The Greek New Testament, Produced at Tyndale House, Cambridge

We recently published the latest issue of our theological journal, Foundations. Editor Martin Salter introduces a review article by Daniel Stevens on The Greek New Testament, Produced at Tyndale House, Cambridge (Crossway, 2017, 540 pp). Also published by Cambridge University Press in various bindings.


It is not every day that an entirely new edition of the Greek New Testament appears on the scene, and the THGNT represents a significant contribution to this somewhat monopolised field. The editor, Dirk Jongkind, associate editor Peter Williams, and the assistant editors Peter Head and Patrick James have made a meticulous, interesting and appealing contribution to the study of the text of the New Testament through this new edition. While the THGNT does not present any world-changing differences in the text of the New Testament, it does succeed in raising questions about the goals and priorities of New Testament textual criticism and extends its editorial concern to the form of the text itself, rather than stopping at what words are present in what order. Since the THGNT is not a composition but rather a critical edition of the text of the New Testament in Greek, this review will focus more on the philosophy, textual content and format of the THGNT than on the conceptual content of the New Testament itself. That is an exercise humbly left to the reader.


The THGNT began as an update of Samuel Tregelles’ (1813-1875) edition of the Greek New Testament with the addition of manuscripts, and particularly papyri, which have been discovered and catalogued since Tregelles’ time. His edition was chosen as a basis because of a shared methodological frame-work, namely “Tregelles’ strong reliance on the testimony of documents and on the principle of proven antiquity” (505-6). However, the work continued to grow and, as it stands, represents an entirely new edition of the New Testament, not just an update of Tregelles’ work.

The editors had three goals that are expressed both in the text and the format of the THGNT:

1) Readability;

2) A documentary focus in which choices must be justified by manuscript evidence, in which at least two manuscripts agree, one of which must be from the fifth century or earlier (with the exception of variants in Revelation);

3) An attempt to present “the best approximation of the words written by the New Testament authors” (505).

Their editing is particularly informed by an extensive study of scribal habits, both generally and of individual manuscripts. This has allowed the editors to exercise a level of letter-by-letter detail that is not usually exhibited in editions of the New Testament, and causes them to be more concerned with aspects of spelling and accentuation than is typical. The aspect of this method that sets the THGNT most apart is the documentary emphasis, which goes on to inform every decision within the text. The THGNT takes this documentary approach rather strongly, refusing evidence from patristic sources as well as from later minuscules and early translational versions, other than for accentuation choices. While the editors acknowledge that a later text can witness an older reading, they do so only in theory, at no point admitting such a reading.

Upon opening the text, readers familiar with other edited Greek New Testaments (or indeed, any modern language translation) will be in somewhat unfamiliar territory. After a table of contents and a two-page preface, the text of Matthew begins and the New Testament continues through to Revelation. Introductory matters are left until the end of the book in an attempt to not interrupt the reading experience. The page itself is uncluttered, with no symbols in the text and minimal textual apparatus at the bottom of the page (more on that later). Further, quotations are not marked within the text at all, causing them to blend seamlessly into the main text.

Paragraphs are more frequent than in all other editions, based on manuscript witnesses more than modern convention, and are marked by ekthesis, projecting the first line into the margin, instead of indentation. This difference in paragraphing has a subtle effect on the reader, both by moving one along a bit more quickly because of the frequent paragraph breaks and by providing a different interpretation of what constitutes units of thought within the text. For example, Rom 6:1-10 is presented as a single paragraph, while Rom 6:18 is a paragraph unto itself. While the paragraphing of the THGNT does not represent the unmediated and universal interpretive traditions of early Christian scribes (as if there were such a thing), the editors have chosen places of agreement between at least two manuscripts for their divisions. Hopefully, in their forthcoming textual commentary the editors will provide some of the evidence leading to these paragraphing distinctions.

Another obvious difference that will meet the reader is the order of New Testament books. Here also the editors were informed by the order found in many manuscripts, and so they arrange the New Testament in the order of Gospels, Acts, Catholic Epistles, the Pauline corpus (including Hebrews at the end) and Revelation. The Pauline corpus follows the familiar order.

