Per syntax criticism, portions of the NT are clearly 'translation Greek'
Also, remarks by Torrey on Revelation.
google translate + light editing, hopefully without changing the meaning
Méhat André, book review in _Revue de l'histoire des religions_ [_Review of the history of religions_], vol 209, number 3 (1992), 304-308
https://www.persee.fr/doc/rhr_0035-1423_1992_num_209_3_2402
Raymond A. Martin, _Syntax Criticism of the Synoptic Gospels_, Lewiston (New York) / Queenstone (Ontario), The Edwin Mellen Press, 1987, 219 p. ("Studies in the Bible and Early Christianity", 10), £69.95.
Id., _Syntax Criticism of Johannine Literature. The Catholic Epistle and the Gospel Passion Accounts_, Lewiston/ Lampeter/ Queenstone, The Edwin Mellen Press, 1989, 185 p. ("Studies in the Bible and Early Christianity", 18), $59.95.
Raymond A. Martin currently teaches at Wartburg Seminary, in Dubuque (Iowa). It is an astonishing fact that the small audience obtained so far by his work on the Semitic substratum of the New Testament. In his last two books, he refers to his earlier publications, of which he summarizes the content. One of the oldest and more accessible was an article from 1964 providing "the syntactic proof of Semitic sources in Act I-XV": ^1 [1: R.A. Martin, "Syntactical Evidence of Aramaic Sources in Acts I-XV"] which had been taken and developed in a small book in 1974^2 [2: R.A. Martin, _Syntactical Evidence of Semitic Sources in Greek Documents_, Scholars' Press Cambridge (Mass.), 1974].
Unlike most authors who had studied one by one the Semitisms of the New Testament, he used a statistical method which is surprising afterwards that it had not been tried before him. He was looking for the frequency of a certain number of facts of language, on the one hand in Greek writers a priori devoid of Semitic influence such as Plutarch or Polybius, and on the other hand in Greek texts translated from Hebrew or Aramaic, mainly the translations of the Septuagint. He thus released criteria from the "original Greek" on the one hand, and from the "Greek of translation" on the other, independent of any literary, historical or doctrinal consideration. In 1974 he identified 17 criteria which, applied to the Acts of the Apostles, had made it possible to establish a clear difference between a first part (chap. I to XV, and not I to XII) and a second, written in "original Greek". The first part presents case frequencies closer to the "Greek of translation", an increase of more than 50% compared to the second, which places it halfway between the pure "original Greek" and the pure "translation Greek". It is enough to rule out the reasons generally advanced to explain the Semitisms of the New Testament: the bilingualism of an author, which is certainly the same in two parts, or the imitation, conscious or unconscious, of the style of the Septuagint which would have produced the same effect in both cases.
The most plausible hypothesis is that the author follows in its first part one or more Semitic sources, which he freely paraphrases (assumption that R.A.M. moves away a little quickly [easily disposes of? doesn't address well? -df]) or that he uses irregularly (_erratic_). Based on the number of verified criteria, the author thought he could isolate 16 unites (_units_) of the first part having, certainly or probably, Semitic sources. Once again, we are surprised that the commentators on Acts have so far ignored these results, even though by chance they cite R.A.M. in their bibliographies.
After a long interval, the author applies the same method to the Synoptic Gospels (1987), then to the Johannine Gospel and to various New Testament writings (1989). Overall the results are quite close to what had been established for the first part of Acts: the frequencies are intermediate between those of translations and those of writings written directly in Greek.
The Gospel of Mark is therefore not, as one might have thought, a pure and simple translation. However, some pericopes have frequencies close to the "original Greek", at home and in the parallel passages of Matthew or Luke (the episode of the Gerasene demoniac, the first multiplication of the loaves, the call of the rich, etc.).
Others have frequencies close to "translation Greek"; the call of the first disciples (Mk 1, 16-20 par.), healing of a paralytic (Mk 2, 1-11 par.), accusation of operating by Beelzeboul (Mk 3, 20-30 par.) certain parables (Mk 4, 13-32), resurrection of the daughter of Jairus (Mk 5, 24-43 par.), second multiplication of the loaves (Mk 8, 1-10 par.), dispute over who is the greatest (Mk 9, 33-37 par.). The other units mostly show intermediate frequencies. These are important facts for history and interpretation of texts.
