Conjunctions and such...

by Yoseph/Joseph Viel

In the following, we will analyze Hebrew and Greek styles for joining together phrases, which is also used semitically describing a flow of events in a time sequence and how this reflects on our critical analysis of the linguistic origin of the New Testament. He will demonstrate:

1. Hebrew frequently uses the letter Waw (w) to mean "and" often implying "and then" (though not necessarily literally meaning that every time), indicating that what follows the "waw" is a continuation of the previous narration, and in most such cases, what comes after the "waw" follows the previous sentence in sequential order. Waw is also used for joining together many things that have nothing to do with time sequence.
2. Waw is usually translated into Greek as "kai", sometimes as "de", and into English as "and".
3. The LXX uses "kai" more than "de", against normal Greek style, mostly because it is a translation from the Hebrew.
4. In Greek, the flow of events are sometimes joined together with "kai", but more frequently with "de", by about a 3 to 1 to 10 to 1 margin in favor of the use of "de" over "kai".
5. Many parts of the New Testament uses "kai" more than "de", against normal Greek style, but similar to what is found in the Hebrew to Greek translation of the LXX.

The end result of this is to demonstrate that there is indeed a Semitic under-pinning to the Greek New Testament.

What's the difference between "kai" and "de"? EW Bullinger's "Critical Lexicon..." says this:

• "kai, the conjunction of annexation, uniting things strictly coordinate"
• "de, conj. of antithesis"
• "kai connects thoughts, de...introduces them" So kai connects things smoothly, de interrupts our thought when it joins them together.

Now if I were to write

John went outside AND walked to the chair AND read a book AND got out of his seat AND went inside

it would be a matter of personal judgement how to translate this into Greek, since one person might see all this as a smooth flow of events, while another person might see each step as a serious interruption in the train of thought involved. Was he going outside with the intention of reading a book? Was he planning on returning a book to the library and then change his mind? Did he start reading the book as he was walking to the chair or did he wait until he got there before reading it? So many factors could enter into our thinking as to whether this is a smooth flow of similar thoughts or an interuption of something different. People could argue ad naseum as to whether this should be translated "kai" or "de".

But Greeks tended to see such a chain of events as introducing new thought, and signifying this by using "de" more than "kai" in connection sentences/phrases together with a conjunction when expressing narration.

1. Hebrew frequently uses the letter Waw (w) for Connecting Events

In English , we occasionally, but infrequently, begin a sentence with the a word like "and", which thought-wise tends to join the sentence with the previous sentence. But one thing that characterizes Hebrew narration is that many sentences in the Tanakh begin with the letter "Waw"/"Vav" , which is often translated "and" , "but", or "then" in English, depending on what seems to make the reading most "smooth" to the translator.

72% of all verses in the Torah begin with the letter "Waw" (w) in Hebrew. That percentage varies in the entire Tanakh from 91% in the book of Ruth to as little as 1% in Song of Songs. In most narrative books, as from Genesis to 2 Chronicles where most of the text is telling a story, the percentage is 76% , as shown in the chart below

The highest percentage of usage of the letter "Waw" at the beginning of a Hebrew sentence tends to appear in the books that are mostly histories. In the mostly Historic books from Genesis to 2nd Chronicles, only Deuteronomy has fewer than 65% of its verses starting with a Waw (it has 47%). For 8 of these books, the percentage tops 80%. In the Aramaic portions of Daniel and Ezra, we see

	Percentage of verses starting with "Waw"	Percentage using "}yd" or "}yd)"	Percentage using either
Ezra 4:8-6:18, 7:12-28	47%	13%	60%
Daniel 2:4-7:28	30%	21%	51%

2. How the Hebrew "Waw" is translated into Greek

How "WAW" it is translated from Hebrew into Greek is a bit more complicated than how it is translated into English (as usually either "and", "then", "and then", "but" or "yet"), but the various forms can be:

• kai is the most common translation. (LXX examples include Gen 1:3-2:3, 2:5,7-9,13-16, etc)
• de is next most common (LXX examples include Gen 1:2, 3:6,10,12,17, 3:1,3,17, 4:5, etc)
• While kai and de represent about 95-98% of Hebrew to Greek translations of "Waw", there are sometimes other translations that represent a Waw, but may be translated as something else as well. Among the numerous other occasional possibilities include:
  • tote may be a translation when "Waw" is joining events in time, but probably not very often. The LXX usually translates the Hebrew "z)" as "tote". "tote" is often used in the Greek NT where "}ydyh" is used in the Aramaic, but in Matthew 26:38 & 27:38 we see a "Waw" used in the Peshitta where the Greek has "tote". tote is not the most common translation of "waw", but it could be used a small minority of the time. LXX Aramaic examples also include Dan 6:21, etc, where tote is also used as a Greek translation for the Aramaic "then" / "}yd)" in places like Daniel chapter 2 (numerous verses). "tote" is translated "then" in KJV 82 times
  • epeita (translated as "then" 10 times by the KJV translators)
  • others

  Because these "other" possibilities exist, but are rare, we can usually ignore their effect for the purposes of anaylzing trends, while still recognizing that they might exist from time to time. For this reason, I will mostly just address the translation of "Waw" into "kai", "de" and occasionally "other".

