How Did Biblical Hebrew Change?

Robert Rezetko has responded to a recent article by Avi Hurvitz in Biblical Archaeology Review on how Biblical Hebrew changed over time. Here’s Robert’s abstract:

s200_robert-rezetkoIn a hot-off-the-press popular article in Biblical Archaeology Review (September/October 2016), Avi Hurvitz discusses “How Biblical Hebrew Changed.” It is certainly true that Biblical Hebrew evolved over time, but the particulars of how that happened are more complex and debated than Hurvitz acknowledges. The example that he discusses, ʾiggeret and sēfer for “letter,” is a case in point.

You can read Robert’s whole article HERE at Bible Interpretation.

This interaction demonstrates yet again how the discussion about the dating of Biblical Hebrew on linguistic grounds is often framed too simplistically. Robert exposes some of the extra issues that are often in a kind of ‘blind spot’ for many participating in the discussion. Yes, Hebrew did develop over time, as every language inevitably does. However, the connection between Standard Biblical Hebrew (aka ‘Classical’ Biblical Hebrew or ‘Early’ Biblical Hebrew) and Late Biblical Hebrew is not one of linear development from one to the other. It isn’t even the standard ‘S’ curve development. These were two styles of Hebrew that were contemporary for quite a long time.

Late Biblical Hebrew is not the child of Standard Biblical Hebrew, but its sibling.



Thinking better about linguistic dating of Biblical Hebrew

Here’s one for the Hebrews and Shebrews.

51i9zzm-zbl-_sx331_bo1204203200_51ooyvykrsl-_sx331_bo1204203200_Conventional wisdom says that Early Biblical Hebrew (aka Standard Biblical Hebrew or Classical Biblical Hebrew) came first, and then Late Biblical Hebrew. But when you actually analyse the evidence, this view starts to unravel. Ian Young, Robert Rezetko, and Martin Ehrensvärd have argued very convincingly that Early Biblical Hebrew and Late Biblical Hebrew were not linear diachronic developments, but rather contemporaneous styles of Hebrew in antiquity. This means that it’s practically impossible to date a biblical text based solely on linguistic criteria. Their compelling argument can found in their two volume work, Linguistic Dating of Biblical Texts, and their more recent Historical Linguistics and Biblical HebrewOnce you “see” their argument, you can’t “unsee” it. They look at the evidence in such a logical way that it makes you wonder why it has taken Hebraists so long to see what is so obvious.

Yet many Hebraists still don’t see it. It almost feels like they’re looking at one of those pictures that have a “hidden” 3D shape (a stereogram, like this). They claim to be finding the 3D shape. And if you can’t see it, it’s because you’re not looking at the right way. Try squinting or staring beyond the page. But the irony is that the picture isn’t one of those 3D shapes! It’s just a normal 2D picture. They’ve been looking at it all wrong, and yet the real picture is there staring them in the face.

So the old and disproven paradigm persists. It seems to be dying a slow death, as evidenced by a few recent articles.


Robert Rezteko

Young, Rezetko, and Ehrensvärd have clarified their position in a paper titled “Do We Really Think That Ancient Hebrew Had No Chronology“.

Robert Rezetko has also put together a few responses to recent studies working with the old paradigm. They are well worth the read:

I hope scholars, especially the younger ones, start just looking plainly at the evidence instead of squinting and forcing a particular paradigm onto it.

What language did Jesus speak?

Today witnessed a very minor verbal exchange between Pope Francis and Israeli PM, Benjamin Netanyahu, over the language Jesus spoke. Reuters reports the incident on the final day of the Pope’s visit to the Middle East:

During his comments about a strong connection between Judaism and Christianity and tolerance towards Christians in Israel, Netanyahu told the [sic!] Francis: “Jesus was here, in this land. He spoke Hebrew.

“Aramaic,” the pontiff interjected.

“He spoke Aramaic, but he knew Hebrew,” Netanyahu shot back.

The difference of opinion reignites a historic debate about the language Jesus spoke two millennia ago.

