Wednesday, November 13, 2019

WM 138: Text Note: Ephesians 3:9

I have posted WM 138: Text Note: Ephesians 3:9. Listen here.

A popular internet apologist (PIA) has recently suggested that Ephesians 3:9 is a “blatant error” in the TR, and, therefore, this poses a supposedly insurmountable “defeater” for the Confessional Text position.

Is this, in fact, the case? Is defense of the TR, in general, and the TR reading in Ephesians 3:9, in particular, completely irrational?

Let’s look at the evidence by posing a series of five questions and responses:

First: What is the supposed “blatant error” in the TR of Ephesians 3:9?

The controversy in Ephesians 3:9 involves the TR reading η κοινωνια του μυστηριου, “the fellowship of the mystery” (as translated in Tyndale, Geneva, KJV). In the modern critical text the reading is η οικονομια του μυστηριου, “the plan of the mystery” (ESV) or “the administration of the mystery” (NASB).

The controversy here is really about one single word:

TR: η κοινωνια

MCT: η οικονομια

We can immediately see that the words are very similar in form to one another, and we can see how there might easily have been scribal confusion between the two words. One has 8 letters and the other 9 letters. Every letter in κοινωνια appears in οικονομια, except one: omega. Both words end in iota alpha.

Second: Why is it argued that the MCT reading is superior to the TR reading in Ephesians 3:9?

A friend shared a FB post from the aforementioned PIA which begins:

There is no evidence to my knowledge (manuscript, patristic, versional, inscriptional) within the first 1000 years of church history of anyone reading Ephesians 3:9 as "the fellowship of the mystery" (ἡ κοινωνία τοῦ μυστηρίου). The reading is unquestioned: it is "the administration/plan of the mystery" (ἡ οἰκονομία τοῦ μυστηρίου).
We should note that this argument against the TR reading is entirely based on the external evidence.
Let’s begin with some analysis of the Greek manuscript evidence, which is generally the most important.
Among current extant Greek manuscripts, of all eras, the Majority reading is indeed η οικονομια. In fact, the external evidence is so overwhelming that the NA28 does not even list any variants at this point in its critical apparatus.
Bruce Metzger, however, offers the following comments on this variant in his Textual Commentary, Second Edition (1994): “The Textus Receptus, in company with a scattering of late minuscules, replaces οἰκονομία with the interpretive gloss κοινωνία (hence AV “fellowship”). The true reading is supported by p46, all known uncials, almost all minuscules, all known versions, and patristic quotations” (535).
Though Metzger, unsurprisingly, dismisses the Greek ms. support for the traditional reading as confirmed only by “a scattering of late minuscules”, he does, at least, acknowledge that this reading is present in the Greek ms. tradition.
This brings up an important related point, which the intrepid PIA seems always to overlook in throwing out random objection passages to the TR like this one. Namely, those who prefer the TR readily and openly acknowledge that it is an eclectic text. It is not based on the Majority text. Many of its readings are found in the Majority text (like the traditional ending of Mark), but some are based on a minority tradition. The PIA seems completely oblivious to this point.
It seems particularly odd for the PIA to reject the TR reading at Ephesians 3:9 based on the fact that it is not the Majority reading since, supposedly, he is not himself an advocate for the Majority text but, instead, embraces an eclectic method (reasoned eclecticism). We might call the TR “providential eclecticism.”

Side Note: There is another variant in Ephesians 3:9 that involves the prepositional phrase at the end of the verse, “through Jesus Christ.” In this case the Majority text and the TR both include the phrase while the MCT rejects it. If the PIA supports the Majority text in the case of the “fellowship/plan” variant, why not accept it here also? Why not follow the Majority text in passages like Mark 16:9-20?

Furthermore, the PIA expresses great confidence in the new CBGM, despite the fact that in the NA28 it favors a reading in 2 Peter 3:10 based on NO extant Greek mss.! There seems to be a problem with consistency.
Third: What about the Greek manuscript evidence for Ephesians 3:9?
When the PIA proclaims that the TR reading does not appear in any manuscripts in the course of over 1,000 years of church history that initially sounds quite overwhelming.
But we should remember the wisdom of Solomon, who said, “He that is first in his own cause seemeth just; but his neighbor cometh and searcheth him” (Proverbs 18:17).
This leads us to another major problem with the PIA’s analysis of this textual variant at Ephesians 3:9: His analysis (or failure to provide sufficient analysis) of the Greek manuscript evidence regarding this reading.
To begin, can we ask the PIA to list for us the minority of minuscules (acknowledged by Metzger) which include the TR reading, along with their suggested dates in order to verify that none of these appear before the eleventh century? If the PIA cannot list these, does this indicate that he has offered this challenge without first doing a proper analytical study of this variant?
Even if he can substantiate his claim, would he not agree that even late mss. sometimes contain the earliest readings? On this see Greg Lanier’s chapter “Dating Myths, Part Two: How later manuscripts can be better manuscripts” in Myths and Mistakes in New Testament Textual Criticism (110-131).
Again, we are not denying that the TR reading is a late minority reading at this point, but this does not mean ipso facto that it is a “blatant error”?
To continue, we might ask for specific information about the earliest Greek manuscript evidence for Ephesians 3:9. When we look more closely, in fact, we find that the total early Greek manuscript evidence for Ephesians 3:9 is extremely thin.
A quick check of the INTF online Liste for Ephesians 3:9 reveals that there are only 6 extant Greek manuscripts containing this verse that are dated pre-AD 800. See this table:
Early mss. containing Eph 3:9
Mss. estimated dates by century

