, that Mary gave birth to Jesus in a stable or barn, because mangers were also found in the main rooms of first-century Judean village houses. Typically, the main room was divided into two sections at different elevations separated by about a meter. The animals were housed in the lower section, the people slept in the upper section, and mangers were located between them.Footnote 64 These village houses, moreover, could have a small room, either on the roof or on the side, which accommodated family members and guests. For example, Jonathan Safrai summarizes his research on the Jewish home and family in antiquity as follows:

From the literary sources and archaeological excavations one finds that most houses had at least two storeys, and sometimes even three. Generally a single owner built a house and its upper chambers; but because of inheritances and sales divided ownership developed… The upper floors were not always full storeys; sometimes they consisted of single rooms on a roof or an attic with its entrance from a ladder inside the house. These attics could be used for a member of the household or as a guest room… Whether or not original plans called for upper storeys, it was common to add rooms or small structures to the roofs of houses and to the courtyards, as it became necessary. The most frequent reason was the expansion of a family; a newly married son customarily brought his wife to live in the family house. The father would set aside a room within the house for the couple or build a marital house (בית חתנות) on the roof. On such an occasion relatives, friends, and neighbours came to assist the father and celebrate the new arrangement.Footnote 65

Accordingly, the element of Luke's narrative that the place where Joseph and Mary were staying had no room to accommodate a newborn or a manger (v. 7) suggests to the reader that they had been staying in one of these small rooms built on top of, or onto the side of, a village family home, and that delivery itself took place in the larger, main room of the house. Since Bethlehem was Joseph's own town as presupposed in the Lukan infancy account, readers of this account could well picture the small apartment they were staying in as attached to the village family home of his close relatives, perhaps even the house he grew up in. Even further, the detail that Joseph brought his betrothed to Bethlehem (v. 5) indicates that their apartment was a marital chamber built for the newly married men of the family.Footnote 66

5. Conclusion

Luke's infancy narrative therefore presupposes the following events. Joseph took his betrothed Mary from Nazareth to Bethlehem (2.5). Bethlehem was his home town (v. 3) and, in accordance with the patrilocal marital customs of the day, it must also have been the place where they finalized their matrimonial arrangements by bringing her into his home. As a newly married man, he no longer would have to sleep in the main room of the village house with his other relatives, but he and his bride could stay in a marital chamber attached to the house until they could get a place of their own. They stayed there for some time until she came to full term (v. 6), and she gave birth to Jesus in the main room of the house rather than in her marital apartment because it was too small, and she laid the newborn in one of those mangers (v. 7) common to the main room of an ancient farmhouse. After staying at least another forty days in Bethlehem (v. 22; cf. Lev 12.2–8), Joseph and Mary eventually moved to Nazareth to make their home together in her family's town (v. 39; cf. 1.26–27).

To be sure, this scenario as presupposed in Luke's infancy account diverges greatly from the conventional Christmas story. There is no inn, no innkeeper, and no stable. But it is grounded in a careful exegesis of the text. The term κατάλυμα has a broad sense of a ‘place to stay’ and the final clause of Luke 2.7 should be translated as ‘because they had no space in their place to stay’. Joseph and Mary's particular accommodations in Bethlehem should be thought of as a cramped, marital chamber attached to his father's or other relative's village house. As unfamiliar and perhaps troubling this scenario may seem to us, we have come a long way from El Brocense's day. Now, nobody expects the Spanish Inquisition.

References

1 El Brocense's proceedings before the Spanish Inquisition were published by Tovar, Antonio and de la Pinta Llorente, Miguel, eds., Procesos Inquisitoriales contra Francisco Sánchez de las Brozas (Documentos para la historia del humanismo español 1; Madrid: Instituto Antonio de Nebrija, 1941)Google Scholar and earlier by Navarrete, Martin Fernandez, Salvá, Miguel, and de Baranda, Pedro Sainz, eds., Colección de documentos inéditos para la historia de España (vol. 2; Madrid: La Viuda de Calero, 1843)Google Scholar.