Adopting the hypothesis of the two sources, the author compares according to the same method the synoptic gospels between them. He gets the following results ("Syntax Criticism of the Synoptic Gospels"; p. 127-128):
a) Luke and Matthew attenuate the Semitic characteristics of parallel Markan accounts;
b) Luke and Matthew are more Semitic in the "Q" sections;
c) In its own sections, Luke is much more Semitic than all the rest of the Gospels.
In a new book, R.A.M. analyzes the Fourth Gospel, the four accounts of the Passion and the Resurrection, and the "catholic" epistles. On the Johannine Gospel, his conclusion (p. 80) joins that already formulated by C.F. Burney (_The Aramaic Origin of the Fourth Gospel_, 1922) : the Gospel of John in its present form dates back to a gospel written in Aramaic. It contains units written in "original Greek". But most are close to the "Greek of translation", some very close, for example the speeches after the Last Supper (chap. 14 to 17), but not the discourse on the Bread of Life (chap. 6). Analyses of the stories of the Passion and the Resurrection draw similar conclusions. Wherever the Fourth Gospel is distinguished from the other three, "it is very Semitic according to the syntactic criteria; the Palestinian Aramaic-speaking environment is very apparent". As for the _Epistles of John_, the first two relate to translation Greek, the third from the original Greek. The _Letter of James_, with the exception of a few small "units" (interpolations?) are from Greek for translation; _I Peter_ as a whole belongs clearly in Greek originally, but three sections are in Greek from translation ; _II Peter_ is in Greek originally, but presents a particular style; the whole of _The Epistle of Jude_ is in Greek originally.
These data should in the future be taken into account by all works relating to the New Testament. However criticisms have been made, which R.A.M, does not entirely reject, insofar as they delimit the field on which the syntactic criticism can be exercised. We could add others, for example concerning the presentation: an enormous place is held by tables crammed with numbers, the off-putting character of which can have discouraged more than one reader; too rarely does the author isolate the significant facts to determine their significance. We can also blame him for the deficiencies of its bibliography. Only the 1989 book contains a summary and very incomplete bibliography. It only contains titles of works written or translated into English. The author seems to ignore for example the work of J. Jeremias on the Semitic background of the New Testament, and the little book by J. Carmignac, _Naissance des Evangiles synoptiques_ [_Birth of the Synoptic Gospels_] (Paris, 1984) and its precious notes method.
The very rigor of R.A.M.'s approach could have done it wrong. It would be a reverse error to consider it infallible. The author is the first to mark the limits of his method. (Cf. p. 163-181 of _S.C. of Joh. Lit._: "The Methodology of 'Syntax Criticism' and Criticisms of it"). We will note two here:
1 The statistical method is only fully valid for large numbers. The more short the sections, the more uncertain the results. With the shortest units (up to 4 lines!), we practically come back to ad hoc observations. The author overcomes the difficulty by making the count of the criteria verified among the 17 it uses. This would assume that they are independent of each other and of equal value. But the first eight for example (use of prepositions) are linked between them. And the proportion of the bonds in καί and in δέ is a fact of language that a good translator can be wary of: he can easily correct the effects, or on the contrary use them systematically to "Do Semitic". On the other hand, the number of criteria is not limited ; we could add others.
2 The existence of Semitic sources is not necessarily detected by the "syntactic criticism". The discernment that it operates is exercised only in a meaning. The Semitisms preserved in Greek prove the existence of a Semitic substrate. But their absence or their rarity is not incompatible with the use of an underlying Hebrew or Aramaic text. An example is given by R.A.M. himself: Flavius Josephus declares that he first wrote the _War of the Jews_ in "the language of his fathers"; however, according to the syntactic analysis, the work presents all the characteristics of "original Greek". It could be the same with certain evangelical pericopes, and that, without reaching the perfection of Flavius Josephus and his collaborators, good translators considerably attenuated the Semitic features from their sources.
Either way, the R.A.M. counts establish as hardly contestable the existence of Semitic sources in the bulk of the Gospels and the first half of Acts. That is to say that these writings take us back to Palestinian origins, and therefore older than is usually admitted. A revision is needed of theories which exaggerate the Hellenic part in Christian origins. Is it because they force this revision that R.A.M.'s work has been given so little consideration so far? After his last two books, it is no longer possible to ignore them.
Finally, let us point out that his books and his method can make great service to all who have to deal with writings suspected to be translations of foreign texts, or recognized as such.
André Méhat.
