Balanced Eclecticism attempts to solve NT textual problems by advocating a cautious approach to external evidence (and particularly, a sceptical attitude toward claims that some manuscripts are of such supposedly excellent character that we must, in cases of doubt, always follow their lead), an evidence-based approach to transcriptional canons, and a presumption in favour of the internal logic of the NT documents (i.e. a high view of their integrity and rationality). It is similar in nature to Reasoned Conservatism, to use D. A. Black's term in his book, New Testament Textual Criticism: a Concise Guide, or the approach advocated by Harry Sturz in his book, The Byzantine Text-type and New Testament Textual Criticism.
To understand what Balanced Eclecticism involves and how it works, here is a simple acronym we can use to help determine the correct reading: PANEL.
Propinquity - the geographical spread of manuscript, versional and patristic evidence for a reading. A reading that has better geographical attestation is more likely to be the authentic reading than a reading found only in witnesses from one geographical location.
We have textual evidence from four main geographical areas:
- North - The areas where the Byzantine text dominated from at least the fourth century onwards (we have no evidence from here before this). This broad sweep of territory from Antioch to the Adriatic was evangelised by the apostle Paul, and included Galatia, Asia Minor (Ephesus, Colosse), Greece (Corinth), Macedonia (Thessalonica and Philippi), and the Mediterranean islands of Cyprus and Crete. The apostles Peter and John also had an influence upon this area as seen Peter's letters to recipients in Pontus, Bithynia, Asia, Galatia and Cappadocia and in John's later ministry in Asia Minor. After the fall of Jerusalem in AD70, these continued to be some of the most important areas in Christendom throughout the first millennium. The north is therefore represented by the Byzantine majority of manuscripts, as well as the Gothic (4th C) and Slavonic (9th C) versions.
- West - The Latin versions (the most important version of antiquity, from the 2nd century onwards)
- South - The Coptic (the third most important version of antiquity, from the third century) and Ethiopian (6th Century) versions.
- East - The Syriac (the second most important version of antiquity, from the 4th century, although some argue that it existed as early as the 2nd century), as well as Armenian (5th Century) and Georgian (5th Century) versions.
How does this information help us determine the correct reading?
The vast majority of readings that are preferred in our currect critical text are not, as some people think, based on the oldest and best evidence. Instead, our current critical text closely follows the readings of a small number of Fourth and Fifth Century Alexandrian manuscripts (Sinaiticus, Vaticanus, Alexandrinus) - see the evidence here. Sometimes these readings have little support apart from Alexandrian manuscripts and the Coptic (i.e. Egyptian) versions. If we look at the map above, this means that they do not have very widespread geographical support, making them unlikely to be the original readings of the New Testament. By the same token, it would be strange if a reading only found among the Byzantine majority of manuscripts was the original reading if all the versional and manuscript evidence from East, West and South were united against it. How would we explain how all the geographically diverse regional witnesses independently hit upon the same alternative yet incorrect reading?
Not all textual decisions are as clear-cut as these cases of geographically extreme readings. What if a reading was supported by the Alexandrian (South) and Latin (Western) evidence, but not by the Byzantine (North) and Eastern versional evidence? Such evidence would call for more careful examination of a textual variant, because we can see from the map above how Alexandrian and Western evidence could have influenced each other (North Africa was a very important area for early Latin Christianity, and was geographically close to Egypt), and how Byzantine and Eastern evidence could similarly be dependent upon each other.
Balanced Eclecticism holds that the Byzantine text-type is neither the original autographic New Testament, but neither is it to be relegated to the rubbish heap (as the contemporary 'standard text' routinely dismisses it). Instead, it is presumed innocent of textual corruption until proven guilty: indeed, studies into scribal habits show that the Byzantine text bears fewer marks of typical scribal corruption than the other text-types (omissions short and long, which in turn produced a terser, more disharmonious, less sensible text). On this basis, in Balanced Eclecticism, the Byzantine text is considered to be at least the equal of any of the other text-types.
Different textual critics give different 'weight' to manuscript and versional evidence according to their personal preferences. Propinquity is therefore a safeguard against according exaggerated influence to any one strand of the manuscript, versional and patristic evidence.
Antiquity - the age of the manuscript, versional and patristic evidence for a reading. A reading that has more ancient evidence is more likely to be the original reading than readings found only in later manuscripts. Thus, some readings in the Textus Receptus or the Byzantine text are only found in late manuscripts, and are less likely to be original than alternative readings found in early manuscripts.