“Jesus was a native Aramaic speaker,” Israeli linguistics professor Ghil’ad Zuckermann told Reuters. “But he would have also known Hebrew because there were extant religious writings in Hebrew.”

Zuckermann added that during Jesus’ time, Hebrew was spoken by the lower classes – “the kind of people he ministered to.”

Pope Francis (R) meets Israel’s Prime Minister Benjamin Netanyahu (L) at the Notre Dame Centre in Jerusalem May 26, 2014 (Photo courtesy of Reuters / Alex Kolomoisky).

Jesus was evidently a native Aramaic speaker. The quotation in Mark 5.41 has Jesus address Jairus’ daughter with the words Talitha koum—an Aramaic phrase meaning, ‘Kid, get up!’ He would also have been very familiar with Hebrew, the language of most of the Jewish Scriptures. His references to the Scriptures on numerous occasions within the Gospels suggests this.

However, Jesus would also have been conversant with Greek. Galilee had one of the highest concentrations of Greek speakers in the Roman Empire outside of the Greek heartland of the Aegean. In fact, within short walking distance of Jesus’ hometown of Nazareth was the city of Sepphoris—a large Greek polis in the heart of Galilee. There is every chance that Jesus not only visited the city multiple times, but he may well have worked there in his profession as a carpenter. It was, after all, one of the largest economic centres in Galilee, and it was in the immediate vicinity of Nazareth.

The Apostle Peter?

Jesus’ disciples were also probably familiar with Greek. The hometown of apostles Simon Peter, Andrew, and Philip, was the town of Bethsaida at the northernmost point of the Sea of Galilee. In c. 2 BC it was granted status as a Greek polis and renamed Julias. Thus these three disciples, two of whom (Andrew and Philip) bore Greek names, grew up in a Greek-speaking environment. It also helps explain how Peter attained the name ‘Peter’. His original name was Simon, a good Hebrew/Aramaic name. Jesus, however, gave him the nickname ‘Kephas’, which is Aramaic for ‘Rocky’ <cue theme music to Rocky>. Paul refers to him as Kephas (or Cephas in modern English versions) in his correspondence with the Galatians and Corinthians. However, elsewhere, his name is easily translated into the Greek equivalent for ‘rock’, Peter (Greek: Petros).

There is a possibility that some of Jesus’ ministry was conducted in Greek. For example, there is good reason to suggest that Nicodemus’ misunderstanding of Jesus’ words in John 3.3 is dependent on an ambiguity in Greek, but not in Aramaic or Hebrew. Jesus tells Nicodemus, ‘Unless someone is born over, he cannot see the Kingdom of God.’ The relevant term used in the Gospel here is ἄνωθεν (anōthen) Nicodemus hears this as ‘born again’, while it seems from the rest of the discussion that Jesus meant ‘born from above’. The rendering ‘born over’ captures something of this ambiguity in English.

Jesus also seems to have conversations with Greek speakers throughout the Gospels. He chats with a Greek woman from Syro-Phoencia (Mark 7.24–30), a Roman centurion (Mark 8.5–13), and the Roman Prefect, Pontius Pilate. On each occasion he might have made use of an interpreter. However, it is more likely that Jesus was able to conduct the conversation personally without an interpreter.

There is even the possibility that the famous ‘camel through the eye of a needle’ saying (Mark 10.25) is a misunderstanding of a saying in Greek. The word for ‘camel’ in Greek is κάμηλος (kamēlos), but the word for thick ‘rope’ (the type used to moor a ship to port) is κάμiλος (kamilos). There is virtually no difference in pronunciation between the two. Did Jesus perhaps say, ‘It is easier for a rope to go through the eye of a needle than for a rich man to enter the Kingdom of God’? If he originally made the statement in Greek, he may well have. If, however, he made the statement in Aramaic, there would be no such ambiguity. I guess we’ll never know.

But all this is to say that Jesus was almost definitely a comfortable speaker of Greek, in addition to his native Aramaic, and the Hebrew of the Jewish Scriptures.