===============================================
"(How) Can We Tell If a Greek Apocryphon or Pseudepigraphon Has Been Translated from Hebrew or Aramaic?" by James R. Davila
https://otp.wp.st-andrews.ac.uk/abstracts-lectures/translated-from-hebrew-or-aramaic/
[endnote] 4. _Syntactical Evidence of Semitic Sources in Greek Documents_ (Cambridge, Mass.: Society of Biblical Literature, 1974); idem, _Syntax Criticism of the Synoptic Gospels_ (Studies in the Bible and Early Christianity 10; Lewiston, New York/Queenston, Ontario: Edwin Mellen, 1987).
The criteria are presented in detail in chapter 1 of _Syntatical Evidence_ on pp. 5-43 and are, briefly,
the relative frequency of eight prepositions in relationship to the preposition en;
the comparative frequencies of _kai_ and _de_ in coordinating independent clauses;
the separation of the Greek definite article from its substantive;
a tendency to place genitives after the substantive on which they depend;
a greater frequency of dependent genitive personal pronouns;
a tendency to omit the article on a substantive with a dependent genitive personal pronoun;
a tendency to place attributive adjectives after the word they qualify;
less frequent use of attributive adjectives;
less frequent use of adverbial participles;
and less frequent use of the dative case without a preposition.
===============================================
_Syntax Criticism of Johannine Literature, the Catholic Epistles, and the Gospel Passion Accounts_ (1989), 187pp., on 7-8 with light edits
https://www.amazon.com/Criticism-Johannine-Literature-Catholic-Epistles/dp/B004QB9GTU/
https://books.google.com/books?id=rGYXAQAAIAAJ&
John's gospel has a total of 1756 lines and a net translation Greek frequency according to Syntax Criticism of -3. As will be seen from Charts 1A and 1B below, this net frequency lies just at the edge of the range of translation Greek for documents of more than 50 lines in length (-4 to -14) and is radically different than the net frequencies of documents known to have been written originally in Greek (+9 to +17).
This indicates that the Fourth Gospel as a whole is most likely translated from a Semitic document, but may have a small number of subsections which were not, but were originally composed in Greek.
....of the 118 subunits in the Gospel, only 10 fall outside the area of translation Greek units of the same size....
On 95, 97:
As Charts XVI and XIX below show, while both 1 John and 2 John fall deeply into the translation Greek area, 3 John falls just as deeply into the original Greek area (Chart XIX).
On 104:
Chart XVI reveals that the net frequencies of the Letter of James fall nearly into the clear translation Greek area and are widely separated from the original Greek documents studied. This suggests that most of the subsections when analyzed will be found to have net translation Greek frequencies as well. This is the case, as can be seen from the classification in Chart XXI below, where only one subsection shows up with clear original Greek net frequencies (no. 9) [i.e. James 2.5-13].
On 109:
The net frequencies of 1 Peter as a whole fall into the clear original Greek area as Chart XVI shows. However, if the net frequencies of each chapter and of the subsections are considered, as can be seen in Chart XXIII below, a significant fact appears: the net frequencies of chapter 5 are more Semitic than any of the others and so are the net frequencies of 4:12-19, the end of chapter 4. When these are combined 4:12-5:14..., the net frequencies become clear translation Greek,
and those of the first part 1:1-4:11 remain clear original Greek (cf. Chart XXIII). One subsection of this larger unit (4:12-5:14) has, however, clear original Greek net frequencies-- 5:1-5.
On 111:
When Jude is analyzed by the criteria of Syntax Criticism, the net frequencies of it as a whole are found to be in the area of clear original Greek (62 lines, +10) as can be seen in Chart XVI. When the individual subsections are studied, they are seen to be conformable to the pattern of such subsections in documents that are known to be original Greek as Charts XIX and XX show, and radically different from the pattern of such subsections in translated writings of the Greek Old Testament.
On 112:
....2 Peter
....from Chart XXIV which follows it will be seen that a number of subsections do show up clearly as original Greek (nos. 2, 6, 8, 9) but none show up clearly as translation Greek.
....thus it is most likely that 2 Peter is, in its entirety and in its subsections, original Greek.
