Category Archives: Biblical Studies

By the Washing of Regeneration

But when the goodness and loving kindness of God our Savior appeared, he saved us, not because of works done by us in righteousness, but according to his own mercy, by the washing of regeneration and renewal of the Holy Spirit, whom he poured out on us richly through Jesus Christ our Savior, so that being justified by his grace we might become heirs according to the hope of eternal life. (Titus 3:4-7, ESV)

This coming Sunday, I have been asked to provide pulpit supply for my pastor who is taking some time to visit his brand new, and very first, grandson. The text I selected was Titus 3:1-7, and in particular, a section in the middle jumped out at me. One phrase caught my attention.

by the washing of regeneration

Paul, writing to one of his successors Titus, is concluding the letter with some instructions. Immediately before this passage, he exhorted the people to obey the governing authorities, to be ready to do good works, and to be charitable all people. (1-2)

Then he grounds his command in the fact that we were once sinners who also needed God to show us kindness and charity. (3)

That brings us to our passage. While we were still in the state described in verse 3, the loving-kindness of the Father appeared. That loving-kindness was Jesus. (John 3:16, Rom 5:8) Contrary to Roman Catholic thought, the Father saved us “not because of works done by us in righteousness.” (3:5a) Rather he saved us “according to his own mercy. (3:5b)

In the second half of verse 5, we come to the contentious phrase which is the subject of our inquiry today.

by the washing of regeneration

This passage has been interpreted variously throughout Church history. Some see it as an obvious reference to the rite of baptism. They use this passage to demonstrate that the washing (baptism) is effectual to bring about regeneration. They read the phrase as though it said something to the effect of “the regenerating washing.” This position the prevailing view among Roman Catholics, Lutherans, and some Anglicans. However, even among early church commentators who affirmed Baptismal Regeneration, this interpretation was not universal.

Strange! How were we drowned in wickedness, so that we could not be purified, but needed a new birth? For this is implied by “Regeneration.” For as when a house is in a ruinous state no one places props under it, nor makes any addition to the old building, but pulls it down to its foundations, and rebuilds it anew; so in our case, God has not repaired us, but has made us anew. For this is “the renewing of the Holy Ghost.” He has made us new men. How? “By His Spirit” (John Chrysostom, Nicene and Post-Nicene Fathers, ed. Philip Schaff, trans. James Tweed and Philip Schaff, vol. 13 (New York: Christian Literature Company, 1889), 538.)

It is interesting to note here, that Chrysostom literally says “we were baptized (βεβαπτισμένοι) in wickedness.” It seems like if he was going to make the point that baptism regenerates us, that this would be a perfect intro. We once were baptized in wickedness, and we are now baptized in righteousness. However, he does not do so. Rather, he points out here that we cannot be purified, but rather we must be entirely rebuilt. That certainly does not sound like an infusion of grace that transforms us such that we are inherently just. I digress.

The Baptismal Regeneration reading is not justified. Rather, we should read the passage such that regeneration itself is the washing. Grammatically, the phrase λουτροῦ παλιγγενεσίας simply does not bear out the adjectival reading above. Rather, the second noun in the construction is better seen as the means or agent of the first. Thus it is better understood as something closer to “the regeneration which washes” or “the washing which comes about because of regeneration.” We see this clearly when we observe the following phrase which is joined with the coordinating conjunction καὶ.

and the renewal of the Holy Spirit

The phrase ἀνακαινώσεως πνεύματος ἁγίου is parallel to λουτροῦ παλιγγενεσίας and thus we may draw a reasonable conclusion that the construction is also parallel. It is evident that the phrase “Holy Spirit” does not describe the word “renewal.” If we take the Baptismal Regeneration view above, and λουτροῦ παλιγγενεσίας means “regenerative washing,” then this phrase here must mean something like “Holy Spirit inducing renewal.” Now, while it is true that those holding to Baptismal Regeneration would agree that baptism indeed brings about the indwelling of the Holy Spirit, I am not aware of a single commentator here that uses this passage to support that. Rather they are of one voice in recognizing that this is telling us that the Holy Spirit brings about renewal. However, the same construct used immediately prior, says the Baptismal Regeneration Advocate, says that washing brings about regeneration. Why the discrepancy?

Instead, we ought to read this passage as though identical constructions function identically. In fact, the two constructions are referring to the same thing. The regeneration which washes is, in fact, an act of the Holy Spirit who renews. The washing described in the first phrase is the renewal described in the second.

If “through” (dia) were used before “renewal,” thus rendering “through the washing of rebirth and through renewal of the Holy Spirit,” it would describe two events instead of one. Simply stated, the text indicates that “washing” is an activity of the Holy Spirit and that this washing involves “rebirth” (palingenesias) and “renewal” (anakainoseos). (Thomas Lea and Hayne Griffin, New American Commentary, vol. 34 (Nashville: B&H, 1992), 323.)

