Cursory Introduction to Textual Criticism and Manuscript Families

The textual critic intends to reconstruct the original autograph as nearly close to perfection as possible. This endeavor is extremely meticulous and daunting. Text reconstruction is a science, which when following well-accepted guideline, allows the critic to accurately determine an original reading—as close as possible—in the face of multiple variants. Textual criticism methods are not mathematically precise, but involve inductive reasoning that results in probable or improbable gleanings. The methods are reliable, and even though we do not have the autograph it can be confidently said that the text—reconstructed today—is the text the early church had.

After the New Testament was penned and distributed to Christians in their local communities, variations began to be introduced into the texts in specific regions. Scholars have noted that certain regions have similar types text types. If a particular manuscript was copied in Greece, for example, and began being distributed in Greece, then we would expect similar variants, errors, corrections etc… to be present, since scribes would be copying from the same manuscripts. The fact that New Testament manuscripts were so broadly distributed geographically is a great help. This allows the textual critic to use manuscripts in different regions as checks and balances against other manuscripts that may have taken a few liberties while copying. We will note the four textual families and their descriptions.

First, there is the Alexandrian text type. It is most noteworthy for Uncials (all capital lettered manuscripts) Sinaiticus and Vaticanus, two of the earliest entire relatively complete Greek New Testaments (Vaticanus has portions missing), dated in the middle of the fourth century. With the discoveries of Bodmer papyri specifically p66 and p75, we know by corroboration that Sinaiticus and Vaticanus extend back to an archetype dated in the second century. The Alexandrian texts has brevity and austerity. They are generally shorter and less “polished” than other types. The majority of scholars recognize the Alexandrian text type as the most reliable. Harold Greenlee posits, that since Alexandrian was the place of classical Greek literature and learning, more sophisticated textual updates (that is updating a word or phrase) sometimes creep into the text. All in all however, Alexandrian text are characterized by rough readings.

Second, there is the Western text type. It was primarily used in Italy. The western text has significant manuscripts, as Bruce Metzger states, “The most important Greek manuscripts…are codex Bezae (D) of the fifth century (containing the Gospels and Acts), codex Claromontanus (D) of the sixth century (containing the Pauline epistles)…[and] codex Washingtonianus (W).” Paraphrasing and additions to the text called “western interpolations” characterize western texts—often words, phrases or clauses are omitted, or altered. In fact, the book of Acts is 10% longer in the Western text than any other family of texts. Generally shorter readings of this text are considered authentic, often in conjunction with reliable corroboration.

Third, we have the Caesarean family. Origen used this text type in Caesarea hence the name “Caesarean.” Some scholars wonder, if this should even be considered a family.[4] At the time there are a group of witnesses which have very similar readings, which are neither, Alexandrian, Western or Byzantine, thus it seems correct to separate them. Noteworthy among this family are f1, f13 W and Q. The synthesis of both Western and Alexandrian style characterizes this text type, though it is “slightly closer to the Western, but it does not generally include long additions and long paraphrases of the Western text or the long additions of the TR [Textus Receptus].”

Finally we come to the Byzantine family. The revered “received text” (the text behind the KJV) is often confused with this family. By far, the majority of our manuscripts are in this family. When Christianity was legalized under Constantine, this text was distributed rapidly, copied, and used by the majority of the churches. Most scholars agree that the Byzantine text is later and inferior to the other text-type—although, there are readings that are preserved in the Byzantine text that reflect original readings. Smooth and clear readings characterize this text (which indicated scribal influence). It has also been noted that harmonization appears in the gospel accounts, this is one of its most common qualities.

The reader may feel that the New Testament was rent piecemeal and scattered toward the four winds of heaven. It need be emphasized: (1) God in his providence always takes care of his people. (2) Even though families have certain readings, it was common for a certain geographic location to receive manuscripts from other locations, so as to temper the spread erroneous copying. (3) No article of the faith is damaged and for all practical purposes Christian had the word of God and were able to live faithful lives. (4) The majority of the manuscripts read relatively similarly to one another, with the exception of some of the above things mentioned.

John 14:14 and the original reading

External Evidence

Scattered group of witnesses, including, a few early version omit verse 14 entirely. However, a substantial amount of manuscripts; the earliest and best include it, as well as the remaining families. The inclusion of this verse is virtually accepted as correct, and not a point of contention. The omission of 14:14 will not be the concentration of the analysis, but whether or not αιτησητε has the direct object me, τον πατερα, or if the verb stands alone with no direct object. In order to determine which reading is best supported by the witnesses we will consider, first, external evidence: (1) The date and quality, (2) geographical distribution and (3), and genealogical distribution of the witnesses. The internal evidence will be discussed in the next section.

Here are the variant readings and the supporting witnesses:

Reading εαν τι αιτησητε me εν τω ονοματι μου ποιησω (Whatever you ask me in my name I will do it)
• Alexandrian P66 aleph B W al Q
• Western
• Caesarean Q f13 (13, 69, 124, 174, 230, 346, 543, 788, 826, 828, 983, 1689, 1709, 1709) 28
• Byzantine D 33 700 al
Reading 2: εαν τι αιτησητε εν τω ονοματι μου ποιησω (Whatever you ask in my name I will do it)
• Alexandrian
• Western
• Caesarean D
• Byzantine A K Y P byz al 1241. 1424
Reading 3: εαν τι αιτησητε τον πατερα εν τω ονοματι μου ποιησω (Whatever you ask the Father in my name I will do it)
• Alexandrian
• Western
• Caesarean
• Byzantine 249, 397