===============================================
_Syntax Criticism of the Synoptic Gospels_ (1987), 220pp.
https://books.google.com/books?id=yRAlAAAAMAAJ&
https://www.amazon.com/Syntax-Criticism-Synoptic-Gospels-Martin/dp/B004QBBCEC/
On 74, information from "Passages in Mark 1-10 Evidencing Translation Greek":
passage; net frequency; verdict
Mark 1.16-20; -1; Translated from Semitic Sources
2.1-12 +1 translated
2.13-17 0 probably translated
2.23-28 0 probably translated
3.7-12 0 probably translated
3.13-19 0 probably translated
3.20-30 +1 translated
4.1-9 +2 probably translated
4.13-20 +2 probably translated
4.21-25 -1 translated
4.26-29 -2 translated
4.30-32 -2 translated
5.24b-34 0 translated
5.35-43 +1 translated
5.21-43 0 translated
6.1-6a 0 probably translated
6.14-29 0 translated
8.1-10 -1 translated
8.31-9.1 -1 translated
9.33-37 0 probably translated
10.35-45 -3 translated
On 95, information from "Q Passages Evidencing Translation Greek in Luke and/or Matthew":
passage; net frequency; verdict
Luke 4.1-13; -1; Translated from Semitic Sources
Luke 6.27-31 +1 probably translated
Luke 6.32-36 -2 translated
Matthew 5.43-48 -5 translated
Luke 6.37-38, 41-42 -1 translated
Matthew 8.5-13 -2 translated
Luke 7.18-23 -1 translated
Luke 10.13-15 +1 probably translated
Luke 10.21-24 0 probably translated
Matthew 11.25-27; 13.16-17 -1 translated
Matthew 6.9-13 0 probably translated
Luke 11.9-13 0 probably translated
Matthew 7.7-11 -1 translated
Luke 11.29-32 0 probably translated
Matthew 12.39-42 -1 translated
Luke 11.33-35 +1 probably translated
Matthew 5.15; 6.22-23 -1 translated
Luke 11.39-44 -2 translated
Luke 11.45-51 +1 translated
Matthew 23.4, 29-36 -4 translated
Luke 12.2-9 -2 translated
Matthew 10.26-33 -4 translated
Luke 12.33-34 +1 probably translated
Matthew 6.19-21 -1 translated
Luke 12.42-46 -3 translated
Luke 14.15-24 -1 translated
Luke 14.26-27 +2 probably translated
Luke 15.4-7 +1 probably translated
Luke 19.11-27 +1 translated
=================================================
Stephen C. Farris, "On Discerning Semitic Sources in Luke 1-2" in _Gospel Perspectives: Studies of History and Tradition in the Four Gospels_, Volume 2 (1981), 375pp., on 207, 210, 213-214 with a little reformatting
https://www.amazon.com/Gospel-Perspectives-Studies-History-Tradition/dp/1592442889/
https://books.google.com/books?id=4HavCwAAQBAJ&pg=PA214&
a later version of which appears in
_The Hymns of Luke's Infancy Narratives: Their Origin, Meaning and Significance_ by Stephen Farris
https://books.google.com/books?id=GyidBQAAQBAJ&pg=PA59&
On 207:
Once more a table may be useful in summarizing the results of such a study.^17
The table shows the range of net results when the criteria are applied to the shorter units of original and translation Greek.
Unit Length ; Original Greek ; Translation Greek
31-50 lines ; +13 to +7 ; +1 to -8
16-30 lines ; +12 to +3 ; +4 to -9
4-15 lines ; +12 to 0 ; +7 to -6
On 210:
When we turn to the infancy narratives and apply Martin's criteria, we find that Luke 1-2 as a whole and also in its various parts consistently displays translation Greek frequencies as the following chart makes clear.^27
Section ; Lines ; Net
Chapter 1 without hymns ; 107 ; -12
Hymns of Chapter 1 ; 30 ; -4
Chapter 2.1-40 without hymns ; 57 ; -5
Hymns of Chapter 2 ; 5 ; -2
2.41ff. ; 23 ; -2
Chapter 1 total ; 137 ; -14
Chapter 2 total ; 80 ; -5
Hymns total ; 35 ; -4
Grand total ; 217 ; -16
Every one of these units, even Luke 2.41ff.,^28 is safely within the translation Greek frequency range.