So, according to the INTF date estimates, there are 0 mss. pre-AD 225, and there is only one ms. pre-AD 300 (p46). The earliest ms. we have of Ephesians is at best c. 250-300 years after the letter was written.
To push a little further regarding the early mss. evidence, I took a look at James R. Royse’s chapter on “The Early Text of Paul (and Hebrews)” in Charles E. Hills & Michael J. Kruger, eds., The Early Text of the New Testament (Oxford University Press, 2012, 2014): 175-203.
Royse notes that there are only 20 extant mss. (19 papyri and 1 majuscule) from the Pauline corpus that are plausibly date before AD 350 (175).
He provides a table of these mss., most of which are highly fragmentary (Table 10, 176-177).
Of these 20 mss. only 3 have any portion of Ephesians:
Mss. with Ephesians
Date according to Royse
c. 200
Eph 4:16-19, 4:32-5:13
Eph 1:11-13, 1:19-21

Regarding p46 Royse describes its textual quality as “free” (Table 10, 177). Later Royse says the text of p46 is “basically Alexandrian, but it often supports readings found in D F G and even the occasional Byzantine reading (such as Eph. 5:9)” (181). Royse suggests “one aspect of a scholarly concern for the text” of p46 is found in its “corrections” (181). He adds: “These number 183, of which possibly 109 are by the scribe, 56 are by the second hand, 14 by the third hand, and 4 by the fourth hand” (181-182). He also notes that it contains 639 “singular readings”, adding, “The overall tendency to omit is clearly evidenced in these numbers” (183). Royse adds that another important aspect of the p46’s copying is “a tendency to harmonize to the content” (183).
So, to sum up, let’s break down the Greek ms. evidence for this variant at Ephesians 3:9 for the first three centuries of early Christianity:
In the first century (up to AD 100): no extant Greek mss. support η κοινωνια, and none support η οικονομια.

In the second century (up to AD 200): no extant Greek mss. support η κοινωνια, and none support η οικονομια.

In the third century (up to AD 300): no extant Greek mss. support η κοινωνια, while one ms. supports η οικονομια.

We readily acknowledge that there is no early Greek ms. support for the TR reading η κοινωνια, but we also recognize that there is almost no early Greek manuscript evidence for the MCT reading of η οικονομια. Yes, η οικονομια is the reading found in the five early uncials and became the Majority reading, but this does not mean that the modern critical method can prove it is the original reading, and, in fact, contemporary text critics would be loathe to say that they can do any such thing, given the evidence.

This illustrates a key point in my lecture presentations at the T & CC which has been completely ignored by the PIA in his responses so far: the fact that we have very little early manuscript evidence for the NT. So little, in fact, that it makes the entire “reconstruction” method suspect. As Wasserman and Gurry state in an illustration I shared in my final lecture at the T & CC, the evidence from the extant NT mss. is more like a “watercolor” than a “topographical map” of the NT and you would not want to rely on it to “find your way out of the forest”!
Aside from the overall shift from modern to postmodern methodology, this acknowledgement of the meager and often fragmentary early Greek NT ms. evidence (including the papyri!) is a major reason that the stated goal of contemporary text critics is not to find the “original autograph” but merely to approximate the “initial text” (Augsgangstext) of the first few centuries.
This makes the PIA’s declaration that the TR reading at Ephesians 3:9 is a “blatant error” all the more inconsistent with the current academic method he supposedly embraces.
Fourth, what about the versional and patristic evidence for the variant at Ephesians 3:9?
Though we have noted that the key evidence should generally be the Greek mss., we should also address the PIA’s charge that the versional and patristic witnesses to this textual variant also serve as a “defeater” for the TR.
Regarding the versional evidence, we should note several key things to keep in mind:
First, the PIA never provides any specific examples from the versions for our comparison and analysis.
Second, the versions were generally produced later and do not provide earliest or direct evidence for the text.
Third, study of the versions also requires more detailed linguistic analysis and comparison. One thing that should be pointed out is that the Greek words η κοινωνια and η οικονομια might have some possible overlap in meaning, so that either word might have been rendered by the same term in the receptor language. Though η κοινωνια is usually rendered as “fellowship” in English, the lexicons remind us that it also has the sense of “association”, “generosity”, or a “gift” given as a “sign of fellowship” (cf. Phil 1:5, etc.). Likewise, the lexicons remind us that the the noun η οικονομια also has the meaning of “stewardship”, as it is used in near context at Ephesians 3:2 where Paul speaks about the οικονομια of the grace of God which has been given to him for the Ephesians.
A thorough study of the versional evidence would require an examination of how each receptor language rendered the Greek terms η κοινωνια and η οικονομια, and whether they generally used two distinct words for each term (as in Latin) or whether the same word was ever used for both terms. If the latter is the case, then it is possible that a version would not, in fact, provide definitive evidence as to which Greek word undergirds the version.
Regarding patristics, we can raise similar concerns. Most importantly, no specific examples are given. How many times do we find references to Ephesians 3:9 in the church fathers? How do we know if the citation was a direct quotation or a paraphrase? Was the father citing from the Greek text or from a translation?
Fifth, why did the Protestant Reformed embrace and affirm the TR of Ephesians 3:9 rather than the Majority reading?
There can be little doubt that the Protestant Reformed knew that “fellowship” was not the Majority reading of the Greek mss., but they consistently recognized this as the true text.
This is one of those texts where the “which TR” argument does not seem to apply in that it is the reading found in the family of printed TR editions.
It is there in the 1516 of Erasmus, the 1550 of Stephanus, and the 1598 of Beza:

Image: Ephesians 3:9 in Erasmus's Greek NT 1516

Image: Ephesians 3:9 in Beza's Greek NT 1598

Interestingly enough, it is also there in the Colinaeus Greek NT of 1534, a text which very often follows readings found in today’s modern critical text, and was the text used by Calvin in his early ministry, before he embraced the TR as his preferred text of the Greek NT.

