For an Answer Home Mars Hill  Index Bibliography Glossary
The Bible Gateway The Blue Letter Bible The Greek New Testament (NA26) Greek & Hebrew Lexicons

 
powered by FreeFind

Mars Hill  Apologetic Discussions

 

 

Rob Bowman and "GK" on Luke 23:43

Regrettably, I do not have a copy of the entire dialog between Robert Bowman and a Jehovah's Witness apologist whose initials are GK.  This dialog took place on CARM's Jehovah's Witness discussion board in late 2004, and continued on Rob's Evangelicals and JWs Yahoo board in 2005.

We begin with GK's reply to a previous post by Bowman.  GK had written a paper in which he attempted to demonstrate from the BDF Greek Grammar that "today" in Luke 23:43 must modify the preceding verb ("I say to you") and not the one following, as in the traditional punctuation.

 

============================

GK's POST TO ROBERT BOWMAN

============================

Dear Rob,

Thank you for your list. One thing that you did not address was my comment about phrases being separated out by conjunctions like ?t?. I also had not included my references to BDF because my original post was made to someone who probably does not have access to that reference work. I have included the reference and also the principles involved in how this word order would be used to disambiguate expressions. Also I have isolated all the verses in the GNT which have a particular pattern that matches Luke 23:43, that of an aµ?? phrase followed by a specific time period. Significantly all these latter phrases are the words of Jesus Christ which makes them more relevant that all the others.

According to BDF §474.2 "An adverb which further defines an adjective (or verb) also takes second position." That rule is not without exception because this word order is for the purpose of disambiguating expressions in order to avoid misunderstanding. Thus the rule is useful particularly when determining which verb an adverb modifies. Luke 23:43 fits this profile.

All eleven occurrences of s?µe??? in the book of Luke follows the verb which it modifies (Luke 2:11, 5:26; 12:28; 13:32; 22:34, 61; 23:43) unless it is governed by a conjunction which forces it into a separate clause (Luke 4:21; 13:13; 19:5, 9.) 

If one includes the nine occurrences in the book of Acts because it was written by Luke, the picture does not change significantly. In five occurrences the adverb follows the verb (Acts 19:40; 20:26; 22:3; 24:21; 26:29) unless governed by a conjunction which forces the term into a separate clause (Acts 4:8-9; 26:26.) One example where the adverb sits in the middle of a participle/infinitive (Acts 26:2) should be disregarded as well as one example which is a quotation of the LXX which follows the word order of the Hebrew text (Acts 13:33.)

Thus all twenty occurrences in Luke/Acts conform to a general rule as published in BDF with the exception of six occurrences which are separated by particles or conjunctions. Even though there are six exceptions which are accounted for by Greek grammar, the following comparison of verbage equivalent to Luke 23:43 of the words of Jesus Christ has no exceptions.

Getting closer to the verbage found in Luke 23:43, when one considers all verses in the Greek Scriptures which contain a phrase like aµ?? ?e?? s?? followed by a definite time period such as s?µe??? these all verses support these observations. (Matthew 26:34 Mark 14:30, John 5:25, Luke 23:43.) Significantly these are all sayings of Jesus which makes them all the more relevant to the question of where the comma goes in English translations of the Greek. With the exception of Luke 23:43, the conjunction ?t? separates the time period (s?µe???, ??a, ???t?) from the verb which precedes it and thus disambiguates the phrase.

Luke 23:43 does not contain a conjunction and therefore s?µe??? should modify the preceding verb. This is supported by BDF, the statistics of all s?µe??? found in Luke-Acts, but more importantly by the words of Jesus himself when all similar verses are considered. These also illustrate how the ?t? would be used to disambiguate these expressions.

~GK

============================

ROBERT BOWMAN'S REPLY TO GK

============================

Dear GK,

You wrote:

>>Thank you for your list. One thing that you did not address was my comment about phrases being separated out by conjunctions like hOTI.<<

That comment has no bearing on the matter at hand. Had John wanted to insert hOTI, he could have done so either before or after SHMERON. That is, he could have written:

“Amen to you I say today that with me you shall be in Paradise.”

Or, he could have written:

“Amen to you I say that today with me you shall be in Paradise.”

However, he chose to write neither of those two things. Either one would have left absolutely no room grammatically for misconstruing which verb the adverb SHMERON modifies. As it stands, we cannot settle the question definitively based on grammar alone.

I will have more to say below about the use of hOTI. Let us now turn to your three arguments.


A. GK’S FIRST ARGUMENT: ADVERBS “TAKE SECOND POSITION”

You wrote:

>>According to BDF <sect.> 474.2 "An adverb which further defines an adjective (or verb) also takes second position." That rule is not without exception because this word order is for the purpose of disambiguating expressions in order to avoid misunderstanding. Thus the rule is useful particularly when determining which verb an adverb modifies. Luke 23:43 fits this profile.<<

Well, let’s look at BDF a bit more closely:

“474. The position of nouns and adverbs. (1) The rule is that an anarthrous adjectival attributive usually follows its substantive. (2) An adverb which further defines an adjective (or verb) also takes second position. (3) Mt particularly has the habit of placing adverbs after imperatives while he places them before indicatives.”

Later on the same page, we read the following note for point 474.2.

“(2) Mt 4:8 hUYHLON LIAN, 2:16 EQUMWQH LIAN, cf. MELAS DEINWS Aelian, NA 1.19, ERHMOS DEINWS 4.27. But also LIAN (om. D) PRWI Mk 16:2, LIAN GAR ANTESTHi, 2 T 4:15.”

BDF gives two examples here of LIAN following the word it modifies and two examples of LIAN preceding the word modified. If this doesn’t give the reader a clue that word order for adverbs is flexible in Greek, the observation in 474.3 that Matthew “particularly” puts the adverbs after imperatives but before indicatives ought to make this clear.

If we back up a bit in BDF, we find additional statements emphasizing the relative fluidity of Greek word order. Even though there is “something like a normal word order,” the positions regarded as “normal” “are by no means mandatory” (sect. 472).

Similarly, Turner, while stating, “An adverb usually follows the adj. or verb which it determines, in NT” (James Hope Moulton, A Grammar of New Testament Greek, Vol. III, Syntax, by Nigel Turner <T & T Clark, 1963>, 16.4, p. 227), cites some exceptions, notes the same pattern in Matthew mentioned in BDF, and points out, “Even in the class. period a considerable flexibility obtained, under the influence of rhythm or emphasis, and the conversational style differed markedly from the rhetorical” (16.4, p. 228).

We should especially notice Turner's reference to emphasis as affecting the position of adverbs. BDF is more explicit: “Any emphasis on an element in the sentence causes that element to be moved forward” (sect. 472). The principle that “the emphatic word comes at or near the beginning of the sentence” (Moule, An Idiom Book of New Testament Greek, 2d ed., 24.1 <p. 166>) is a commonplace of Greek syntax.

Robertson also observes “the freedom of the Greek” especially “in the order of words in the sentence” and points out that emphasis “is one of the ruling ideas in the order of words” (417). Like BDF and Turner, Robertson states, “The adverb generally has second place...but not always” (419); and he actually cites BDF and some of its examples.


