The New Testament was originally written in Greek. This claim is not particularly controversial among biblical scholars, though some have argued that parts of the New Testament were originally written in Hebrew or Aramaic.
Some popular writers and religious groups, however, have claimed that much or all of the New Testament was originally written in Hebrew or Aramaic. In this article, we will survey the evidence and arguments that lead the vast majority of scholars today to believe that the original language of the New Testament was Greek.
The history of spoken and written languages in first-century Palestine
In order to identify the original language of the New Testament, it is important to understand the language situation in the first century AD. Before the exile of Judah in the early sixth century BC, Hebrew was the main spoken and written language in ancient Israel and Judah, and most of the Old Testament was originally written in Hebrew.
Under Persian imperial rule in the sixth through the early fourth centuries BC, Aramaic became the official language of government and most scribal education, and it gradually became the most common spoken language in the region. Hebrew and Aramaic are closely related languages of the so-called “Semitic” branch (of the Afroasiatic family), and they mixed and influenced each other to a large extent during this period. Some portions of the Old Testament were originally written in Aramaic, especially parts of Daniel and Ezra, as were some of the Dead Sea Scrolls and other early Jewish literature.
With the conquest of Alexander the Great in the late fourth century, Greek gained prominence as the common language of government, trade, and elite culture throughout the Eastern Mediterranean, including Judea and Galilee. By the third century BC at the latest, the Jewish expatriate community in Egypt had largely lost the ability to speak Hebrew and/or Aramaic, and so they translated the books of the Hebrew Bible into their then-native Greek. These translations—collectively called the “Septuagint”—became the main Scriptures used in the Jewish diaspora, the Jewish people living outside of the land of Israel. Many Jewish works from the diaspora, as well as some from Judea, were also written in Greek in this period.
Thus, by the time of the first century AD, the language situation in Palestine was very complicated and multilingual. Aramaic appears to have been the most common spoken language, especially among the working classes. Hebrew continued to be used for prayer and to compose religious texts, such as many of the Dead Sea Scrolls. And there is some evidence that Hebrew also continued to be used as a spoken language in some circles. Greek was widely spoken in the area as well, especially by Jews from higher socio-economic classes and those who had reason to interact with their Gentile neighbors. Thus, the language used for any spoken or written interaction depended on who was involved and the purpose of the conversation, and Greek allowed for the widest possible dissemination of written works throughout the Mediterranean world.
Beyond this general linguistic background, the manuscript evidence is a crucial part of addressing the question of what language the New Testament was written in. Here, the evidence is unequivocal: the New Testament is a corpus of Greek compositions. The Institute for New Testament Textual Research has documented over five thousand Greek manuscripts containing parts (or all) of the New Testament, ranging from the second century AD into the early modern era. This Greek tradition ultimately was and is the source for all known translations of the New Testament into other languages, ancient and modern. This includes ancient translations into Latin, Coptic, Syriac, Ethiopic, Armenian, and Georgian. It also includes, of course, modern translations into countless languages around the world.
In sharp contrast, there are no ancient Hebrew manuscripts of the New Testament whatsoever. Some late-antique Jewish polemical works do include Hebrew translations of parts of the New Testament, but the earliest extant Hebrew version of a complete New Testament book is the fourteenth-century version of Matthew included in a polemical work by the Jewish scholar Shem Tov. This Hebrew version likely pre-dated Shem Tov, but it has many elements from Latin and medieval vernacular languages that prove that it is a late translation ultimately derived from the known Greek Matthew, rather than reflecting an original Hebrew version of the book.1 (The books of the New Testament have also been translated into Hebrew on multiple occasions in modern times, but these are irrelevant for the question of the original language of the New Testament.)
The situation with Aramaic is more difficult, since there are ancient copies of the New Testament in different dialects of Aramaic. Even after the first century AD, Aramaic continued to be widely spoken in the Eastern Mediterranean, Mesopotamia, and surrounding areas in a wide variety of local dialects. These dialects cannot exactly be considered “Jesus’ mother tongue,” because they changed considerably over time and varied significantly from place to place. Given the growth of Christianity in the East, it is no surprise that both the Old and New Testaments were translated into these dialects and revised multiple times between the second and the seventh centuries. These versions are usually called the “Syriac,” which is one of the most widely used and well-documented dialects of Aramaic. Another noteworthy translation was made into the Christian Palestinian Aramaic dialect, which has more Palestinian influence than the other versions. While these Aramaic New Testament versions were made already in antiquity, the scholarly consensus is clear that they were translations, mostly from the Greek, into later Aramaic dialects. They were not original Aramaic versions of the New Testament books.2 The fact that even the earliest Syriac translators had to rely on Greek manuscripts is a good indication that Hebrew or Aramaic copies of the New Testament were unavailable already in the early centuries AD. The manuscript tradition thus strongly indicates a Greek origin for all of the books of the New Testament.