Image: Ephesians 3:9 in Colinaeus's Greek NT 1534
We might add that the TR reading also clearly departed from the Latin Vulgate which followed the Majority Greek text and read “dispensatio sacramenti” (cf. Erasmus’s Latin above: “communio mysterii”).

On what basis did the Reformed men affirm “fellowship” here as the true reading, over against the Majority Greek ms. tradition? We do not know. It is certainly possible that they had access to Greek mss. which are no longer available to us.
Those who scoff at this notion (like the PIA), should consult Jacob W. Peterson’s recent contribution to Elijah Hixon and Peter J. Gurry, eds. Myths and Mistakes in New Testament Textual Criticism. Peterson’s chapter three is titled “Math Myths: How Many Manuscripts We Have and Why More Isn’t Always Better” (48-69). In a section on the loss of manuscripts, Peterson notes, “Another factor affecting our manuscript count is loss” (54). He notes that even in the INTF Liste 136 mss. are listed as “Besitzer unbekannt” or “owner unknown”, adding, “there are a number of ways this happens, ranging from accidental to illegal” (54). He also observes, “Manuscripts are lost through more natural causes such as fires, floods, and insects” (55). Manuscripts 1257-1259 from a school in Izmir are listed as “burnt” (55). Manuscripts like 241 and 2039 were damaged or destroyed in the firebombing of Dresden in WW2 (55). Peterson adds that “numerous early manuscripts, such as 062, catalogued in Damascus, Syria” are listed as “owner unknown” (55).
Then consider that Peterson is only addressing manuscripts that were once known by modern scholars and appear on the current Liste that have been lost or destroyed. What about all the ones that were never catalogued or photographed in modern times?
The PIA does not seem to acknowledge the fact that the printed editions of the TR may serve as witnesses to mss. that are no longer extant.
In the end, we can only be sure that in the providence of God the reading “the fellowship of the mystery” was that preserved in the TR. It was the Greek text that became the basis for the Protestant translations of Europe that brought the Reformed faith to the masses. It was the text studied, taught, and preached in the Reformation and Post-Reformation eras, and it remains the preferred text of Scripture embraced by countless thousands of faithful churches and Christians today.
So Calvin would write in his commentary on Ephesians 3:9: “The publication of the gospel is called a fellowship, because it is the will of God that his purpose, which had formerly been hidden, shall now be shared by men.” 
It is only in the modern era that “Reformed” men have abandoned the traditional text for the modern reconstructed text. In so doing they have embraced a religious epistemology that abandons stability, continuity, and consistency.
We do not believe, in the end, that it is irrational or irreligious or irresponsible to embrace the traditional Protestant text of the Christian Scriptures, rather than the ever-changing, every-evolving modern critical text based on an empirical method with origins in the Enlightenment (“Enlightenment Text-Onlyism”).

Summary on Ephesians 3:9
  • The TR is an eclectic text and is not based merely on the reading of the earliest or the Majority of the extant Greek manuscripts. Therefore, the fact that texts like Ephesians 3:9, are based on a later minority reading, is not necessarily a “defeater” for the TR position.
  • There is, in fact, very little early Greek manuscript evidence for the Pauline epistles, for the book of Ephesians, and, especially, for Ephesians 3:9.
  • The Confessional Text position rejects the reconstructionist method of modern text criticism, in part, because there is not enough extant Greek manuscript evidence to justify this approach.
  • Reformed pastors and scholars of the Reformation era, based on evidence and reasoning that may no longer be available or discernable to us, providentially affirmed “the fellowship of the mystery” in Ephesians 3:9 as the fitting reading of the received text.
  • The printed editions of the TR may serve as witnesses to Greek mss. of the NT that are no longer available to us.
  • There is no compelling reason to abandon the TR in our times and many convincing reasons as to why it should continue to be affirmed by faithful Christians instead of the Enlightenment text.


A. J. MacDonald, Jr said...

The Complutensian Polyglot has οικονομια at Ephesians 3:9.

Jeffrey T. Riddle said...

Thanks for the comment AJ and for this insight on the Complutension Polyglot. That makes sense given that the Latin Vulgate also follows this reading. It makes it all the more interesting that the Protestants preferred "fellowship" here, in distinction from the RC view. This adds to the list of places where the modern critical text supports the Latin Vulgate (as with the doxology of the Lord's Prayer in Matt 6:13b).


Mark said...

I trust that the following link will be of use to you -

Jeffrey T. Riddle said...

Thanks Mark. That is an excellent book. I did include it in the annotated bibliography I shared in 2011:

Nice that it is available for free via Chapel Library. Great resource!

Elijah Hixson said...

Thanks for this. It's good to see someone is reading Myths and Mistakes (also correction: it's Greg Lanier, not David). This is only a response to one part. You say that "Metzger, unsurprisingly, dismisses the Greek ms. support for the traditional reading as confirmed only by 'a scattering of late minuscules'". I'm not sure what Metzger meant by that, and honestly, I can't verify it. I'm not at my desk at the moment, so I can only check what I can via electronic access, but the only reference I can find to anything other than a marginal reading in 69 is a vague note in Tischendorf. Tregelles and Tischendorf both mention a marginal reading in 69, but the hand is late and matches a series of notes in the manuscript that are almost certainly copied from a printed TR, so it's not a Greek minuscule witness to the antiquity of the TR reading at all, just a copy of a TR. Appealing to it would be like appealing to 0319 to prove that readings in D06 were ancient—0319 is a copy of D06, so such a claim would be nonsense. I didn't see any minuscule support for the TR reading listed in von Soden, either. Are there *any* manuscripts that support the TR here? There might be, but I couldn't find them named explicitly in Tischendorf, Tregelles, von Soden or the CNTTS apparatus, and only Tischendorf gives any indication that there might be some out there. Appealing to Metzger's comment without verifying it or giving specifics (when the specifics are not given in Tisch. Treg. Sod. or CNTTS) seems like appealing to evidence that you don't have. I found a claim online ( that manuscripts 31 and 57 have it, but 31 is a Gospels codex that doesn't even have Ephesians, and while 57 does have Ephesians, it clearly has οικονομια (page view 408, end of the 5th line from the bottom: Why not just admit head-on that the evidence is decidedly against the TR here but say that's ok because it's not an evidence-based argument? (And a slightly unrelated question, but why trust Metzger implicitly when he says something that could be taken to support the TR but have reservations about things he says when they don't support the TR?)

