Here’s a thought experiment. Say you’re a strong Christian. You’re comfortable arguing for your position. You don’t think much of the atheist arguments that you’ve seen so far, and you’ve seen quite a few.

Now you come across a holy book from some other religion. It’s an English translation from an original written in a long-dead language from a foreign culture.

What could the book possibly say that would convince you that its miracle claims are correct? That it—and not your own Christianity—is correct? It’s just words on paper. What possible combination of words would be compelling?

We’ll make the case for the ancient religion as strong as possible

• Oral history. The earliest New Testament books were Paul’s epistles, written more than two decades after the life of Jesus. The gospel of John was written roughly six decades after, and Revelation and others possibly later still. The books of our imaginary religion were written roughly contemporaneously so that the period of oral history is almost nonexistent.
• Translation. The Christian story came from Jewish culture and the Aramaic language, but the original New Testament documents came from a very different culture (Greek) and language (Greek). Our imaginary religion will not have this extra level of translation.
• Copies far removed from originals. Our oldest copies of the books of the New Testament were written centuries after the originals. We’ll make ours just decades after. No—what the heck—let’s make them the originals themselves, 100% complete.
• Eyewitness testimony? Some of the noncanonical gospels claimed to have been written by eyewitnesses. For example, the Gospel of Peter says, “But I Simon Peter and Andrew my brother took our nets and went to the sea.” The Infancy Gospel of Thomas begins, “I Thomas, an Israelite, write you this account.” The Gospel of Thomas begins, “These are the secret sayings which the living Jesus spoke and which Didymos Judas Thomas wrote down.” Historians doubt that they are actual eyewitness accounts, but at least they claimed that they were. The four canonical gospels make no such claim—we don’t even know who wrote them—but the books in our imaginary religion will.
• Contradictions. The four gospels tell an inconsistent story. Was Jesus on earth 40 days after the resurrection (Acts) or one day (Luke)? Was there an earthquake followed by the dead rising from their graves? Only Matthew makes this claim. How many women were there? In what ways did Peter deny Jesus three times? And so on. The typical apologist’s response is to harmonize any differing accounts into a clumsy whole, but our imaginary religion will have multiple mutually supporting accounts with no contradictions.

In short, any positive feature of the evidence for the Jesus story will be matched or exceeded by our imaginary religion.

How are we doing so far, my Christian friend? Do you have any remaining objections where you see better evidence in favor of the gospel claims? Apply those features to improve our imaginary religion. How many independent accounts of the miraculous events do you want? Add that to the list. Does an ancient book sound more in touch with cosmic truth, or would you prefer a more recent and verifiable book? Would you prefer the documents to be written in an ancient language or modern English? A living language or a dead one? Do incidental elements in the story make it sound more authentic? Add it to the list. The only rule is that the evidence itself must be natural, as is the case with Christianity. No microfiche or holograms or levitating tablets or anything else outside of the technology of the time. No modern English written on a papyrus dating to 2000 years ago.

Given all this, would you become a believer? After a few days kicking the tires to verify that scholars indeed did agree to these claims, would you switch your allegiance?

If not, then evidence can’t be particularly important to you. If you say that your belief is based primarily on something besides objective evidence—personal experience or the religion you were taught as a child, perhaps—then don’t imagine that this will convince anyone else. If evidence doesn’t underlie your belief, why should I be convinced?

Instead, make clear that you believe because of this personal experience and not from evidence. Don’t raise evidence-based arguments if you believe in spite of the evidence.

The invisible and the non-existent
look very much alike.
— Delos McKown

(This is an update of a post that originally appeared 11/6/13.)

Photo credit: Wikimedia

A popular Christian argument declares that historians have roughly 25,000 manuscripts of New Testament books, far more than any other book from ancient history. Compare that with 2000 copies of the Iliad, the second-best represented manuscript from history. Even more poorly represented are the works of Herodotus, Thucydides, Aristotle, Julius Caesar, Pliny, Tacitus, and other great figures from history, for which we have more like a dozen manuscripts each.

Do we conclude that our records of Gallic Wars by Julius Caesar or Histories by Tacitus are so unreliable that they can’t inform our understanding of the past? Of course not. But if that’s the case, we must then accept the far-better attested New Testament manuscripts—or so the popular argument goes.

The first problem is that more manuscripts at best increase our confidence that we have the original version. That doesn’t mean the original copy was history—just like the original copy of The Wizard of Oz or the Arthurian legends wouldn’t be a record of history.

Consider the claim of 25,000 manuscripts. The originals of every New Testament book were written in Greek, but three-quarters of these manuscripts are translations into other languages. We can avoid the extra layer of interpretation imposed by a translation by focusing on just the 5800 Greek manuscripts.

Now consider when these manuscripts were written.