Church tradition of Hebrew or Aramaic originals
Given the manuscript evidence, why do some argue for Hebrew or Aramaic originals for at least parts of the New Testament? One of the strongest reasons is church tradition. The ancient Christian historian Eusebius cited a Christian writer from the second century named Papias who claimed, “Matthew collected/arranged the sayings [of Jesus] in the Hebrew dialect/manner, and everyone translated/interpreted them as they were able.”3
This was understood by many early Christian writers to mean that the Gospel of Matthew was originally written in Hebrew and later translated into Greek. Irenaeus, Origen, Eusebius, Augustine, Jerome, and others interpreted Papias’s statement this way. But as many modern scholars have pointed out, this brief statement contains many ambiguities that make it difficult to understand and assess its veracity:
- What does it mean that Matthew “collected/arranged the sayings”? Did he simply write a document collecting sayings of Jesus that was used as a source for composing the complete Gospels? Or does it mean that he wrote in Hebrew the entire Gospel of Matthew as we know it today, including all of the narrative?
- What does “Hebrew dialect/manner” mean? Some scholars interpret this as meaning simply that Matthew arranged the sayings in a typical Jewish way; and indeed Matthew is often thought to reflect a very Jewish–Christian understanding of the life and words of Jesus. But even if Papias was referring to composition in a Semitic language, it is not entirely clear whether it would refer to Hebrew or to Aramaic. In ancient texts, the word “Hebrew” could also be used to refer to the language we now call Aramaic, with “Hebrew” perhaps better understood as “the language typical of the Hebrews/Palestinian Jews.”
- And, finally, what does the word “translated/interpreted” mean in this context? In Greek literature it is used to refer to interpretation and exposition, sometimes but not always including translation from one language to another.
Thus, while this early tradition undoubtedly merits attention due to its antiquity, it is far from obvious what Papias meant and whether he had accurate information.
Indeed, there is no evidence that any early Jewish or Christian writer ever actually had access to a Hebrew or Aramaic original of Matthew, nor does anyone cite from such a hypothetical text. In the fifth century, the famous Christian textual scholar Jerome was sent to the Holy Land and tasked with translating the Old and New Testaments from their original languages into Latin. He came across an Aramaic gospel used by the Jewish–Christian sects of the Nazarenes and Ebionites, which he subsequently translated into Greek.4 These sects claimed that their gospel—also called the Gospel according to the Hebrews—was the original version of Matthew, written in Hebrew. Though initially intrigued, Jerome appears ultimately not to have been convinced, and he translated his Latin version of Matthew instead from the canonical Greek version.
Other church fathers also discussed the contents of the Gospel according to the Hebrews, which is in many respects very far removed from the canonical Greek Matthew received in the orthodox Christian churches. At one point, Jerome cites a supposed saying of Jesus from the Gospel according to the Hebrews that cannot have been from an original Hebrew version, but seems to be dependent upon the Greek version of Matthew.5 In the Greek Gospel of Matthew, the author quotes John the Baptist saying that Jesus does not need to be baptized, and Jesus saying that it is necessary “to fulfill all righteousness” (Matt 3:14–15), pre-empting the theological question of why Jesus would need to undergo John’s baptism of repentance if he was without sin. The Gospel according to the Hebrews cited by Jerome adds a non-canonical saying of Jesus in which he explicitly denies his need for baptism because of his sinlessness. This statement provides a clear safeguard responding to the theological problem first raised in the Greek Matthew, and it is implausible that such an important statement would have been left out in translation if the Gospel according to the Hebrews really had been the Hebrew original of Matthew. Thus, this “Hebrew gospel” can by no means be considered the Semitic original behind our canonical Greek Matthew; it appears to be a separate apocryphal gospel dependent upon Matthew.