B. GK’S SECOND ARGUMENT: “TODAY” ALWAYS MODIFIES THE VERB PRECEDING EXCEPT WHEN A CONJUNCTION FORCES IT TO MODIFY THE VERB FOLLOWING

You wrote:

>>All eleven occurrences of SHMERON in the book of Luke follows the verb which it modifies (Luke 2:11, 5:26; 12:28; 13:32; 22:34, 61; 23:43) unless it is governed by a conjunction which forces it into a separate clause (Luke 4:21; 13:13; 19:5, 9.)

If one includes the nine occurrences in the book of Acts because it was written by Luke, the picture does not change significantly. In five occurrences the adverb follows the verb (Acts 19:40; 20:26; 22:3; 24:21; 26:29) unless governed by a conjunction which forces the term into a separate clause (Acts 4:8-9; 26:26.) One example where the adverb sits in the middle of a participle/infinitive (Acts 26:2) should be disregarded as well as one example which is a quotation of the LXX which follows the word order of the Hebrew text (Acts 13:33.)

Thus all twenty occurrences in Luke/Acts conform to a general rule as published in BDF with the exception of six occurrences which are separated by particles or conjunctions.<<

First of all, you seem to have skipped Acts 27:33. (And I even supplied you with a nice list!) In Acts 27:33, the verb preceding SHMERON is a verb of speech, LEGWN, and there is no intervening conjunction, but no one would ever construe SHMERON as modifying LEGWN. So, this verse flatly contradicts your claim that Luke uniformly places SHMERON after the verb it modifies unless a conjunction preceding SHMERON forces it to modify the verb that follows.

Your citation of Acts 26:26 does not appear to be correct. SHMERON does not appear in Acts 26:26 in the UBS, NA, or WH Greek texts. You also cited one verse incorrectly as Luke 13:13; the correct reference is 13:33. In Luke 13:33, the adverb SHMERON sits between DEI and the infinitive POREUESQAI.

Your assumption is that in texts where the modified verb follows SHMERON, without the conjunction we would most naturally construe SHMERON as modifying the verb that precedes it. But this is not so. If Luke 4:21 had read, “And he began to say to them today this scripture has been fulfilled,” the sense would clearly have been still that “today” modified “has been fulfilled.” Similarly, if we had Luke 19:9 without hOTI, we would not construe it to mean that Jesus spoke to Zaccheus “today”; we would still understand “today” to be part of the words Jesus spoke to him. In Luke 19:5, had Luke omitted the conjunction GAR, we would hardly have concluded that Jesus was telling Zaccheus to come down quickly “today”! In Acts 4:9, the conjunction EI serves to introduce the subordinate clause; its absence (were that possible) would not make it possible to construe SHMERON with any verb other than the one that follows, “are being examined.” (Again, no one would ever think to construe it with “said” in verse 8.)

It is true that in most of the Lukan texts in which SHMERON precedes the verb, a conjunction precedes SHMERON. But there is a simple reason for this phenomenon: in these texts, SHMERON is at or near the beginning of the sentence, and Luke begins most of his sentences with a conjunction!

Likewise, in none of the texts where SHMERON follows the verb it modifies is there any ambiguity. In Luke 2:11, the verb following SHMERON is marked as belonging to a new clause by the relative pronoun hWS (“who”), though the intervening direct object SWTEROS would make this clear. SHMERON is clearly the last word of the sentence in Luke 5:26 (verse 27 begins, “And after these things…”). In Luke 12:28 and 13:32, SHMERON follows the verb it modifies, but then the temporal adverb AURION (“tomorrow”) appears before its verb. These two texts show that the placement of the temporal adverb can be more a matter of style or emphasis than of grammar. In Acts 24:21, the verb following SHMERON is marked as belonging to a new sentence by the conjunction DE (but we would not need it to know how to construe SHMERON). In Acts 26:29, we know that the word SHMERON modifies the verb that precedes it because of the sense of the sentence, not because of word order (“I would wish to God that whether in a short or long time…all who hear me today might become as I am except for these chains”). The temporal qualification “whether in a short or long time” makes it clear that “today” qualifies “all who hear me.” In Luke 22:61, SHMERON is textually dubious; but even here, the sense makes it clear that SHMERON modifies “crows” and not “deny”; and we would know this with certainty from verse 34.

On the other hand, in Acts 20:26 the omission of the conjunction hOTI might result in some ambiguity. Here, though, the reader might mistakenly construe SHMERON as modifying the verb that *follows*, not the one that precedes: “Therefore I testify (MARTUROMAI) to you on this day today that I am innocent (EN TH SHMERON hHMERA hOTI KATHAROS EIMI) of the blood of all people.” Grammatically and semantically, if one omitted hOTI, there would be no problem construing Paul as saying, “Therefore I testify to you, I am innocent on this day today of the blood of all people.” Usage and idiom, not grammar, would be our only basis for rejecting such an exegesis. Yet it is worth noting that the one place where hOTI makes the relation of SHMERON clearer, it clarifies that SHMERON modifies the verb that precedes it, not the verb that follows it.


C. GK’S THIRD ARGUMENT: THE CONJUNCTION hOTI ALWAYS SEPARATES “AMEN I SAY TO YOU” FROM REFERENCES TO A TIME PERIOD FOLLOWING, EXCEPT IN LUKE 23:43

You wrote:

>>Even though there are six exceptions which are accounted for by Greek grammar, the following comparison of verbage equivalent to Luke 23:43 of the words of Jesus Christ has no exceptions.<<

As I have demonstrated, the so-called “exceptions” would be clear without the conjunctions on which you place such emphasis, and at least one of these texts (Acts 27:33) has no conjunction preceding SHMERON.

You wrote:

>>Getting closer to the verbage found in Luke 23:43, when one considers all verses in the Greek Scriptures which contain a phrase like AMHN :EGW SOI followed by a definite time period such as ¦Ň¦Ç¦Ě¦Ĺ¦Ń¦Ď¦Í these all verses support these observations. (Matthew 26:34 Mark 14:30, John 5:25, Luke 23:43.) Significantly these are all sayings of Jesus which makes them all the more relevant to the question of where the comma goes in English translations of the Greek. With the exception of Luke 23:43, the conjunction ¦Ď¦Ó¦É separates the time period (SHMERON,WRA,NUKTA) from the verb which precedes it and thus disambiguates the phrase.<<

(I have retained the typographical oddities, but I think the three odd-looking words are LEGW, SHMERON, and hOTI.)

You cite here a grand total of three (!) texts in which a reference to a time period follows the AMHN formula with hOTI intervening. Even if what you said about all three texts were correct, three instances is simply too small a data pool on which to base a grammatical generalization.

In Mark 14:30, Jesus says, “Amen I say to you (AMHN LEGW SOI) that you, today (hOTI SU SHMERON), this very night (TAUTHi THi NUKTI), before a rooster crows twice, will deny me three times.” In this text, the pronoun SU would make it clear that SHMERON is part of a new clause even without the use of hOTI. In fact, Mark may have used hOTI here for the stylistic reason of separating SOI and SU. (Mark uses SU, and places it first after the conjunction, for emphasis.) Then again, Mark usually uses hOTI after the AMHN introductory formula (nine times, Mark 3:28; 9:1, 41; 11:23; 12:43; 13:30; 14:18, 25, 30; four times without hOTI, 8:12; 10:15, 29; 14:9).