This chart shows the number of Greek manuscript copies by century. (The data is from Wikipedia, with manuscripts categorized on the cusp of two centuries put into the earlier century.) We have zero manuscripts from the first century and eight from the second. The twelfth century has the most, with 1090 manuscripts. The printing press was invented in the middle of the fifteenth century, which explains much of the drop on the right of the chart.

I recently explored the three most famous additions to the New Testament (the Comma Johanneum, the woman caught in adultery, and the long ending of Mark). The scholarly analysis for whether some of these passages are authentic or not turn on just a few manuscripts, and this chart shows why. The vast majority of the manuscripts, from perhaps the sixth century and after, never enter the conversation.

Our 25,000 manuscripts became 5800 Greek manuscripts, but those have now dwindled to just those few in the first few centuries after the crucifixion.

There are one hundred manuscripts in the first four centuries, and many of these are just tiny scraps. Consider papyrus P52 above—yes, that is considered a “manuscript.” It is a tiny fragment of John just 9 cm long. It is our oldest New Testament manuscript and dates to the first half of the second century, or perhaps later. Three more manuscripts (P90, P98, and P104) are also scraps of a similar size and date to the second half of the second century. (Though it’s probably obvious, I’ll emphasize that these dates are all just approximations, and arguments can be made for different dates.)

Another handful of manuscripts date to around 200 CE. Six of them (P4, P32, P64, P66, P77, and P103) are scraps, but in this group we get our first substantial manuscripts. P46 (part of the Chester Beatty collection) has much of nine epistles. P66 contains most of John. P75 (the Bodmer Papyrus) has a substantial fraction of Luke and John.

The record looks fairly good when you look at the dates of our earliest fragments of the various books in the New Testament—John in the second century, Matthew and Luke around 200, Mark around 250, and so on. But, again, the emphasis should be on the word fragment. Only when you get to the oldest complete (or nearly complete) texts—the Codex Sinaiticus and Codex Vaticanus from the fourth century—do you get all the missing pieces. (I’ve written more about this centuries-long dark age here.)

The “best attested by far!” claim for the New Testament is not only irrelevant, it’s not even true. You can find Egyptian hieroglyphics and Mesopotamian clay tablets that are both older and original. A single version, the original, beats enormous numbers of copies.

It’s not all that surprising that a handful of early documents from a popular religion in a dry climate were preserved until today, and let’s acknowledge that that’s impressive and historically important. But that we have 1090 manuscripts in the original Greek from the twelfth century is not much more helpful in recreating the originals than that we have 100 million new copies printed each year. What matters are the earliest copies—perhaps the hundred from first four centuries. And the hundred dwindle down to just a relevant handful of copies that are larger than scraps.

25,000 New Testament manuscripts? Big deal.

It is error alone which needs the support of government.
Truth can stand by itself.
— Thomas Jefferson

(This is an update of a post that originally appeared 11/4/13.)

Photo credit: Wikipedia

In Part 1, we saw the evidence and resolution of two disputed passages in the New Testament. Let’s look at one more and see what lessons we can draw.

Success #3: Long ending of Mark

Mark ends with three women going to the tomb and seeing it already open. (I’ve written more about the many contradictions in the gospel accounts of the resurrection here.) Inside the tomb, there is no Jesus but instead a man in white who tells the women the good news. But, “trembling and bewildered, the women went out and fled from the tomb. They said nothing to anyone, because they were afraid” (Mark 16:8).

The End.

Most readers find this an unsatisfying ending. How does the marvelous story get out if the women told no one? Are there no final words of comfort and direction from Jesus?

That disquiet must have been felt by early scribes as well, because they created the Long Ending (Mark 16:9–20), used in the King James Bible.

Jesus says,

These signs will accompany those who believe: In my name they will drive out demons; they will speak in new tongues; they will pick up snakes with their hands; and when they drink deadly poison, it will not hurt them at all; they will place their hands on sick people, and they will get well. (Mark 16:17–18)

Speaking in tongues, invulnerability to snakes and poison, and healing by touch—those claims are both startling to many Christians today and relevant to many more. Pentecostal and charismatic churches, which rely to varying degrees on the Long Ending, have half a billion members worldwide.

Justin Martyr, Tatian, and Irenaeus reference the Long Ending in the second half of the second century. The Codices Sinaiticus and Vaticanus, our oldest copies of Mark that include the ending, don’t have it, but they’re from the fourth century. Sounds like strong evidence for it being authentic.

On the other hand, the Long Ending appears tacked on and doesn’t flow well, and the consensus of scholars is that this is not original.

Nevertheless, that the Long Ending isn’t original doesn’t mean that Mark ending at 16:8 is. In fact, there are a total of five manuscript traditions for the ending of Mark, and the consensus is that none of them are the original ending.