CC said...

Thanks for this WM. I was able to find a quote from Tertullian in Against Marcion book 5 chapter 18 where he says, "To himself, the apostle says, last of all was the grace given of making all men see what is the dispensation of the mystery which from the ages has been hidden in God who created all things." I'm not entirely convinced of κοινωνία, but I do not believe the meaning is greatly affected either way.

Jeffrey T. Riddle said...

Elijah, thanks for your feedback. “Greg” Lanier is corrected. I think I had a childhood friend named David L. Smiles.

As for Metzger’s note, I’m not sure what he meant by it either. It is interesting that the second edition is different from the first. The first edition entry here begins, “The Textus Receptus, in company with merely 31mg and a few other minuscules replaces…” I also noted that 31 only has the Gospels, so this was an error in the first edition corrected in the second, but he still mentions “a few other minuscules” (?). Metzger did love to revise and produce corrected editions! I’m not surprised that some of the other scholarly sources you mentioned did not provide much info on the TR reading at Eph 3:9, since none of them were particularly interested in finding evidence in support of the TR. The INTF Liste has over 200 mss. with Ephesians 3:9 if I recall correctly (with only six pre-800, as noted). I suppose one would have to examine each of them to figure this out.

Regarding appeal to Metzger without validation/verification, that was exactly the point I was making about the PIA. This conversation began with his sweeping statement about the TR of Eph 3:9 having no support in the first 1,000 years. I think the ball is in his court to substantiate this claim. My post was simply a response to his claims. Maybe he will be able to answer this question for us (or at least reveal to us the source for his definitive pronouncement). My suggestion that the TR reading is supported by at least some Greek ms. support was based not only on Metzger’s mention (who was, after all, a pretty well-respected scholar—even if I do not agree with his method and conclusion), but also on the fact that it appears in the printed editions of the TR.

BTW, I just checked Wilbur Pickering’s The Greek Text of the New Testament According to Family 35 and he says this TR reading at Eph 3:9 is supported by 10% of the mss. Again, I supposed we’d have to examine the mss. in order to verify this.

I think I made it pretty clear here that TR supporters acknowledge that the TR of Eph 3:9 is not supported by early or Majority text evidence, but I don’t think I would say that this defense is “not an evidence-based argument” all together. How does one make any argument with absolutely no evidence/information?

Jeffrey T. Riddle said...

CC, thanks for this Tertullian quote. I am assuming that behind the English "dispensation" is the Latin "dispensatio" here, which would presumably support the MCT. It is an interesting question as to how the meaning might be affected here by this variant. As noted above, perhaps it is significant that the Protestant TR departs from the Latin Vulgate here. Perhaps there is an underlying polemic regarding sacramental theology.

Elijah Hixson said...

Jeff, thanks for your reply. My suspicion—this is reasoned speculation, and I admit that—is that "31mg" in Metzger is either a misreading or a misprinting of "37mg", which is how both Tregelles and Tischendorf referred to the marginal reading in GA 69 (before manuscript numbers were standardised). I wouldn't dismiss the testimony of Tregelles, Tischendorf, von Soden and the CNTTS apparatus though simply because they aren't interesting in finding support for the TR. Tischendorf does mention that a few others (al pauci) have the TR reading, but he doesn't list which ones. The others don't appear to know of any other manuscripts that have the TR reading there. Pickering does report "[10%]", but it's important to note how he gets that number (this is from the final footnote to Ephesians; in his preface he directs the reader to see the last footnote in each book for information on percentages given in the apparatus for that book) :

"In the statements of evidence I have included the percentage of manuscript attestation for each variant, within either ( ) or [ ]. I have used ( ) for the evidence taken from [Text und Textwert], which I take to be reasonably precise. For the variant sets that are not covered there I had to revert to von Soden and the apparatus of N-A27, supplementing from other sources where possible (Scrivener and Tischendorf)—the percentages offered, I have used [ ] for these, are extrapolations based on a comparison of these sources."


Elijah Hixson said...


For Pickering's 10%, it isn't coming from TuT, it isn't in von Soden or NA27, and I couldn't find anything on it in any of Scrivener's works that I just checked. That leaves Tischendorf, according to the sources Pickering claimed to have used. It seems to me that Pickering's "[10%]" is nothing more than a guess based solely off Tischendorf's comment "al pauci". So, again, is there *any* manuscript that independently supports the TR here (i.e. isn't copied from a printed TR)? Maybe there is, but maybe there isn't. Pickering admits his number is just an "extrapolation", and Tischendorf is the only source he claims to have used who said anything about it. My guess on Metzger is that, given that I can't find anything other than Tischendorf's comment is that Metzger's comment is probably based on Tischendorf's comment. Still, appealing to the existence of the reading in the TR as evidence that some manuscripts probably have it still falls a little flat. If you're making an evidence-based argument (and that is one), you must consider all the evidence. Given the presence of the non-TR reading in every manuscript we can name (and, if Metzger is right, "all known versions and patristic quotations"—I can't verify it, but I haven't come across anyone citing a version or a father who supports the TR here either), and given the similarities of κοινωνια and οικονομια, why not entertain the possibility that Erasmus, for whatever reason, made a mistake here, and later TRs followed him the same way Pickering (and possibly Metzger) followed Tischendorf?

