Papyrus P38: Acts 18:27–28 in a Greek Witness
Michigan Papyrus 1571, or P38, is a three-piece leaf from about AD 300 preserving Acts 18:27–19:6 and 19:12–16 in a distinctive text closely related at points to Codex Bezae.
- Artifact date
- 4th Century
- Findspot
- Egypt; exact ancient findspot is not stated in the consulted manuscript catalogs
- Current location
- Ann Arbor, Michigan, University of Michigan Library, Inv. Nr. 1571
What archaeologists can actually examine
P38 preserves Acts 18:27–19:6 and 19:12–16, considerably more than the two verses in its title. The text moves from Apollos' ministry into Paul's encounter with disciples at Ephesus, their baptism and reception of the Spirit, and later extraordinary healings and the failed exorcism by sons of Sceva. University of Michigan inventory 1571 consists of three fragments from a one-column papyrus codex leaf, usually dated around AD 300. Its wording includes conspicuous agreements with the expanded or 'western' tradition represented by Codex Bezae, while also differing from Bezae. The papyrus demonstrates early textual plurality in Acts; it does not identify one complete version as automatically original.
Acts 18–19 survives in one leaf
The fragments preserve Acts 18:27–19:6 and 19:12–16, joining Apollos, Ephesian disciples, healings, and Sceva's sons.
Bezae-like readings are early
Several expansions and rearrangements align with Codex Bezae, showing that related wording circulated in Greek around AD 300.
Similarity is not identity
P38 also differs from Bezae and should not be described as its copy, ancestor, or a uniform representative of every western reading.
Provenance remains unresolved
The papyrus was purchased in Cairo in 1924; Sanders explicitly lacked evidence to assign its origin beyond Egypt.
Follow the evidence from discovery to debate
Move through the find in a deliberate order: establish its provenance, examine the surviving object, place it in time, and then weigh the scholarly interpretation.
How the find came to light
Michigan Papyrus 1571 was bought in Cairo in 1924 as part of a large purchase allocated among contributing institutions by H. I. Bell. Henry Sanders reported that it arrived in three dirty pieces, one only partly unfolded, already identified as Acts and assigned to the fourth century. He cleaned, joined, and published it in the Harvard Theological Review in 1927. Many documents in the lot came from the Fayum, but Sanders explicitly said there was no further evidence for this fragment's origin. Later writers sometimes convert that possibility into provenance; the dossier should not. The manuscript remains at the University of Michigan, where modern photography supplements the early facsimile and edition.
What you are looking at
The extant joined pieces are roughly 15 by 10 centimeters from an original leaf estimated near 27 by 14 centimeters. A single column probably contained thirty-five or thirty-six lines, of which about twenty-one survive in part. A page number, often read as 59, appears at the top of one side and contributed to Sanders's reconstruction of a single-quire Acts codex, though calculations about original extent depend on assumed page density and format. Writing runs on both sides as continuous Acts. Tears remove line beginnings and endings, and dirt initially obscured letters. The artifact is a working codex remnant, not a freestanding lection or magical text despite the dramatic exorcism passage.
Where it belongs in history
Date proposals have ranged from the third century to the fourth. Sanders argued roughly 200–250; Clark preferred the fourth century; current catalogs often use early fourth century or about 300. Because no colophon or dated layer survives, 'late third to early fourth century' best expresses the evidence. The manuscript comes from a period when more than one form of Acts circulated in Greek, Latin, Syriac, and Coptic traditions. Its possible Fayum connection remains unprovenanced market inference. A page number and reconstructed format can illuminate the parent codex, but they do not establish who commissioned it, whether it contained only Acts, or whether its unusual wording originated in Egypt.
Consensus, questions, and debate
Sanders called the text thoroughly western, and later Aland classification placed it in a free or Category IV stream. A. C. Clark immediately analyzed its relationship to Codex Bezae, and subsequent scholarship notes striking shared readings in Acts 18:28; 19:1, 5, and the Sceva episode. Yet P38 also disagrees with Bezae, so it is not a direct copy or simple ancestor. 'Western text' is a conventional label for a complex group of longer, rephrased, or differently ordered readings, not a proven western-European edition. Scholars debate whether some expansions preserve early tradition or explanatory rewriting. P38's strongest contribution is demonstrating that Bezae-like readings existed in an Egyptian Greek papyrus around AD 300.
The biblical record beside the artifact
Acts 18:27–28 describes Apollos crossing to Achaia and powerfully refuting opponents from Scripture. Acts 19:1–6 narrates Paul's arrival at Ephesus, questions about the Holy Spirit, John's baptism, baptism in Jesus' name, laying on of hands, tongues, and prophecy. Acts 19:12–16 recounts healings associated with Paul and the disastrous attempt of Sceva's sons to invoke Jesus as a formula. P38 preserves these passages in a form that sometimes expands or rearranges familiar wording, particularly around Apollos and the Ephesian disciples. It witnesses an early textual form of Acts and ancient copying of charismatic episodes; it does not authenticate modern exorcism practice or independently prove the miracles.
Acts 18:27–28
27And when he was disposed to pass into Achaia, the brethren wrote, exhorting the disciples to receive him: who, when he was come, helped them much which had believed through grace:
28For he mightily convinced the Jews, [and that] publickly, shewing by the scriptures that Jesus was Christ.
A responsible historical conclusion
P38 makes textual history visible where Acts itself describes movement across places, languages, and communities. The leaf carries Apollos, Ephesus, baptism, Spirit, healing, and failed ritual power, while its wording shows that the narrative also traveled in more than one textual form. Its close-but-not-identical relationship to Bezae prevents a simplistic choice between a single short and long edition; copying history could branch, revise, and recombine. The page number and reconstructed codex invite questions about whole-book design, while the market-era provenance demands restraint. For readers, P38 is valuable not because every expansion must be original, but because it proves early Christians encountered Acts in materially and verbally varied books.
Authoritative sources
Institutional catalogs, excavation reports, and academic publications let you examine the evidence beyond this summary.
A Papyrus Fragment of Acts in the Michigan Collection
Harvard Theological Review, Cambridge University PressSanders's object-based publication records the 1924 purchase, three-piece condition, uncertain provenance, original date assessment, text, and facsimile context.
The Michigan Fragment of the Acts
Journal of Theological StudiesClark's early independent analysis tests Sanders's readings and evaluates the manuscript's relationship to Codex Bezae and allied versions.
Manuscript P38
Center for the Study of New Testament ManuscriptsConfirms Michigan custody, inventory, approximate date, material, three-fragment format, line estimate, passages, and exact images.
Acts, Michigan Papyrus of
Claremont Coptic Encyclopedia Digital LibrarySummarizes dimensions, contents, competing dates, Bezae relationship, Egyptian significance, inventory, and scholarly bibliography.