What Paul here is describing is the regeneration and conversion of a Christian. He goes on to say that the purpose of the washing of regeneration and the renewal of the Holy Spirit by which we were saved is “so that being justified by his grace we might become heirs according to the hope of eternal life.” (7) He also notes that the Holy Spirit who brings about this regeneration and renewal is “poured out on us richly through Jesus Christ our Savior.” (6) This act is an act of the triune God from start to finish.

I would be remiss if I failed to note that this is not precisely the same as what is advocated by many in the Reformed tradition. Both Matthew Henry and John Calvin associate the phrase “washing of regeneration” with baptism. Henry does so more strongly than Calvin, but it is important to note that both draw the conclusion that the combination of phrases “washing of regeneration” and “renewal of the Holy Spirit” deny Baptismal Regeneration. What they are saying is not all that far off from what I’m saying. Rather than understand this passage as advocating Baptismal Regeneration, instead what we see is that the sign (baptism, here called the washing of regeneration) is here directly associated with that which is signified (regeneration itself here called the renewal of the Holy Spirit). While I disagree with them that baptism is in view here, I fully affirm the theology they are putting forward.

SBLGNT, Robinson/Pierpont, and the Majority Text

One thing that I commonly hear repeated in discussions between Textus Receptus and Critical Text advocates is the idea that modern text critics always, or nearly always, favor the readings found in the oldest manuscripts as opposed to the readings found in the majority text.

Now, it’s not an exact study, but this generally means that the Robinson/Pierpont text represents the majority text while Critical Texts represent a different reading.

I contacted Michael Holmes, who is arguably an heir of Bruce Metzger, and the publisher of the recent Society of Biblical Literature Greek New Testament. He is also a friend and was my Greek professor in college.

The SBLGNT apparatus is slightly different from other critical apparatuses in that rather than explain which manuscripts contain which readings, it instead shows which major Critical Texts contain which readings. The four major Critical Texts he compares are Westcott-Hort, Tregelles, NA28/27, and Robinson/Pierpont. Generally speaking the RP favors the majority/Byzantine text type. The WH, Treg, and NA28/27 tend to favor the Alexandrian text type. For this reason the RP tends to be closer to the TR, while the other three usually depart from the TR.

Back to the original claim. The original claim is that modern Text Critics always, or nearly always, favor the oldest texts. This is the case, so says the TR advocate, even when there is a vast majority of readings in the manuscript traditions which differ.

However, as is always the case with sweeping statements, a single exception disproves a universal claim. While it is true that the careful TR advocate will not claim this as a universal fact, I have far too often interacted with TR advocates who are not quite so careful.

Back to Dr Holmes and the SBLGNT. Because of his unique apparatus, it is quite easy to see instances where Holmes chooses the same reading as the Robinson/Pierpont do, as opposed to WH, Treg, and NA28/27. There are 56 such instances.

Now, I know that this doesn’t prove conclusively that Holmes ever favors the Majority/Byzantine text type over the Alexandrian (which presumably represents the earliest manuscripts in most cases) it does serve as a handy response to the sweeping claims of the TR advocate. If Holmes and RP chose the Byzantine, then Holmes —IE a modern text critic in the tradition of Metzger— has chosen the Byzantine text type over the Alexandrian. If Holmes and RP chose the Alexandrian text type, then WH, Treg, and NA28/27 —IE modern text critics in the tradition of Metzger— have chosen the Byzantine text type over the Alexandrian. In either case it disproves the sweeping claim.

Although most of the instances consist of a single word, of particular importance is Romans 16:24 where WH, Treg, and NA28/27 omit the verse entirely while Holmes and RP include it. According to Bruce Metzger.

The earliest and best witnesses omit v. 24.[1]

So here we have an explicit instance of a modern Text Critic (Holmes) siding with the Majority/Byzantine text as opposed to simply siding with the earliest witnesses.

Beyond that, there are also a number of cases where one of the three critical texts agree with RP against the other two, and 46 instances where Holmes prefers a reading that all four other editions reject.

This conclusively disproves the claim that modern Text Critics always opt for the oldest reading, even in the face of overwhelming numbers of manuscripts in the Byzantine text type.


  1. Roger Omanson, A Textual Guide to the Greek New Testament (Stuttgart: Deutsche Bibelgesellschaft, 2006), 324.

Text Criticism and Matthew 27:34

I recently had a discussion in an online group with a Textus Receptus (TR) advocate regarding Matthew 27:34. The nature of the discussion was a textual variant. The passage reads as follows:

they offered him wine to drink, mixed with gall, but when he tasted it, he would not drink it. (ESV)

They gave him vinegar to drink mingled with gall: and when he had tasted thereof, he would not drink. (KJV)

I’ve quoted both the ESV and the KJV in order to demonstrate the difference in English (I’ve placed the word in question in bold).