I think the thing that illustrates why I have serious misgivings about many TR claims is that one argument commonly used against me is that I'm saying that the church didn't have God's Word for 1800 years, but instances like this show the same thing. If the TR reading is correct, then God in his providence has left us abundant evidence that almost every single Bible (and until someone can figure out which specific manuscripts Tischendorf was referring to, *every* single Bible) that survived for the first 1500 years of the Church has lost God's Word here. Now, if God is not the author of confusion, and if it is necessary for the Church to have all God's Words (and not merely to have sufficient access), then the relevant question for me is not "Why did God not preserve evidence that the Church had his Word at Eph. 3:9 until the 1500s?" but rather "Why did God preserve *so much* evidence that the Church did *not* have his Word at Eph. 3:9 until the 1500s, if the TR reading is correct?"

Jeffrey T. Riddle said...

Elijah, thanks for your diligent work. This includes sharing your “suspicion” and “reasoned speculation” on Metzger. You may be right about the 31/37 confusion, but it is interesting that he does not withdraw his assertion that the TR reading is in “a scattering of late minuscules" in the second edition of his Textual Commentary.

Of course, I would not dismiss the research of Tregelles, etc. (I often say that we should “plunder the Egyptians”—that is, benefit from the scholarship of those who do not agree with us), but I think it would also be naïve to overlook the animus against the TR and the desire to topple it, especially in the 19th century, and how this might have had an impact on their research. We all have presuppositions right?

Thanks for the insights on Pickering’s “[10%].” I guess in his case since he is still living (?) and could be directly consulted to see why he gave this estimation and what “other sources” he might have consulted. If he was only dependent on Tischendorf’s “al pauci” one might think he would have given a lower estimation (maybe 1% rather than 10%) (?).

Your speculation on Metzger’s dependence on Tischendorf is interesting, but I’m sure you’d agree it remains a speculation, and it is also possible that Metzger knew of minuscules that included the TR reading (as he states). Whether or not those can now be located is another question. As you also admit, you cannot “verify” Metzer’s assertion that the TR reading is absent from all versions and patristic witnesses.

You suggest that “appealing to the existence of the reading in the TR as evidence that some manuscripts probably have it … falls a little flat.” On the other hand, I’m sure you’ll agree it is also at least possible. In Beza’s annotations, for example, he justified his TR reading at Rev. 16:5 by saying he had found it in “an ancient hand-written codex.”

You also suggest the possibility that “fellowship” entered into the TR through an error by Erasmus confusing the two words. I’m sure you’d agree that it is also at least possible that this confusion happened at a very early stage in transmission. Against the notion of it simply being an Erasmus confusion would be the fact that Erasmus and the later Protestant scholars would have known that “plan/administration” was in the Complutension edition and other mss., but they persisted in following “fellowship” in their printed editions of the TR.


Jeffrey T. Riddle said...

To Elijah 2/2

Thanks for sharing your “misgivings” about the TR in a charitable manner. Given your highly specialized training in the academy and investment in the reconstruction method I’m hardly surprised you do not find the TR position convincing.

I have not been privy to discussions you have had with other TR advocates, so I’m not sure what was said on either side.

I would take exception to the way you framed some things in this last paragraph. For example, you wonder at why if “the TR reading is correct….then almost every single Bible… that survived for the first 1500 years of the Church has lost God's Word here.” First, I think you’d agree that the adjective “extant” should be added to “every single Bible.” Second, as I also point out in the blog, there is actually very little early evidence in general for Ephesians 3:9, only six mss. pre- AD 800, and, as you’ve pointed out, the evidence is far from clear as to the later minuscule evidence.

This whole discussion reinforces for me the weaknesses of the reconstruction method. There are limits to what any clever person (or even group of clever persons) can do to comprehend completely the empirical evidence. In this case, for example, if one wanted to examine the evidence without consulting secondary sources, one would need to examine all 200+ mss. that contain Eph 3:9 on the Liste, and then track down every version and patristic source that contains it, and then one would only have access to the extant material and would be in the dark about what is lost.

And this is just one minor variant (regarding one word). There is also a question about what significance this variant has with regard to the verse’s meaning (as already mentioned here in these comments, especially a possible distinction between a Protestant and RC theological difference here).

I’m all for historical research on the text, but I think believers are better off to stick with the text of the Reformation.

The view you present here on the providential preservation of Scripture is not consistent with my understanding of its articulation in WCF/SD/2LBCF 1:8. I also noted this when I read the statement in the introduction to your new book: “Nor do we think that God has preserved the original text of the New Testament equally well at every point in history or at every place in the world” (20).

Anyhow, thanks for the engagement. I’d be curious to know your views on more significant variants like Mark 16:9-20 and John 7:53—8:11. Or, what your view is on the NA28 conjecture at 2 Peter 3:10, but I do not think that this comment section is the best place to carry on that conversation. I look forward to reading a positive presentation of your views when you present them on your blog or in scholarly articles and books.

Blessings, JTR

CC said...

Ah, I found a manuscript that reads κοινωνία: 2817 from the 11th century (according the Scrivener and the INTF). It's in the INTF's NTVMR, page ID 5190. Apparently this is where Erasmus got the reading from.

Jeffrey T. Riddle said...

Thanks CC. I'll check it out. Curious about its location.


Jeffrey T. Riddle said...

Yes, I see it is in Basel. Makes sense.

Elijah Hixson said...