The limitation of historical analysis

Let me highlight a couple of points. First, note that the enormous number of New Testament manuscripts that apologists like to point to (roughly 6000 in the original Greek and 19,000 in other languages) don’t matter. What matters are the best handful. Each of the debates above is resolved with just a few ancient manuscripts. The discovery of a thousand 12th-century manuscripts would likely advance scholarship less than the discovery of a single second-century papyrus.

Second, this analysis requires two versions. The scholar needs two traditions (a short Mark and a long Mark, for example, with at least one manuscript to represent each) to judge between the two. With only one tradition represented, there’s nothing to judge between.

It’s like a card game. You throw down two fourth-century Bible manuscripts that demonstrate your point—that a particular passage isn’t authentic, let’s imagine. I might be able to trump that with my third-century church father who quotes the passage from the gospel in question.

A single older manuscript reflecting your version might tip the balance in your favor. And—who knows?—such a manuscript might yet be found that would overturn the conclusions for one of our three examples. For example, Daniel Wallace claims to have a first-century copy of Mark, the analysis of which will soon be released. This might turn out to be like the tiny fragment called the Gospel of Mrs. Jesus, which after more thorough analysis didn’t turn out to be all that old. But suppose this first-century Mark had the last chapter. Could it overturn the consensus of scholars about the ending of Mark?

Apologists like to claim that manuscript variants cause them no lost sleep. Look at these three questionable sections, they’ll say. They’re all resolved to everyone’s satisfaction, so where’s the problem?

Here’s the problem. Imagine a historical fork in the road for a particular document. The correctly copied version goes down path A, and the one with a significant error goes down path B. Now imagine that one of those traditions is completely lost to us. That’s hardly surprising. We have only a few tiny scraps of papyrus manuscripts within the first century after the original authorship.

Historians today would have a single, consistent tradition, but is it A or B? Worse, they wouldn’t even know that there had been a fork in the road.

For how many errors in our Bible do we not even know to ask what the original was?

See also: How Long from Original New Testament Books to Oldest Copies?

Those who do not know their opponent’s arguments 
do not completely understand their own.
— Anon.

(This is an update of a post that originally appeared 10/31/13.)

Photo credit: Wikipedia

Historians use textual and historical criticism (also called lower and higher criticism, respectively) as they sift through mountains of Bible manuscripts to decide how old they are, where they came from, who wrote them, and what the original probably said. Despite their power, however, these techniques are hobbled by an important and obvious limitation.

Let’s first consider some of the impressive successes by looking at three important New Testament passages. Scholars are in general agreement that these were not in the original manuscripts.

Success #1: the Comma Johanneum

This “comma” (that is, phrase) is the bold part in the epistle of 1 John below:

For there are three that testify in heaven: the Father, the Word and the Holy Spirit, and these three are one. And there are three that testify on earth: the Spirit, the water and the blood; and the three are in agreement. (1 John 5:7–8)

This phrase is the clearest biblical statement in support of the concept of the Trinity. (I’ve written more on the long and bumpy road to the Trinity here and here.) This fundamental doctrine became an important point of debate, and it was resolved only at the First Council of Constantinople in 381. And that’s an important clue. The Comma would’ve surely been a central piece of evidence in the debate, if it existed at that time, but it was never brought forth. The early church fathers clearly didn’t know about it. Our oldest manuscript evidence is from a seventh-century Latin manuscript, and it’s in none of the early Greek copies.

With the best manuscripts not having the Comma and early fathers curiously silent, this one is easy to resolve.

Success #2: The woman caught in adultery

In this story (John 8:1–11), the Pharisees bring to Jesus a woman accused of adultery. The Law says that such a woman should be stoned to death, so what does Jesus say? Jesus replies, “Let any one of you who is without sin be the first to throw a stone.” While Jesus draws in the dirt, the men drift away. When they are gone, he admonishes the woman to sin no more.

The evidence is similar to the case of the Comma Johanneum. The story is absent from copies of John from the early third century and from the earliest complete New Testaments (the Codex Sinaiticus and Codex Vaticanus, from the fourth century).

The story is referenced by third- and fourth-century church fathers, but the first copy of John to include it is from the fifth century. The New Testament manuscripts that include it don’t always put it in the same place, as if it were a cherished story that scribes found a home for in various spots.

Though Augustine in the fourth century thought that the story might’ve been omitted to avoid the impression that Jesus sanctioned adultery, the consensus is that this story is not original.

We will look at one final example and the limitations of this method in part 2.

Be hard on your opinions.
A famous bon mot asserts that opinions are like a**holes
in that everyone has one.
There is great wisdom in this but I would add
that opinions differ significantly from a**holes in that
yours should be constantly and thoroughly examined.
— Tim Minchin

(This is an update of a post that originally appeared 10/30/13.)

Photo credit: Wikipedia