Scribal Methods & the Hebrew Bible - Scribes & the Old Testament - Part 3

Welcome back to our series on the text of the Hebrew Bible. If you haven't already watched the first two videos that come before this one, I'd encourage you to do so. Now by way of introduction to this video, let me tell you a little personal story.

When I went to seminary, I had the privilege of studying Hebrew and textual criticism under Dr. Peter Gentry. After my first course with him, I quickly realized that I wanted to learn as much as I could from him.

So I ended up taking every single class he offered at Southern Seminary. Fast forward to today, and now he has become something of a YouTube star with his short video on the Nephilim, which now has almost 5 million views.

But his most important contribution probably to biblical scholarship has been his excellent book, Kingdom Through Covenant. And the shorter, less technical version of the book called God's Kingdom Through God's Covenants, which I hope all of you get a chance to read someday.

Anyway, I asked Dr. Gentry if he would be willing for me to make one of his presentations on the text of the Old Testament into a video, and he kindly agreed. So what follows is an adaptation of his article, which will be linked in the description, and will have all the footnotes not included in this video.

Some of it will overlap with what we've already talked about in the previous videos, but repetition and review is always helpful when talking about a complex subject. For those who believe in divine revelation mediated by authorized agents, the central questions are number one, which writings come from these agents authorized to speak for God?

And number two, have their writings been reliably transmitted to us? Although this presentation is focused on the latter question, the former is logically prior. How one answers the first question will determine evaluation of evidence relating to the second.

What defines a canonical text according to Nahum Sarna is a fixed arrangement of content and the tendency to produce a standardized text. Since the very first biblical text constituted a covenant, this automatically implies a fixed arrangement of content and a standard text.

I'm referring to the covenant at Sinai, a marriage between Yahweh and Israel. A marriage contract doesn't have a long oral prehistory. Its content is fixed from the start. The current view today is that the content and text of the Old Testament was not standard until the second century AD.

So Jesus could not really know for sure what writings were inspired by God, nor did he have a stabilized text. This is what I am calling chaos theory. Analysis of the evidence has led me to conclude that the text of the Old Testament in content, arrangement and stability was fixed probably at the beginning of the fourth century BC by Ezra and Nehemiah.

It is the history of this text that I attempt to treat in what follows. The authors of the Old Testament produced their work between the 15th and 4th centuries BC. How can we know that the final form of the text, regarded as canonical by the second century BC, has been transmitted to us in a reliable and trustworthy manner?

The answer to this question can be provided, number one, by describing the sources that have survived, whether they are copies of originals in Hebrew or Aramaic, or whether they are ancient translations or versions.

And number two, by understanding the history of the transmission of the text. The word understanding here is all-important because the data are not self-interpreting. In broad terms, three stages can be discerned in the history of writing the biblical text in Hebrew.

First, only consonants were used to represent the language in the earliest stage of writing. This is a reliable way of writing since context determines readings that are uncertain. Israeli newspapers still use only consonants.

Correct pronunciation of the biblical text, moreover, was passed down orally from priest to priest and from scribe to scribe. Second, beginning sometime in the 9th century BC, the letters Hevav and Yod, and later also sometimes Aleph, were given a double function to represent long vowels as well as consonants.

This system, however, was not consistent or systematic and moreover did not represent all the vowels. Thirdly, during the period 600 to 1000 AD, Jewish scholars called Masoretes developed a system of dots and squiggles to go over and under the letters.

The dots represented all the vowels and also the accents. Early Hebrew writing employed a script similar to that used by the ancient Phoenicians. Later, under the influence of the Chaldean kings of Babylon, scribes switched to using the Aramaic script.

Here's an example of Genesis 1 -1 in archaic Hebrew script. And here's what it looks like in Aramaic square script. Finally, here's what it looks like with Masoretic vowels and accents. As already noted, the Masoretes devised a system of signs to represent the vowels and committed the reading tradition handed down orally to writing.

At the beginning, only a few vowels were shown. Later, full vocalization was shown under the influence of Syriac and Arabic literature. Secondly, they also developed a set of diacritical signs to mark the accents according to the chanting of the text in the synagogue.

