A favorite topic of modern critical scholars is the role of apocryphal books in early Christianity. How often were these books used? And did Christians regard them as Scripture? Bart Ehrman’s book Lost Christianities is typical in this regard. Ehrman explores a number of books that did not make it into the canon and argues that Christians originally regarded these books as part of God’s word.
One critical piece of evidence for Ehrman is that the Shepherd of Hermas and the Epistle of Barnabas were both included in one of our earliest complete NT manuscripts, the fourth-century Codex Sinaiticus (p.245). He also appeals to the fact that 1&2 Clement were included in the fifth century Codex Alexandrinus (p.142).
Now, to be sure, the inclusion of these apocryphal books in Sinaiticus and Alexandrinus is interesting. And it certainly attests to the popularity of these books. But is it necessarily evidence of their scriptural status? One key fact that Ehrman does not mention in his discussion is the location of these apocryphal books within the codices. They all occur at the end of the codex, even after Revelation. This is certainly noteworthy since Revelation is uniformly regarded as the final book in the NT corpus.
Moreover, if a book like 1 Clement or the Epistle of Barnabas were regarded as scriptural, then surely it would have been placed among the other General/Catholic epistles, which contain letters from a variety of different authors. Even Hebrews, with its disputed authorship, is not placed after Revelation, but was included (in these codices) amongst Paul’s letters.
But, if these books are not regarded as Scripture, then why are the included in these codices at all? One possible explanation, suggested by William Horbury, is that early canonical lists would include all the received books first and then, at the end, it would name “disputed” books and/or books that were deemed useful or edifying but not necessarily canonical. This pattern is visible in the Muratorian fragment which includes the Apocalypse of Peter and the Shepherd at the end. Whereas the former is regarded as disputed, the latter is rejected outright and is mentioned only because it can still be read in the church (for edification, not as Scripture).
Athanasius’ Festal Letter (c.367) also lists both Old and New Testament books and then, at the end, lists the books which are not canonical but nevertheless useful, such as the Wisdom of Solomon and the Shepherd amongst others. Likewise, Eusebius, Codex Claromontanus, and Epiphanius, exhibit a similar structure in their canonical “lists.”
Thus, it seems that Codex Sinaiticus and Alexandrinus are simply following the standard structure of canonical lists in their time period. Given the location of these apocryphal books in these codices, they are best understood not as Scripture, but as helpful/edifying books that can be read in the church.
 W. Horbury, “The Wisdom of Solomon in the Muratorian Fragment,” JTS 45 (1994): 149-159.
 Hist. eccl. 5.8.8, also joins the Wisdom of Solomon and the Shepherd together.
 Pan. 76.5.
 Horbury, “The Wisdom of Solomon,” 152-156.