In the following, I will demonstrate that the frequency these Greek words appears in the Gospels at a rate which we would expect that is well above normal Greek writing styles, but perfectly in line with a Semitic writing style, had the Greek text been translated from an earlier Semitic text or even if there was a Semitic sub-thought existing beneath the Greek expression. The statistics obtained in this discussion came from computer generated calculations.

3. The LXX translated "waw" (w) to "kai" more frequently than "de"

54.3% of all verses in the Tanakh begin with a Waw. 86% of the time it was translated into Greek as a "kai", and only as "de" about 14% of the time. And the more narrative the book, the more likely a "waw" is translated as "kai" rather than as "de". Here's a breakdown on how it was translated...

Book/Section	% of time a Waw begins a verse	% of time it is translated as "kai"	% of time it is translated as "de"
Torah (first five books)	71%	74%	26%
Joshua thru 2Chronicles (Joshua, Judges, Ruth, Samuel, Kings & Chronicles)	81%	98%	2%
Prophets (Isaiah/Yesh, Ezek, etc) Note: This figure includes Daniel, which should probably be treated separately, so consider these figures preliminary.	45%	~94%	~6%
Writings of Poetry (Psalm/Teh, Prov, Qoh, SOS, Lamentations)	10%	~75%	~25%

So in the narrative books, the preference to translate "Waw" into "kai" over "de" is even stronger. The only 2 books were "de" is used more than "kai" was found in the Poetric Writings of Proverbs (40 "de"'s compared to 22 "kai"'s) and Job (308 "de"'s to 60 "kai"'s). A detailed study of Job finds that the statistics for Job are skewed by the frequent use of "Waw" near the start of a verse, but not at the start, as well as by a less literal translation that sometimes interprets Job more frequently than other translations. This added "interpretative" element caused Job to be "more naturally Greek" than most of the other books.

In the Aramaic sections of the Tanak we see

So here, we see a slight shift towards using "kai" slightly more frequently than "waw" is used in the Hebrew, largely because in some cases we see "ADYN"/"}yd)" translated as "kai" (Dan 3:3, 24, 26, 4:16 , plus other verses), thus "kai" sometimes comes from "Waw" and sometimes comes from "ADYN". A combination of other grammatical words in Aramaic and Greek are also involved in getting from the Aramaic to the Greek that isn't as straightforward as our analysis of Hebrew to Greek translations and some of these statistics can only be explained with a more in-depth analysis of Aramaic that would only explain minor trends and not the major trend being focused on in this study. The re-ordering of words has also skewed some of these statistics, but the trend is still intact of preferring "kai" over "de" against normal Greek style.

So we see here that in both Hebrew and Aramaic, the translators had a preference for translating "waw" as "kai" over "de", even though "de" is used more frequently in Greek. We see this bias towards "kai" as small as 5-3 in Genesis, and as large as almost 10-1 in Ruth. It seemed the translators of Genesis struggled between a literal translation that would express the "waw" as a "kai", thereby helping to preserve for the reader what word it was translated from in Hebrew, and a more natural expression of the use of "de". Often, the more natural expression of "de" won out, while "kai" was used where Greek thinking would tolerate the rendering from Hebrew.

4. Various Comparisons - Greek uses "de" more than "kai"

In seeking to compare the Greek NT to normal Greek style, I selected a few works at semi-random as well as having done a comparison to a wide body of online Greek texts. I was looking for work that were highly narative , since this would most closely match the expected style of the Gospels, Acts, And Revelation, which were the focus of this study since they are the most narative books of the NT and therefore would most frequently exhibit the characteristic under study.

Josephus would be a good comparison to make since it was a Greek writing by a Jew in the first century AD that, like the Gospels, Acts, and Revelation, is highly narrative. Josephus tells us he wrote a first draft of his works in Aramaic first, and then put it in Greek. We don't really know if he 'translated' his Aramaic work to Greek or if he wrote a fresh Greek expression of the same thoughts, or some combination of the two. But his use of expressing "and" or "then" more as "de" than "kai" seems to be in line with most native Greek speakers. He probably did not worry about literally translating his own work, but was more concerned with expressing it in a way that would sound natural in Greek.