2 According to the declaration of Juan Collado, El Brocense was originally circumspect about this matter: ‘Dijo: que en la dicha leçión el dicho maestro dijo que lo que se dice en la scriptura de que nuestro Señor estuvo en el pesebre, que no se avía de entender como comúnmente se piensa sino de otra manera: e que de la manera que se avía de entender no lo declaró’ (Tovar and Llorente, Procesos Inquisitoriales, 9–10). El Brocense's actual position as summarized above became clearer during the proceedings.

3 El Brocense's exegesis of Luke 2.1–7 has been published in Tovar and Llorente, Procesos Inquisitoriales, 51–3, and Fernandez Navarrete et al., Colección, 51–3. Credit for bringing this to the attention of contemporary scholarship belongs to Yubero, Dionisio, ‘Una opinión original del “Brocense” sobre Luc. 2,7’, CB 22 (1954) 3–6Google Scholar.

4 The least convincing of his arguments would have to be his appeal to the ‘house’ (οἰκίαν) of Matt 2.11. Not only would this now be considered an improper harmonization to the Matthean account by modern standards, but it also falls short by his own standards since he argued in another context that the Magi did not arrive in Bethlehem for another a year or two: ‘Magos Christum Dominum adoraturos post annum unum vel duos potiùs venisse’ (Tovar and Llorente, Procesos Inquisitoriales, 53).

5 Tovar and Llorente, Procesos Inquisitoriales, 52: ‘Diversorium autem non hic accipitur pro eo quod vulgò dicimus meson, sed pro quavis habitatione privata ut lib. [3.o Regum c. 18], et [D. Lucæ c. 22]: ubi est diversorium ubi pascha cum discipulis meis manducem? Græcè κατάλυμα’. (The Biblical citations are corrected from Fernandez Navarrete et al., Colección, 52, and appear to refer to 1 Kings 18.27 and Luke 22.11 more precisely.) The instance of κατάλυμα in Luke 22.11 where it does not mean ‘inn’, is routinely noted by scholars.

6 Tovar and Llorente, Procesos Inquisitoriales, 52: ‘Venit igitur Joseph aut in domum suam (erat enim civis Beleemita de domo David) aut certè in domum alicujus propinqui, si propria domus erat inquilinis locata’. Less persuasively, El Brocense goes on to cite Theophylact's view that Bethlehem was also Mary's hometown. That Joseph could count on the hospitality of his relatives has often been pointed out in the literature. See, for example, Malina, Bruce J. and Rohrbaugh, Richard L., Social-Science Commentary on the Synoptic Gospels (Minneapolis: Fortress, 2d ed. 2003) 376Google Scholar: ‘If close family was not available, mention of Joseph's lineage would have resulted in immediate village recognition that he belonged and space in a home would have been made available’. Also Grassi, Santi, Luca (Commenti biblici; Rome: Borla, 1999) 98Google Scholar: ‘Con tutta probabilità i genitori di Gesù erano ospiti nella casa natale di Giuseppe o presso parenti’; and Miguens, M., ‘“In una mangiatoia, perchè non c'era posto…”’, Bibbia e Oriente 2 (1960) 193–8Google Scholar.

7 Tovar and Llorente, Procesos Inquisitoriales, 52: ‘Nec tamen necesse erat eodem die omnes adesse: satis enim fuit intra præscriptum aliquem diem profiterentur’. In fact, those subjected to the census had an entire year to register. See Llewelyn, S. R., ‘§15 “And everyone went to his own town to register”’, New Documents Illustrating Early Christianity: A Review of the Greek Inscriptions and Papyri published in 1980–81 (vol. 6; Macquarie University: The Ancient History Documentary Research Centre, 1992) 121Google Scholar; and Hombert, Marcel and Préaux, Claire, Recherches sur le Recensement dans l'Égypte Romain (Papyrologica Lugduno-Batava; Leiden: Brill, 1952) 79–80, 108Google Scholar.