Matthew 26:34 has a simpler version of the same sentence: “Amen I say to you (AMHN LEGW SOI) that (hOTI) on this very night (EN TAUTHi THi NUKTI), before a rooster crows, you will deny me three times.” It is likely that Matthew has simplified Mark’s more cumbersome sentence while retaining his use of hOTI. (All but one of Matthew’s “Amen” sayings that lack hOTI are in material not found in Mark; Matthew 21:21 is the only exception.) Thus, the use of hOTI in Matthew 26:34 does not demonstrate any concern about making the role of the adverbial phrase “this very night” grammatically unambiguous.

On John 5:25, a noun (hWRA, “hour”), not an adverb, expresses the “time period,” and the sentence would be unambiguous without the conjunction hOTI: “Amen, amen I say to you, an hour is coming and now is (ERCETAI hWRA KAI NUN ESTIN).” This text is by far the weakest of the three in terms of relevance to your argument.

Thus, of these three texts, only one would be grammatically ambiguous without the conjunction hOTI (Matt. 26:34), and even this text would not be semantically ambiguous. The other two are grammatically unambiguous, and one is irrelevant to boot (since the expression of time is a noun, not an adverb).

If we are going to consider all similar sayings of Jesus, let us really do so. There are some 74 AMHN LEGW sayings of Jesus in the Gospels. Half of these sayings use hOTI following the AMHN LEGW formula and half of them do not. In none of these 74 sayings (unless one counts Luke 23:43) does any adverb appear that modifies LEGW.

Expanding our study, we may consider all of the other sayings of Jesus using some variation of LEGW SOI or LEGW hUMIN (without AMHN) as an introductory expression. There are some 70 of these sayings. In none of them does SHMERON or any other temporal adverb function as part of the introductory expression, modifying LEGW. The only possible exception is Matthew 19:24, where Jesus begins his saying with the words, “But again I say to you” (PALIN DE LEGW hUMIN). However, PALIN here functions as a conjunction—and it precedes the verb rather than following it. As Turner says (with reference to Mark), when PALIN “occurs at the beginning of its phrase, it may be reasonable to take it as a mere conjunction” (Turner, Syntax, 16.4, p. 229). In any case, if Matthew 19:24 is an exception, it is one where the adverb appears *before* the verb.

The only other LEGW sayings of Jesus in which any sort of adverb modifies LEGW are four texts in which Luke uses ALHQWS (“truly”) or EP’ ALHQEIAS (“in truth”) as part of Jesus’ introductory speech form (Luke 4:25; 9:27; 12:44; 21:3). (In three of these, the adverbial precedes LEGW, once again shifting the evidence from the position of the adverb in favor of the conventional translation of Luke 23:43.) The expressions “truly” and “in truth” are unmistakably paraphrases for Jesus’ own idiomatic use of AMHN. We know this for two reasons. First, the Greek AMHN is simply a transliteration of the Hebrew word ’amen, which meant “so be it” or “truly.” In fact, modern English translations of Jesus’ AMHN sayings (including the NWT) commonly translate AMHN using the English word “truly.” Second, in three of these four texts the saying is paralleled in Matthew or Mark or both, where the saying uses Jesus’ customary AMHN and not ALHQWS or an equivalent (Luke 9:27, cf. Matt. 16:28 and Mark 9:1; Luke 12:44, cf. Matt. 24:47; Luke 21:3, cf. Mark 12:43). Thus, in these four texts, we know to construe ALHQWS or its equivalent as part of the introductory formula because it is Luke’s paraphrase of Jesus’ idiomatic use of AMHN.


REVIEW OF THE ARGUMENTS TO DATE

You wrote:

>>Luke 23:43 does not contain a conjunction and therefore SHMERON should modify the preceding verb. This is supported by BDF, the statistics of all SHMERON found in Luke-Acts, but more importantly by the words of Jesus himself when all similar verses are considered.<<

A nice review of your arguments; here is a review of my rebuttals:

1. BDF and other grammars state that adverbs usually follow the verb they modify, but other factors, especially emphasis, may cause the adverb to move in front of the verb.

2. In Luke-Acts, SHMERON appears both before and after the verb it modifies, and usage, idiom, and the sense of the passage always make it clear which verb it modifies.

3. One Lukan text, Acts 27:33, which you overlooked, falsifies your claim that Luke uniformly places SHMERON after the verb it modifies unless a conjunction preceding SHMERON forces it to modify the verb that follows.

4. In the only Lukan text in which hOTI makes the relation of SHMERON clearer, it clarifies that SHMERON modifies the verb that precedes it, not the verb that follows it (Acts 20:26).

5. Although you claim support from “all similar verses” of the sayings of Jesus, in fact you cited only three sayings of Jesus, not enough on which to base an inductive argument. Moreover, your claim that these three verses use hOTI to “disambiguate” which verb a temporal expression modifies is false. Two of these verses would be grammatically unambiguous in this respect even without hOTI, and the third would be contextually unambiguous without hOTI.

6. A review of all 144 “I say (LEGW) to you” sayings of Jesus in the Gospels reveals that he never uses SHMERON to modify LEGW. Jesus never uses any adverb to modify any of the 74 AMHN LEGW sayings. The only occurrences of an adverb of any kind in any of Jesus’ 144 introductory LEGW sayings are PALIN in Matthew 19:24 (which is probably to be construed as a conjunction) and the four times that Luke uses “truly” or “in truth” to paraphrase AMHN. Of these five possible adverbial elements in Jesus’ introductory LEGW speech forms, four of them precede the verb LEGW. These data constitute a solid body of evidence for the conclusion that we should construe the adverb SHMERON in Luke 23:43 with the verb that follows it.



In Christ's service,
Robert M. Bowman, Jr.
Center for Biblical Apologetics
http://www.biblicalapologetics.net


============================

THE DEBATE CONTINUES...

============================

This was the last post dealing with the substantial issues of the debate on CARM as of January, 2005.

In March 2005, I was contacted by GK, saying that he had revised his paper, based on his dialog with Rob.  He later submitted it to Rob's public Evangelicals and JWs discussion board for Rob's response.  

Click here for a copy of GK's revised paper.  

I should note that GK has revised his paper numerous times over the course of this discussion.  The copy I have linked is current as of March 15, 2005.  I cannot promise to keep current with GK's continuing revisions, as it seems that whenever Rob or another poster points out flaws, GK quickly attempts to shore up his argument with additional qualifications and limitations.  I would comment here Bowman has already demonstrated several exceptions to GK's "rule," and continues to do so in his reply, below.  But GK's "rule" would be unpersuasive even without these examples.  It is simply exegetically illegitimate to demand parallel examples which are so narrowly specified that one would not expect to find many, if any, examples.  Nothing in the nature of NT Greek requires that SHMERON (or any other adverb, for that matter) modify verbs in its clause in one way and not another.  GK has based his exegetical judgment on an artificial distinction that has no real significance in the actual use of the language.  This point is driven home both by Bowman's response, below, and by Luis C. Reyes, here.

And now, Robert Bowman's response:

============================

ROBERT BOWMAN'S NEXT REPLY TO GK

============================

GK,

We have gone back and forth on this subject of Luke 23:43 many times now.
You have revised your grammatical definitions and rules so many times that I
have lost count; I think it’s probably in the neighborhood of half a dozen
times or more. As you plug up one hole in your argument, one or two more
always seem to spring a leak. That pattern is still holding.