A manual count of the first 221 sentences of Josephus' Antiquity of the Jews revealed that 21 sentences began with kai, 97 use de, making a total of 118 or 53% of his first 221 sentences. (Due to slow download and a manual counting effort, an exhaustive calculation was not done, but counting quit after a stable percentage seemed to have been reached.)

In the following, partial counts are for texts that were not easy to download, while full counts tended to be for texts I was able to download in full and run software against in order to get a complete count.

Now the above compositions were chosen because they were expected to be highly narrative, and therefore similar to what would be expected in the Gospels, Acts , and Revelation. All works yields very similar results, and according to Perseus, these results are typical for Greek in general. Perseus gives a frequency of occurence for each word that is useful to compare how typical these samples are. the frequency is based on frequency anywhere in a sentence, not at the beginning as we are analyzing here, but this is still useful in comparing our sample with Greek in general.

Perseus gives the following frequencies for "de" in Josephus and Greek in general as:

So Josephus' use of the word "de" is very typical of Greek in general with almost 0% variance from "normal". For kai we see...

...which again, is very reasonably close to normal, within 11% of normal. Josephus used "kai" about 10% more than most other Greek writers in our sampled survey, and Perseus shows him about 10% above normal here as well.

As a control, one might still be interested in what happens when we examine non-narrative works. In such cases, the preference of "de" over "kai" is still evident, but occurs less frequently. The use of "kai" tends to remain about the same, but the use of "de" drops in proportion to the lack of narration. Thus, "de", more than "kai", tends to be the more natural expression of connecting events through a time sequence in Greek.

Examples are included in the following table from works that contain few to no narration

5. The Greek New Testament Bias: tendency towards use of "kai" over "de" indicates a possible translation.

The most narrative parts of the New Testament are the Gospels, Acts, and Revelation. The following chart shows the result of a computer generated count of what percentage of verses in the various Greek manuscripts begin with a Greek equivalent to the Hebrew "Waw" or a word that "Waw" might be translated to into Greek from an original "waw" source/thought. The percentage of verses in the New Testament rivals the Tanakh for a frequency of the use of "and" or it's equivalent/near-equivalent in Greek, with as many as 1975 of the 2900 verses (or 68%) of the verses in the Synoptic Gospels of the Westcott-Hort text beginning with something that might translate to a "waw" / "AND". The book of Revelation also begins about 70% of its verses in this Semitic fashion. Only the Gospel of John is significantly less than the others, with only about a third of it's sentences beginning this way.

Now compare this to most Greek writings of the first century, or before, and in general, you have to go several sentences or even paragraphs between sentences that begin this way. So this is NOT typical Greek style. Even the book of John exceeds what is normal for native Greek writings.

One of the most eye-popping results of this is that the book of Mark shows a significant change in style from the Westcott-Hort (a majority reading of early 4th-6th century manuscripts) and the Textus Receptus (based on 5th to 16th century manuscripts).

Another significant item is that the story of the adulterous woman in John 8 is not in most early manuscripts of John. Also, it departs from John's style, in that John 8:1-11, 10 of the 11 verses begin with "kai" or "de" near the beginning, with "de" used near the beginning in 9 cases. So the story of the adulterous woman contains this style in 91% of its verses. John uses a significant amount of verses that use oun as a conjunction, which is rare in the other Gospels.

Now let's go back and compare the New Testament with our typical Greek narratives:

All Tanakh texts show an above average percentage of the frequency of beginning a sentence with the word "kai". This is also true of the synoptic Gospels, which use kai and de in more proportion to the translated works of the Tanakh than typical Greek writings. Revelation is so badly skewed towards abnormally high use of "kai" and abnormally low use of "de", one has to wonder seriously if it is not a translated Hebrew work based on this evidence. Acts is reasonably closely aligned with other typical Greek narrations, in that the proportion of the use of the words "kai" and "de" is in line with typical Greek usage. But the Synoptic Gospels and Revelation show a bias against typical Greek and more towards what is observed in Greek that is translated from Hebrew and/or Aramaic.

NT Differences

There are numerous places where we see "kai" used in the WH (Westcott-Hort version of the Greek NT) and "de" in the parallel passage in the TR (Textus Receptus). For example...

Luke 23:44...

WH : "kai hn hdh wsei wra ekth ..."
TR: "hn de wsei wra ekth ..."

Mark 1:14...

WH : "kai meta to paradoqhnai ..."
TR: "meta de to paradoyhnai ..."

Mark 4:5

WH: "kai allo epesen ...."
TR: "allo de epesen ...."

Why? Does this represent two different translations of the same Semitic source?

Some might theorize that scribes "smooth out" a 'rough' reading of "kai" in earlier manuscripts (WH represents earlier texts than the TR) to make it sound more natural in Greek. But the statistics don't support that in that such alleged "smoothing" is not randomly distributed, as will be shown in the following statistics.