8 Tovar and Llorente, Procesos Inquisitoriales, 52–3: ‘Itaque dicit Evangelista: quia non erat eis locus in diversorio, id est, quia in illa domo nec erant cunæ, nec alius commodior locus ubi collocatur puer, in præsepio posuerunt eum. Solet enim fieri multis in regionibus (quod sæpè videmus et in nostris) ut in eadem parte domus et domini et boves et jumenta commorentur’. Nowadays scholars bolster the observation by pointing to Near Eastern homes: e.g., Bailey, Kenneth E., Jesus Through Middle Eastern Eyes: Cultural Studies in the Gospels (Downers Grove, IL: IVP Academic, 2008) 28–30Google Scholar; and Dalman, Gustaf, Sacred Sites and Ways: Studies in the Topography of the Gospels (trans. Levertoff, Paul P.; New York: Macmillan, 1935; Germ. ed. 1924) 41–3Google Scholar.

9 Bell, Aubrey F. G., Francisco Sanchez El Brocense (Hispanic Notes and Monographs 8; Oxford: Oxford University, 1925) 30–1Google Scholar, makes the case that El Brocense was reprimanded instead of being imprisoned because he was under the protection of Pedro de Portocarrero, who later became Grand Inquisitor in 1594. Five days after the latter's death on September 20, 1600, the Inquisition again moved against El Brocense, who died the following December under house arrest at the age of 78.

10 The most thorough recent studies include Bailey, Jesus, extending the observations of Bailey, Kenneth E., ‘The Manger and the Inn: The Cultural Background of Luke 2.7’, NETR 2 (1979) 33–44Google Scholar; and Benoit, Pierre, ‘“Non erat eis locus in diversorio” (Lc 2,7)’, Mélanges bibliques en hommage au R. P. Béda Rigaux (ed. Descamps, Albert and de Halleux, R. P. André; Gembloux: Duculot, 1970) 173–86Google Scholar.

11 A case in point is Trudinger, L. Paul, ‘“No Room in the Inn”: A Note on Luke 2.7’, ExpT 102 (1991) 172–3Google Scholar, who argues that the innkeeper was actually ‘compassionate’ and ‘sensitive’ for turning away the family from a place infested with ‘thieves and cut-throats’. Trudinger did not extrapolate his thesis, however, to the story of the Good Samaritan, who arranged for the beaten traveler to be put up in an inn (Luke 10.25–27).

12 So Danker, Frederick, ‘κατάλυμα’, BDAG (2000) 521Google Scholar: ‘The sense of inn is possible in Lk 2.7, but in 10.34 Lk uses πανδοχεῖον, the more specific term for inn. κ[ατάλυμα] is therefore best understood here as lodging…or guest-room, as in 22.11; Mark 14.13, where the contexts also permit the sense dining-room’ (citations omitted). Also favoring ‘guest-room’ include Fabris, Rinaldo, I Vangeli: Luca (Assisi: Cittadella, 2003) 61Google Scholar: ‘stanza degli ospiti’; Bailey, Jesus, 32–3; Green, Joel B., The Gospel of Luke (NICNT; Grand Rapids, MI: Eerdmans, 1997) 129Google Scholar; Witherington, Ben III, ‘Birth of Jesus’, Dictionary of Jesus and the Gospels (Downer's Grove, IL: InterVarsity, 1992) 60–74Google Scholar; and Byrne, Matthew, ‘No Room for the Inn’, Search 2.2 (1982) 37–40Google Scholar.

13 So Legrand, L., ‘The Christmas Story in Lk 2.1–7’, ITS 19 (1982) 289–317 at 308Google Scholar: ‘the use of the article (the kataluma) implies that the text does not speak of any inn but of a well defined place’.