CC, well done; we have one manuscript! It is a catena manuscript (not all of these are given GA numbers because at some points it is unclear as to whether they should count as manuscripts; some of them, like 2817, are more like a Study Bible, but others are more like a single-volume commentary on a book of Scripture that contains the text the comments are about). But you're right; it's clearly there, and it is also given in the lemma for the note on the same page. I checked Theodora Panella's thesis on the Pseudo-Oecumenian catena on Galatians, and—admittedly this is Galatians and assumes no block mixture, she places 2817 in a group with 2007 (12th cent.) and 1914 (11th cent.). If she's right and assuming I've read her correctly, then these three manuscripts should be very closely related. The format of 1914 is almost identical to 2817. However, I checked them both, and both of these manuscripts have οικονομια—page view 4540 (f. 224v) in 2007 and page view 3100 in 1914. If these three manuscripts are a closely-related group, then common sense would dictate that 2817 is the one with the error rather than that both 2007 and 1914 are in error—especially at this point in history when copyists made fewer mistakes than earlier.

Jeff, thank you for your comments, but I apologise if I was unclear about Erasmus. I did not say that he confused the two words, but "that Erasmus, for whatever reason, made a mistake here". True that allows for him confusing the two words, but it also allows for him following the wrong manuscript and printing the wrong word, which I think, with CC's identification of 2817, is what happened.

I agree though that there are limits to what a single person can do, which is why we should have more people collating Byzantine manuscripts! I am glad though that God's promises to preserve his word sufficiently for us are stronger than the limits of any one person or group of people, such that we can't mess it up in a way that thwarts God's promises or purposes.

Matthew M. Rose said...

I would hope/think that if Metzger and the U.B.S. committee figured out that he/they mistook (somehow) 37mg in Tischendorf for 31mg they would have placed 69mg(i.e.37mg) in the 2nd edition of the Textual Commentary--especially considering that ms.69 is more "important". Has anyone checked ms.31?

Perhaps the verse in question (Eph.3:9) is placed within the margin of a pertinent location within the Gospels (somewhere) as a topical reference or annotation (?).

If not, Dr. Hixson's suggestion is most assuredly correct. -MMR

A. J. MacDonald, Jr said...

Pastor Riddle, I mention the Complutensian Polyglot (CP) has οικονομια at Ephesians 3:9 because the CP was the first (1514) printed edition of the Greek New Testament. In fact, I would say the CP was the pinnacle of Renaissance biblical scholarship. Erasmus' edition was only popular because it was small and cheap. The CP was far superior. I bring this up because James White probably don't realize that a 16th century printed edition has οικονομια at Ephesians 3:9. White, like most people, completely ignore the CP and go straight to Eramsus. I'm disappointed that you make it seem as though the scholars who produced the CP were following the Vulgate at Ephesians 3:9, as though the CP were a back translation of the Vulgate. My point was that the editors of the CP had access to Greek manuscripts with οικονομια at Ephesians 3:9 and included this reading in their 1514 edition of the Greek New Testament.

Jeffrey T. Riddle said...

Just getting around to moderating comments for this post.

Yes, we have a manuscript! So, there is at least some evidence, no matter how meager, for the TR reading at Eph 3:9 in the Greek ms. tradition.

Sounds like this sent you scrambling for an explanation. Smiles. Thanks for pointing us to Panella’s thesis. Very interesting. Of course, in her conclusion she concedes that much of her research remains “preliminary” (208). A lot seems speculative. What is the relationship between 2817 and 2007 and 1914? I’m not so sure I agree that “if” these three are related, “then common sense would dictate that 2817 is the one with error rather than both 2007 and 1914 are in error…” If the three are closely related (and we are dependent here on a secondary source for this suggestion) and 2817 and 1914 are nearly identical (and both are dated c. 11th century), then is it not possible that 2007 (12th century) simply followed 1914 or the Majority tradition, rather than 2817? How is this necessarily “common sense” proof that 2817 is “the one with error”? It seems just to be proof that 2817 has a different reading at this point than 1914 and 2007.

Not sure I understand your “apology.” I think I understood you correctly when you said that “Erasmus, for whatever reason, made a mistake here.” One possibility is (as I noted) that he might have confused the two words, and another is that he followed a supposedly flawed ms. like 2817, or some of the other minuscules (Metzger) which contain the TR reading. But can either of these supposed errors be definitively proven empirically? On the other hand, can it be definitively disproven by empirical evidence that the TR does preserve the true reading (though it is a minority)?

In the end, I just don’t think that having more people collate Byzantine mss. is going to get us any closer to a definitive original text of the NT using empirical methodology. Hasn’t the academy also pretty much acknowledged that this is no longer the goal of text criticism?

It seems to me that the current evangelical view of preservation via modern empirical reconstruction first promoted by Warfield is a novelty. It is not consistent with the providential preservation view of the Reformation men and the Reformed orthodox. Can you cite anyone pre-Enlightenment who held this view of preservation?


Jeffrey T. Riddle said...


Yes, interesting that if the error in the first edition was caught by Metzger and if he had meant to refer, instead, to 37 (69mg) that this correction was not made in the second edition. I have not checked 31, but I don't think anyone is disputing the error in the first edition. The question, it seems to me, is what were the other "scattering of later minuscules' he mentions in the second edition. We've found one, 2817, but we don't know if this was one to which Metzger was making reference.


Jeffrey T. Riddle said...


I certainly did not mean to suggest that the CP was not important. It often supports Protestant TR readings. I also did not mean to suggest that it "back-translated" here in Eph 3:9 from the Vulgate (and I certainly did not claim that). Clearly, here it follows the Majority text. My point was that as a RC edition it would have been perceived as advantageous that it here follows the Latin Vulgate.

