I appreciated the title of a recent essay (June 9, 2014) by Seth Sanders on Religion Dispatches entitled, “Why the Argument Over Jesus’ Language is More Complicated and More Interesting Than Media Experts Have Claimed.” In fact, many of Sanders’ points that elucidated the complexity of the language situation in the first century resonated with me. He even cited the article that I had written for the Israeli newspaper Haaretz, “Why Jesus really was a Hebrew speaker.”  However, within his essay there were some erroneous or misleading “facts” that need further clarification if the public is going to learn more of what went on in the linguistic landscape of first-century Galilee and Judea.

Ἑβραϊστί  means Hebrew

When turning to the data on rabbouni, Sanders seems out-of-touch and somewhat prejudicial when he says:

In John 20, Mary Magdalene is described as calling the resurrected Jesus “rabbuni,” which, every standard translation tells us, is Hebrew and means “teacher.” This is fascinatingly wrong. It’s actually the only time in the New Testament that an Aramaic form of the word is used; every other time Jesus is addressed with a similar term it’s the Hebrew “Rabbi.”  Indeed, all the comprehensible words in John labeled “hebraisti” (translated “Hebrew”), like Golgotha, are Aramaic…

For starters, rabbouni occurs not once but twice in the New Testament—once in Mark 10:51 and once in John 20:16. In John, rabbouni is called “Hebrew,” while in Mark, there is no language appellation. Ironically, in some translations (e.g., NIV and footnote of NRSV) rabbouni has now been called “Aramaic,” prejudicially, as will be seen below. Sanders also calls the word Aramaic, and later in his essay, he refers to the word as “obviously Aramaic.” Sanders seems to be unaware of the clarifying study done by E.Y. Kutscher fifty years ago. Since I had to cut my original offering to Haaretz down from 3300 words to 1200 words, I was not able to fully describe the complexity of rabbouni. From the recently published article “Hebraisti in Ancient Texts: Does Ἑβραϊστί Ever Mean ‘Aramaic’?”:

Traditionally, it has been argued that the Greek ραββουνεί [1] comes from Aramaic רִבוֹנִי rather than Hebrew רַבִּי, a word more widely known among commentators. Yet, this understanding is too simplistic and probably shows a tendency in the eyes of New Testament scholarship to attribute anything different from a basic understanding of Hebrew or Biblical Hebrew to Aramaic.[2]  Kutscher has demonstrated that  רַבוּנִי and רַבּוֹנִי vs. רִבּוֹנִי represents a difference between Western and Eastern pronunciations of Hebrew and Aramaic rather than a Hebrew vs. Aramaic distinction. Both languages show the same West/East distinction. Texts such as the early Hebrew Mishnah Taanit 3:8 (according to Codex Kaufmann) and later Aramaic Palestinian Targum fragments from the Cairo Geniza [3] show that רַבוּנִי with pataḥ is found in Western Semitic texts.[4]  Eastern texts, such as the Aramaic Targum Onkelos (passim), use the form רִבוֹנִי, “riboni.” Kutscher has speculated that Targum Onkelos has caused the textual corruptions in later printed texts of both Hebrew and Aramaic.[5]  Since the word ραββουνεί was used in both Hebrew contexts and Aramaic contexts, John must be recognized as correct when he calls rabbouni “Hebrew,” and it cannot be used as evidence that Ἑβραϊστί means “Aramaic.”[6]

Even more, Sanders’ statistics are actually a sleight-of-hand, since there are not over a hundred examples of rabbouni in Targum Onkelos. Sanders quoted statistics for the use of ribbon, a form also common in mishnaic Hebrew (e.g., ribbono shel `olam “lord of the universe”). The New Testament is actually the oldest attestation of the western pronunciation rabboun-. To further complicate matters, we have another near-synonym of this word, rabban, again found in both Hebrew and Aramaic. However, to assume that rabbouni is “obviously Aramaic” is a circular faux pas.