14 Blass, F., Debrunner, A., and Funk, Robert W. [hereinafter ‘BDF’], A Greek Grammar of the New Testament and Other Early Christian Literature (Chicago: University of Chicago, 1961) 102 § 189(1)Google Scholar; Robertson, A. T., A Grammar of the Greek New Testament in the Light of Historical Research (New York: Hodder & Stoughton, 4th ed. 1923) 541Google Scholar.

15 E.g., Benoit, ‘Non erat’, 184.

16 Noted, e.g., by Bailey, Jesus, 32.

17 Brown, Raymond E., The Birth of the Messiah: A Commentary on the Infancy Narratives in the Gospels of Matthew and Luke (ABRL; New York: Doubleday, new upd. ed. 1993; 1st ed. 1977) 400Google Scholar.

18 Brown, Birth, 400.

19 Brown, Birth, 400, also arguing for the first option: ‘Yet Luke's use of the definite article seems to preclude his referring to a totally unidentified home’.

20 Specifically, Brown, Birth, 400, dispels modern misconceptions about such an inn and appeals to a khan near to Bethlehem mentioned 500 years earlier in Jer 41.17. Other recent scholars supporting the ‘inn’ interpretation include: Miller, Robert J., Born Divine: The Births of Jesus and Other Sons of God (Santa Rosa, CA: Polebridge, 2003) 56, 58Google Scholar: ‘travelers’ shelter'; Louw, Johannes P. and Nida, Eugene A., Greek-English Lexicon of the New Testament Based on Semantic Domains (New York: United Bible Societies, 2d ed. 1989) 1.82Google Scholar; Goulder, Michael D., Luke: A New Paradigm (JSNTSS 20; Sheffield: Sheffield University, 1989) 253Google Scholar: ‘a single public hostelry for travellers’; and Fitzmyer, Joseph A., The Gospel According to Luke (I–IX) (AB 28; New York: Doubleday, 1970) 408Google Scholar: ‘a public caravansary or khan, where groups of travelers would spend the night under one roof’.

21 Brown, Birth, 400.

22 Brown, Birth, 400. Benoit, ‘Non erat’, 185, whom Brown cites, also suggests that it is perhaps wrong (due to our excessive historicizing and psychologizing of this text) for us to want to specify too much what Luke wanted to put there: ‘A côté de cela, le κατάλυμα a peu de poids, et on a peut-être tort de vouloir trop préciser ce qu'il a voulu y mettre’.

23 For a discussion of the distinction between sense and reference, see, e.g., Osborne, Grant R., The Hermeneutical Spiral: A Comprehensive Introduction to Biblical Interpretation (Downers Grove, IL: InterVarsity, 1991) 77Google Scholar; and Silva, Moisés, Biblical Words and Their Meaning: An Introduction to Lexical Semantics (Grand Rapids, MI: Zondervan, rev. ed. 1994) 102–8Google Scholar.

24 See, e.g., Danker, et al. , ‘καταλύω’, BDAG (2000) 521–2Google Scholar. This sense development is hardly unique; in Biblical Aramaic, for example, the verb שׁרא, ‘to loosen’, also means ‘to abide’, as does the Syriac verb ܫܪܐ.

25 Smyth, Herbert Weir, Greek Grammar (Cambridge, MA: Harvard University, 1920) 231–2 § 841Google Scholar; BDF, 58–9 § 109(2).

26 Husson, Geneviève, Oikia: Le vocabulaire de la maison privée en Égypte d'après les papyrus grecs (Papyrologie 2; Paris: Sorbonne, 1983) 133–6Google Scholar. The usage in the Roman period is much less clear, however; it occasionally refers to some kind of agricultural building.