I'm curious as to how your views on text are evolving. I know you have written some on "ecclesiological" affirmation of the text.


Jeffrey T. Riddle said...


Is your Greek good enough to provide a translation of the paragraph of commentary in 2817 just below the τις η κοινωνια του μυστηριου citation?

I'd be curious to know what it says.


CC said...

Take another look at P46 in the NTVMR. It seems to me that it only reads κονομια. A case could be made that the exemplar had κοινονια (ο rather than ω) and the ι got squished into the Ν and the second Ν became μ. The transcription includes the οι in red brackets, but those letters aren't actually on the page. Of course I'm not entirely convinced that κοινωνια is correct, and I think there are other issues in certain TR editions such as ημων rather than αυτων in Revelation 11:8 (only found in 2814 as opposed to all other versions, patristic citations, and all other Greek mss including the 2814's sister manuscript 2186). Other than in a few minor points, I will defend the TR in everything.

Jeffrey T. Riddle said...

CC, you strike again. Very interesting.

Thanks for this insight.


Elijah Hixson said...

It’s odd because the same note shows up in manuscripts that have οικονομια as well, and relatively unchanged. It looks like 2817 has an error in the note (that is also in 2007 but not in 1914 or one other catena manuscript I checked). I say error because it’s the omission of a couple of words, and to fix it, the scribe put a masculine ending on a feminine word. I suppose that would be intelligible, but it’s not standard (this gives me even more reason to suspect that 2817 is not preserving an early text). I put in brackets the bits that are omitted in 2817, and there are a couple of other places where I am unsure of the endings (i.e. 2817 has αυτον at one point, but other manuscripts have abbreviations that are accented as αυτων or αυτου).

A rough translation: “[It is] that: for on the one hand the Gentiles are called. Some had recognized through his Spirit. But on the other hand that concerning [επι could mean a lot of things, so take ‘concerning’ with a grain of salt in absence of better context] such as these so that they became his first-fruits, they did not recognize the Word of God (“God” here in the accusative, not Genitive as I would expect, but punctuation seems to follow “word” not “God” in more than one manuscript). Therefore, this (he says) is the mystery that is hidden.”

I think the gist of the comment is that during OT times, some Gentiles came to faith through the Holy Spirit, but others did not recognize the Hebrew Scriptures as God’s Word (or possibly the proclamation of the prophets), and this is how the mystery is hidden. I’m no expert in Byzantine theology though.

Interestingly, this is the same note that’s there in the manuscripts I checked regardless of whether the biblical text has κοινωνια or οικονομια.

On P46: My suspicion is that the letters οι may just be rubbed off—there is a lot of that in that area of the papyrus, and note the line endings—they are consistently further to the left than on this line. The lines above and below this line end 1–2 letters to the left, and on up the page lines are even longer. For this line to end on η would be an anomaly in itself. If it did omit οι (rather than that the papyrus is just currently too abraded there), it was almost certainly just an omission of οι. It’d be a one-syllable omission at a line break—not at all unusual. No need to come up with complicated and multi-layered explanations to explain what’s happening here.

Elijah Hixson said...

I wouldn’t say “scrambling for an explanation.” As soon as I saw the number, I remembered this is one of the ones Erasmus used, which is the obvious first place to look. Still, I like to check things, so I checked it and noticed it’s a catena manuscript. I get the impression that a lot of people who don’t work explicitly with them don’t know a lot about them. I certainly didn’t until a year and a half ago when I started looking into a group of catena manuscripts of Luke. A lot of them are closely related—nearly half of the examples of direct-copy relationships that Alan Taylor Farnes lists in his book are pairs of catena manuscripts, so these are often great places to check scribal habits since we have a number of examples where we know what the scribe wrote and what he was looking at when he did. It makes sense for many to be very closely related, because it’s easier to copy a commentary from a a commentary (and maybe to supplement or trim what comments are included) than to compile anew every time you want a commentary manuscript.

I’m happy to stick by my position that 2817 contains an error. Of course we cannot *prove* that 2817 was the one of its group of three that erred, but if these three do make a group, the simpler explanation is that one of them erred than two of them. A step back from that would be to look at a lot of catena mss (such as taking Panella’s list on pp. 55–57 and going through them at Eph. 3:9). This would give us the text of more manuscripts that are probably more like cousins than brothers—the extended family. If 2817 continues to be alone in this broader group, is it reasonable to claim that it alone preserves the original reading against its closest relatives AND against the broader tradition at this one point?

As to pre-Enlightenment preservation-by-modern-reconstruction—I don’t think that’s an accurate description, but that’s a conversation for another time. I’d argue (and my ETS paper next week does argue) that it has always been the position of Christians before the printing press that the text we ‘receive’ is authoritative and sufficient to the extent that it is what the prophets, evangelists and apostles wrote—and that we do not always have certain and perfect access to what they wrote. We see that in comments by Augustine, Jerome, Erasmus, and the copyists themselves who made the manuscripts (most of my time will be spent with Δ037 and 1424). Augustine will suffice though—In a letter to Jerome written probably around A.D. 405, Augustine wrote that he has “respect and honour only those books of the Scriptures now referred to as canonical. I firmly believe that none of the authors of these books has erred in writing . . . .” Immediately next, Augustine continues: “…and if I should find fault with anything in them which appears to conflict with the truth, I am sure that the reason must be that there is some textual error or that the translator did not follow what was said or that I do not understand it properly” (Ep. 82). In De Doctrina Christiana (3.1), he indicates that textual criticism is necessary *before* exegesis: “The student who fears God [and] earnestly seeks his will in the holy scriptures…with the assistance of reliable texts derived from the manuscripts, with careful attention to the need for emendation, he should now approach the task of analyzing and resolving the ambiguities of the Scriptures.” Over 1000 years later, Erasmus appealed to the Vulgate to justify his text-critical work: “When Damasus asked Jerome to undertake the business of correcting the New Testament from the Greek sources, certainly the church already had a text it was reading and perhaps had had it for some centuries. If it was correct, what need was there for Jerome to emend it? If it was corrupt, then it is clear that the church was temporarily using a text that needed to be improved” (from Contra morosos).