On 213-214:
....Romans and Galatians... From these books I have chosen a passage of doctrinal exposition, Romans 5, and a passage which contains biographical narrative, Galatians 1-2.5.^34 Finally, there are two texts of somewhat different character, Revelation 3 and 4-5.10. The former is the concluding section of the letter to the seven churches, while the latter forms the beginning of the apocalyptic vision itself. This latter section contains narrative, direct discourse, and hymnic materials, as do the infancy narratives of Luke. Of the author of Revelation R.H. Charles wrote, 'while he writes in Greek, he thinks in Hebrew'.^35
Revelation, it appears, may serve as our example of 'poor quality Greek influenced by Semitic idiom'.
The results of the analysis are as follows:
Text ; Lines ; Net Result
Luke 5.12-6.11 ; 91 ; 0
Mark 1.40-3.6 ; 89 ; +2
Luke 12.13-13.9 ; 109 ; -2
Lukan passages (total) ; 200 ; 0
Romans 5 ; 50.5 ; +2
Galatians 1-2.5 ; 56 ; +3
Paul (total) ; 106.5 ; +3
Revelation 3 ; 59 ; -2
Revelation 4-5.10 ; 61 ; -5
Revelation (total) ; 120 ; -5
================================================
Charles C. Torrey, _The Apocalypse of John: Introduction, Excerpts, and a New Translation_ (1958), was at the now-defunct https://www.preteristarchive .com/1958_torrey_the-apocalypse-of-john/
For the Apocalyptist the language of the New Dispensation of the Christian Church was Aramaic only. It is most significant that the numerous hymns and doxologies sung or recited by the saints and angels in heaven, in chapter after chapter of the book, are composed in Aramaic (wherever it is possible to decide), not in Hebrew; though the writer could have used either language. ....
The language of the "Revelation of John" has offered one of the most perplexing problems of biblical study. Its author is, on the one hand, a master of Greek and a man of learning: on the other hand, one who writes in an idiom which is not Greek but Semitic, and whose work swarms with major offenses against Greek grammar.
As for the mastery of Greek, it seems at the outset to be shown by the vocabulary of the book. See the details given in Swete's Apocalypse of St John, pp. cxv ff. It is a large vocabulary containing many unusual words, all handled with certainty: employing a great variety of verbs and verbal compounds, all so used as to give the precise shade of meaning desired in each case. As modern commentators agree, the author of the book was a master of Hebrew and made his own translations for that language in his multitudinous allusions to the Old Testament. These translations, without exception, are the work of a scholar who evidently was at least as thoroughly at home in Greek as in Hebrew. They are as accurate, concise, and skillfully fashioned, word on word, as those of any translator. ....
To all appearance, thus far, the Greek of Revelation is the work of a linguist of skill and erudition; why, then, the astonishing "offenses against grammar"? This peculiarity has been a source of great interest and perplexity since the earliest times. ....
For the truth is that the Book of Revelation was written in a Semitic language, and the Greek translation which alone has been preserved is a remarkably close rendering of the original.
Reason for extraordinary faithfulness on the part of a translator is given in 22:18f.:
I testify to every man that hears the words of the prophecy of this book: If any man shall add to them, God shall add to him the plagues which are written in this book; and if any man shall take away from the words of the book of this prophecy, God shall take away his part from the tree of life, and out of the holy city.
Here is the warning to any interpreter. The last degree of exactness in reproducing both words and idioms is plainly prescribed, and this is what we see attempted in our curious Greek. In fact, underlying all of the amazing solecisms is seen the wording of the Semitic original. The grammatical monstrosities, recognized in their true nature, testify to the execution of a definite purpose carried through with remarkable consistency. When they are examined, they are found to show grammatical appreciation rather than the lack of it. But it is Aramaic grammar!
(Nevertheless, the ideal of a thoroughly accurate translation was incapable of realization, as we know to our sorrow. No Greek translator of an unpointed Semitic text of the extent of this apocalypse could possibly come through without his considerable sheath of mistranslations. We have no knowledge of any such faultless―or even nearly faultless―achievement.*)
What the Greek translator of Revelation does, in the effort to be exactly faithful, is merely an exaggeration of what is regularly and constantly done in Matthew, Mark, Luke, and John. The translators rendered as they did because of the conception of their task. They were handing down works of high importance, and would assume no unnecessary responsibility. What they―each and all―aimed at was to produce a text which could be understood by the Greek reader and at the same should mirror faithfully every word and phrase of the sacred original. This, the original, was the all-important thing, and the fact was always kept in view. The style of the translation was of no consequence; it was not Greek, nor ever intended to be.
*See Our Translated Gospels [by Torrey], chapter 1; The Four Gospels [by Torrey], pages 265-74.