27 Husson, Oikia, 135: ‘c'est le logis où l'on descend, où l'on séjourne pendant quelque temps’.

28 Studies of the Septuagintal usage of κατάλυμα include: Iglesias, Salvador Muñoz, Los Evangelios de la Infancia III: Nacimiento e infancia de Juan y de Jesús en Lucas 1–2 (vol. 3; Madrid: Biblioteca de Autores Cristianos, 1987) 93Google Scholar; dell'Acqua, Anna Passioni, ‘Ricerche sulla versione dei LXX e i papiri’, Aegyptus 61 (1981) 171–211, at 203–5Google Scholar; Benoit, ‘Non erat’, 179–80.

29 The Matthean parallel at 26.18 is even less specific: πρὸς σὲ ποιῶ τὸ πάσχα (‘with you [or, at your house] I will do the Passover’).

30 This is what caused Brown, Birth, 400, to back off from translating κατάλυμα as ‘inn’. Goulder, Luke, 253, on the other hand, merely sees πανδοχεῖον as synonymous.

31 Surveyed by, for example, Benoit, ‘Non erat’, 181 n. 2.

32 The Curetonian manuscript, containing the other Old Syriac version, unfortunately no longer preserves Luke 2.7.

33 Lewis, Charlton T. and Short, Charles, ‘deversorius’, A Latin Dictionary (Oxford: Clarendon, 1879) 565Google Scholar.

34 Tovar and Llorente, Procesos, 52.

35 Lewis and Short, ‘stabulum’, 1749–50.

36 Bailey, Jesus, 32, states that the literal meaning of κατάλυμα is a ‘place to stay’, and then argues that it refers to a guest room. He does not, however, offer a specific translation of 2.7.

37 So McNamara, Elmer A., ‘“Because There Was No Room for Them in the Inn”’, The Ecclesiastical Review 105 (1941) 433–43 at 435–6Google Scholar: ‘That some such inn was meant by St. Luke, is attested to by his use of the definite article with the noun, i.e., there was no room for them in the inn. He supposes the inn was well known, probably because it was public and very likely the only one since Bethlehem was a small town’.

38 See generally, Wallace, Daniel B., Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament (Grand Rapids, MI: Zondervan, 1996) 215–16Google Scholar; Robertson, Grammar, 684; Smyth, Greek Grammar, 287 at § 1121: ‘The article often takes the place of an unemphatic possessive pronoun when there is no doubt as to the possessor’.

39 E.g., Bovon, François, Luke 1: A Commentary on the Gospel of Luke 1.1–9.50 (trans. Thomas, Christine M.; Hermeneia; Minneapolis: Fortress, 2002) 86Google Scholar; and Brown, Birth, 399.

40 Bovon, Luke, 86: ‘It contradicts the text to say that the parents found no room only for the child and accordingly laid it in a manger; it says there was no room for them, not him’ (italics original). See also Cadbury, Henry J., ‘Lexical Notes on Luke-Acts: III. Luke's Interest in Lodging’, JBL 45 (1926) 305–22 at 318Google Scholar.

41 BDF, Grammar, 102 § 189(1); Robertson, Grammar, 541. Cf. Wallace, Greek Grammar, 149–50; Smyth, Greek Grammar, 341 § 1476. See also Schürmann, Heinz, Das Lukasevangelium (HTKNT 3; Freiburg: Herder, 2d ed. 1969; repr. 1982)Google Scholar 1.105 n. 57.

42 On the use of τόπος to mean ‘room’ or ‘space’ in Luke, see 14.22, ‘And the slave said, Lord, what you have ordered has happened, and there still is room (καὶ ἔτι τόπος ἐστίν)’.

43 Much has been written about the practice of the Roman census; leading treatments include: Llewelyn, New Documents, 112–19; Benoit, P., ‘Quirinius [Recensement de]’, Dictionnaire de la Bible Supplément (vol. 9; Paris: Letouzey & Ané, 1979) 693–720Google Scholar; Schürer, Emil, The History of the Jewish People in the Age of Jesus Christ (175 b.c.–a.d. 135) (ed. Vermes, Geza and Millar, Fergus; Edinburgh: T&T Clark, English rev. ed. 1973) 1.411–13Google Scholar; Braunert, Horst, ‘Der römische Provinzialzensus und der Schätzungbericht des Lukas-Evangeliums’, Historia 6 (1957): 192–214Google Scholar; and Hombert and Préaux, Recherches, 67–70.