The dates of the above manuscripts are as follows: Manuscripts: P66 (2nd century), aleph (fourth century), B (fourth century), A (fifth century) D (fifth century) W (fourth/fifth century) L (eighth century) K l 844 (late eighth century) 017 (ninth century), D (ninth century), Q (ninth century), P (ninth century), Y ((ninth/tenth century) 33 (ninth century), 1424 (ninth/tenth century) 28 (eleventh century), 700 (eleventh century), 1241 (twelfth century), f13. When consider external evidence one always notes manuscripts, ancient translations, and church father quotations in that order. Under manuscript evidence (the most important), we examine papyri (the early writings after autographs, which include a few verses to entire accounts), uncial (All upper-case manuscripts), minuscule (lower-case later manuscripts, and lectionaries (church service readings of scripture)

Summarizing the information above awards us the following analysis. Reading three has the weakest external support, being supported with only two later minuscule manuscripts. This narrows the readings down to reading one and reading two. As far as geographical distribution is concerned; three different families attest the first reading, where as the second reading only has two. Reading one is better distributed geographically—that is—more family support its variant. The Byzantine family best supports reading two, as manuscript A K and P have a better genealogical relationship than D (9th century) of reading one. Ultimately, however, the overall genealogical relationship of reading one is better. The weight of evidence for reading one is strong, being supported by the earliest papyrus containing John—which contains the “me”. Furthermore, reading one is supported by aleph and B which are recognized as our best two manuscripts in extant. Caesarean witnesses also support the second reading. Reading one has the best external supports, I give it grade A.

Internal Evidence Criteria

External evidence alone does not determine a correct variant. It must be weighed against the internal evidence as well. The criteria for weighing evidence internally, varies from external methodology, here are a few generally guidelines: (1) The shorter reading is generally favored, (2) the more difficult meaning is to be preferred, (3) determine which reading gave rise to the other readings, (4) Determine which reading is more appropriate in the context (the immediate context, the context of the book, and the literary and stylistic idiosyncrasies of the author, spelling errors, and historical context)[9]. Of course this is not a fail-safe criteria—it is a safe guideline but often other factors must be used. Explaining how the various errors (unintentional or intentional) crept into the text is outside of the scope of the article, though reference will be made to some.

John 14:14 variant readings

Reading 1: εαν τι αιτησητε me εν τω ονοματι μου ποιησω (Whatever you ask me in my name I will do it)
Reading 2: εαν τι αιτησητε εν τω ονοματι μου ποιησω (Whatever you ask in my name I will do it)
Reading 3: εαν τι αιτησητε τον πατερα εν τω ονοματι μου ποιησω (Whatever you ask the Father in my name I will do it)

One might argue that reading two meets the criteria from guideline #1 (it is the shortest of the three). But this conclusion would be hasty. On certain occasion there are good reason not to go with the shorter reading: when homoioteleuton (an omission caused by two words or phrases that end similarly), or homoiarchton (an omission caused by two words or phrases that begin similarly), or haplography occur (writing a letter or word once when it should have been written twice). In other words when there is an omission the shorter reading is not to be preferred.

Notice how an uncial manuscript including John 14:13-14 would have looked like to a copyist (no spaces).



The first reading seems to best explain the others. It is more reasonable to conclude the scribe omitted me accidentally. Although it could be argued that a scribe would insert “me” or “my father” in order to clear up ambiguity—as was apparently done for the third reading. This conclusion should be rejected on a two grounds. First, “me” is not added to Christ’s saying in (v 13), why not harmonize both verses? Since intentional correction does not seem like an adequate explanation we must opt for an unintentional mistake. Namely, the scribe omitted “me” from verse 14, because of its absence in verse 13. Perhaps the scribe’s eye drifted to the previous verse (13) as he was copying αν αιτησητε. This phenomenon has a common occurrence in copying.

The next criterion—the reading, which best explains the origin of the other readings is probably the original reading; and training closely behind it; the more difficult reading—that is for the scribe is to be preferred. It is more difficult to explain the reading that includes me, on the grounds that it appears to contradict (John 16:23)—where Jesus says, “you will ask me nothing”. A scribe would attempt to correct this apparent contradiction, by either removing “me” or adding “the father” (though it is possible that the scribe accidentally copied “the father” from verse 15). Furthermore, in Johannine literature, Jesus describes the person who is answering a request as the person being requested (John 11:22; John 15:16 John 16:23). Thus, when Jesus states, “he will do it” we have reason to believe that he is the one being entreated (and I might add whether the “me” is present or not).

Summary of internal evidence

Reading one has the strongest internal evidence. It is the more difficult reading and best describes the origin of the others. It fits best with the idea that the omission was more than likely an unintentional omission as opposed to a malignant scribe mowing through the text—cutting away what he pleased. I give it a B+ rating


In this brief study we have given a brief overview of the purpose of textual criticism and the strengths and weaknesses of each textual family. Upon surveying the evidence we can conclude that reading one is the very probable original reading. Combining both internal and external evidence we grade reading one with an A-. Jesus did authorize his disciple to pray to him in his name and the early church understood this to be the case (Acts 7:59; 1 Corinthians 16:22, 1 Timothy 1:12).


Greenlee, Harold “Introduction to New Testament Textual Criticism”

Metzger, Bruce “A Textual Commentary on The Greek New Testament”

Nestle-Aland “Novum Testamentum Graece 27th Edition”

Robinson, Maurice “The Case for Byzantine Priority”

Wegner, Paul “The Journey From Text To Translation”