Your paper is now entitled, “The adverb SHMERON in relation to its verb in
Biblical Greek when found in Direct Discourse.” This title and your argument
presupposes what you never substantiate, namely, that there is some clear
difference in the syntactical possibilities for SHMERON when used in direct
discourse than when it is not so used. Merely cutting down your data pool to
this subset of usage of the adverb in order to try to salvage a ‘rule’ is
not sound linguistic procedure. This is only one of several serious problems
with your argument.


YOUR “GENERAL RULE”

After some introductory comments, you present a “general rule” on this
subject as a place to begin, even though you recognize the rule has
exceptions:

<< Greek grammars are not silent on the subject. In Koine Greek
sentences[10] an adverb which further defines a verb takes second
position.[11] Thus when an adverb directly follows a verb it generally
modifies the verb it follows[12]. >>

The statement that “an adverb which further defines a verb takes second
position” is virtually a quote from Blass-Debrunner-Funk (BDF), which you
quote as follows in footnote 11:

<< BDF §474. *_The position of nouns and adverbs_*. … (2) An adverb which
further defines an adjective (or verb) also takes second position. >>

Your explanatory comment is the actual rule that you derive from BDF,
namely, that “when an adverb directly follows a verb it generally modifies
the verb it follows.” The term “directly” is a crucial qualification
because, as you explain in a separate post, you once took “second position”
in BDF to mean that the adverb follows the verb one or more words later, but
(after this form of the rule didn’t hold up) you later concluded that by
“second position” BDF means that the adverb follows “directly” or
immediately after the verb with no other words intervening (except
postpositives and similar words of relatively inflexible syntactical
position).

Unfortunately, BDF does not offer a definition for its expression ‘second
position.’ I can see why you interpret it as you now do (and of course why
you would *want* to interpret it that way). I am not sure what BDF means by
that expression. However, if it means what you say it does, that
generalization simply does not hold up for the adverb SHMERON. Not only are
there exceptions to that ‘rule,’ the fact is that the ‘rule’ as you
understand it more often than not does *not* apply.

Suppose we take the statement in BDF 474.2 as you suggest and further that
we qualify it as a generalization, meaning that if true it would apply, at
the very least, more often than not. Then BDF 474.2 would mean that more
often than not, when an adverb modifies a particular adjective or verb, it
follows immediately after that adjective or verb. In other words, the ratio
S/T would by this definition be more than 1/2 (0.5), where S and T have the
following definitions:

S = Number of adverbs immediately following the verb or adjective they
modify
T = Total number of adverbs modifying a verb or adjective

Now, what are the facts in this matter? You stated in your paper that in the
corpus of biblical Greek the word SHMERON occurs 332 times in 314 verses.
This is not quite right. (For what follows, I am dependent on Bible Works
5.0. A note to those checking biblical references: all references given here
use the Greek text versification as found in Bible Works, which is often
different from the English Bible in such books as the Psalms and Jeremiah.)
If you ask Bible Works 5.0 (which you apparently also use) to generate a
list of all occurrences of SHMERON in biblical Greek, you will indeed get
332 hits in 314 verses, but these count eight duplicate references. These
include seven variants in Judges and Daniel (Jda. 9:18; 11:27; 19:9; 21:3,
6; Dat. 3:37, 40) and the one occurrence assigned to 1 Sam. 4:15 and 4:16 in
different versions. These eight duplicates need to be dropped, yielding 324
occurrences in 306 verses. Of these, 25 occurrences do not modify any
expressed verb:

Gen. 19:37, 38; 26:33; 40:7; 42:13, 32; Num. 22:30; Deut. 8:18; 11:4; Josh.
14:10; Ruth 4:9; 1 Sam. 9:12; 12:5, 17; 1 Ki. 5:21; 8:15, 56; Sir. 10:10;
Isa. 37:3; Bar. 3:8; Matt. 16:3; Acts 19:40; Heb. 3:13; 4:7a; 13:8

Since these 25 occurrences have no relevance to our topic, I set them aside,
leaving us so far with 299 occurrences in 282 verses (since we must still
count Heb. 4:7b) to consider. Since Luke 23:43 is disputed, I will also set
it aside for now, realizing of course that we want to determine just how to
categorize its placement of SHMERON. Now we are down to 298 undisputed
occurrences of the adverb (in 281 verses).

I have analyzed these 298 occurrences into several categories that need to
be distinguished if we are to assess the evidence properly and completely:

Category 1: SHMERON immediately follows the verb it modifies. In order to
accommodate your argument, I will include in this category instances where
SHMERON modifies two verbs, one of which it immediately follows. I also
count two occurrences that you dropped because they were not in direct
discourse (this also helps you out at this point of the argument). Based on
these parameters, I come up with a list of 72 occurrences in this category
(note, though, this list differs somewhat from your list of 72 occurrences
in footnote 18 of your paper). Here is my list:

Gen. 22:14; 24:42; 30:16; 41:9; Exod. 2:18; 16:25a; Deut. 1:10, 39; 2:18;
4:38, 39; 5:3; 9:1, 3, 6; 11:2; 20:3; 26:3, 17; 31:2; Jos. 14:11; 22:16,
18a, 18b; Jdg. 21:3, 6; Ruth 2:19a, 19b; 1 Sam. 9:20; 10:2; 14:44; 17:36,
45; 21:6; 24:12, 20; 26:8; 27:10; 2 Sam. 3:39; 6:20a, 20b; 11:12; 19:6a, 7a,
7b, 8, 21, 36; 1 Ki. 1:25, 48; 2:31; 22:5; 2 Chr. 18:4; 35:21; Neh. 9:36;
Est. 5:4b; 1 Ma. 3:17; 5:32; 6:26; 9:30, 44; Isa. 10:32; Jer. 41:15; Matt.
21:28; 27:19; Lk. 12:28; 13:32; 22:34, 61; Acts 22:3; 24:21; 26:2

I have 2 Samuel 19:6a, 7a, 7b, whereas you have 19:6a, 6b, 7; this may be
simply a versification difference. Let me make some other comments comparing
the above list to the list in footnote 18 of your paper.

EXCLUSIONS: Judges 11:27 does not count in category 1 because the participle
KRINWN that appears immediately before SHMERON actually modifies the noun
KURIOS (“The Lord, the judging one”); SHMERON does not modify KRINWN but
rather KRINAI, which appears before KURIOS. You counted Judges 11:27, 21:3,
and 21:6 twice each by counting the variant textual tradition labeled
“Judges (A)” in Bible Works (as explained above). I do not count Esther 5:4a
because SHMERON must be treated either as the subject of the verb ESTIN
(“Today is my great day”) or as the subject complement (“My great day is
today”); either way, SHMERON does not modify the word it immediately follows
(the adjective EPISHMOS, “great”). And again, for now I am setting Luke
23:43 aside as debated. The other five should definitely be dropped from
your list in footnote 18.

ADDITIONS: As I noted above, I am including occurrences not in direct
discourse (Gen. 22:14; 2 Sam. 19:8). I count 2 Samuel 11:12 because ENTAUQA
KAI GE SHMERON is a compound adverbial phrase immediately following the
verb. I count 1 Kings 22:5 and 2 Chronicles 18:4 because DH (which precedes
SHMERON) is a particle with adverbial force. Isaiah 10:32 must be counted
because the verb it modifies actually is the last word in verse 31 (I almost
missed this!). These last four occurrences are “freebies” that you can add
to your list in footnote 18.

