My last post was intended to give some context to the work that modern Biblical scholars do in order to give us the Bibles that you and I read. In short, I noted that, because writings in the ancient world had to be copied by hand, mistakes and intentional alterations were commonly made while the books that made it into our NT were copied. This leaves us with thousands of various manuscripts and fragments, many of which differ from one another in more than a few areas.
What we have in our English NT are translations of copies that were made over a century after the originals were supposedly written. The earliest copy of Galatians that we have, for example, is dated to around the year 200 CE, some 150 years after the apostle Paul wrote the original.
Thus, the problem that biblical scholars are attempting to solve with much of their work could be summarized as follows: How can we know what the originals of the NT books said considering what we have today are copies of copies of copies that were written much later than the originals and most of which differ from one another due to errors (or deviations from the originals) made by the copyists?
I want to highlight some of the tools that scholars employ in order to help them solve this problem. (The following is summarized from Ehrman’s Misquoting Jesus).
There are two categories of evidence that scholars look for when attempting to determine which readings are more likely to reflect the originals: external and internal. External evidence has to do with examining everything about the manuscripts themselves rather than what they actually say. Internal evidence looks into the content of each manuscript.
- Number – How many manuscripts support a given reading of a text? This criterion is rather straight forward. If more manuscripts support one reading over another then this counts as evidence for the reading that has greater representation being more original. However, in isolation this criterion offers us rather weak evidence. The fact that more manuscripts reflect a given reading over another does not ipso facto mean that that reading is more likely original. It simply means that the given reading was copied more than others.
- Age – How old is the manuscript? Older manuscripts are generally believed to reflect the original better than later manuscripts because it’s thought that texts get changed more with the passing of time. However, this criterion, like the ‘Number’ criterion above, must not be applied uncritically. Sometimes manuscripts that are in fact from a later time better reflect the original because they were copied from manuscripts that predate our oldest surviving manuscripts as well as the copies that the earlier manuscripts used. For example, a manuscript from the 8th century may have been dependent on a non-extant manuscript from the 3rd century whereas an earlier manuscript we have from the 5th century may have been dependent on one from the 4th. In this case, the 8th century manuscript would be considered more reliable despite its later date.
- Geography – Where was the manuscript written and how does it compare to manuscripts that originated in other areas? Readings that are attested to in multiple geographic regions are thought to be more reliable. If a number of manuscripts from Antioch, for example, support one reading whereas manuscripts from Alexandria, Rome and the region of Asia Minor support another reading then the reading from Antioch is thought to likely represent a local variation rather than the original.
- Reliability of the manuscript – Has the manuscript proven to be reliable in the past? After rigorously applying the above criteria as well as others to the manuscripts that we have for many years, scholars have been able to show that some manuscripts are more trustworthy than others. For example, if a manuscript from Rome represents one reading and a manuscript from Palestine represents another then scholars will take into consideration how both of these manuscripts have fared in the past with other variant readings. If the manuscript from Rome contains more variants that scholars believe to be reliable then this counts as evidence in favor of the reading in the Roman manuscript. Ehrman’s illustration is helpful: “When you know that a person is prone to lying, then you can never be sure that he or she is to be trusted; but if you know that a person is completely reliable, then you can trust that person even when he or she is telling you something you can’t otherwise verify” (kindle location 2089).
- Intrinsic Probabilities – What was the original author of this text most likely to have written based on writing style, vocabulary and theology? If a given reading contains words that can be found nowhere else in this author’s corpus or if the theology seems to contradict what’s known of this author’s theology from his other works then the reading under consideration is more than likely a later scribal addition.
- Transcriptional Probabilities – Which reading is likely to have been the result of a scribe’s redaction? The more problematic reading is likely to be original based on the fact that later copyists would want to harmonize or simplify any difficult passages they were copying. This criteria, which many scholars believe to be the most reliable text-critical principle, is summarized well in the following adage: The reading that best explains the existence of the others is more than likely to be original.