Illegitimate circularity raises its head again when Sanders discusses the meaning of the term ebraisti ἑβραϊστεί “Hebrew.” It has been commonplace within New Testament scholarship over the last 400 years to attribute hebraisti to both Aramaic and Hebrew, as if ancient writers were referring to Aramaic when using the term. That was seen as a necessity as long as Hebrew was considered a dead language. Sanders has recognized that Hebrew remained in use in the Second Temple. He may find it interesting, even surprising, to carefully sift through the data on the word hebraisti. The numbers of errors and misreadings that have been allowed to echo within secondary literature are quite humbling to all of us in the field.[7]

Superficial logic has helped modern scholars to misread ancient authors. For example, Josephus showed that he meant Hebrew when he called σαββατα sabbata (“sabbath”) Hebrew and said that it meant “to rest.” Scholars have sometimes alleged that Josephus must have meant Aramaic because sabbata ends in “-a.” However, the Aramaic-friendly citation form, sabbata, only shows the preferred way to refer to the term in Greek, following a centuries-long interface between Greek and Aramaic in connection with Hebrew in the Middle East.[8]  The citation form does not make Josephus’ σαββατα “Aramaic.” Josephus was referring to the Hebrew word when he called the word “Hebrew” because Aramaic did not have a verb shabat but used a different verb, naH, for “rest.” The attested Aramaic word for sabbath in Egypt and Israel was shabba [sic]. I would recommend that someone read through the article, “Hebraisti in Ancient Texts: does Ἑβραϊστί Ever Mean ‘Aramaic’?,” before continuing the conversation.[9]  This is a case were an assumed fact on the ground turns upside down. The result of the study also leads one to recognize that John most probably meant Hebrew when he correctly called rabbouni Hebrew. The same apparently goes for betesda, gabbata, and golgota (John 5:2; 19:13, 17).[10]

Dating the Aramaic Targums

Another item that needs to be flagged in Sanders’ article regards the Aramaic Targums.  He says:

By the second century B.C.E. Palestinian Jews were writing major religious texts in all three languages, and translating the Hebrew scriptures into both Aramaic and Greek.

Sanders’ statement goes beyond the evidence, especially since Tobit is not part of the Hebrew canon. We have Hebrew and Aramaic copies of Tobit among the Dead Sea texts, and not a few non-canonical Hebrew and Aramaic texts. But the only sure copy of an Aramaic Targum is an Aramaic translation of Job, with linguistic traits that show that the targum had been imported from the East. There is also a piece of Leviticus 16’s “Day of Atonement,” which may testify to a copy of the whole book, or it may be a reading that was intended for the many foreign pilgrims who would visit on the high holidays. What is remarkable is that we have lots of Hebrew Bible and Greek Bible among Judean texts, but virtually no Aramaic Bible, and the primary example is an import (11Q Targ Job).

Just like Alexandria is credited with the impetus in creating a Greek Bible, we may assume that the Eastern diaspora is where an Aramaic Bible first gained ground. If so, Qumran Targum Job is all we have. The later targumic traditions are a complex mix. Onkelos is a 2-3rd century CE(!) mix of west and east, and the Palestinian targum tradition likewise stems from the 2-3rd century CE, although a local product. Applying them to the Second Temple period is anachronistic without further support, although unfortunately commonplace in scholarship.

Hebrew Story Parables

Finally, Sanders’ comments on proverbs missed the point about “story parables” and missed the source of the observations. Sanders says:

Both Buth and his Jerusalem School for Synoptic Research colleague Steve Notley add the same peculiar argument that, as Notley wrote in the Times of Israel: “Outside of the Gospels, story-parables of the type associated with Jesus are to be found only in rabbinic literature, and without exception they are all in Hebrew. We have not a single parable in Aramaic.”  Even if this were true (Jews were already reading the Aramaic parables of Ahiqar in the 5th century BCE at Elephantine in Egypt), it is unclear what it would mean.

It was Moshe H. Segal [11] who made the observations about story parables. Steven Notley and I merely concurred and highlighted the ramifications. While anecdotal stories occur in both languages in rabbinic literature, and while maxims and proverbs can occur in both languages, the “story parable” is a genre that did not exist at Elephantine in Aramaic and it is primarily a shared genre between Jesus and rabbinic literature. Rabbinic story parables are timely evidence since they include 400 parables among the tannaim and tannaitic literature (commonly attributed to the first three centuries CE and not to be lightly ignored). Many questions remain to be asked and answered as to why the situation became as it did. The data needs to be explained and dealt with in a comprehensive way. I have some ideas that I am working into a forthcoming study. What is unacceptable is the former practice within New Testament scholarship where Aramaic parables were simply assumed, and retroversions from Greek into Aramaic were presented without acknowledging the lack of Aramaic parables among either Tannaim or Amoraim (early and later rabbinic teachers). The field is so unaware of the problem that argumentation turns data on its head.

