This is a topic that has interested me a long time. I do agree that the NWT's use of punctuation here is incorrect. But it should be recognized that it is technically permissible by the text (without hoti "that" in the sentence, the parsing of the clause is ambiguous) and the Society does provide arguments in support of their rendering. First they cite precedents for the NWT rendering in certain modern translations: "Verily I say unto thee this day: With me shalt thou be in Paradise" (Rotherham), "Truly I say to you today, You will be with me in Paradise" (Lamsa), "Verily, to you am I saying today, with Me shall you be in paradise" (Concordant Literal New Testament), "Indeed today I say you, you shall be with Me in the paradise" (Tomanek), etc. This does not amount to much, but it shows that they are not the first to render the passage that way. They also find an ancient witness for this parsing (versional support) in the Curetonian Old Syriac (fifth century AD): "Verily I say to thee to-day that with me thou shalt be in the Eden's garden". This parsing is also alluded to in a number of other sources from Late Antiquity and the Middle Ages:
(1) Macarius (fourth century AD) wrote: "How altogether vile are those who twist his words into a mere promise for the future, 'Verily I say to you today,' then punctuating, 'you shall be with me in paradise' (amèn legò soi sèmeron, heita diasteilantes, esè met' emou en paradeisò)" (Apocriticus 3.91.9); (2) (Pseudo?)-Hesychius (date uncertain, fifth century or later) similarly noted: "Some indeed read this way, 'Truly I tell you today' (amèn legò soi sèmeron), then they add (heita epipherousin), 'that you will be with me in paradise' (hoti met' emou ese en tò paradeisò)" (Quaestiones 47); (3) The Gospel of Nicodemus (sixth century AD) phrased the promise: "Today I tell you the truth that in my kingdom I should have you in paradise with me (sèmeron legò soi alètheian hoti en tè basileia mou hina se ekhò eis ton paradeison met' emou)" (10.2,1.18, recension m1), "Today I tell you, you will be with me in paradise (sèmeron legò soi met' emou esè en tò paradeisò)" (10.2,1.18, recension m1-m2); (4) A scolion in three miniscule MSS from the tenth-eleventh centuries AD states: "Others say it must be understood with a comma (dein hupostizontas anaginòskein), 'Truly I tell you today (amèn legò soi sèmeron),' and then adding the expression this way (eith' houtòs epipherein to), 'You will be with me' (met' emou esè)" (Catena in Lucam, typus B, 169.2); (5) Theophylact (twelfth century AD) also noted: "Others insist on the expression, punctuating after 'today' (stizontes eis to sèmeron) so it reads, 'Truly I tell you today,' and then 'You will be with me in paradise (amèn legò soi sèmeron, heita to, met' emou ese en tò paradeisò)" (Enarrationes in Lucam, 23.43).
The Society, following the lead of E. W. Bullinger, also finds OT support for their reading of the text. "Truly I tell you today" is reckoned to have a model in the Deuteronomistic formulae "I declare to you this day" (ha`îdotî bakem hayyôm), "I command you this day" (m e tsauka hayyôm), "I am speaking to your ears this day" (dober be'azenêkem hayyom), etc. which has a similar tone of solemnity (Deuteronomy 4:26, 40, 5:1, and some 40 other examples). In the LXX these are translated in such forms as lalò en tois òsin humòn en tè hèmera tautè "I speak to your ears this day" (5:1) and enetellomai soi sèmeron "I command you today" (6:6). Finally the Society supports their parsing by noting that Jesus could not have promised the thief that he would go to Paradise that same day if they both would be going to Hades as the same author states in Acts 2:31.