Differences in the text and apparatus

So far all this has been how the THGNT differs in format, but what differences are found in the text and apparatus? That is, in what ways does the THGNT differ from other critical editions in terms of the words on the page?

The Apparatus

The textual apparatus is noticeably smaller than that in other editions, and represents only a minority of the choices made and the evidence supporting those choices. Because of the editors’ desire to have a clean and easily readable page, the evidence in the apparatus has been reduced to three categories: First are those variants which the editors believed were close contenders with the reading provided in the text. These are marked in the apparatus by a black diamond. Second are variants with “high exegetical importance” (515), that is, those that the editors feel comfortable deciding against, but would have a significant enough impact on the meaning of the passage if correct. Third, and finally, are variants which illustrate scribal habits, a particular emphasis in the editors’ method.

While reduced, the apparatus of the THGNT stands out as particularly helpful on two counts: First, its comments are almost entirely in English, not Latin. (The main exception is vid, for ut videtur, “apparently”.) Second, for most vid readings, that is for readings of manuscripts in which some or all of the letters of a given variant are missing due to damage, the apparatus provides a best reconstruction of the reading in round brackets. In such reconstructions, letters which are incomplete in the manuscript and must be partially reconstructed are represented with a dot beneath them, and those which are completely absent and must be entirely reconstructed are presented in square brackets. This is a great step forward in making vid readings more transparent, and puts the relevant evidence in the hands of the reader.

The Text

As for the text itself, the greatest overall differences are in the area of orthography. Spelling decisions were made on the basis of manuscript evidence, with emphasis on spellings testified in manuscripts of the fifth century and earlier. Accentuation decisions had to depend on later manu-scripts or scribal hands, however, since the earlier uncial manuscripts did not initially have accentuation. The editors admit that these spellings and conventions may not reflect the actual spellings of the first century in general or the New Testament authors in particular, but do represent the most common spellings of the earliest evidence we have. While much of the text approximates the presentation of the text in the manuscripts, some ancient conventions were not followed: Lower-case letters are used throughout with minimal capitalization; spaces between words and punctuation are added, though the editors attempted to be sparing in their use of punctuation; nomina sacra, abbreviations of the names of holy persons or things, are not used, though they may be in later editions after a more full study of their use in manuscripts.

Christos is not capitalised at all, even when used like a title. Iota subscripts are omitted when not necessary for identifying the form, which means that outside of the dative singular for nouns and adjectives and the active subjunctive of verbs, they tend to be absent, even when etymologically correct. In compounds involving syn, the nu is occasionally not assimilated to the following letter, which may prove an inconvenience for readers attempting to look up the terms in a lexicon if unaware. Within this tendency, syng only appears once (1 Cor 7:6), synl is usually not preferred to syll, whereas synm is slightly preferred to symm as is synch to sygch, and synk is heavily preferred to sygk.

The representation of Semitic names follows the majority of manuscripts, which yields some irregularities, such as a rough breathing at the beginning of Abraham’s name, resulting in Habraam, and not placing the accent on the penult as other editions do. In a somewhat more controversial move (at least among the subset of people who care deeply about Greek accentuation), the editors followed the evidence of later accentuation that sometimes placed an accent on indefinite tis. In addition to patterns of emphasis, the editors have appealed to stress patterns across Indo-European languages regarding the interrogative and indefinite pronouns etymologically related to tis (those beginning with *kw-) to justify this move. Again, however, the editors emphasise that accentuation choices reflect a period centuries removed from the writing of the New Testament, and cannot necessarily always be projected backwards. This is certain to spark further debate on how Greek accentuation is understood and represented in edited texts.

The most widespread and obvious orthographic variant is the inconsistent use of the two letters epsilon and iota for long ī. This, again, is an attempt to represent the manuscript tradition in which epsilon iota was substituted for long iota. I call this usage inconsistent because the manuscript evidence is itself inconsistent, with some scribes preferring epsilon iota for ī (such as the scribes of Vaticanus), some not, and some switching between the two. Because the editors wished to follow this tendency in general (they made their attestation rule stricter by requiring it to be present in two manuscripts, one of which from the fifth century or earlier, and to appear in other instances of spelling that particular word), the THGNT sometimes features this variant, and sometimes does not. For example, geinomai is provided for ginomai “become” in Mark, Luke, two verses in John (!), and Romans-Colossians, but not elsewhere. Similarly, geinōskō is put for ginōskō “know” in Mark, Luke, John 10:14-14:17, and 1 Corinthians-Philippians. Here is a key point where the editors’ twin goals of documentary evidence and presenting the words written by the authors of the New Testament may come into conflict. Luke and Acts have different tendencies on the usage of epsilon iota, for example, leading one to think that this evidence is more relevant to the study of scribal habits in the New Testament manuscripts than the spellings of the authors themselves. The editors themselves acknowledge this possibility, and throughout their introduction these two goals exist in an unresolved tension.

