date to the text. While dates as late as the fifth century were proposed at the beginning of the twentieth century, these proposals were debunked with the discovery of the third- or fourth-century Bodmer Miscellanous Codex in 1952, which serves as our earliest manuscript of the Protevangelium.26 In support of the earlier dating of the text scholars have looked for indications of knowledge of the text in the works of early church writers.27 As noted above, Origen (ca. 185–254) cites a “Book of James” or the “Gospel of Peter” as a source for the tradition that Joseph was previously married and had children before his engagement to Mary (Comm. Matt. 10.17). This belief is articulated several times throughout the narrative: Joseph’s protest against taking on Mary is based on his already having children (Prot. Jas. 9:9); Joseph makes reference to his sons and specifically names Samuel as one of his children when contemplating how he will register for the upcoming census (17:2); and Joseph leaves his sons to care for Mary while he searches for a midwife (18:1). Clement (ca. 150–215), an older contemporary of Origen, mentions the tradition of a midwife who aids Mary at the birth and attests to her virginitas in partu (Strom. 7.16). Although Clement does not provide a source for this knowledge, he does relay that it is a widely held belief by most people. Other witnesses to Mary’s post-partum virginity include Irenaeus of Lyon (Epid. 54), Ignatius of Antioch (Eph. 19.1), the Ascension of Isaiah 11:1–16, and the Odes of Solomon 19.28 Finally, the Carthaginian author Tertullian (ca. 160–225) vehemently refutes Mary’s eternal virginity (non virgo quantum a partu; Carn. Chr. 23) as well as the possibility that Jesus’ brothers and sisters were Joseph’s children from a previous marriage (Carn. Chr. 7).
P. A. van Stempvoort has drawn on additional evidence to propose a precise date of composition between 178 and 204 CE.29 The terminus a quo of ±178 CE is based on van Stempvoort’s reading of Celsus’s True Doctrine (preserved in Origen’s Contra Celsum), which attacks Mary’s character most viciously. Celsus questions the legitimacy of Mary’s virginity, respected lineage, high social background, and livelihood. Reading the Protevangelium as an apology, van Stempvoort asserts that the text functioned specifically to counter these slanders. For example, he cites the Protevangelium’s reference to Mary’s wealth and royal lineage as a rebuttal of Celsus’s accusation that Mary was a poor village girl who had to spin for a living (Cels. 1.28–32). His highly scandalous accusation that Mary’s child, born in secret, was the product of an adulterous relationship with a Roman soldier named Panthera (Cels. 1.32) is countered by the emphatic statements and physical proof of her maintained virginity found throughout the Protevangelium. Van Stempvoort adds that Origen knew a Biblos Iakobou and that many of his contemporaries had knowledge of the Protevangelium’s content, thus strengthening the earliest possible date of 178 CE. For his specific terminus ad quem date of 204 CE, Van Stempvoort looks to Hippolytus’s homily on Susanna (in Comm. Dan.). Seeing strong parallels between the depiction of Susanna and the two major female figures in the Protevangelium, he suggests that the parallels are indicative of a shared compositional time period. Van Stempvoort’s incredibly precise dating may give some readers pause,30 but his proposal shines important light on sources that may indicate knowledge of the Protevangelium. If Origen’s, Clement’s, and Tertullian’s source for all these unique details is indeed the Protevangelium, its terminus ad quem can be placed reasonably at the beginning of the third century CE.
Another ambitious attempt at dating the text has been made by George Zervos who has proposed a very early date based on his determination of a literary dependency between the Protevangelium and Justin Martyr (d. 160), who makes reference to Mary giving birth in a cave outside of Bethlehem (1 Apol. 1.33). Zervos proposes a date no later than 150–160, the period in which Justin was actively writing, and a date no earlier than 80–90 when Matthew and Luke were discussing the virgin birth.31 Zervos’s argument is based in part on the work of Émile de Strycker, who suggested that the Protevangelium knew Justin’s work grounded on four concordances he found between the two texts.32 Zervos reverses the direction of dependency.33 While Zervos has not been able to persuade many, his theory is intriguing and speaks to the range of proposed dates for our text.