The history of the Masoretes correlates with different groups of Jewish scholars. First, a large-scale immigration to Babylon occurred from 132 to 136 AD after the Third Jewish Revolt. Later, conquest of Palestine by Islam in 638 AD made possible a return to Palestine of Jewish scholars and a revival of textual work in Tiberias.

As a result, there are different systems of vocalization. Tiberian, which is sublinear. Palestinian, which is superlinear. Babylonian, which is also superlinear. And Expanded Tiberian, which is a system that combines elements of the standard Tiberian vocalization with features from the Palestinian tradition, which can be seen in Codex Reuchlin from 1105 AD.

There are two famous families of Tiberian Masoretes. Number one, Ben Asher and number two, Ben Naftali. The text of the Ben Asher family is universally accepted as the most faithful preservation of the text.

These medieval Masoretic manuscripts are accurate witnesses to an ancient, consonantal text of the highest quality. Now the following chart lists important early manuscripts and you can pause the video to read through it.

As many as 3 ,000 manuscripts are known from the Middle Ages. All of them attest the same textual tradition with only minor variation. These manuscripts have not been consulted systematically since Kennecott and de Rossi analyzed them in the 18th century.

And that's a fascinating story in and of itself for another time. Now, as we saw in previous videos, before 1900, we had no Hebrew manuscripts prior to the Masoretes around 1000 AD. And two discoveries changed all this.

And we are just beginning to evaluate the new materials. And these were number one, the text from the Judean desert, which we'll cover at length in a future video. And number two, the Cairo Geniza fragments.

Now, at this point, we can also mention eight manuscripts from the 3rd to 7th centuries, which I'll put on the screen so you can pause it. These witnesses cast enormous light on the early history of the Masoretic text.

Now, at this point in Gentry's article, he discusses the ancient versions of the Hebrew Bible, like the Septuagint and the Samaritan Pentateuch. But since we covered these in the first video, we'll skip this section.

But you can benefit from it if you read the original article. The discovery of the Dead Sea Scrolls has highlighted the fact that before the 2nd century AD, differences are attested between our earliest preserved copies of the text, as well as between the parent texts of the earliest translations.

What are these differences like? And what do they tell us about the history of the transmission of the text? We can classify our earliest witnesses to the text according to two types. Manuscripts or translations that represent a simple, straightforward copying and transmitting of the text exactly and precisely as received.

And two, manuscripts and translations that represent scribes revising and updating the text to make it relevant and understood to the current circumstances or generation. James A. Sanders labels the former the repetition factor, and the latter the resignification factor.

Such a classification is extremely helpful in evaluating the apparent chaos in the witnesses. Andrew Teeter uses the terms conservative and facilitating to describe the two types of approaches taken by copyists and translators.

Allow me to quote his description of the evidence. The evidence from the period demonstrates a general distinction between two scribal models defined by the effort either to produce an exact copy, the primary goal being fidelity to the letter, or to produce a copy which facilitates understanding, the primary goal being readability or comprehension of meaning, a goal which authorizes a certain latitude with regard to textual intervention, above all in matters of linguistic updating and interpretive expansion.

A spectrum of manuscripts produced by both models coexisted in Palestine in the late Second Temple period. Both were in widespread use, demonstrated on the one hand by the broad attestation of exact or conservative manuscripts among the discoveries at various sites in the Judean desert, including Qumran, and on the other hand by the facilitating texts represented by the Samaritan Pentateuch, the Septuagint, other scriptural manuscripts and citations within the literature of the Dead Sea Scrolls, by the forelogging of several rewritten Bible compositions, by various New Testament attestations, as well as by a variety of echoes in rabbinic tradition.

At this point I want to insert a few modern personal examples of how scripture can be used in different ways to help us grasp the idea of resignification or facilitating texts. Now over the years I've written a lot of songs that are either pure scripture or scripture-based.

Let's take this example from Psalm 89. Listen to how the chorus goes and compare it to the actual.

ESV text it was based on. Now did you notice all the little differences? Some things were added.

And some things were skipped. And why? Because this was an artistic adaptation of the Psalm, not an attempt to reproduce it verbatim for the sake of preserving the text for future generations or to produce an accurate Bible.

A thousand years from now a scholar might look at my song and say oh my goodness this is proof that the canonical text had not yet been established in 2024. There was still great fluidity in what was perceived as an authoritative scripture.