So, my list excludes six texts that you listed (Jda. 11:27; 21:3, 6; Jdg.
11:27; Esth. 5:4a; Luke 23:43) and includes six texts that you did not list
(Gen. 22:14; 2 Sam. 11:12; 19:8; 1 Ki. 22:5; 2 Chr. 18:4). Hence my list has
72 occurrences as compared to your 72, but partly because our lists have
different parameters and partly because I disagree with some of your
inclusions.

Please notice that out of 298 occurrences of SHMERON modifying one or more
verbs, only 72 place SHMERON immediately following a verb that it modifies.
That is a ratio of .24, or a little less than one-fourth, and thus a long,
long way shy of the .5 ratio needed to confirm your generalization with
regard to adverbs usually taking second position in relation to the verbs
they modify. Even if we added Judges 11:27 and Esther 5:4a and included Luke
23:43 in the pool of texts for the sake of argument, the ratio would be 75
out of 298, which is just short of .25.

Please note that in offering this statistical analysis I am not formulating
a statistically-based exegetical argument of my own. I am merely responding
to your argument and laying down a comprehensive view of the syntax of
SHMERON.

Category 2: SHMERON follows two or more words after the verb or verbs it
modifies. That is, the verb appears, then one or more words (not counting
postpositives), and then SHMERON. Here is the list I came up with for this
category:

Gen. 4:14; 21:26; 24:12; 25:31, 33; 30:32; 31:43, 46; 41:41; 47:23; 50:20;
Exod. 5:14; 14:13a, 13b; 16:25b; 19:10; 32:29; Lev. 10:19b; Deut. 4:1, 2, 8,
26, 40; 6:2, 6, 24; 7:11; 8:1, 11, 19; 10:13; 11:7, 8, 13, 22, 26, 27, 28,
32; 12:8, 11, 14; 13:1, 19; 15:5; 19:9; 26:18; 27:1, 4, 10; 28:1, 13, 14,
15; 29:9, 11, 14a, 14b; 30:2, 8, 11, 15, 16, 18, 19; 31:21, 27; 32:46; Jos.
7:19, 25; 24:15, 27, 31; Jdg. 6:17; 9:18; 11:27; Ruth 3:18; 4:10, 14; 1 Sam.
4:3, 7, 16; 9:19, 27; 14:28, 30, 38, 41; 15:28; 16:5; 17:10, 46; 20:27;
21:3; 24:11, 19a, 19b; 25:32, 33, 34; 26:19, 23, 24; 30:13, 25; 2 Sam. 3:8a,
8b; 11:12; 18:31; 19:6b, 7c, 23a; 1 Ki. 1:51; 8:28, 56; 21:13; 2 Ki. 4:23;
6:28, 31; 1 Chr. 29:5; 2 Chr. 6:19; Neh. 1:6, 11; Jdt. 12:18; 1 Ma. 4:10;
7:42; 10:20; 3 Ma. 5:20; 6:13; Odes 7:37, 40; Sir. 38:22; Jer. 1:10; Sus.
1:55; Dan. 3:37, 40; Matt. 6:11; Lk. 2:11; 5:26; Acts 26:29

This is by far the largest category, with 140 occurrences (nearly half). If
we add categories 1 and 2 together, we get 212 occurrences out of a total
298 occurrences. That is a ratio of .71, or somewhat less than
three-fourths, which though not overwhelming is well over the minimal .5
threshold ratio to make the generalization in BDF defensible with regard to
SHMERON. Again, I’m not speculating as to what the authors, translators, and
editors of BDF had in mind; I’m simply saying that with regard to SHMERON,
it is credible to say that it (usually) follows somewhere after the verb it
modifies, but it is frankly wrong to say that it (usually) follows
*immediately* after the verb it modifies.

Category 3: SHMERON is part of a prepositional phrase that follows the verb
it modifies. In such constructions, SHMERON does not directly modify the
verb, nor of course can it directly follow the verb. Still, it is part of a
word group that modifies a verb, so we can consider that usage. I count 36
occurrences in this category:

Gen. 35:4, 20; Deut. 4:4; Josh. 4:9; 6:25; 9:27; 10:27; 13:13; 22:29; 1 Sam.
29:6; 2 Chr. 35:25; 1 Es. 8:74, 86; Neh. 5:11; Jdt. 6:2; 7:28; 8:12, 28;
13:17; Tob. 7:12; 1 Macc. 10:30a, 30b; 13:39; 16:2; Sir. 47:7; Isa. 58:4;
Jer. 1:18; Ezek. 2:3; 20:29, 31; 24:2; Matt. 11:23; 27:8; 28:15; Acts 20:26;
Rom. 11:8

Categories 1, 2, and 3 exhaust all of the occurrences where SHMERON follows
the verb that it, or a phrase of which it is a part, modifies. These three
categories add up to 248 out of 298 occurrences, or a ratio of .83, about
five-sixths.

Category 4: SHMERON immediately precedes the verb or verbs that it modifies.
I find 30 such occurrences:

Lev. 10:19a; Josh. 22:31; 1 Sam. 10:19; 14:45; 22:15; 25:10; 2 Sam. 14:22;
15:20a, 20b; 16:3; 19:23b, 23c; 1 Ki. 2:24; 18:15; 2 Ki. 2:3, 5; Esth. 5:4a;
Jdt. 13:11; Tob. 6:11; 1 Macc. 2:63; Ps. 2:7; Prov. 7:14; Sir. 20:15; Matt.
6:30; Luke 4:21; Acts 4:9; 13:33; Heb. 1:5; 5:5; James 4:13

It is worth noting that when SHMERON is immediately adjacent to the verb it
modifies, 70% of the time it follows the verb and 30% of the time it
precedes the verb (72 occurrences in category 1 and 30 occurrences in
category 4). Given these statistics, one simply cannot assume that if
SHMERON is positioned between two verbs it must if at all possible modify
the verb it follows. This may be true a majority of the time, but the
evidence weighs against assuming it is true all of the time.

Category 5: SHMERON precedes the verb or verbs that it modifies, with one or
more words intervening between SHMERON and the verb. I find 10 occurrences
in this category:

Lev. 9:4; 11:13; Esth. 1:18; Ps. 94:7; Luke 19:5, 9; Acts 27:33; Heb. 3:7,
15; 4:7b

Note that I am not counting Luke 23:43, even though I would place it in this
category, so as not to beg the question.

The 40 occurrences in categories 4 and 5, out of 298, make a ratio of .13,
or a bit more than one-eighth. That is enough to invalidate any
*presumption* that SHMERON modifies a verb that it follows. We could also
count the occurrences where SHMERON modifies verbs both preceding and
following (already counted in category 1). This would add at least four more
occurrences (Josh. 14:11; 2 Chron. 35:21; Luke 13:33; Acts 26:2) and bump up
the ratio to 44 out of 298, or .15.