44 E.g., Brown, Birth, 396. Also, Green, Luke, 126; and Bock, Darrell L., Luke 1.1–9.50 (BECNT 3A; Grand Rapids, MI: Baker, 1994) 204–6Google Scholar.

45 Brown, Birth, 396. Brown, Birth, 549, also warns: ‘It is dangerous to assume that he described a process of registration that would have been patently opposed to everything that he and his readers knew’.

46 It is sometimes argued that the Romans might have accommodated their census practice in Palestine to Jewish customs (e.g. Olshausen, Hermann, Biblical Commentary on the New Testament [vol. 1; trans. Kendrick, A. C.; New York: Sheldon, 1863Google Scholar; orig. German ed. 1830] 237). Even aside from the lack of any affirmative evidence for this particular accommodation, Schürer et al., History, 412, point out that it would have been unworkable.

47 Brown, Birth, 396.

48 The Greek text used for the AV was an edition of the Textus Receptus, which was based on late and inferior manuscripts in the Byzantine textual tradition.

49 Robertson, Grammar, 690. Though Robertson did not also list Luke 2.39 in conjunction with 13.19 and 19.13, the relevance of Robertson's examples to 2.39 has been noticed by Smith, Jay E., ‘1 Thessalonians 4.4: Breaking the Impasse’, BBR 11 (2001) 65–105 at 80 n. 65Google Scholar.

50 Another example of exegetical inertia in Brown's argument is his use of the term ‘returning’. That rendering is also fine for the Textus Receptus, which read ὑπέστρεψαν, ‘they returned’, instead of the blander ἐπέστρεψαν, ‘they went back’, in accordance with the best manuscripts and the Nestle-Aland critical text of the NT.

51 Malina and Rohrbaugh, Social-Science Commentary, 376 (cited in n.6); Green, Luke, 129: ‘Mary and Joseph, then, would have been the guests of family or friends’.

52 There is also much literature on ancient Jewish marriage customs. Some of the most useful modern treatments include: Satlow, Michael L., Jewish Marriage in Antiquity (Princeton: Princeton University, 2001)Google Scholar; Ilan, Tal, ‘Premarital Cohabitation in Ancient Judea: The Evidence of the Babatha Archive and the Mishna (Ketubbot 1.4)’, HTR 86 (1993) 247–64CrossRefGoogle Scholar; and Archer, Léonie J., Her Price is Beyond Rubies: The Jewish Woman in Graeco-Roman Palestine (JSOTSS 60; Sheffield: Sheffield Academic, 1990)Google Scholar. Roman marriage practices were generally similar in this respect except that it was easier to break off the engagement. See generally, Treggiari, Susan, Roman Marriage: Iusti Coniuges from the Time of Cicero to the Time of Ulpian (Oxford: Clarendon, 1991)Google Scholar.

53 The clearest recognition of this aspect of the Lukan infancy account I could find is Box, G. H., The Virgin Birth of Jesus: A Critical Examination of the Gospel-Narratives of the Nativity, and Other New Testament and Early Christian Evidence, and the Alleged Influence of Heathen Ideas (London: Pitman, 1916) 214Google Scholar: ‘If Joseph's home was in Bethlehem, by taking Mary, his betrothed, with him when he left Nazareth for his home-town, he was performing the central and public act which proclaimed the marriage’.

