Out of the Ashes

It’s not often that two of my interests – computer technology and the Bible – intersect, but they did in a science discovery announced this week. A team of computer specialists at the University of Kentucky were able to read the writing on a 1700-year old Torah scroll that had been turned into charcoal by a fire in 600 CE. How this miracle of science was accomplished and what it uncovered is fascinating.

The Spring of the Young Goat

The story begins on the Western shore of the Dead Sea, a barren desert without tree or shrub except for those around oases such as the one at En-gedi, 35 miles southeast of Jerusalem. En-gedi (the name means “spring of the young goat”) is mentioned a few times in the Bible, most notably as one of the spots where David fled from Saul (1 Sam 24). It was the site of a Jewish community beginning in the 8th century BCE and by the 4th century CE, it was a large Jewish city with at least one synagogue. The city was destroyed by fire in the 7th century during the Arab conquests.

Archaeological excavations in 1970 uncovered the Holy Ark in the ruins of the city’s synagogue. The Holy Ark is a cabinet in a synagogue that holds the scrolls used for synagogue readings. Inside the En-gedi Ark were the charred fragments of its Torah scrolls. The blackened lumps of charcoal fell apart at the touch and there was no way to unwrap them, so the Israel Antiquities Authority (IAA) preserved them on the chance that some future technology would allow them to be read.

Start with a Good Scan

Did the charred scrolls contain any writing at all or were they a total loss? Sefi Porat, one of the excavators who worked on the En-gedi synagogue, asked his colleague, Pnina Shor, head of the Dead Sea Scrolls (DSS) Project at the IAA, if she could scan the En-gedi scrolls as part of the project to create images of all DSS material.

Modern X-ray-based micro-computed tomography (micro-CT) scans of the En-gedi scrolls can reveal there was writing on the layers inside because the ink, containing metals like iron or lead, is denser than the parchment on which it was written and provides a good contrast. The micro-CT scan, designed for high-resolution scanning of biological tissues, can also visualize the layers in the scroll.

But it was impossible to decipher the writing. The Hebrew letters appear on a 3D cylindrical image and cannot be read without conversion to a 2D image. It would be as if you wrote a message on a piece of paper, rolled it up and tried to read it by holding it up to a light. The letters on one layer would be superimposed on the letters on the other layers.

Computer-Assisted Archaeology

This is where the University of Kentucky comes in. Similar charred scrolls were unearthed in the library of Julius Caesar’s father-in-law in Herculaneum, destroyed along with Pompeii in 79 CE. W. Brent Seales, professor and chairman of the computer science department at the University of Kentucky, has spent the last 13 years working on ways to read such damaged scrolls. Dr. Seales and his team developed a technique to “virtually unwrap” the charred scrolls once they had been scanned.

Learning of Dr. Seales’ work, Dr. Shor sent him the micro-CT scan of one of the En-gedi scrolls and asked if he could analyze it.

The program developed by Dr. Seales’ team builds a triangulated surface mesh to identify the layers of material in the scroll. The team identified five complete revolutions of parchment in the scroll. Once the layers had been identified and modeled, the program renders readable textures on the layers by assigning a “brightness” value from the appropriate layer of the CT scan to each point on the segmented surface in the computer model.

Video from The New York Times describing the technique.

Because the layers were rolled-up in the scan, the next step was to flatten the segments. This was done via common 3D-to-2D mapping techniques. The final step was to merge the flattened segments into a composite image. The final image revealed the scroll to be two columns of text that include 35 lines (18 preserved and 17 reconstructed) of Hebrew text from the first two chapters of Leviticus, the third book in the Torah.

A composite image of the completed virtual unwrapping of the En-gedi scroll (source: B. Seales)

The Findings

Radiocarbon results date the scroll to 3rd-4th century CE. This puts the En-gedi scroll a century or two after the period of the Dead Sea Scrolls (3rd century BCE to 2nd century CE) and centuries before medieval biblical fragments found in Cairo dating to the 9th century CE.

The text on the En-gedi scroll is identical to the Masoretic Text (MT). The MT is the text printed in most modern Hebrew Bibles and used as the source for translations into English. This is interesting because the DSS written just one or two centuries earlier reveal some variations in the biblical texts. This discovery pushes back the date for the standardization of the MT to the 3rd-4th century CE.

The work of Dr. Seales and his team was financed by a government grant. The suite of software programs, called Volume Cartography, will be released as open source once the grant ends. Use of the software on the damaged scrolls from the Herculaneum library could re-discover long-lost works of Latin and Greek literature. It is a remarkable accomplishment.

