Was Matthew's Gospel first written in Hebrew?

Type: supporting argument (off-timeline) · Assessment: contested — a genuine standoff between a unanimous early tradition and the internal evidence of the text. Papias, Irenaeus, Origen, Eusebius, and Jerome all say Matthew wrote first in Hebrew (or Aramaic); but our Greek Matthew reads as a Greek composition that uses the Greek Mark and the Greek Old Testament, and Papias’s key term, logia, is ambiguous. Strong by the lights of patristic testimony, weak by those of source criticism, with no clear winner.

Related arguments: the Pantaenus “India” report (Eusebius, Ecclesiastical History 5.10) — the sixth-hand corollary to this testimony — a Hebrew Matthew found in the East, left by Bartholomew — carries the elastic “India” that unsettles two apostolic missions: see also Did Bartholomew preach in India or in Armenia? · Did Thomas evangelize India?

The claim

— Papias of Hierapolis (c. 125), in Eusebius, Ecclesiastical History 3.39.16. The claim under test is that the Gospel of Matthew was originally composed in a Semitic language — Hebrew or the Aramaic often called “Hebrew” — for Jewish believers, and that our Greek Matthew is a translation of that lost original.¹

The evidence for

The external testimony is early, wide, and remarkably consistent. Papias, writing perhaps within a lifetime of the apostles and gathering the traditions of “the elders,” says Matthew composed the logia “in the Hebrew language.” Irenaeus (c. 180): “Matthew also issued a written Gospel among the Hebrews in their own dialect.” Origen (c. 230): the first Gospel was written by Matthew the publican “for the converts from Judaism, and published in the Hebrew language.” Eusebius (c. 325): Matthew, preaching first to the Hebrews and about to go elsewhere, “committed his Gospel to writing in his native tongue.” Jerome (393) repeats it and claims more — that the Hebrew original survived in the library at Caesarea, and that he had handled a copy the Nazarenes of Beroea used. Five independent-looking witnesses across two centuries, and a sixth-hand corollary in Eusebius’s report that Pantaenus found a Hebrew Matthew in the East: a Semitic Matthew was simply the common knowledge of the early church. That such a text once existed is not intrinsically improbable — Matthew wrote for a Jewish-Christian readership, and a Palestinian gospel in Aramaic or Hebrew would suit that audience.²

The evidence against

The internal evidence of the Gospel we actually have pulls hard the other way. Greek Matthew is not written like a translation. It reproduces large stretches of the Gospel of Mark — itself a Greek work — often word for word, which is what one expects of a Greek author using a Greek source, not of a translator rendering a Hebrew original; and where it quotes the Old Testament in narrative it frequently follows the Greek Septuagint rather than the Hebrew. A Gospel translated from Hebrew would not naturally quote the Greek Bible or copy a Greek Mark. Papias’s words are ambiguous. His logia (“oracles,” “sayings”) need not mean the Gospel at all: it has been read as a lost collection of Jesus’ sayings (a source like the hypothetical Q), or as a book of Old-Testament testimonia (messianic proof-texts) that Matthew assembled — either of which a later church could have mistaken for the whole Gospel. And his phrase that “every one interpreted them as he was able” fits a set of sayings needing translation better than a finished Gospel. The chain may be one link. If every later Father is ultimately repeating Papias, the “unanimous” tradition is a single ambiguous sentence multiplied. Jerome’s copy was probably a different book. The text he saw among the Nazarenes is widely judged to have been the Jewish-Christian Gospel of the Hebrews, a distinct work — so his “Hebrew Matthew” is evidence for that gospel’s existence, not for a Semitic original of the canonical Matthew.³

Adversarial assessment

What holds. The tradition is not nothing. It is early — Papias reaches back toward the apostolic generation — and it is unembarrassed and consistent across writers who had no obvious motive to invent a Hebrew Matthew. It may preserve a real memory: that the apostle Matthew stood behind a body of Semitic material (sayings, testimonia, or a short gospel) for the Jewish church, later drawn into or associated with the Greek Gospel that bears his name. On that modest form the claim is defensible, and a purely dismissive verdict overreaches.

What it cannot reach. The strong form — that our canonical Matthew is a translation of a Hebrew original — runs into the text itself, which behaves as a Greek composition dependent on Greek Mark and the Greek Old Testament. Most modern scholarship therefore rejects the translation-hypothesis while leaving open what Papias meant. The decisive weakness is the ambiguity of logia and the likelihood that the later witnesses echo Papias rather than confirm him independently; the confusion of the Nazarene gospel with a “Hebrew Matthew” shows how easily the tradition could grow. So the two disciplines genuinely divide: the patristic testimony is real and early, the source-critical case against a Hebrew canonical Matthew is strong, and neither simply defeats the other. That is what makes the question contested.⁴

Assessment

Contested. That Matthew wrote first in Hebrew is affirmed by an early and unanimous line of Fathers and is not improbable for a Jewish-Christian gospel; but the Greek Matthew we have reads as a Greek work built on Greek sources, Papias’s “oracles” may not mean the Gospel, and the later witnesses may merely echo him. A real dispute between patristic testimony and internal evidence, with no settled resolution.

Sources