54 E.g., Bock, Luke, 205: ‘It does not suggest that Mary is not yet married to Joseph, since this trip in a betrothal situation would be unlikely’. Marshall, Luke, 105: ‘it is unlikely that she would have accompanied Joseph had she been merely betrothed to him’. Schmid, Josef, Das Evangelium nach Lukas (RNT 3; Regensburg: Pustet, 1960) 65Google Scholar: ‘Wäre Maria in diesem Zeitpunkt erst (gegen Mt 1, 24) die Verlobte Josephs gewesen, so wäre es ein grober Verstoß gegen die Sitte gewesen, wenn er mit ihr zusammen nach Bethlehem gereist wäre und dort mit ihr zusammenwohnt hätte’. Plummer, Luke, 53: ‘Had she been only his betrothed (i. 27; Mt. i. 18), their travelling together would have been impossible’. None of these commentators adduce any evidence for their claims.

55 Archer, Price, 197–8; and Satlow, Jewish Marriage, 150–1.

56 This may be another case of exegetical inertia. Both the Textus Receptus and the Vulgate read the apparently contradictory ‘betrothed wife’ in Luke 2.5 (τῇ μεμνηστευμένῃ… γυναικί and desponsata…uxore, respectively), which had prompted many exegetes to weaken the force of ‘betrothed’.

57 E.g, Bock, Luke, 206; Schmid, Lukas, 65; Leaney, A. R. C., A Commentary on the Gospel According to St. Luke (HNTC; New York: Harper & Brothers, 1958) 23Google Scholar; Lattey, Cuthbertus, ‘“Ad Virginem desponsatam viro” (Lc 1,27)’, VD 30 (1952) 30–33 at 32Google Scholar; and Lagrange, M.-J., Évangile selon Saint Luc (Paris: Lecoffre, 8th ed. 1948) 70Google Scholar.

58 Schaberg, Jane, The Illegitimacy of Jesus: A Feminist Theological Interpretation of the Infancy Narratives (San Francisco: Harper & Row, 1987) 91Google Scholar.

59 Brown, Birth, 397.

60 Fitzmyer, Luke, 407.

61 Times have changed. Now, many couples prefer to honeymoon in hotels.

62 See generally, Ilan, Tal, Jewish Women in Greco-Roman Palestine: An Inquiry into Image and Status (Tübingen: Mohr Siebeck, 1995) 116–19Google Scholar; Levin, S., ‘Obstetrics in the Bible’, Journal of Obstetrics and Gynaecology of the British Empire 67 (1960) 490–8CrossRefGoogle ScholarPubMed; and de Vaux, Roland, Ancient Israel: Its Life and Institutions (trans. McHugh, John; New York: McGraw-Hill, 1961; French orig. 1957) 42–3Google Scholar. For childbirth in Greco-Roman contexts, see also French, Valerie, ‘Midwives and Maternity Care in the Roman World’, Rescuing Creusa: New Methodological Approaches to Women in Antiquity (ed. Skinner, Marilyn; Helios ns 13/2; Lubbock: Texas Tech University, 1987) 69–84Google Scholar.

63 Bailey, Jesus, 26; and Levin, ‘Obstretics’, 494. This was still true well into twentieth-century Palestine; see Granqvist, Hilma, Birth and Childhood among the Arabs: Studies in a Muhammadan Village in Palestine (Helsinki: Söderström, 1947) 56–8Google Scholar. Even the Protevangelium of James 18.1, which recounts a miraculous birth for Jesus, includes a midwife.

64 Discussed by, for example, Bailey, Jesus, 28–31; and Dalman, Sacred Sites, 41.

65 Safrai, Jonathan, ‘Home and Family’, The Jewish People in the First Century (ed. Safrai, Samuel et al. ; CRINT 1/2; Assen: Van Gorcum, 1976) 730–31Google Scholar (footnotes omitted).

66 Cf. Büchler, A., ‘The Induction of the Bride and the Bridegroom into the חופה in the First and Second Centuries in Palestine’, Livre d'hommage à la mémoire du Dr Samuel Poznański (Warsaw, 1927) 82–132Google Scholar, who argues that such a marital room built for the bridegroom corresponded to the חופה (chuppah) in its first- and second-century guise.