There are so many variants that it must be coming from a source text very different from the Masoretic text. But that scholar would be dead wrong because he doesn't realize that there are many creative uses of scripture for liturgical purposes and teaching and much more.

Moreover some of my other songs combine verses from different books of the Bible into one place creating a kind of scriptural potpourri. Another simple example of scripture use that could be called resignification is my book Praying the Bible Together which turns passages of scripture into prayers.

A lot of changes to the original text had to be made. So here's a short example from just two verses. Have mercy on us for we have plowed iniquity, we have reaped injustice, we have eaten the fruit of lies, we have trusted in our own way.

Do not destroy us. Now notice how the same two verses are molded into a much shorter passage as a prayer. This isn't because it's an attempt to adulterate, deceive, or rewrite scripture. No, rather it's simply an effort to facilitate the spiritual discipline of praying the Bible.

It doesn't claim to be the Bible nor is it evidence of a wildly different underlying source text. Now back to Gentry's article. Let's take some time to grasp and illustrate both of these approaches or scribal models repetition and resignification.

First an example of a conservative or repetition approach which copies the parent text exactly and precisely in every way is the Masada Psalm scroll from the last third of the first century BC. The Masada Psalm scroll has a precise format and layout.

As we all know the book of Psalms is written in poetry and Hebrew poetry is based on couplets of parallel lines and each column of this scroll has approximately 29 to 30 lines and one couplet is placed on each line with an appropriate space between the parallel lines of the couplet.

Only about 10 of the manuscripts from the Judean desert are carefully laid out in this way. The one manuscript at Qumran which most closely resembles the Masada Psalm scroll is 4QPSB although it has only half a couplet per line in the column of text and it has only 16 to 17 lines per column compared to 29 to 30 lines in the Masada Psalm scroll.

The format and text of 4QPSB are also not as close to the later MT as the Masada Psalm scroll. So the Masada Psalm scroll is a model scroll. We can compare the Masada Psalm scroll with both earlier and later traditions.

First its text agrees almost completely with the Aleppo Codex and the divisions marked by the blank spaces and line breaks in the Masada Psalms agree very closely with the Masoretic terminal markers, accents, and puzzle forms.

The Aleppo Codex also employs a system of division by blank spaces but this doesn't correspond well with meaningful breaks or the pattern in the Masada Psalm scroll. This suggests that the Masoretic tradition of the Psalter retained the visual concept of the line layout of earlier scribal praxis but without necessarily preserving the ancient content divisions.

The differences in layout between the Masada Psalms scroll and the Aleppo Codex are largely due to changing the book format from scroll to codex and using additional symbols for accents and vowels to mark what was indicated earlier by spacing in the manuscripts.

Otherwise the text a thousand years later is identical. The stichometry or layout of parallel lines of poetry in the Masada Psalms scroll agrees closely with the layout of lines evident in the Greek codices Sinaiticus and Vaticanus.

This proves a common tradition going back much earlier than the Masada Psalms scroll at least to the third century BC. Therefore the textual tradition in the Masada Psalms scroll is old. Next let's look at examples of the facilitating scribal model which is used for revising a text.

These are changes made in the copying process to help a community, a next generation, or a reader understand the text. Such changes might involve revising or updating the script. They might entail linguistic updating in terms of grammar, spelling, and vocabulary.

Geographical names can change over time and places are called by a new name. Aesthetic or stylistic improvements might be made and expansions are frequently inserted or parallel passages are harmonized.

Many of these types of changes have been made to the King James Version since it was first published in 1611. First consider the change in script from 1611 to the script we use today. They use a symbol like an F for an S and much of spelling is also different.

Second consider how Psalm 4 -2 looks in the 1769 edition often reprinted. We are familiar with this kind of type but the language is archaic and old. No one says ye anymore and what does it mean to seek after leasing?

It turns out that this is a word that meant lying in 1611. It has nothing to do with renting a car or house and you can check out the New King James Version of 1982 where this archaic language is modernized and if you want many more examples of this check out Mark Ward's YouTube channel.