There is a similar claim that the second-century Christian philosopher and missionary Pantaenus found a Hebrew version of Matthew in India.6 The local Christians claimed that this was the original Hebrew of Matthew passed on by the apostle Bartholomew, but this is probably no more credible as evidence for a now-lost Hebrew original than the Ebionites’ Gospel according to the Hebrews. The latter example proves that such biased claims cannot be accepted uncritically, and there is no reason to think that Pantaenus was linguistically competent to evaluate the evidence himself. Thus, the evidence from tradition for a Hebrew original remains almost entirely dependent upon the ambiguous claims of Papias and assertions by certain sects about the priority of their own preferred gospels. And this applies only to the book of Matthew, not the other books of the New Testament, for most of which there is no early church tradition of a Semitic original at all.
One other church tradition worthy of mention supposes that the letter to the Hebrews was first written by Paul in Hebrew or Aramaic and only later translated into Greek.7 But the early church’s theories about the origin of the letter to the Hebrews are conflicting and debated. Indeed, Origen, one of the few church fathers who actually knew Hebrew, already argued that the book’s high literary Greek style implies that Hebrews was first written in Greek by a very skilled writer.8
Language evidence for Hebrew or Aramaic
Is there any linguistic evidence for the use of Hebrew or Aramaic in the New Testament? The answer is yes, especially in the sayings of Jesus in the Gospels. Like most Jews of the time in Palestine, Jesus’ first language was undoubtedly Aramaic. Often Jesus’ sayings are still preserved in Aramaic in the Greek New Testament and even modern translations. For example, in Mark 5:41, Jesus tells a little girl, Talitha qum(i), which is Aramaic for, “Little girl, get up!” In Mark 7:34, Jesus commands Ephphatha, “Be opened,” and the deaf man’s ears are opened. In Mark 15:34, Jesus cries out on the cross, Eloi, Eloi, lema sabachthani, “My God, my God, why have you forsaken me?”—which is an Aramaic translation of the Hebrew of Psalm 22:1. In John 1:42, Jesus calls Simon Cephas, which is the Aramaic word for “rock” and equivalent in meaning to the Greek word Petros (hence “Peter”). Jesus’ Aramaic sayings are not limited to personal interactions with others in stories, but are also evident in his teachings. In Matthew 5:22, Jesus commands his followers not to insult each other with the Aramaic word Raqa “empty-head.” In Matthew 6:24, Jesus says that one cannot love God and Mammon, the Hebrew/Aramaic word for “money/wealth.”
The fact that these and other words of Jesus are preserved in Aramaic, even in otherwise Greek New Testament texts, suggests two things. First, they establish that they were indeed probably originally spoken by Jesus in Aramaic. Second, they suggest that the Greek texts of the Gospels are not all translations from a Semitic original. If the entire text of a Gospel had been in Aramaic, why would a translator have preserved these phrases in Aramaic instead of replacing them with their Greek translations like the rest of the text? If the Gospel was originally written in Greek, on the other hand, it makes sense why Aramaic sayings would occasionally be quoted in the original—often accompanied by translations—to provide the reader with a more authentic-feeling access to Jesus’ original words and to increase credibility by demonstrating the author’s ability to bridge the language gap between Jesus and the Gospel’s intended audience.
If the entire text of a Gospel had been in Aramaic, why would a translator have preserved these phrases in Aramaic instead of replacing them with their Greek translations like the rest of the text?
Additionally, scholars have noted examples of wordplay that suggest that certain sayings were originally delivered in Hebrew or Aramaic. For example, Matthew 1:21 says that Jesus would be so-named, because “he will save his people from their sins.” In Hebrew and Aramaic, “he will save” sounds like and has the same root (yasha’) as Jesus/Yeshua’. This reflects the common Old Testament practice of naming individuals with names that match the circumstances of their birth, but the obvious wordplay cannot come across in Greek. In another more subtle example in Matthew 3:9, John the Baptist says,
And do not presume to say to yourselves, “We have Abraham as our father,” for I tell you, God is able from these stones to raise up children for Abraham.