Jeffrey T. Riddle said...

EH, thanks for the translation. I was thinking it might provide some insights into a theological rationale for either "fellowship" or "plan/administration" but apparently, it does not.

Jeffrey T. Riddle said...

EH, thanks again for your diligence. For me this illustrates the limits of the reconstruction method. You think 2817 is in error but concede this cannot be definitively proved. It must be a possibility, or, as you think, a high probability. Still, it's not certain. Of course, this is a minor variant. What about more significant ones with weightier significance for doctrine and practice? The weakness of the reconstruction method is epistemology.

What if this was the ms. that Erasmus used? I guess we can't assume that. Whether he had this one or another or others, his reading at Eph 3:9 was "fellowship" and other Protestant scholars affirmed this as the reading and maintained it, even though they likely knew that it was not the Majority and the respected Complutension did not follow their reading. At this point I want to move beyond naturalistic attempts to understand this reading to a providential approach.

Yes, the discussion on providence may have to wait till another day. I'd be happy to get a copy of the ETS paper or outline if you're sharing it. Interesting quotes, but I still think it would be anachronistic to assume Jerome, Augustine, etc. were thinking along the lines of modern reconstructionist text criticism. This was certainly not the way the Puritans and Westminster divines were thinking about preservation as is clear from WCF 1:8 and as R. Muller, I think, definitively shows in PRRD, vol. 2. I also find it interesting that men like Calvin (who was a "student" of the church fathers, so A. N. S. Lane) and John Owen (who was also steep in them) did not follow this line of thinking. As R. Brash has argued they saw a "practical univocity" between the autographs and the apographs (as reflected in the printed TR).


MarichD said...

Thank you Pastor Jeff and thank you for the link.

Appreciating your ministry via sermon audio and WM too!


WritetheVision said...

My thoughts on the upcoming debate over the reading in Ephesians 3:9.

Regarding "the fellowship of the mystery" (Ephesians 3:9) there is no need for any other evidence other than the context and proper interpretation of scripture, precept upon precept. “Fellowship” is well defined in the context of the epistle itself as in Ephesians 3:6, "That the Gentiles should be fellow-heirs, and of the same body...," (a fellowship), whereas in contrast, in Ephesians 3:2 "dispensation" is clearly more of a time frame. Then yet again, the context in Ephesians 1:10 reads, "That in the dispensation of the fulness of times he might gather together in one all things in Christ, both which are in heaven, and which are on earth; even in him"

Further the mystery itself in context is defined clearly by Paul as "This is a great mystery: but I speak concerning Christ and the church" Ephesians 5:32 (KJV). The Jews, the Gentiles, and all men are now in 'fellowship' with Christ, "the fellowship of the mystery" revealed in God's timing; in that after Christ ascended into heaven; we have entering into a new dispensation, the time “of the dispensation of the grace of God..." Ephesians 3:2 (KJV).

The debate over "fellowship" based on TR's κοινωνία as opposed to "dispensation" based on Maj, Byz, NA/UBS's οἰκονομία gets us off the train tracks of what the context of the whole of Ephesians is clearly telling us. In addition to that, where is the consistency such as “all men” (πάντας) (Eph 3:9) in the very same verse, with “all” omitted in the NASB, even with TR, Maj, Byz, Sinaiticus, and P46 support? Where is the consistency in the science there? The #1 FIRST RULE if we are to examine any portion of scripture, whether by textual means or any other means, is “precept must be upon precept, precept upon precept, line upon line, line upon line, here a little and there a little” (Isaiah 28:9-13). Without following God’s own guidelines for examination of the scripture, we may find ourselves drifting off into left field “doting about questions and strifes of words” (I Timothy 6:4). The scripture, if “rightly divided” (II Tim 2:15; Isa 28:9-13), in Eph 3:9 does not need any help or any other evidence as we see the consistency and the continuity in the ‘context’ of the KJV’s ‘fellowship’ reading. When using the proper formula of context and a proper interpretation of scripture, many times the critical readings defeat themselves by violating the first rule in giving a passage or a word its very own “private interpretation” by ignoring “Knowing this FIRST” (II Peter 1:20). If we follow the FIRST rule of knowing what God said in a passage such as Ephesian 3:9, we must FIRST consult the entirely of SCRIPTURE and the readings context. After all, is not that what the debate is about, to know what God said here in Ephesians 3:9? And let us not ignore the omission of “Jesus Christ” in this very same verse, also having TR AND Majority support, for without Jesus Christ there would be no ‘fellowship of the mystery.’

The KJV reading of ‘fellowship’ follows this FIRST rule and is consistent with the context of the epistle of Ephesians and proper interpretation of scripture. The critical reading of Ephesians 3:9 is not consistent by claiming ‘fellowship’ should be ‘dispensation,’ or administration,’ because it is not a majority reading, yet; “all men” (omitted NASB) and “Jesus Christ” have both TR and majority support, yet they remain omitted. This critical text method lacks continuity and consistency; whereas the epistle of Ephesians has its very own flow and continuance! The reading for Eph 3:9 has already been resolved by its other readings within the same epistle.

That my friends is “the key of knowledge” (Luke 11:52) for those who want to get off the milk and dig into the meat of the Word, we must know what comes FIRST, in having a God given “knowledge….and to understand doctrine” (Isaiah 28:9-13). KJV.