Of course, some readings only found in later manuscripts are in reality ancient (as the discovery of the papyri proved) and some ancient readings cannot be followed because they are simply early idiosyncratic errors. Nevertheless, this criterion provides a safeguard against the adoption of evidence that is only found in late manuscripts.
Number - the number of manuscripts supporting a reading. A reading found in more manuscripts is more likely to be the original reading than a reading only found in few witnesses. This does not mean a simple majority is determinative, of course (it would spoil the acronym to use the word Majority instead of Number). Rather, this criterion is a warning against the adoption of a reading found in an extreme minority of manuscripts. Just because a reading has the most manuscripts on its side does not mean it is the genuine reading of the text, but this criterion guards against the opposite tendency current today for editors to prefer idiosyncratic readings only found in a few, closely related, manuscripts.
Explanatory - a reading that better explains the origin of the alternative readings on the grounds of transcriptional probability. A reading that scribes would tend to create from other readings, in line with known scribal habits, is more likely to be the original reading.
Sadly, however, much of what the textual handbooks tell us about scribal habits is wrong. Accumulating evidence from recent studies into scribal tendencies show (in line with common sense), that scribes tended to omit material (the easiest mistake to make), to make the text more terse and hence less polished (by short omissions of material), to introduce errors into the text making it more difficult to understand, and to disharmonize originally parallel passages (by short omissions and introductions of novel difficulties).
Logic - a reading that fits more logically into the author's thought-flow or argument, and more in keeping with his theology and style. This is actually the way most people practice a rudimentary form of textual criticism when they error-correct an email from a friend (or a letter, if any people still write handwritten letters): they look at the sentence, try to sum up what it was trying to say, and auto-correct as they read.
In Matthew 5:22, Jesus said, 'whoever is angry with his brother without a cause shall be in danger of the judgement'. A small number of Greek manuscripts omit the words 'without a cause' (P64, 01, B, plus two other late manuscripts). P64 is the oldest manuscript at this place, but these three manuscripts are all Alexandrian - that is, they are textually related to each other, so it is not surprising that they have the same reading. The vast majority of manuscripts contain the words. When we look at the versional evidence, not only do the Old Latin and the Syriac versions contain the words, but even the Coptic versions contain them too, breaking ranks with the three Alexandrians, their stablemates.
In terms of transcriptional probability, the omission is only one word in Greek, and the fact that scribes tended to omit single words rather than add them means that the omission is far more likely to be the result of scribal oversight, especially when there is a case of homoeoarcton involved, like here, where the following word starts with the same letter as the word in dispute.
In terms of theology, the Bible makes a clear distinction on the subject of anger: it is sometimes right and proper for us to be angry, while anger can also sometimes be sinful. Thus, Jesus was angry (Mark 3:5), God is angry with the wicked every day (Psalm 7:11), and the Christian is told to 'be angry and sin not' (Eph. 4:26). The words 'without cause' in Matthew 5:22 seem to nicely capture the right balance and the retention thus seems more in keeping with what the Bible elsewhere teaches.
Thus, although antiquity favours the omission, on the other hand majority, propinquity, theology and transcriptional probability strongly favour the retention of the words. It is hard to see how the words, if they were originally missing, could have spread to so many different geographical regions in exactly the same form if they had been added in by inventive scribes taking liberties with the text. It is more likely that one extremely limited stream of related evidence accidentally omitted the word.
Romans 8:28 reads that 'we know that all things work together for good to those who love God' in the majority of manuscripts, but four Greek manuscripts (P46, A, B and 81) instead read 'we know that God works all things together for good to those who love God' (see the NASB and NIV for this reading). Among the versions, only one part of the Coptic (the Sahidic) and the Ethiopic version follow the minority reading, which is hardly surprising considering the minority manuscripts are Alexandrian. All the other versions, including the Latin, the Syriac and part of the Coptic tradition (the Bohairic) follow the majority of Greek manuscripts. Propinquity thus overwhelmingly favours the majority reading.
In terms of transcriptional probability and internal logic, neither reading clearly stands out as superior. The minority reading seems more clumsy, mentioning God twice in the one phrase, but on the other hand the minority reading also states more explicitly the Christian belief that God is sovereignly guiding us through the turbulent events of life. Because scribes tended to omit rather than add, the minority reading seems more probable, however the fact that the extra words are found in such an extreme minority of manuscripts suggests that the addition could be a scribal addition.