The Greek words in question is οἶνον which I will transliterated as oinos (Critical Text) as opposed to ὄξος which I will transliterated as oxos (Textus Receptus).

Now, it should be said here that I am not a Textual Scholar. I have very limited training in Text Criticism, having only had a basic introduction as part of my Greek studies. That said, I do have some training in how to read the Critical Apparatus for the NA27 Greek New Testament, as well as some training in the methodology used by Text Critics.

I’m not going to get into the history of everything, but here is a brief (and not comprehensive) listing of the manuscript evidence for each reading. I’m summarizing the information provided in the Apparatus of the NA27. To my knowledge, this information was not changed in the NA28.

oxos – Alexandrinus (5th century), W Fragment (4th or 5th century) 0250 Fragment (8th Century), 0281 (7th or 8th Century), the Byzantine text (the majority of texts, but generally from the 12th century and on), Latin fragments from the 12th, 8th, 5th, and 7th centuries, Philoxeniana (a Syriac text from the early 6th century), a Boharic manuscript (a Coptic text from the 3rd century), and a Middle Egyptian manuscript (a Coptic text from the 3rd century)

oinos – Sinaiticus (4th century), Vaticanus (4th century), Bezae Cantabrigiensis (5th Century), K Fragment (9th Century), L Fragment (8th or 9th century), Theta Fragment (6th century), all Latin manuscripts besides the ones mentioned above as well as the Vulgate (4th and 5th centuries) and the Old Latin Translation (4th Century), as well as some Boharic manuscripts (3rd century), Harklensis (a Syriac text from AD 616), and the Sahadic (a Coptic text from the 3rd Century)

Now, Text criticism is not as simple as picking the oldest manuscript, or even looking at the average age of a manuscript with a given reading and picking the oldest. Likewise, it is not as simple as simply counting the number of manuscripts and picking that. There are many factors which play into the decision which text critics make. A helpful guide toward understanding how and why a given reading was selected for the NA27 or UBS4 texts is Omanson, Roger. A Textual Guide to the Greek New Testament. Stuttgart: Deutsche Bibelgesellschaft, 2006. It provides a narrative explanation for why a given reading was chosen, as well as a very helpful introduction which lists common reasons why a scribe may have copied a given word incorrectly.

However, this doesn’t help us too much here, because the committee assembling the UBS4 did not include this variant in their apparatus and no such companion exists for the NA27. However, just at a surface level and with a relatively untrained eye… what we see is that the Greek testimony for oinos are on average earlier, as is the Latin testimony, the Syriac text, and the Coptic text. So generally speaking, both when looking at the totality of the manuscripts, as well as within each language group of texts, the earliest reading is oinos.

Beyond just the manuscript information, Text Critics also look at what is called internal evidence. This includes the immediate context, in the case of parallels in the Gospels how the two readings compare, as well as additional factors. They also look at other external evidence, such as how the text is quoted throughout the ages. The apparatus for the NA27 does not include the Church Fathers, and the apparatus for the UBS4 does not include this verse… so we get no help there.

However, as far as the internal evidence goes, we get a pretty significant help from Mark. Both Matthew and Mark recount the crucifixion account and include two distinct times that Christ was offered something to drink. The only variant is in the verse in Matthew that is in question.

Christ is first offered some kind of drink, mixed with a bitter agent (called gall in Matthew, and myrrh in Mark). Christ refuses this drink. Later, Christ is offered some kind of drink which he accepts.

In the Critical text, this first drink is called oinos in both Matthew and Mark. The second is called oxos in both Matthew and Mark.

In the Textus Receptus, the first drink is called oxos by Matthew, but oinos by Mark. The second is called oinos by both Matthew and Mark.

Now, just prima facie, it makes more sense to think that Matthew and Mark use the same word to describe the same thing. That would be a standard assumption. To argue that they use a different word to describe the same thing would require some kind of explanation (I won’t get into it here, but the difference between gall and myrrh in this passage is an example. Why did Matthew use gall, and Mark use myrrh?). However, to assume they used the same word to describe the same thing requires no explanation.