In both Hebrew and Aramaic, the words for stones (‘abanim/‘abnayya) and children (banim/bnayya, lit. “sons”) look and sound very similar. The meaning of this saying is adequately translated into Greek, but the poetic quality of the likely underlying Semitic wordplay is lost in translation. Because Hebrew and Aramaic were so closely related and intermixed in first-century Palestine, it is often difficult or even impossible to decide on which of the two languages sayings like this were originally produced.
In other cases, scholars have noted that differences between the Gospels can be explained by different translations of a Semitic original. For example, in Matthew 23:26, Jesus commands the Pharisees to “cleanse” the inside of the cup and the plate. But in the parallel saying, Luke 11:41 quotes Jesus as saying that they need to “give alms.” This difference is puzzling to explain in Greek, but long ago Julius Wellhausen observed that in Aramaic “cleanse” is dakko and “give alms” is zakko, which not only sound very similar but may also have been spelled the same way in Aramaic. In this case, it is likely that Matthew and Luke differ because of different interpretations of the underlying Aramaic expression.
The grammar and language of the Greek New Testament also reflect the influence of Semitic forms in other ways. One particularly striking example is how writers sometimes add an extra (redundant) pronoun where it is common in Hebrew and Aramaic, but not considered good Greek. For example, Mark 7:25 in the Greek literally reads “whose her daughter had an unclean spirit,” which is as awkward in the Greek as it is in English. Mark also begins most of his paragraphs with the word “and,” which is very odd for Greek texts but normal in Semitic usage. Other common grammatical structures in the Greek New Testament also seem to have been influenced by Semitic grammar, like:
- “Answering, Jesus said” (Matt 11:25) means “Jesus answered”
- “King of kings” (Rev 19:16) means “the highest king”
- “And it happened when he returned” (Luke 19:15) means “When he returned”
- “I desired with desire” (Luke 22:15) means “I eagerly desired”
These kinds of expressions, even if not totally impossible in native Greek speech and writing, reflect the Semitic background of the writers of the New Testament. They do not indicate that the New Testament books were translated from Hebrew or Aramaic, however. Many of the New Testament authors were bilingual (or potentially even multilingual), so it would be no surprise for one language to interfere with another. And most of these Semitic features in the Greek of the New Testament were well-known to Greek-speaking Jews and God-fearing Gentiles through literal translations in the Septuagint. In a sense, they formed a sort of mixed biblical idiom that shaped how the New Testament writers spoke and wrote religious texts in the Greek language. This is similar to how a modern believer’s prayers or speech about religious topics might be influenced by the King James Version in ways that would sound odd to most modern English speakers (“How Great Thou Art”). Even more to the point, when modern worship leaders cry out “Hallelujah!” “Hosanna!” or “Amen!” it by no means implies that they know Hebrew or Aramaic.
When modern worship leaders cry out “Hallelujah!” “Hosanna!” or “Amen!” it by no means implies that they know Hebrew or Aramaic.
To summarize, there is much evidence for an underlying Semitic language (especially Aramaic) for parts of the Gospels, particularly in the sayings of Jesus. There is also considerable Semitic influence in the Greek language of many New Testament authors. But there is no linguistic reason to suppose that any of the complete Gospels or other New Testament books were originally written in Hebrew or Aramaic in their entirety.
Language evidence for Greek
Why then do most modern scholars insist that the New Testament was originally written in Greek? We have already discussed how the presence of Semitic influence in the language of the New Testament need not indicate that the Greek texts were translated from Semitic languages. These “Semitisms” were very familiar to readers of Greek translations of the Old Testament and became thoroughly ingrained in their own Greek writing. Furthermore, it is important not to exaggerate these Semitic elements. While they may stand out in sharp contrast to what would have been considered proper classical Greek, the discovery of many thousands of Greek papyri in the past 150 years has radically changed scholars’ understanding of everyday Greek in the first century AD. While earlier scholars sometimes emphasized the uniqueness of Jewish Greek against that of classical authors, scholars can now see how the Greek of the New Testament fits comfortably within the common (or Koine) Greek spoken and written throughout the Eastern Mediterranean. Greek, as the lingua franca of the region, was often used in a simpler form than the classical authors and often reflected the influence of local customs and languages, but still remained recognizably Greek. This is exactly what we see in the Greek writings of the authors of the New Testament and other Jewish works of the period. In short, the books of the New Testament look just like other Greek writings from the period.