“Gospel of Jesus’ Wife” May Be A Fake

The tiny piece of papyrus that recently sparked controversy for supposedly portraying Jesus talking about his wife is now shrouded in doubt as to its authenticity. Recent suggestions are that the Coptic text on the papyrus was forged by an amateur using a Coptic-English interlinear edition (written by Michael Grondin) of the Gospel of Thomas, since it seems to duplicate a typographic error from that book. The Harvard Theological Review, which was originally going to publish the first article announcing the papyrus’ find, has decided to pull the article from its first 2013 issue.

You can read the fuller story HERE.

UPDATE: Andrew Bernhard has put together a paper outlining how the fragment might have been forged. Click HERE for the paper.

The “Gospel of Jesus’ Wife” Papyrus Fragment

A Rejoinder by Young, Rezetko, and Ehrensvärd on Linguistic Dating of Biblical Texts

Last week, Bible Interpretation published an article by Ron Hendel taking to task Ian Young, Robert Rezetko, and Martin Ehrensvärd for a summary of their work in Linguistic Dating of Biblical Texts. This week, Bible Interpretation has published a response to Hendel’s article by Young, Rezetko, and Ehrensvärd. Here is just a little of what they say:

There are weaknesses in our books, and there are things we wish we had emphasized more or stated differently but, honestly, we are becoming tired of knee-jerk reactions to our work that reflect a superficial understanding of its content and purpose. For example, we have seen the claim repeated that the aim of our arguments is to prove a late date of all biblical literature. This is a truly bizarre claim. We are pretty clear that we are saying that all linguistic dating arguments, both for early or late dates, don’t work.

You’ll find the article by clicking here. Ron Hendel has also posted a couple of responses in the comments.

An Unconvincing Cautionary Tale about “Unhistorical Hebrew Linguistics”

I have in the past pointed out and recommended the two volumes by Ian Young, Robert Rezetko, and Martin Ehrensvärd, titled Linguistic Dating of Biblical Texts (Equinox, 2008). In these two volumes the authors question the traditional view of the diachronic development of ancient Hebrew that sees the language move from ‘Early Biblical Hebrew’ (EBH) to ‘Late Biblical Hebrew’ (LBH). Instead, they put forward a compelling case that these two forms of Hebrew were actually different styles that existed concurrently.

Their theory has caused quite a stir in linguistic circles, as evidenced again by the recent objection from Ronald Hendel (University of California, Berkeley) in his article ‘Unhistorical Hebrew Linguistics: A Cautionary Tale’ at Bible Interpretation.

After reading Hendel’s short article, I do not believe he has adequately grasped the issues as Young, Rezetko, and Ehrensvärd present them. In fairness to Hendel, he has only written a very short article, so it was never going to be a comprehensive critique. However, it appears from Hendel’s argument that he is more interested in establishing a text via text criticism, than engaging with the both the linguistic data and the epistemic problems that the traditional diachronic model presents. Perhaps the major problem that Hendel does not address is the grounds on which biblical texts are dated. There are a number of assumptions inherent in his short analysis about when books are dated, and this depends on the traditional diachronic model, which is then used to support the dating of texts. It’s a circular argument that he does not appear to step out of. One of the good things about the work of Young, Rezetko, and Ehrensvärd is that they are able to highlight this epistemic circularity and actually demonstrate that the traditional diachronic model is trying to hold itself up by its own bootstraps. And it is this very point that many linguists appear unable to come to grips with and, indeed, continue to perpetuate.

Again, I commend the work of Young, Rezetko, and Ehrensvärd, and urge those interested in understanding the development of the Hebrew language to consider their arguments more closely. This should begin with an actual reading of the two volumes of Linguistic Dating of Biblical Texts, rather than a potted summary, upon which Hendel seems to have relied. Yes, they are offering a paradigm shift in our understanding of Hebrew, but as one works through the data and its implications, one realises how many assumptions have been implicitly at work in our previous understanding, and how tenaciously many of us seem to be clinging to them unawares. Their work deserves close study, and although I don’t agree with every point they make, they do put forward a largely compelling case. Hendel has only served to convince me of this by demonstrating the classic faux pas that Young, Rezetko, and Ehrensvärd attempt to expose.