So there is some grounds to support the NWT rendering. But exegesis involves an evaluation of the preponderance of evidence to determine which interpretive option best accounts for the evidence, and the majority opinion is against the parsing found in the NWT for a number of reasons. First of all, the OT parallels are not very close since none of them utilize the verb legein or the solemnity expression amèn. It is clear that amèn (sometimes doubled) + legein + "you" is a fixed expression in the canonical gospels, occurring some 70 times elsewhere (plus a few more cases in Luke where alèthòs "truly" is used), and nowhere else does a time expression form part of the formula, much less sèmeron specifically. This suggests that the similarity with the OT formulae (which always contained hayyôm) is coincidental. In contrast, the OT expression occurs once in the NT in a form very close to what is found in Deuteronomy: "I testify to you today" (marturomai humin en tè sèmeron hèmera) in Acts 20:26 (cf. diamarturomai humin sèmeron in Deuteronomy 4:26 LXX). Some Watchtower apologists list Luke 2:11, 4:21, 5:26, 19:9, 22:61 as instances where Luke inserts hoti in order to separate sèmeron from preceding speech, but none of these are syntactically similar to Luke 23:43 (where ambiguity would result without the hoti). But there are a number of instances where a time expression does immediately follow legò humin/soi "I say to you" without an intervening hoti, and in each case they do not belong to the introductory formula: (1) Matthew 5:18: "For truly I tell you, until heaven and earth passes away, not one stroke of a letter, will pass from the law until all is accomplished", (2) Matthew 26:64: "But I tell you, from now on you will see the Son of Man seated at the right hand of Power and coming on the clouds of heaven", (3) Luke 17:34: "I tell you, on that night there will be two in one bed, one will be taken and the other left", (4) John 21:18: "Verily verily I tell you, when you were younger you used to fasten your own belt and go wherever you wished". Morever, there are two additional examples of a time expression following the introductory formula with hoti, and in both cases the time expression does not belong with the formula: (1) Matthew 26:34: "Truly I say to you that (hoti) during this night, before the cock crows three times, you will deny me", (2) Mark 14:30: "Truly I say to you that (hoti) you today, this night, before the cock crows three times, you will deny me". So the NT usage of this formula supports strongly that sèmeron in Luke 23:43 belong with the second clause.
There are also contexutal considerations. Notice the way Luke 23:43 answers the question posed by the repentant thief: "Remember (mnèsthèti) me whenever (hotan) you may come (elthès) into your kingdom". Here (1) the verb "remember" implies the passage of some time (cf. Luke 22:61, 24:6), (2) hotan "whenever" in particular has in view an indefinite time in the future, and (3) the aorist subjunctive of elthès "you may/might come" also implies an indefinite future when used with hotan. In the reply, sèmeron "today" is the counterpoint to this indefiniteness; it won't be some vague, indefinite time in the distant future but "today". The wording of the preceding verse thus strongly supports the traditional parsing of Luke 23:43. The idea of being "with" Jesus at death also finds parallels in Paul: " Therefore we are always confident and know that as long as we are at home in the body we are away from the Lord. ...We are confident, I say, and would prefer to be away from the body and at home with the Lord (endèm è sai pros ton kurion). So we make it our goal to please him, whether we are at home in the body or away from it " (2 Corinthians 5:6-9), " If I am to go on living in the body, this will mean fruitful labor for me. Yet what shall I choose? I do not know! I am torn between the two: I desire to depart and be with Christ (sun Khristò einai), which is better by far; but it is more necessary for you that I remain in the body " (Philippians 1:22-24); cf. also 1 Thessalonians 4:17: " We who are still alive and are left will be caught up together with them in the clouds to meet the Lord in the air. And so we will be with the Lord (sun kuriou esometha) forever". The promise to the thief would then resemble the Pauline expectation if "today" is understood as belonging to the second clause. Finally, there might be a faint OT echo in 1 Samuel 28:19. There the dead prophet Samuel prophetically declares to Saul that he would die the next day: "Tomorrow you (and your sons) will be with me" (i.e. in the underworld), and the wording in the LXX is quite reminiscent of that in Luke 23:43: aurion su kai ho huios sou met' emou. Here a promise is made about an immediate postmortem presence with the one speaking, and the time element forms part of the promise.