On the other hand, the books of the New Testament do not look like translations from Hebrew or Aramaic versions of the books. Scholars actually know a lot about what Jewish translation from Hebrew or Aramaic into Greek looked like, because all the Hebrew Scriptures and many other works were translated into Greek before or around the first century AD. For most of these books, we know the text in both the original language and the target language (in this case, Greek) and can compare them word by word for many thousands of words to develop detailed profiles of how the translators worked. These “Septuagint” Greek translations are now a well-developed field of study, covering examples ranging from more to less literal and encompassing the idiosyncrasies of many individual translators. Even in the most free of these ancient translations, the Semitic base texts constrain and influence the translations to a much greater degree than could be supposed based on the Greek texts of the New Testament. The translations regularly mimic the simple sentence structures of their Semitic sources, whereas the New Testament texts tend to have the more complex sentences that are characteristic of Greek literary writing. Translations tend to have a limited range of Greek vocabulary and grammatical particles that conventionally correspond to Semitic words and usage, whereas the New Testament texts frequently have more varied vocabulary and use particles and expressions in ways that have no exact Semitic equivalents. Known translations also generally have a fairly consistent profile of translation technique throughout the entire translated book. In this regard, the greater evidence for Semitic language in the sayings of Jesus than in the surrounding narrative actually reinforces the argument that the entire Gospels could not have been written in Hebrew or Aramaic. If the Gospels had been translated in their entirety from Semitic originals, both the sayings and the narrative would have had a more consistent profile, rather than fluctuating back and forth between Semitic-sounding and proper Greek. Put simply, though the various books of the New Testament differ from each other in terms of style and level of writing, none of them looks like the known Greek translations from Hebrew or Aramaic texts from the period.
Another important indication regarding the original language of the New Testament is that the authors often used Greek sources. Many books of the New Testament are filled with Old Testament quotations, and these are almost always cited from the Septuagint translations or revised Greek versions of the Old Testament texts. We cannot easily explain these away as just borrowing from previous biblical translations while translating the whole book, because the larger context sometimes depends on using the Greek text instead of the Hebrew. For instance, in Hebrews 10:5, the author cited Psalm 40:6 in the Greek tradition that has the word “body,” instead of the Hebrew text which reads “ears.” This reading from the Greek is important in the context of Jesus’ body being the perfect sacrifice, which shows that the author must have been composing the text in Greek.
Furthermore, scholars have long argued that the similarities in Greek wording between the synoptic Gospels (Matthew, Mark, and Luke) are so close and extensive that some of them must have used the Greek text of one or more of the others as a source. The most common explanation is that both Matthew and Luke used Mark as a source, which is particularly important here, since Matthew is the one Gospel for which there is ancient speculation about a lost Hebrew or Aramaic original. If, as most modern scholars suspect, Matthew used the Greek text of Mark as his main written source and incorporated most of Mark in his Gospel, it makes little sense to think of the present Gospel of Matthew as a translation of a Hebrew or Aramaic gospel. A similar situation can be seen in the relationship between Jude and 2 Peter, one of which used the Greek text of the other as a main source for composing the text. The use of Greek sources by the New Testament authors is thus another strong indication that these works were originally written in Greek, rather than Hebrew or Aramaic.
Scholars have also pointed out literary features in some passages that seem to work only in Greek. For example, in John 3 the dialogue between Jesus and Nicodemus hinges on the ambiguity in the Greek word anothen, which can mean either to be born “from above” or “again.” Jesus seems to be implying that this rebirth needs to be of heavenly origin by the power of the Spirit, a nuance which Nicodemus misses with his response about being born again to an earthly mother. The meaning of this text turns on the nuances of the Greek language it must have been spoken and/or written in, and it is perhaps not coincidental that here Jesus interacts in Jerusalem with a high-status Jewish leader with a Greek name who likely would have been very comfortable speaking Greek.