Second, the proximity of similar words and concepts in Mark seems to lend itself to a sort of contrast. The wine mixed with a bitter agent being refused by Christ contrasted with the unmixed vinegar which Christ accepted. Christ would not drink the oinos, but he would drink the oxos. What exactly that contrast means is a different discussion, but it provides a reasonable explanation for the use of two different words which may (or may not) describe the same substance. Matthew, telling the same account —according to the Textus Receptus— uses the word oxos to describe both the drink that Christ refused, as well as the drink which Christ accepted. Textus Receptus advocates have postulated that in Mark’s account it was regular oinos mixed with myrrh which Christ refused, but then it becomes oxos (sour wine or vinegar) because the souring agent was mixed with it, making it sour wine instead of just wine. This may explain why Mark used different words, but it does not account for why —on their reading— Matthew did not.

However, if we understand that Matthew and Mark are both recounting the same events, and likely were making the same contrast (oinos mixed with a bitter agent in the former, oxos in the later… as the Critical Text reads in both Matthew and Mark) then we have no confusion or conflation.

To wrap it all up, the oldest manuscripts of every language group testify to oinos. The use of oinos first then oxos second requires no significant explanation as the different words are describing different things, and this explanation can be applied to both Matthew and Mark (as opposed to just Mark in the Textus Receptus reading), and it represents a clear literary function in both Matthew and Mark which may have theological significance (Especially in light of Matthew’s preceding chapter).

At the end of the day, this textual variant bears little or no difference on the body of Christian doctrine as a whole, however it does serve as a useful and relatively simple example of Text Criticism.

Jesus as the fulfillment of the Law in Luke 2

My devotional reading over the past 2 days has brought me to the birth, circumcision, and dedication of the baby Jesus found in Luke 2. Concurrently I have been reading Introducing Covenant Theology by Michael Horton. As I’ve read Luke 2, I’ve been amazed at how clearly the text presents Jesus as fulfilling all the requirements of the Law, even though as an infant he had no direct control over this reality.

The first instance was the application of the covenant symbol of Circumcision to Jesus at the age of eight days. Interestingly enough, the text does not indicate that this is in fulfillment of the Law. This is because, strictly speaking, it is not. The sign of circumcision was the covenant symbol of the Abrahamic covenant. Without going into a lot of details, the Abrahamic covenant was a Royal Grant treaty, in which the Great King simply declares a positive benefit for his vassal, with no conditions. Circumcision was not a stipulation of the covenant, it was simply a sign to recall the covenant. In being circumcised, Jesus is associated with the covenant, as he is in his later Baptism.

As we progress through the chapter, we immediately see that in verse 22, the Law is fulfilled in that Mary waited the appropriate time of purification to bring Jesus to the temple. Next, in verse 24 we see that the appropriate sacrifice was made “according to the Law of the Lord.” After a brief description of Simeon we see in verse 27 that Mary and Joseph bring Jesus to the Temple “to do for him according to the custom of the Law.” Finally, we see again in the summary of verse 39 that Mary and Joseph only return to Nazareth after “they had performed everything according to the Law of the Lord.”

In the course of 20 or so verses, Luke explicitly tells us 4 times that Jesus’s early life was already explicitly fulfilling the law.

As we continue to the account of the boy Jesus in the Temple (sidenote: This is commonly preached as though Jesus was TEACHING in the Temple, which has little or no solid exegetical proof in my opinion. The text presents Jesus respectfully learning from the teachers there, who are astounded by his knowledge and answers. Although the text does present Jesus as having special knowledge of the Scriptures and knowledge of who his Father is, this account is enveloped in an inclusio which begins and ends with an explicit statement that Jesus was growing in “wisdom and stature.” The text clearly presents a Jesus who develops intellectually and increases in knowledge, not some freakish docetic Jesus who is fully aware of his identity and all knowledge from the womb.) we see that the poor Mary and Joseph make the pilgrimage to Jerusalem for Passover every year (verse 41). Although not explicitly associated with “the Law,” the text does indicate that they make this long trip “according to custom.” (verse 42) Although it is possible that this is simply their personal custom, it is more likely that this is a reference to the customs proscribed by the Mosaic Law (possibly a reference to Oral Tradition. Compare the usage of ἔθος / ethos in Acts 6:14).

It is often said that Luke’s purpose is to show that Jesus Christ is the savior of not just Israel, but also of the gentiles and other outcasts. However, we must not ever forget that even though this is true, Jesus saves us by fulfilling the stipulations of the Covenant of Works, of which the Mosaic Covenant is a republication. Luke makes clear in this chapter that from the beginning of Jesus life, and all through his childhood, that Jesus lives a life of active obedience to the requirements of the Law. It is the blessings of the Covenant of Works that Christ conveys to us through the Covenant of Grace.


Update 10/28/2016: It is worth noting that I no longer hold that the Mosaic Covenant is a republication of the Covenant of Works, at least not in the sense that I implied here. However, the point I was making with this post still stands, that Christ fulfills the Covenant of Works which is formally —but not materially— republished at Sinai. See WCF19 and the recent OPC Report of the Committee to Study Republication for more information.