Details of the “virtual unwrapping” technique are detailed in an article published in the 21 Sept 2016 issue of the journal Science Advances.

Begin the Begats

There is much in the story of Cain and Abel to perplex a casual reader of the Bible. First, when YHWH sentences Cain to exile in the land of Nod, Cain complains in Gen 4:14c, “Anyone who meets me may kill me!” According to the story, there is no one else in the world beside his mother and father, so why should Cain fear that he will be hunted down and killed? A couple of verses later, in 4:17, we read that Cain “knew his wife,” she bore him a son (Enoch), and he built a city which he named after his son. Where did his wife come from? And how could there be enough people in the world, let alone the “land of Nod,” that a city would be needed to hold them all?

Based on clues like these, biblical scholars speculate that the story of Cain and Abel was once an independent story that was attached to the story of Adam and Eve. The story of Cain and Abel presupposes that there were more people in the world than just Adam and Eve and their two sons. Unless there were neighbors around to witness the deed, why else would Cain be concerned about taking Abel out into the field to kill him? And why else would he be concerned about his safety when he was expelled from the community? It also explains how he was able to find a wife and need to build a city.

There’s another clue in the genealogies in Gen 4 and 5. Gen 4:17-24 gives a list of Cain’s descendants: Enoch, Irad, Mehujael, Methusahael, Lamech. The closing verses list the three sons of Lamech: Jabal, Jubal and Tubal-cain (and one daughter, Naamah). Vv. 25-26 describe the birth of Seth to replace Abel, and the birth of Seth’s son, Enosh (which means “man” in Hebrew).

The Cain Family tree according to Gen 4:17-24

Gen 5:1-32 begins with Adam begetting Seth and proceeds to Enosh, Kenan, Mahalalel, Jared, Enoch, Methuselah, Lamech, and Noah. Noah, it is noted, had three sons: Shem, Ham and Japheth. The name of the Seth line is eerily similar to the Cain line. Both genealogies end with the naming of 3 sons instead of the usual one. Biblical scholars believe that we have two traditions citing the line of descent from Adam, one through Cain and one through Seth.

The writing styles differ, too. The genealogy in Gen 5 follows a strict formula, just like the creation story in Gen 1 followed a strict formula. Biblical scholars attribute Gen 1 and 5 to the “Priestly” writer (or “P”) because of his interest in priestly matters like legal codes, genealogies and rituals. The genealogy in Gen 5 cites the name of the father, states he was X number of years old when he begat his first son, he lived Y number of years after that, and then after living X+Y years, he died. This formula is not broken except for Enoch who, instead of dying, was “taken by God.” If we only read Gen 5, we would have no indication that Cain, rather than Seth, was Adam’s firstborn son.

Compare this to the genealogy in 4:17-26 and it is a bit more free-form. We get details about the various accomplishments of the individuals mentioned: Cain built the first city, Lamech took two wives, Jabal was the ancestor of tent-dwellers, Jubal of musicians, and Tubal-cain of blacksmiths. A couple of wives and a daughter are also named. This genealogy is showing us how civilization and technology spread in the primeval period. Also, violence as Lamech brags about how he takes extreme vengeance on any who slights him in the least way. Lamech’s story prepares us for the increasing amount of sinfulness in the world that will lead God to want to destroy it in a great flood. Scholars attribute the author of this genealogy in Gen 4 (along with Gen 2-3) to the “Yahwist” writer (or “J”), who gets his name from his consistent use of the name YHWH for God.

Another difference between the two is that J’s genealogy does not provide ages for the various ancestors. P’s genealogy provides precise numbers and, using those numbers and others like them, we could calculate a date around 4000 BCE for the creation of the world. The problem for chronologists is that the numbers of years separating Adam from Noah are different depending on whether you consult the Hebrew Masoretic Text (1656 years), the Greek Septuagint (2242) or the Samaritan Pentateuch (1307). It quickly becomes apparent that the ages were not in the original text, but were added later.

It is also apparent that there is a desired number each tradition is aiming for and the ages are adjusted to meet that pre-determined date. One proposal, for example, is that the total of the years from the creation of earth to the dedication of the second temple in Jerusalem is 3600 according to the MT. When you consider that the Bible was edited during the Babylonian Captivity and the Babylonians used a base-60 mathematical system, 3600 is 60 x 60, a nice round symbolic number in base-60. Perhaps the numbers are different in the SP because they were counting to founding of the Samaritan temple on Mount Gerizim. We don’t really know for sure, other than that the numbers were definitely manipulated for some symbolic purpose.