Book Review: A Basic Introduction to Biblical Hebrew

I recently reviewed a new introductory Hebrew grammar by Prof. Jo Ann Hackett of the University of Texas at Austin. Prof. Hackett is perhaps best known for her work on the Balaam texts from Deir Alla. My review of her grammar book, A Basic Introduction to Biblical Hebrew (Hendrickson, 2010) is now up at Review of Biblical Literature, along with reviews by two others (Bernard Levinson and James Robson).

In short, I think this new grammar is a marked improvement on other older texts, especially in its attempt to be a true introductory grammar, rather than an undercover reference grammar. However, Prof. Hackett does make some choices that I do question.

Just after it became publicly available, Prof. Hackett contacted me about the review. With her permission, I’ve made some of her comments available here. As far as I know, no journal actually publishes an author’s response to reviews, so I’m glad to provide her with a public right of reply here. It’s best to read the review first, and then her comments below.

Jo Ann Hackett: I have just read your review of my textbook, and I must say that most of your negative reactions have been mirrored by others.  And I remain unconvinced, of course!  I do want to question, however, your suggestion that I treat what you call the “vav consecutive” as if it simply reverses the tense of the verb, because that is something I most certainly do not do, and I’m not quite sure how you could assume that I do.  I would suggest you read 15.3, where I specifically reject that point of view, and 15.4, which I hope shows that I think the historical explanation is correct.

In 16.8, where I say that some people believe that the vav of the ve-qatal form exists as an imitation of the wayyiqtol vav, I suppose you might read it the way you did, but even there, the explanation depends to a certain extent on the historical explanation of the wayyiqtol form.  So I don’t understand that part of the review, and to be honest, I really hate it that such a thing is out there about me, being read by people who don’t know any better.

Let me add a less serious aside.  The objections I’ve gotten to the fact that we call the verb by the 3ms SC form, even though I don’t start there, have all come from teachers.  I’ve taught this to all sorts of students, and I’ve yet to have a single one confused by it.  The same goes for the 1st-person first decision.  Teachers often hate it because we all learned it the other way.  Students, who don’t know any better, never even notice.  I have to admit that it’s hard for me, too, because I have them recite the paradigms and have to remember that they’re doing it in the way I taught them, not the way I learned it.

As you will see in her first paragraph, Prof. Hackett takes me to task for misrepresenting her on the ‘vav consecutive’. Having had my own arguments misrepresented in reviews, I can understand the frustration this can cause an author. However, in this case, I think there is some ambiguity inherent in the wording used within the grammar book that explains why I received the apparently wrong impression. In the review I state that it is unfortunate that Hackett opts for viewing the waw (or ‘vav‘) prefix of the wayyiqtol as a ‘waw conversive’—that is, an element that somehow converts or reverses tense. I quote §15.3 (p.90) of the book in its entirety:

15.3 Other Names For the Consecutive Preterite

The form we are calling the consecutive preterite is usually called “converted imperfect,” as if the addition of the וַ somehow “converted” a future-tense verb into a past-tense verb. It is also called the “imperfect with vav consecutive,” a better choice, but also misleading, since the basic verb is not the “imperfect” (our prefix conjugation) but rather the jussive.

You will see from this paragraph that there is no specific denial of the concept of a converted imperfect. The first sentence is stated without challenge. It is most likely that Prof. Hackett intended the ‘also’ in the phrase ‘but also misleading’ (second sentence) to signify a denial of the veracity of both the ‘converted imperfect’ and ‘imperfect with vav consecutive’ terminologies. However, since the concept of a ‘converted imperfect’ is not labelled misleading immediately after it is described, the impression I gained was that ‘also misleading’ applied only to the use of the word ‘imperfect’ in the terminology. In other words, I read the paragraph as saying that the wayyiqtol (or ‘consecutive preterite’ as Hackett terms it) can also be conceptualised as a ‘converted imperfect’ or, even better as an ‘imperfect with vav consecutive’, even though the wayyiqtol happens to be using the jussive, rather than the imperfect (yiqtol). Thus, although Prof. Hackett meant to put these other terms forward so as to deny their veracity or usefulness, there is no specific and unambiguous statement to that effect, and the paragraph can plausibly be read as though it were a mild endorsement (albeit with slight correction) of these conceptualities.