Now let's consider one more example from English literature before looking at examples from the Dead Sea Scrolls. Consider Canterbury Tales by Geoffrey Chaucer written between 1387 and 1400. The current text is based on just 84 manuscripts and four in cannabula which are early books printed before 1540.

55 of these manuscripts are thought to have been complete and 28 are extremely fragmentary. Variants are due to copyist errors in some cases and others they are due to revisions by Chaucer himself. Here's a quote from the Merchant's Prologue.

Weeping and wailing care and other sorrow I know enough in the evening and in the morning said the merchant and so do many others who have been married. Here I've not shown what an early printed typeface would have looked like or what the original spelling might have been.

Even using a modern typeface the language is almost unintelligible. Now if we have difficulty reading an English text from only 500 years ago, remember that parts of Hebrew scriptures were already a thousand years old by the 2nd century BC.

Many copies of the biblical text entail updating in script or spelling or changes in forms syntax and vocabulary. A minimal type of updating involved changing the script from the Phoenician style used in the time of Hezekiah to the Aramaic square script beginning to be used in the 5th century BC.

At some point a scribe said to himself, if I don't change the bible from the script I learned in school to the script my children are using in school, my children won't be able to read the bible. About a dozen or so of the scrolls from the Judean desert are in the old style script and most of them are scrolls of the Torah.

While many differences are due either to copying mistakes or due to revision and updating involved in re-signifying the text, some types of facilitating or re-signifying were more radical. Sidney White Crawford in her 2008 monograph entitled Rewriting Scripture in Second Temple Times characterizes texts at Qumran on a continuous spectrum from biblical texts of the Pentateuch in the pre-Samaritan tradition to a text that is called reworked Pentateuch to the book of Jubilees, the Temple Scroll, the Genesis Apocryphon, and finally to 4q 252 which is a commentary on Genesis.

This spectrum moves from conflation, harmonization, and modification through new compositions closely related to the source text to commentary involving citation plus comment. She concludes that both canon and text were fluid and not standardized at this time.

What is helpful is that her study shows the graduated continuum from biblical text to paraphrase to commentary. Her conclusions however do not follow from the analysis of the evidence. The evidence from Qumran must be put within the larger picture of all the scrolls from the Judean desert which in the end is only the evidence of one sect within the widely variegated Judaism of the Second Temple.

In the larger picture there is a central stream dominated by the proto-Masoretic texts. The fact that most of the texts described by Crawford employ as a base a modernized text similar to that in the pre-Samaritan tradition is revealing.

She is describing the path of resignification at this time but this is only part of the larger picture. This is no different from a Christian or a Jewish bookstore today and should not be interpreted to show that the text was fluid or non-standardized.

So for example here is a list of Bibles in a modern bookstore in 2008 that you can pause and read. So the same categories used to classify texts at Qumran exist in Bible editions currently published. Bibles that offer a standard text unadorned and uninterpreted, and Bibles that adorn and decorate, paraphrase, interpret, and arrange the text for the audience and culture of our times.

Do we conclude from this that both canon and text are fluid? Hardly. Scholars studying the ancient scrolls have not paid sufficient attention to the function of these texts. There are many reasons why a person might resignify the biblical text.

An example is 4Q Deuteronomy N. This is an exerted and harmonized text. The term exerted means that certain passages have been taken out of the biblical text and put together for another purpose. This manuscript has the text of the Ten Commandments.

Now as you may know the text of the Ten Commandments is slightly different in Exodus 20 and Deuteronomy 5. In particular the reason for the Sabbath differs in these two texts. The reason in Exodus 20 is that God created the earth in six days and rested on the seventh.

The reason in Deuteronomy 5 however is that the Israelites were slaves in Egypt and God gave them rest from slavery so they should give rest to their slaves as well. In 4Q Deuteronomy N the person who extracted this text to teach the Ten Commandments harmonized the text and then used Deuteronomy 8 as a historical introduction to his Bible study pamphlet.

This clearly shows it was not a Bible. If you're interested in seeing the differences with your own eyes I'll put the three different texts on the screen and you can pause the video to read each of them.

Now it should be noted that with conservative copying on the one hand and facilitating texts on the other it's possible that both types of texts preserve original readings. One example is in Isaiah in the Isaiah scroll from Qumran Cave 1.