Category 6: SHMERON is part of a prepositional phrase that modifies a verb
that follows that phrase (compare category 3). I find 10 of these
occurrences:

Ex. 13:4; Josh. 5:9; 22:3; 1 Sam. 26:21; 2 Chr. 10:7; Jdt. 8:29; Odes 11:19;
Isa. 38:19; 2 Cor. 3:14, 15

If we add categories 4, 5, and 6 together (as we did categories 1, 2, and 3
above), we have 50 occurrences out of 298 in which SHMERON precedes the verb
that it, or a phrase of which it is a part, modifies. (I am not counting the
four occurrences that also count in category 1.) That works out to a ratio
of about .17, or roughly one-sixth.

Here is a tabulation of these five categories:

Category 1: 72
Category 2: 140
Category 3: 36
Category 4: 30
Category 5: 10
Category 6: 10

Subtotal 298
No verb 25
Luke 23:43 1

Total 324


Now, your inference from BDF’s ‘rule’ is that “when an adverb directly
follows a verb it generally modifies the verb it follows.” If we emphasize
the word “generally,” this statement is unobjectionable. After all, about
five-sixths of the time, the adverb SHMERON follows the verb it modifies
(sometimes immediately after it, more often not). However, contextual
information can always ‘trump’ this generalization; it cannot serve as the
basis for a ‘rule’ that supposedly holds true inflexibly and from which we
can then infer which verb SHMERON modifies in a text like Luke 23:43. A
statistical analysis can tell us what we usually find, but not what we will
find in any particular instance.

Thus, you are jumping beyond the evidence when you say:

<< If we apply this general rule to the exegesis of Jesus’ statement at Luke
23:43 the adverb SHMERON in second position to the verb LEGW should indicate
that the adverb “today” modifies the first part of Jesus’ saying, thus
supporting the rendering of the New World Translation at Luke 23:43. >>

From the generalization that when SHMERON immediately follows a verb it
*usually* modifies that verb, you cannot infer that when it immediately
follows a verb in a particular instance that fact “should indicate” that it
modifies the preceding verb. That particular instance may be one of those
pesky exceptions to the generalization. I think you understand this logical
point, because you proceed in your paper to acknowledge that this “rule”
(which is actually an overstatement) has “exceptions” and then to try to
formulate an iron-clad ‘rule’ on which to base your conclusion.


YOUR ARGUMENT FOR AN INFLEXIBLE RULE

Regarding the 72 occurrences listed in footnote 18 of your paper, you write:

<< These examples are the subset which parallels the syntax found in Luke
23:43 where SHMERON also follows a verb in direct discourse. When these are
analyzed it is apparent that when SHMERON takes second place to a verb in
the first position in the same independent Greek sentence of direct
discourse, the relationship between this verb[1]-adverb[2] pair is not
flexible. The adverb SHMERON in position two always modifies the verb in
position one so long as they are in the same sentence. >>

This argument is flawed, as I will explain. However, let us assume for the
sake of argument that your ‘rule’ is correct. Can this tell us anything
about which verb SHMERON modifies in Luke 23:43? No, it cannot. After all,
SHMERON MET’ EMOU ESHi EN TWi PARADEISWi is a complete sentence of direct
discourse. Only by *assuming* that SHMERON is in the same sentence as AMHN
SOI LEGW can you use your ‘rule’ to show that SHMERON takes second place (in
the same sentence) to LEGW and therefore modifies LEGW.

Now, you can respond that AMHN SOI LEGW is part of the sentence that
includes MET’ EMOU ESHi EN TWi PARADEISWi; that is, you can argue that AMHN
SOI LEGW SHMERON MET’ EMOU ESHi EN TWi PARADEISWi is a single sentence. In a
sense, of course, it is. The fact is that what we have here is a sentence
within a sentence. The sentence AMHN SOI LEGW SHMERON MET’ EMOU ESHi EN TWi
PARADEISWi contains the sentence SHMERON MET’ EMOU ESHi EN TWi PARADEISWi
(or, as you would prefer, MET’ EMOU ESHi EN TWi PARADEISWi). The shorter
sentence functions grammatically as the direct object of the verb LEGW. So
in that shorter sentence within Luke 23:43, SHMERON might very well be
included, in which case it would not be in ‘second position’ (as you define
it) in the same sentence as LEGW.

A serious problem for your argument is that you give no reason why the
“general rule” that you extract from BDF should be narrowed to form a rule
applying only to sentences of direct discourse.

Take a look at 2 Samuel 16:3.

“Then the king said, ‘And where is your master’s son?’
And Ziba said to the king, ‘Behold, he is staying in Jerusalem, for he said,
“Today the house of Israel will restore the kingdom of my father to me.”’”

In this verse, SHMERON is the first word in a complete sentence of direct
discourse. In a broader sense, it is part of the larger sentence that
includes “for he said” (hOTI EIPEN). Those words are *also* part of ‘direct
discourse,’ since they are being reported as the words spoken by Ziba.
According to your use of your ‘rule,’ we should take SHMERON as modifying
EIPEN, since they are both words in direct discourse. However, we don’t,
because we recognize that SHMERON in fact starts a new sentence within that
complex of direct discourse.

The above text is a good example of the inadequacy of your “specific rule”
to prove that SHMERON in Luke 23:43 modifies LEGW:

<< When the Greek adverb SHMERON takes second position to a verb in a
separate sentence of direct discourse it always further modifies the verb in
the first position, without exception, in the corpus of the Greek Septuagint
and Greek New Testament. >>

I could agree to this rule and then stipulate that SHMERON takes first
position in a complete, separate sentence of direct discourse in Luke 23:43,
as it does in 2 Samuel 16:3. If you come back and tweak your rule to exclude
Luke 23:43 by linguistic fiat, your new rule will be subject to the same
objection as all your other rules have been: they beg the question by
assuming that Luke 23:43 must follow the same syntactical pattern that you
‘discover’ by searching for an otherwise exceptionless rule.

You concluded your paper with yet another formulation:

<< Or, simply: When SHMERON follows a verb in Koine where Greek syntax
allows for it to modify the verb it follows, it always does. >>

I think this means “follows immediately.” Assuming that is your claim, in 2
Samuel 16:3 Greek syntax certainly *allows* for SHMERON to modify EIPEN, but
it does not. There goes yet another formulation of your rule!

There are other examples. In Proverbs 7:14, we know that SHMERON is the
first word of the second line (since SHMERON APODIDWMI TAS EUCAS MOU is a
quite literal Greek rendering of the Hebrew of that second line). Yet,
grammatically and syntactically, it would be *possible* to take SHMERON as
modifying the verb ESTIN from the first line. Moreover, SHMERON follows
immediately after SHMERON. Thus, by your reasoning, we ought to take SHMERON
with the verb that immediately precedes it, because Greek syntax allows for
it to do so. But we know that happens to be wrong in this instance.

In James 4:13, Greek syntax would surely *allow* us to construe the text to
say, “Come now, you who say today or tomorrow, ‘We will go into such and
such a city...’” But this is just plain wrong (as I hope you agree).
Likewise, in Hebrews 3:7 and 4:7, it is syntactically possible for the words
to be construed, “Therefore, as the Holy Spirit says today, ‘If you hear his
voice,’” and, “...saying through David much later, as already quoted today,
‘If you hear his voice...’” But we happen to know this to be an erroneous
way of construing the text.