Provenance
The Protevangelium’s provenance is one of the more debated issues surrounding the text’s origin, most likely because it is intricately related to questions regarding its relationship to Judaism (see section on relationship to Judaism below). Frequent dismissals of proposed geographical areas largely depended upon the text’s knowledge of Judaism or lack thereof. The problematic assumption that knowledge of “Jewish tradition and customs” would necessitate a Palestinian setting while the lack of this knowledge would require a different location functioned as the dominant determinant for how the text’s provenance should be discussed. Diasporic Jews who engaged in pilgrimages to Jerusalem offer a clear example of why such an approach is flawed, inasmuch as such travels gave them intimate knowledge of the geographical space of Jerusalem, despite not being inhabitants of the area.34
Other contentious criteria traditionally used for determining locale involve references to or knowledge of geographical and/or environmental details. De Strycker is often cited in this regard, as he rejected Palestine because of the author’s apparent confusion about the distance between Jerusalem, Judea, and Bethlehem with regard to Mary’s and Joseph’s travels. He argues instead for an Egyptian provenance based on linguistic features and the influence of Coptic elements in the author’s writing, but also on the description of the mountains and wilderness at Prot. Jas. 1:9; 4:5; and 16:4–5, which he argues is characteristic of the geographical landscape of Egypt.35 Also problematic is the close proximity of the gates of Jerusalem to the desert (Prot. Jas. 4:4–5), which persuaded de Strycker even further that the text was not composed in Palestine.36 É. Cothenet also supports Egypt as the place of origin, but his reasoning is based on Origen’s and Clement’s knowledge of the text.37
Not everyone has so readily rejected Palestine as a possible place of composition. Malcolm Lowe reevaluated the passages traditionally viewed as demonstrating ignorance of Palestinian geography and proposed instead an alternative interpretation of the details. He suggests that Joseph’s reference to being “in,” “near,” or “around” Bethlehem before preparing to “depart for Judea” (Prot. Jas. 21:1) may be influenced by John 3:22, where Jerusalem is described as distinct from Judea. In this case, Jesus is described as going forth into Judea from Jerusalem. Lowe also offers other examples from Jewish literature (e.g., Ezra 1:2, 1:3, 2:1; m. Ket. 4.12), which acknowledge that a distinction can be made between Jerusalem and Judea.38 In response to the widely accepted proposals by de Strycker over the location of mountains and deserts, Lowe also argues that Palestine cannot be so easily dismissed if one considers the heavy rainfall on the western slopes that would allow for the desert to start on the eastern slopes of the mountains that borders Jerusalem. To further support his claim, Lowe notes that the author calls the people not Ἰουδαῖοι (Judeans), but “Israel,” that is, the term commonly used by writers living in Palestine. The only reference to ᾿Ιουδαῖοι is made by one of the Magi, a non-Israelite, in his questioning of Herod over the location of the newborn King of the Jews. While Lowe’s investigation offers legitimate pause for the exclusion of Palestine, several scholars have noted that his explanation for Joseph going “from Bethlehem to Judea” at Prot. Jas. 21:1 is still not adequate.39
While the debate on provenance remains unsettled, perhaps the strongest contender for the text’s origin is Syria. Much like the arguments put forth for Egypt, however, traditional proposals have also relied on dubious assumptions about the proposed location’s geography and environment. H. R. Smid, for instance, rejected an Egyptian origin because he was unconvinced that the geographical details reflected a specifically Egyptian landscape. Opting for Syria instead, he cites the popular laurel trees and gardens of Syria as having inspired Anna’s garden scene.40 More recent trends have seen Syria as the more persuasive locale, but the criteria for determining such a conclusion have moved away from geographical details to a focus on literary parallels and the concerns and interests of the texts that may be more popularly reflected there. Ron Cameron notes that the Protevangelium’s harmonization techniques parallel the popular use of gospel harmonies in Syria.41 Drawing comparisons between the Protevangelium’s depiction of Mary’s unique virginal status and the birth