For example, in the Parable of the Vineyard (Matt 21:33-46, Mark 12:1-12, Luke 20:9-19) there is a scripture citation that plays on the ‘son’/ ‘stone’ motif. That is a classic Hebraism. Some scholars, even recently (names withheld to protect the guilty), have rejected the connection of the scripture with the parable because “the wordplay doesn’t work in Aramaic.” Imagine if someone were to use the same yardstick for a teaching by Yochanan ben Zakkai, a first-century sage who lived in Arabba, a village less than a day’s journey due north of Nazareth. Would anyone have a problem if Yochanan ben Zakkai were alleged to teach a story parable in Hebrew? Problems only seem to arise when this is attributed to Jesus. Instead, scholarship should be assuming that Hebrew would have been an expected medium for telling a parable in the first century. At least, we need to recognize that the burden of proof would be on those who would assume common Aramaic story parables, since such parables do not seem to exist.

By the way, let me close on a point of agreement with Seth Sanders. Things are definitely more complex and more interesting than the media (and sometimes scholarship) are aware of. It is hard to overestimate this. In 2009, a remarkable document was published from the first or second century CE which aptly exemplifies this complexity. This document from Beit-Amar begins in Aramaic though inserts a Hebrew place name.  The document then switches to Hebrew for a large part of the text.  Yet, in the midst of the Hebrew text, a couple of Aramaic terms are inserted.[12]  This is another example of “linguistic pluralism” which should cause 21st century scholarship to be reassessing old, accepted “facts” about the languages of the first century and Jesus.


[1] The Greek texts have ραββουνει [B], ραββουνι [א, Byz], ραββωνει [D], ραββωνι [Θ], et al. They consistently record an [a] sound in the first syllable and an [i] in the final syllable according to Koine Greek phonology.

[2] For an example of the trend, and needed correction, see the note 40 on ὡσάννα in Buth’s “The Riddle of Jesus’ Cry from the Cross,” pages 408-409 in the present volume, where it is noted that the Hebrew הושע-נא is often called Aramaic in commentaries; also in agreement on this point is Jan Joosten, “Aramaic or Hebrew behind the Gospels?,” Analecta Bruxellensia 9 (2004): 88–101 (91) states: “hosanna (said by the crowds) and amen, are in fact Hebrew and not Aramaic.”

[3] Michael L. Klein, Geniza Manuscripts of Palestinian Targum to the Pentateuch, vol. 1 (Cincinnati, Ohio: Hebrew Union College Press, 1986). See, e.g., רַבּוּנִי at 1:133 (col. 2, line 3—Gen 44:18), where the vocalization is clear but the consonants [ני] are in a lacuna. At line 5 of col. 2, the vocalization רַבּוּנִי is attested but the top parts of the consonants are missing.

[4] E. Y. Kutscher, “Language of the Sages” (Hebrew), in Ben-Hayyim, Dotan, and Sarfatti, eds., Hebrew and Aramaic Studies, 95–98.

[5] Ibid., 98.

[6] Randall Buth and Chad Pierce, “Hebraisti in Ancient Texts: does Ἑβραϊστί Ever Mean ‘Aramaic’?,” in Buth and Notley, edd., The Language Environment of First Century Judaea [Brill, 2014, pp 66-109], 99-100.

[7] See for example, Matthew Black, An Aramaic Approach to the Gospels and Acts, 3rd ed., [Clarendon, 1967] p 48, on Letter of Aristeas 11-12.

[8] Other examples of this phenomenon are πάσχα “Passover” (Mt 26:17; Mk 14:12; Lk 22:7) and σίκερα “beer” (Luke 1:15), loanwords in the NT that are already found in the Septuagint translation of the Hebrew Bible.

[9] Randall Buth and Chad Pierce, “Hebraisti in Ancient Texts: does Ἑβραϊστί Ever Mean ‘Aramaic’?,” in Buth and Notley, edd., The Language Environment of First Century Judaea [Brill, 2014, pp 66-109].

[10] Bet-esta and Gabbata may not even be Semitic in etymology. The writer of John’s Gospel did not say what those two names meant, only that they were the local names in Hebrew. Again, see the article, “Hebraisti in Ancient Texts: does Ἑβραϊστί Ever Mean ‘Aramaic’?,” for the details.

[11] Moshe H. Segal.  A Grammar of Mishnaic Hebrew [Oxford: Clarendon Press, 1927] 4-5.

[12] See Haggai Misgav “Notes on the Document from Beit-Amar” [Hebrew: הערות לתעודה מבית עמר] Cathedra 148 (2013):179-184.