The genealogies are considered by many to be some of the “boring” parts of the Bible, but there are some interesting aspects to them. Once you know where to look.

An Inerrant Translation?

The King James Version (KJV) of the Bible is, without a doubt, the most influential translation of the Bible in the English language. Commissioned by James I of England and completed in 1611, it became the standard for quoting scripture. Written in the language of Shakespeare, millions can quote passages of it from memory: “man shall not live by bread alone,” “for whither thou goest, I will go,” “the Lord is my shepherd; I shall not want,” and so on.

Since 1611, however, many discoveries either gave us older versions of the manuscripts that were then available to the KJV translators or shed light on the meanings behind the Hebrew and Greek words. Yet, the KJV had become so beloved over the centuries that release of the Revised Standard Version (RSV) in 1952 was greeted with accusations of blasphemy and book burning. It was only inevitable that, after having been used almost exclusively for almost 350 years, there would be a contingent of bible believers who conclude that the KVJ was the only English translations that Christians should read: “If it was good enough for St. Paul, it’s good enough for me.”

But it is one thing to prefer the KJV for the beauty of its language or its familiarity. It is quite another to fetishize it as a divinely inspired translation, based on specific Hebrew and Greek texts that are also divinely inspired. To be clear, many evangelicals believe the original Hebrew and Greek autographs – originals which everyone agrees are no longer in existence – are divinely inspired or inerrant (or both). But there is a certain cadre of fundamentalists who argue that the Hebrew Masoretic Text and the Greek Textus Receptus underlying the KJV are superior to the Hebrew and Greek texts today that have been reconstructed using older and better manuscripts.

If you turn to Gen 4:8 in the KJV, you will read: “And Cain talked with Abel his brother: and it came to pass, when they were in the field, that Cain rose up against Abel his brother, and slew him.” Now, let’s compare to the RSV: “Cain said to Abel his brother, ‘Let us go out to the field.’ And when they were in the field, Cain rose up against his brother Abel, and killed him.”

Cain’s words “Let us go to the field” (only two words in Hebrew) do not appear in the KJV. In the RSV (and other modern translations that have the line), there is a textual note: “Sam Gk Syr Compare Vg: MT lacks Let us go out to the field”. This note is telling us that the Masoretic Text (MT), the authoritative Hebrew text for rabbinic Judaism, is missing Cain’s actual words, but we are able to fill in this gap by looking at other ancient texts such as the Samaritan Pentateuch (Sam), the Greek Septuagint (Gk, also abbreviated LXX), the Syriac Peshitta (Syr), and the Latin Vulgate (Vg). All of these texts were copied or translated from a Hebrew text much older than the oldest complete Hebrew text we possess today and serve as a check against the MT. 

The relationship between the various ancient manuscripts of the Old Testament, according to the Encyclopaedia Biblica

In the case of Gen 4:8, it turns out that the MT appears to have lost a couple of words and the ancient versions were able to supply them. And Gen 4:8 is not an isolated example. Despite the care of the scribes copying the texts, over the centuries mistakes were made. Sometimes the eye of the scribe would jump over a line and miss copying it.

When you have several ancient manuscripts that agree with each other against the MT, what most likely happened is that the words were originally there in the root Hebrew text. When the other versions branched off they copied or translated the words that were there. But in the branch that led to the MT the words accidentally dropped out at some point.

While logical, this is not an acceptable explanation for the KJV-only crowd. Their doctrinal belief is that the MT on which the KJV is based is without error because God would not inspire the human authors to write down his words, only then to allow those holy words to succumb to scribal copying errors. No, he would divinely preserve his words through the centuries. Any “gaps” in the MT are only apparent gaps. The other ancient versions independently (and erroneously) concluded there were words missing from Cain’s speech and just happened to supply the same words to plug the hole. The KJV translators were divinely inspired to perceive that, in fact, there were no missing words and thus translated the passage exactly how God intended.

It only makes sense if you are already predisposed to the conclusion.

I appreciate that ancient texts can correct mistakes that were made somewhere down the line when a copy of a copy of a copy was being copied. I welcome new discoveries of non-biblical texts that shed light on how biblical words should be translated. We should always be striving to improve our understanding, to make known that which was once hidden.

Or as the KJV itself put it in Luke 8:17, “For nothing is secret, that shall not be made manifest; neither any thing hid, that shall not be known and come abroad.”