Gospels versus letters
To summarize, there is considerable evidence for Semitic languages in the Gospels, primarily in the sayings of Jesus, most of which were likely originally delivered in Aramaic (or sometimes possibly Hebrew). There is also significant Semitic influence in the Greek of the Gospels, but of the type that would be expected for Jewish authors who were bilingual and/or familiar with Greek translations of the Old Testament, rather than indicating that the Gospels themselves were originally written in Hebrew or Aramaic. Thus, while Semitic traditions—whether oral or written—undoubtedly lay behind the Gospels, it is very unlikely that any of the four canonical Gospels were originally written in Hebrew or Aramaic in anything like their current forms. Only Matthew has support from church tradition for a Hebrew original, but this tradition is ambiguous at best and dubious at worst. Luke and Acts are also sometimes supposed to have been written (at least partially) in Hebrew because of frequent Semitisms, but this is highly unlikely, since most of these were well-known from Greek translations of the Old Testament. Luke’s Gospel and Acts are actually written in a relatively high level of smooth Greek overall, and they are explicitly written for a Greek-speaking Christian with the Greek name Theophilus (Luke 1:3; Acts 1:1). Furthermore, like Matthew, Luke clearly used Greek sources, and he positioned himself among the best in the Greek tradition of history writing (cf. Luke 1:1–4).
Only Matthew has support from church tradition for a Hebrew original, but this tradition is ambiguous at best and dubious at worst.
In short, claims that any of the canonical New Testament Gospels as we know them were originally composed in Hebrew or Aramaic are baseless and contrary to the linguistic evidence of the texts themselves. Any Semitic sources that may have been used by the Gospel writers appear to have been limited in scope and must have been thoroughly reworked in the composition processes of these works. Perhaps a good parallel to this might be the Jewish War by the Jewish historian Josephus. In his prologue to the Greek text, Josephus claims to have written about the war previously in Aramaic for his compatriots, whereas now he has written his book in Greek. Much like the New Testament, however, the Greek book we have today is so thoroughly Greek that it must have differed considerably from any lost Aramaic precursors.
Hebrew or Aramaic originals for the rest of the New Testament books are even more implausible than for the Gospels. Though these books also contain Semitisms and sometimes even Hebrew or Aramaic words—for example, Maranatha (“Our Lord, come!”) in Aramaic in 1 Corinthians 16:22; Abaddon (“Destruction”) in Hebrew or Aramaic in Revelation 9:11—their texts are so thoroughly Greek in structure and idiom that there can be no serious doubt about their original language. Furthermore, the epistles were explicitly written to Christian believers in mixed Jew–Gentile congregations outside of Palestine, who could hardly have been expected to understand Hebrew or Aramaic. Many of the individuals mentioned had Greek or Latin names, and Paul’s secretary who helped him write the letter to the Romans had the Latin name Tertius (Rom 16:22). The letter “to the Hebrews”—which some ancient authors supposed was originally written in Hebrew—depends on Greek Old Testament translations and is written at such a high level of literary Greek that no serious linguistic case can be made that it is a translation from a Semitic original. The book may have been written primarily to Greek-speaking Jews (“the Hebrews”), but it was almost certainly not composed in the Hebrew language.
In light of all of this, it is clear that the scholarly consensus that the New Testament was written in Greek is correct. The Aramaic (and perhaps sometimes Hebrew) words of Jesus and his earliest followers still grace some pages of the New Testament, either explicitly or implicitly. But by no means can they bear the weight of argument necessary to claim that any of the canonical Gospels as we know them today was originally written in Hebrew or Aramaic. Whatever the underlying sources, the language of the canonical Gospels—and even more so the rest of the New Testament—is Greek through and through. This makes perfect sense in historical context, given Jesus’ mission to take the good news to all the nations.
- William L. Petersen, “Some Observations on a Recent Edition of and Introduction to Shem-Tob’s ‘Hebrew Matthew,’” TC Online 3 (1998).
- For an introduction to the ancient translations, see Bruce Metzger, The Bible in Translation: Ancient and English Versions (Grand Rapids: Baker Academic, 2001).
- Eusebius of Caesarea, Ecclesiastical History, eds. T. E. Page et al., trans. Kirsopp Lake et al. (London: William Heinemann, 1926), 3.39.16.
- Jerome, On Illustrious Men, trans. Thomas P. Halton (Washington, DC: Catholic University of America Press, 1999), 3.
- Jerome, The Dialogue against the Pelagians, in Dogmatic and Polemical Works, trans. John N. Hritzu (Washington, DC: Catholic University of America Press, 1965), 3.2.
- Eusebius, Ecclesiastical History, 5.10.3.
- Eusebius, Ecclesiastical History, 3.38.1–3; 6.14.2–3.
- In Eusebius, Ecclesiastical History, 6.25.11–14.