In comparison with the Masoretic text many of the variants represent linguistic modernizing and updating. Although it doesn't lay out the text in parallel poetic lines in a precise manner as we saw in the Masada Psalms scroll in one place it uses special spaces to show where the line division would be.

This is an image of Isaiah 61 10 -11. Notice the larger spaces. These spaces match the line division of the Masoretic text. There are also places where it preserves the original reading and the later Masoretic text doesn't such as in Isaiah 53 8 which is what we'll talk about at length in the next video to give you a taste of what the process looks like when addressing a specific variant.

Another example is 11q psa the famous Qumran Psalms scroll. This scroll is best described as a compilation. It's a selection of biblical psalms arranged with non-biblical hymns and songs probably for use as a liturgy in synagogue worship.

It is not a bible. It does not lay out parallel lines and couplets with appropriate spaces. It runs everything together as in prose texts. Yet in Psalm 145 it contains a verse missing from the later Masoretic text.

We know the MT is missing a verse here because Psalm 145 is an acrostic poem. Each verse begins with a successive letter of the Hebrew alphabet and in the MT the verse beginning with the letter N is missing.

The Septuagint has this missing verse but now a manuscript from Qumran that is not particularly carefully written also has the missing verse. Now if you're curious as to all the details and analysis of this verse it'll be in the next video in the series.

So after everything we've seen so far it's important to recognize that different publications or texts have different functions within the community of faith. In the case of the Septuagint since translation by definition is focused on explaining a text it's natural to use a facilitating text as a Hebrew parent text for the Septuagint.

This also explains why there are many agreements between the Samaritan Pentateuch and the Septuagint because the Samaritan Pentateuch was by and large a popularized text that was updated in various ways especially grammatically and lexically.

Both conservative and facilitating scribal models coexisted in Palestine. In fact recent research on the handwriting of the scribes has shown that the same scribe produced both types of manuscripts. This demonstrates that the two models are complementary.

The desire to transmit the ancient form of the text requires facilitating texts if the faithful are to understand and in turn the facilitating texts presuppose a standard. Of course there are other attempted explanations of the pluriformity in the late second temple period but they cannot be substantiated.

We'll address some of this at length in the next video but for now let's just say that the central issue is whether or not each of these different texts represents a conservative approach or a facilitating approach or a mixture of the two.

If one of the Dead Sea Scrolls supports the Septuagint this may indicate only that both are facilitating texts. In any case both approaches are complementary and presuppose a standard text. Instead of comparing the Dead Sea Scrolls with MT, LXX, and Samaritan Pentateuch we should assess the extent to which any of these witnesses represents a conservative or a facilitating model of scribal copying.

After the fall of Jerusalem in the Hebrew textual transmission there was only repetition and no longer any resignification. This gives the impression that the text only became standardized at this time and before was fluid but this is an incorrect conclusion.

Let me be absolutely clear the consensus view that the text was standardized in the first century AD is wrong. Rather what was dominant before the fall of Jerusalem in terms of repetition was likewise dominant after the fall of Jerusalem, the Proto-Masoretic text.

Since there was no longer any resignification it only appears that the text became standardized and wasn't before the first century. Two important reasons support this reconstruction. First after Jerusalem was destroyed Judaism was no longer variegated but rather dominated by one sect, the Pharisees, the precursors of the rabbinic tradition.

Their approach to the text restricted transmission to repetition. Now second the period from the first to fourth centuries AD is the period in which the Aramaic Targums developed. Hebrew was no longer a living language by the second century AD.

Jewish people spoke Aramaic. They continued to provide facilitating texts but they were in Aramaic and no longer in Hebrew. As we saw in the first video the Targums exhibit exactly the same types of resignification that we see at Qumran.

Thus there was resignification after the fall of Jerusalem but it was in Aramaic and in the Targumic tradition and therefore separate from the textual transmission of the Hebrew Bible. So analysis of the surviving witnesses shows complementary approaches to copying the text, conservative and facilitating.

A conservative approach requires producing facilitating texts and in turn facilitating texts presupposes a standard and the evidence of the Targums explains why no facilitating texts in Hebrew are found after the fall of Jerusalem.

There was a standard text all along. Thanks for watching and I look forward to working through some textual variants with you in the next video.