There's 34 places in the NT where the WH reads "kai" and the TR reads "de" at the start of a verse. Most (28 of them) appear in the Gospel of Mark. It also happens 4 places in Luke (Luke 9:57, 12:42, 23:24,44) and 2 places in Matthew (20:10 & 28:9). The WH Mark has an abnormally high rate of "kai"'s at the beginning of a verse - much more so than any other book. But Mark stands out as being abnormally low in significant differences between the WH and TR versions (only 5 SIGNIFICANT verses made my list of omissions, etc, verses 16 in John/Yoch, 26 in Luke/Luqa, and 25 in Matthew). But most of Mark is paralleled in other Gospels, so doubtful passages could have been checked against parallel accounts in other gospels easier than for the other Gospels.

There's also 18 places where a "kai" appears in the WH and no corresponding word appears in the TR at the start of a verse. Such occurences are reasonable spread out among the various books (once in Matthew, 3 in Mark, 4 in Luke/Luqa, 5 in John/Yoch, 3 in Revelation, etc). So this was probably either a scribal omission or a "smoothing" to better agree with normal Greek expression.

There's 30 places where "de" appears in the WH and "kai" is in the TR at the start of a verse. They are reasonably spread out among the narrative books (Gospels, Acts, Revelation), but rare anywhere else. It occurs 10 times in Matthew, 8 in Luke, 5 in Acts, 3 in Mark, 2 in John/Yoch, and once in Romans and James/Ya'acov. In 9 of these cases, the Aramaic uses "dk". In 7 cases, the Aramaic has no "Waw" and in 15, either the Peshitta or an OS mss omits a "waw" while the other(s) includes it.

More Translational Evidence

Even more evidence that the Aramaic manuscripts preceeded the Greek manuscripts of the NT is the way "de" is paralleled in the Aramaic manuscripts. The Hebrew and Aramaic "Waw" tend to be translated into various languages as

• "Kai" in Greek, and sometimes "de"
• "And" in English
• "et" in Latin

And the Aramaic "dyn" might get translated a number of ways, but "de" would be preferred over "kai", while "autem" is preferred over "et" in Latin.

We know the Latin Vulgate and the 5 "disputed" books of the Harklean Peshitto (Revelation, 2&3 John/Yoch, 2Peter/Kefah, Jude/Yehudah) where translated from Greek sources. In the 1700+ places where "de" appears in all versions of the Greek mss (WH, TR & Byzantine Majority Text all agree on using "de"), we see...

When All 3 Greek versions Uses "de"

Vulgate uses et 113 times (6%) autem 1373 times (75%) neither 119 times (18%) Harklean Peshitto (in 5 disputed books) uses Waw 0 (0%) DYN 25 (96%) neither 10 (4%)

Now if the Peshitta and Old Syriac where a translation from the Greek, we might expect to see something similar to the Harklean in the Peshitta and Old Syriac. But it's not even ballpark. Instead, here's the full chart showing that the Peshitta and Old Syriac don't even come close. You'd come closer to predicting what frequency the Peshitta would use "dyn" over "waw" by using the Vulgate, and even that would be way off the mark!

When Greek Uses "de"

Vulgate uses et 113 times (6%) autem 1373 times (75%) neither 119 times (18%) Harklean Peshitto uses Waw 0 (0%) DYN 25 (96%) neither 10 (4%) Old Syriac S uses Waw 355 (42%) DYN 252 (30%) neither 148 (28%) Old Syriac C uses Waw 252 (45%) DYN 163 (29%) neither 58 (26%) Peshitta uses Waw 605 (34%) DYN 976 (54%) neither 139 (12%)

In the above, "neither" includes HYDYN (}ydyh) 6 times in the Peshitta, twice in the OSS and once in the OSC. Most of the time, there's no word there that correlates to the "de".

In translating from Greek to Aramaic, we see the Harklean give a clear preference to using "DYN" over "Waw", in near unanimous style. But we see the Old Syriac preferring "Waw" here and the Peshitta prefering "DYN" by only a small margin compared to the Harklean. However, we do see it as normal to translate "Waw" into either "kai" or "de" depending on which fits better in the Septuagint. So it seems an Aramaic to Greek translation more plausibly explains this correlation than a Greek to Aramaic translation.

The distribution of "Waw" versus "DYN" in the Old Syriac is in line with the Aramaic parts of the Tanakh. The Peshitta is in a slightly different dialect and distribution of use, but both are within a reasonable range of each other and drastically different from results of the Harklean Peshitto.

So overall, we see that the use of conjunctions, focused mostly on when it is used to connect time events, is used in a proportion that is not normal for Greek, but very normal for Greek translated from Aramaic or Hebrew. The Aramaic Peshitta and Old Syriac use conjunctions in a proportion that is not normal for a Greek to Aramaic translation.