Therefore, it seems that on the grounds of propinquity and number, we should stick with the reading found in the more diverse textual witnesses.
In Ephesians 5:9, the textual variation involves whether we should read 'for the fruit of the Spirit is in all goodness, righteousness and truth' (as in the KJV and NKJV) or 'for the fruit of the light is in all goodness, righteousness and truth' (as in most other modern Bibles). The first reading 'Spirit' is found in our oldest witness at this point (P46, c. A.D. 200) as well as the majority of Greek manuscripts. However, the second reading is found in about twenty manuscripts, including the early P49 (3rd. cent.). But crucially, this second reading is found in the majority of versions, including the Latin, Syriac and Coptic versions from the east, west and south. Only one later version (the Syriac Harklean version) aligns with the first reading. It seems less likely that the transmission of the NT text into all the different versions of antiquity would independently hit upon the same wrong reading than that the transmission of the majority of Greek manuscripts was corrupted at this point.
The other big point in favour of the second reading is that it fits with the context. All through the first part of Ephesians 5, Paul is writing about how Christians should be 'light in the Lord' (5:8), that we should not 'have fellowship with the unfruitful works of darkness' (5:11), that 'all things are exposed by the light' (5:13), and that 'Christ will give you light' (5:14). The 'fruit of the light' seems more in keeping with what Paul is writing about than the 'fruit of the Spirit', which instead seems to be a harmonization to Galatians 5:22.
Therefore, on grounds of propinquity, internal logic, and transcriptional probability (the possibility of harmonization to Gal. 5:22 here), we should strongly prefer the reading found in the more geographically diverse witnesses, 'the fruit of the light'.
Lastly, in 2 Peter 1:3, the majority of manuscripts, including the earliest witness (P72), read about 'him who called us through glory and virtue'. About 25 Greek manuscripts, however, read 'him who called us by his own glory and virtue'. Crucially, this second reading is found in the all of the geographically diverse early versions into Latin, Syriac and Coptic, while the first reading is found in none of them.
Added to this is the fact that the second reading makes more sense: Christ called us by his own personal glory and virtue. We can understand why someone like Peter was attracted to Christ - there was something totally compelling about following a brilliant person like Christ (and 'virtue' is a word Peter likes to use in his letters). By comparison, the first reading seems like vague and colourless religious waffle: what exactly does it mean that we were 'called by glory and virtue'?
Therefore, we should prefer the reading that makes more sense in context and that is found in more diverse witnesses.
The Problem with Reasoned Eclecticism
One of the major differences between Balanced Eclecticism and Reasoned Eclecticism (the 'mainstream' method of textual criticism practiced today) is worth noticing.
New Testament textual criticism has been plagued, increasingly so over the last few decades, by the fact that Prefer the Harder Reading has always clashed with Prefer the Reading that Makes More Sense in Context. This has resulted in an internal contradiction at the heart of Reasoned Eclecticism's textual decision-making (or, if you prefer, a 'Mexican stand-off'). A reading that made more sense in context, and a reading that made less sense were equally defensible. Textual Critics were forced to personally intervene to break the deadlock, and were able to advocate for any reading however eccentric or poorly supported by external evidence (although they were usually happy to plump for their preferred external evidence, whether it was Alexandrian, Western or even Byzantine).
Now, however, with evidence that shows that scribes disproportionately introduced far more 'harder readings' than 'easier readings', the transcriptional canon Prefer the Harder Reading needs to be abandoned. The result of this is that textual criticism is freed from the subjectivism and personal preferences that the textual stalemate between Intrinsic Probability (Prefer the Reading that Makes more Sense) and Transcriptional Probability (in particular, the flawed and evidentially-baseless canon, Prefer the Harder Reading).
Thus, with increasingly accurate information about scribal habits, the three main arguments that textual critics make, instead of being internally contradictory, should all point in the same direction:
1. Prefer the Reading with the Best External Evidence (Propinquity, Antiquity and Number of Witnesses)
2. Prefer the Reading that makes More Sense in Context (Logic)
3. Prefer the Reading that Best Explains the origin of the Alternative Readings (Explanatory Power)
The acronym PANEL reminds us that, just like a dashboard with multiple dials, a decision to choose one reading over its alternatives should not depend upon one factor on its own. Instead, the combination of a number of factors pointing in the same direction leads to the choice of one reading over another.