The most weighty objection to this parsing is the problem with the location of Jesus in Paradise at the same time he was supposed to be in Hades. Some early church fathers solved this theologically by claiming that if " all the fullness of the Deity lives in bodily form" in Christ (Colossians 2:9), and if " in him all things hold together" (1:17), and if " all things in heaven and on earth are united under Christ" (Ephesians 1:10), then Jesus could be in both heaven and Hades at the same time. But this is hardly what the author of Luke had in mind, and the deutero-Pauline concept has in mind Jesus' glorification through resurrection rather than his lowly status as a slave prior to his resurrection. Another solution is to locate Paradise in the realm of the dead. This has some contextual support from Luke. The saying in Luke 23:43 represents a Lukan redaction and the Parable of Rich Man and Lazarus is also uniquely Lukan material. In that story, the rich man went to a place of torment in Hades in his postmortem state while Lazarus was taken to the "bosom of Abraham", but both were in the same realm where despite being separated by a great chasm Abraham and the rich man were able to talk with each other (Luke 16:19-31). The closest parallel to this in the Book of Watchers in 1 Enoch 22, where the souls of the dead are gathered into a place at the far extremities of the earth, and the "souls of the righteous are separated by a spring of water with light upon it" from the sinners who are punished in agony "until the great day of judgment" (v. 9-12). It should be recalled that the Garden of Eden was located in the east and 1 Enoch (likely under influence from the Epic of Gilgamesh and Greek conceptions of Hades) locates Eden and the entrance to the underworld at the extremities of the world (where the sun sets and rises and where it must pass through the underworld in order to rise the next day in the east). A later notion was that Paradise was located in heaven, specifically "third heaven" (cf. the Greek Life of Adam and Eve, 2 Enoch 8:1-3, and 2 Corinthians 12:2). It was indeed at the same extremities of the earth where the gates to heaven are located. "And beyond these mountains is a place, the edge of the great earth, where the heavens come to an end, and I saw a great chasm among pillars of heavenly fire .... beyond this chasm I saw a place where there was neither firmament of heaven above, nor firmly founded earth beneath it" (18:10-12), "From there I proceeded to the west, at the ends of the earth, and I saw there three gates of heaven open, as I saw in the east, the same number of gates and the same number of outlets" (35:1). It was at this place at the extremities where the underworld and heaven are mutually accessible, and it was the garden of Eden "wherein the elect and the righteous ones dwell" (60:8). The proximity between Paradise and Hades is apparent in a number of texts. In the Greek Life of Adam and Eve, when Adam dies, his soul washes up at the Acheron (a river in Hades) and Michael the archangel then takes him "back to Paradise which is in third heaven" (48:40), where "the flowers of Paradise with their sweet fragrance" are still to be found (v. 2). Similarly, in the Testament of Abraham, Michael the archangel takes Abraham "toward the east, to the first gate of heaven" (11:1), and there are two portals there, one leading the righteous into Paradise and the other leading the sinners to eternal punishment (v. 10-11). The same book also states that in Paradise "there are the tents of my righteous ones and where the mansions of my holy ones, Isaac and Jacob are in his [Abraham's] bosom" (20:14), this language very closely parallels what is found in the Parable of Rich Man and Lazarus in Luke. The proximity of Hades and Paradise is also apparent from 2 Enoch, where both are located in third heaven. First, Enoch was set down in Paradise (8:1-8), and then he was taken "to the northern region and they showed me there a very frightful place, and all kinds of torture and torments are in that place, cruel darkness and lightless gloom" (10:1-2), and this is where the wicked dead are tortured. Both places are lower than fourth heaven, which is where the sun and moon rise and set on chariots placed on celestial tracks (ch. 11). So when we examine how the cosmology of world was conceptualized in early Judaism, which had both heaven and the underworld meeting together at the ends of the earth, the problem discussed above is more apparent than real.
Finally, there is the matter of the early interpreters. The readings found or referenced in the Curetonian Syriac, Macarius, (Pseudo?)-Hesychius , the Gospel of Nicodemus, etc. are exegetically motivated by the very issue discussed in the last paragraph: How can Jesus be in both Hades and Paradise after he dies? The cosmology that allows this (discussed in the last paragraph) co-existed with others that strictly envisioned earth as intervening between heaven above and Hades below, without any grey areas. That these readings were exegetically motivated was sometimes remarked upon. With regard to the reading given above, (Pseudo?)-Hesychius noted that it eased the following problem: "How can the Lord immediately fulfill his promise to the thief, 'Today you will be with me in paradise', if indeed after his crucifixion Christ was in Hades setting free the dead?" It is also apparent from the patristic references that the NWT-like parsing was a minority position and not generally accepted. The readings in the Gospel of Nicodemus are also problematic because other recensions of the same book lack the reordering of sèmeron, e.g. amèn legò soi s è meron met' emou es è en t ò paradeis ò (10.2,1.18, 26.1.16, recension m3). One may easily cite other citations of the passage in the church fathers that parse the text in accordance with the majority position: (1) amèn legò soi (labe kai su to am è n, h ò l è sta, ho s è meron l è st è s), s è meron met' emou es è en t ò paradeis ò (John Chrysostom, In Sancta et Magna Parasceve 50.816.18), (2) amèn am è n gar legò soi hoti s è meron met' emou es è en t ò paradeis ò (Hesychius, Homilia in Sancta Pascha, 1.3.4.6-7), (3) amèn am è n gar legò soi, ph è si, s è meron met' emou es è en t ò paradeis ò (Cyril of Alexandria, Commentarii in Joannem 1.684.4, De Adoratione et Cultu in Spiritu et Veritate 68.245.5), etc. And of course in modern translations, the versions cited by the Society are clearly in the minority.