Thus, you are incorrect when you assert (in your other post) that if SHMERON
in Luke 23:43 does not modify LEGW “then this is the only example in the
entire corpus of the LXX/GNT where SHMERON does not modify the action of the
verb in postion#1 when it is grammatically possible.” This claim is
factually incorrect: note the examples given above where it is grammatically
possible to construe SHMERON with an immediately preceding verb and yet we
should not do so.

The fact is that it is too common in biblical Greek to begin a sentence,
including a sentence within a sentence, with a temporal adverb such as
SHMERON, to exclude this possibility in Luke 23:43 on the basis of your
‘rule.’ (Remember that SHMERON immediately follows the verb it modifies only
24% of the time in biblical Greek.) Not counting Luke 23:43, there are 20
such sentences using SHMERON (including sentences introduced by a
conjunction or other words):

“Today the Lord will appear to you” (Lev. 9:4).
“Today we know that the Lord is among us...” (Josh. 22:31).
“Today the Lord has accomplished salvation in Israel” (1 Sam. 11:13).
“Today servants are increasingly turning away...” (1 Sam. 25:10).
“Today your servant knows that I have found favor...” (2 Sam. 14:22).
“Today the house of Israel will restore to me my father’s kingdom” (2 Sam.
16:3).
“Today no man in Israel shall be put to death” (2 Sam. 19:23b).
“Today Adonijah shall be put to death” (1 Ki. 2:24).
“Today I will appear to him” (1 Ki. 18:15).
“Today the Lord is taking your lord away from your head?” (2 Ki. 2:5).
“Today we will spend the night with Raguel” (Tobit 6:11).
“Today they will be exalted...” (1 Macc. 2:63).
“Today if you hear his voice, do not harden your hearts...” (Ps. 94:7-8).
“Today I pay my vows” (Prov. 7:14b).
“Today he lends and tomorrow he collects” (Sir. 20:15).
“Today this Scripture has been fulfilled in your hearing” (Luke 4:21).
“Today I must stay in your house” (Luke 19:5).
“Today salvation has come to this house” (Luke 19:9).
“Today if you hear his voice, do not harden your hearts...” (Heb. 3:7-8;
repeated, 3:15; 4:7b).
“Today or tomorrow we will go to such and such a town...” (Jas. 3:14).

It will not refute the point here to say that in Luke 23:43 the direct
discourse does not begin with the word SHMERON but with the words AMHN SOI
LEGW. Such a difference has no bearing on the syntax involved. Furthermore,
in many of the above texts, the sentence that begins with SHMERON is
prefaced by other words in the direct discourse. Sometimes these preceding
words are separate sentences (Lev. 9:3-4; 1 Sam. 11:13; 25:10; 2 Sam.
19:23b; 1 Macc. 2:63; Prov. 7:14; Luke 19:5). In some cases the preceding
words make the sentence beginning with SHMERON a sentence within a sentence
(as in Luke 23:43):

“And now, as the Lord lives...today Adonijah shall be put to death” (1 Ki.
2:24).
“As the Lord of hosts lives...today I will appear before him” (1 Ki. 18:15).
“Do you know that today the Lord is taking your lord away from your head?”
(2 Ki. 2:5).

In biblical Greek, AURION (“tomorrow”) can also have this initial position
in a sentence:

“Tomorrow turn and depart for the wilderness by the Red Sea way” (Num.
14:25).
“Tomorrow at this time I will put them to flight...” (Josh. 11:6).
“Tomorrow salvation will be yours by the time the sun is hot” (1 Sam. 11:9).
“Tomorrow we will come out to you” (1 Sam. 11:10).
“Tomorrow is the new moon” (1 Sam. 20:18).
“Tomorrow I will let you go” (2 Sam. 11:12).
“Tomorrow go down against them” (2 Chr. 20:16).
“Tomorrow they will not be found” (1 Macc. 2:63).
“Tomorrow I will give” (Prov. 3:28).
“Tomorrow he collects” (Sir. 20:15).
“Tomorrow we die” (Isa. 22:13).
“Tomorrow will worry about itself” (Matt. 6:34b).
“‘Tomorrow,’ he said, ‘you will hear him’” (Acts 25:22).
“Tomorrow we die” (1 Cor. 15:32).

One can find examples with other adverbs as well, such as the following:

“In times past (ECQES KAI TRITHN, “yesterday and the day before”) you sought
David to be king over you” (2 Sam. 3:17).
“Yesterday at the seventh hour the fever left him” (John 4:52).

The point is that this pattern of speech is so familiar in biblical usage
that it is a mistake to construct some rule of syntax that would literally
‘rule out’ a biblical speaker employing this speech pattern if he happened
to use a verb just before the adverb SHMERON. The fact that you can delimit
your data pool to a group of texts that happen to fit your ‘rule’ (more or
less) does not prove that your rule reflects syntactical reality.


THE ORDER OF “AMHN SOI LEGW”

In every version of your paper, you have included a lengthy footnote
critiquing my handling of the “Amen” sayings of Jesus in the Gospels. Your
current version softens the criticism somewhat but seeks to make essentially
the same point (footnote 2):

<< Evangelical apologist Rob Bowman criticizes the rendering found in the
New World Translation and offers his own “exegesis” by statistically
comparing the word order and placement of commas in 74 English renderings of
Jesus’ AHMN LEGW sayings (e.g., Matt. 5:18; 16:28; Mark 3:28; Luke 4:24 and
Matt. 5:26; 26:13,21,34; Mark 8:12; 14:9,18,25,30; Luke 11:51; 21:32; John
1:51; 21:18). While none of the 74 English renderings (e.g. “Truly I tell
you”) on the surface look different than Jesus’ saying at Luke 23:43 an
analysis of the Greek shows that order of the verb and personal pronoun is
reversed only at Luke 23:43. (Compare AMHN GAR LEGW hUMIN; AHMN DE LEGW
hUMIN; AMHN LEGW SOI; AMHN LEGW hUMIN; AHMN AHMN LEGW hUMIN; AHMN AHMN LEGW
SOI with the AMHN SOI LEGW SHMERON of Luke 23:43.) If the words of Jesus at
Luke 23:43 conformed to the 74 examples Robert Bowman provides, the adverb
“today” (Greek SHMERON) would follow the verb which would then confirm his
conclusion. The evidence that the word order in Luke 23:43 of AHMN SOI LEGW
SHMERON places the verb in the first position in its clause relative to
SHMERON thus distinguishing it from the 74 examples is overwhelming. >>

Your second to last sentence in the above footnote is confusing: you seem to
be saying that if the words of Jesus at Luke 23:43 had the same word order
(with LEGW preceding the pronoun hUMIN or SOI) as the other “Amen” sayings
of Jesus, then SHMERON “would follow the verb” and that would somehow
support my conclusion. I think you’re missing the word ‘not’ from that
sentence: I’m guessing you mean that the word SHMERON would not immediately
follow the verb LEGW. Feel free to clarify this point.

Admittedly, I did refer to an English translation of these verses: the New
World Translation. I did so in order to use a version that the Jehovah’s
Witnesses respected so that they could see the point I was making without
prejudice. My point was that the AMHN LEGW form is consistently construed,
even in the NWT, as an introductory formula in Jesus’ sayings, with
everything following that formula as part of the distinct saying that Jesus
is introducing. The only exception in the NWT is our text, Luke 23:43.