I am quite relieved to learn from Prof. Hackett that she does not endorse the ‘waw conversive’ or ‘consecutive imperfect’ views of the wayyiqtol. Yet, that makes the ambiguity inherent in the wording of §15.3 all the more unfortunate. I’m quite glad, however, that this issue could be raised and clarified here, and hope that instructors using the book are aware of the ambiguity and can take action to ensure a proper understanding of this particular section the way Prof. Hackett intended it.

Can we date biblical texts on linguistic grounds?

It’s an interesting question: Do the linguistic features of Biblical Hebrew allow us to figure out the date of biblical texts? Traditionally, the answer has been ‘yes’. And so Hebrew has been divided into ‘Early Biblical Hebrew’ (‘EBH’ — a.k.a. ‘Standard Biblical Hebrew’, or ‘SBH’) and ‘Late Biblical Hebrew (‘LBH’). As the terms suggest, EBH was viewed as an earlier stage of the language, usually dated to the pre-exilic era (i.e. before 587 BC), while LBH became more prevalent after this time.

Recently, however, a two-volume study, Linguistic Dating of Biblical Texts (London: Equinox, 2008), by Ian Young, Robert Rezetko, and Martin Ehrensvärd, has called this hypothesis into question.

The hypothesis of Young, Rezetko, and Ehrensvärd is that the data usually shown as evidence for a linguistic transition from an early form of Hebrew to a later form can and should be reinterpreted. They argue that instead of being a linear development, both EBH and LBH were concurrent ‘styles’ of Hebrew that coexisted. As a result of this, biblical texts can’t really be dated in the manner previously done. In other words, it is a mistake to think that a text written in EBH is necessarily earlier than a text written in LBH. That would have to be established on other grounds beyond linguistics.

The ramifications of this for our study of Hebrew language and biblical texts are actually quite significant. For starters, if Young, Rezetko, and Ehrensvärd are correct, then we would need to privilege non-linguistic data in determining (or at least trying to determine) the date of particular texts.

But are they correct?

Significant debate has ensued since the publication of LDBT in 2008. Some of it has been carried out in journals and academic conferences. But some of it has been conducted through online forums and blogs. You can find one such exchange being carried out between the authors of LDBT and two critics of their new take, John Cook and Robert Holmstedt. You will find the exchange on the Ancient Hebrew Grammar blog of John Cook and Robert Holmstedt, here:

Cook and Holmstedt disagree with the method and conclusions in LDBT. The discussion is quite heated at times, but that at least makes for interesting reading. You’ll also find John Hobbins over at Ancient Hebrew Poetry weighing critically into the debate.

For my part, let me lay my cards on the table in this debate. I was one of the proof readers for LDBT, and I have to say I found the arguments logically convincing. The critique of the linguistic approach of Avi Hurvitz was, especially, quite persuasive. I agree with Young (who was my PhD supervisor), Rezetko, and Ehrensvärd that too much has been made of linguistic data in the past, and that we cannot really date biblical texts based solely on linguistic grounds. I also agree that building a history of the language based on biblical texts is seriously undermined by the difficulties of thinking about redaction and scribal transmission.

However, there are numerous other questions that arise. If EBH and LBH are not successive stages of the Hebrew language, but rather concurrent ‘styles’, what exactly do they represent? Are they purely dialectal? Was ancient Israel/Judah/Samaria/Yehud a place where diglossia occurred? Is the linguistic divide between LBH and EBH based on geographic, social, or literary grounds—or even other grounds entirely? Do we need new terms to describe these two ‘styles’ of Hebrew?

It will be interesting to see how discussion develops.