All these orthographic changes may seem relatively minor to most readers, but they are worth thinking about for at least two reasons. First, the editors of the THGNT present them as a significant move toward the documentary evidence, and as such they represent an explicit part of what the editors set out to do. Second, they reveal a fundamental tension in the method of the THGNT between a desire to represent the documentary evidence of NT manuscripts and to provide the closest text possible to the original composition. While the editors’ study of scribal habits is extensive and no doubt accurate, they admittedly run the risk of presenting later convention as the oldest text form.

As far as concerns textual decisions, it would be impossible for this review to include every variant passage, but I will provide a few particularly salient textual decisions. There is a textual break after Mark 16:8, followed by an explanatory note from minuscule 1 in Greek, with an English translation in the apparatus, “In some of the copies, the evangelist finishes here, up to which (point) also Eusebius of Pamphilus made canon sections. But in many the following is also contained.” Mark 16:9-20 then follows. The Freer Logion of codex W after Mark 16:14 is not mentioned in the apparatus. The Pericope Adulterae in John 7:53-8:11 is entirely relegated to the apparatus, with the text continuing from 7:52 directly to 8:12. The apparatus records that 1 Cor 14:34-35 occur after 14:40 in codex D. In the catholic epistles there will be more pronounced difference from the NA28, because of diverging editorial methods.


Since so much of the THGNT’s difference is in issues of formatting, it is worth noting that the overall effect of this is a rather pleasant one. The clean pages without intrusive symbols or references are far more approachable, and facilitate an easy reading experience. The use of ekthesis for paragraphing is no less pleasing than normal indentation, and I find that it helps one scan through sections of the text more quickly. The font is Adobe Text, a visually appealing serif font. In all, it is a pleasure to read, and the editors do reach their goal of readability.

Overall thoughts

The THGNT represents a significant contribution to the editing of the text of the Greek New Testament. It brings readers into closer contact with the evidence of the early manuscript traditions than any edition prior, and even takes a step closer to approximating the reading experience of those manuscripts. The level of detail provided by the editors is to be commended, and the evidence of spelling and accentuation convention is a worthy addition to the scholarly analysis of the Greek of the New Testament. The THGNT’s greatest strength, however, may also be its weakness, in that it is bound to the early documentary evidence of Greek New Testament manuscripts, and thus is closed to the testimony of later manuscripts, versional evidence, or patristic citations of the New Testament. At times, then, there is the possibility that the information provided from the scribal habits of the first few centuries is ultimately just that, information about scribal habits rather than about the spelling and pronunciation of first century authors. That being said, however, this does not overshadow the benefits of the approach, and the Tyndale House Greek New Testament is well worth the work and the reading. It is a useful tool in the study of the New Testament, and works well in conjunction and comparison with other editions. Even at times when its evidence may speak more about the manuscript tradition, it is indirect evidence to the earliest text, and provides fascinating linguistic data about how Greek could have been pronounced and how it was transmitted, especially in the fourth and fifth centuries.

Jongkind, Williams, Head and James have produced an edition of the Greek New Testament that will benefit students of the New Testament, and I look forward to further editions and their textual commentary, both of which are sure to spark further conversation, study and interest.

Note: A gratis copy of the Tyndale House Greek New Testament (THGNT) was received at its launch at Tyndale House, Cambridge, for which I am grateful.

Daniel Stevens is a Gates Cambridge Scholar and PhD candidate in Theology and Religious Studies (New Testament) at the University of Cambridge. 


There are currently no comments on this post

Post a Comment

Your comment will have to be approved by a site administrator before it is shown on the site so please be patient.