I must disagree with your characterization of the argument in my book as
“statistically” grounded. A statistical argument would be of the form, ‘X is
so most of the time; therefore, X is true (or probably true) in particular
instance _a_.” My argument did not take that form. Rather, I began by
explaining the semantic nature and linguistic form of the AMHN LEGW
expression as an introductory formula used only by Jesus in the Gospels (p.
99), noted that this expression operates *uniformly* (not most of the time)
as an introductory formula with everything that follows it being a distinct
statement (pp. 99-100), asserted that some overriding contextual evidence
would be needed to construe the expression differently in Luke 23:43 (pp.
100-101), and then proceeded to discuss the textual and contextual evidence
(pp. 101-108).

You are correct in observing that Luke 23:43 stands out from the other
occurrences of the AMHN LEGW sayings in its placement of the pronoun SOI
before the verb instead of after it. At least, this is true for the reading
favored by most modern critical editions of the Greek NT. Two of the
earliest manuscripts containing this verse -- the Bodmer Papyrus (P75, dated
ca. AD 200) and the Vaticanus (B, dated in the fourth century) -- have AMHN
SOI LEGW. However, a majority of the Greek manuscripts have AMHN LEGW SOI.
It is impossible to be sure which reading is correct, and it is therefore
unwise to base any argument solely on one reading over the other.
Admittedly, the argument for the earlier reading is fairly strong: not only
is it attested in two early manuscripts, it is the harder of the two
readings. It is unlikely that a copyist would deliberately change the usual
order and put SOI before LEGW if the manuscript from which he was working
had SOI after LEGW. On the other hand, it would not be surprising for a
copyist to move SOI from before to after LEGW in order to conform to the
usual word order. (Such a change might even have been done accidentally, and
text-critical principles on balance favor accidental changes over deliberate
ones.) Moreover, it is not impossible for Luke to have put SOI before LEGW,
since in other introductory expressions using LEGW (but not AMHN) the
pronoun appears before LEGW in several instances (SOI, Luke 5:24; 7:14; cf.
Matt. 16:18; Mark 2:11; 5:41; hUMIN, Luke 6:27; 11:9; 16:9). On balance,
then, the reading more likely to be correct is AMHN SOI LEGW, though again,
we cannot be sure enough to make the position of SOI the all-important
deciding factor in exegeting the verse.

Now, assuming for the sake of argument that the preferred reading AMHN SOI
LEGW is correct, the question is whether this word order disambiguates the
relation of SHMERON to the rest of the text in favor of its modifying either
LEGW (as you argue) or ESHi, or has no such disambiguating function.

Several considerations lead me to disagree with your suggestion that the
reversal of the usual word order of LEGW SOI to SOI LEGW is significant
because it unambiguously makes SHMERON modify LEGW (which now immediately
precedes it).

To repeat myself, your generalization about adverbs in the ‘second position’
modifying the verb that precedes them is an insufficient basis for reaching
a verdict in the matter. That argument of yours presupposes that SHMERON is
not in the first position in a distinct sentence, SHMERON MET’ EMOU ESHi EN
TWi PARADEISWi. In other words, your argument begs the question.

In texts not using AMHN but using LEGW with either SOI or hUMIN, there is no
discernable difference in meaning arising from the placing of the pronoun
before instead of following LEGW. For example, it is difficult to see any
difference in meaning between “I say to you” (LEGW hUMIN, Luke 11:8) and
“And I (myself) to you say” (KAGW hUMIN LEGW, Luke 11:9). It may be that the
usual word order was reversed in Luke 11:9 for the purely stylistic reason
of avoiding the wording KAGW LEGW, since the two words rhyme and look almost
identical. The reversal does not appear to have any substantive
significance.

It is a commonplace that moving a word forward in a Greek sentence from
where it customarily would be positioned has the effect of placing some
emphasis on that word. Such emphasis with pronouns often has indexical or
deictic (‘pointing’) significance. In Luke 6:27, where the expression “But
to you I say” places hUMIN before LEGW, the position of the pronoun probably
has such indexical significance. Most likely, the emphasis on the pronoun
reflects the fact that 6:27-38 is directed to a different group (Jesus’
disciples) than the preceding verses, which are directed to evil people who
are rich and comfortable now (6:24-26). Likewise, it is possible that in
Luke 23:43 the pronoun SOI has been moved in front of LEGW to indicate an
indexical emphasis on that pronoun. Perhaps Jesus was singling the repentant
criminal out and making emphatic that his promise is directed to him and not
to the verbally abusive criminal (see Luke 23:39-42). Or perhaps Jesus was
being emphatic in promising the repentant thief that *even he* would be with
him in Paradise. Either way, *at most* the difference is one of emphasis,
not of meaning.

Such an emphatic, indexical understanding of the significance of the unusual
position of the pronoun has much more to commend it than the speculation
that the pronoun was moved to make LEGW unambiguously the verb modified by
SHMERON. For one thing, if that was the reason for putting SOI first, it
didn’t work: the vast majority of readers of the passage (even that minority
that favors construing SHMERON with LEGW) missed this supposedly
all-important clue. (In fact, to the best of my knowledge you are the first
person to suggest it.) Understanding the relative position of the pronoun as
one of emphasis rather than covert syntactical coding is consistent with the
general tenor of syntactical variations in Greek. Moreover, we have other
examples using LEGW with pronouns in Luke where such variations can fairly
easily be seen to have emphatic import, as illustrated above.

There is an interesting irony in your appeal to the position of the pronoun
before the verb in Luke 23:43 to establish that SHMERON belongs to the
introductory formula (“Amen to you I say today”) rather than to the
statement that follows. Most Jehovah’s Witnesses and others taking this view
of SHMERON appeal to the Deuteronomic sayings that use “today” (SHMERON in
the Greek OT) as a biblical ‘idiom’ that they argue Jesus was using in Luke
23:43. You have not been making this claim in the various versions of your
paper; I don’t recall if you used this line of argument in the past. The
argument was popularized by E. W. Bullinger and is found, for example, in
Greg Stafford’s book _Jehovah’s Witnesses Defended_ (2d ed., 550-51).

Well, there are a number of problems with the above argument. I won’t go
into all of them now (I can if necessary), but I do want to draw your
attention to one interesting dissimilarity between those Deuteronomic
sayings and Luke 23:43 of some relevance to our discussion. In those
Deuteronomic sayings, the pronoun consistently stands *after* the verb of
speech, not before it. The most frequent formula looks like this:

“which I am commanding you today”
hOSA EGW ENTELLOMAI SOI SHMERON
4:40; 6:6; 7:11; 8:1, 11; 10:13; 11:8, 13, 27, 28; 13:19; 15:5; 19:9; 27:1,
4, 10; 28:1, 13, 14, 15; 30:2, 8, 11, 16

Curious, is it not, that in all of these texts, SHMERON does not follow
immediately after the verb? And more curious still, if Luke 23:43 is echoing
this ‘idiom,’ why doesn’t SOI appear between the verb and the adverb, as is
the regular pattern in the samples of that idiom?

So, I think your focus on the position of SOI has added another nail in the
coffin of the theory that Luke 23:43 reflects the idiom of those
Deuteronomic sayings using SHMERON.

In Christ's service,

Robert M. Bowman, Jr.
Center for Biblical Apologetics
Online: http://www.biblicalapologetics.net