Tov: Greek Punctuation

This appendix analyzes several scribal features displayed in early witnesses of Greek Scripture with an eye toward discovering links with early Jewish scribal traditions such as those known from the Hebrew scrolls from the Judean Desert. Special attention is paid to the indication of verses, sections, paragraphoi, ekthesis, and the writing of the divine names. The parameters of this investigation are as follows:

• The coverage of the Greek texts is intentionally vague (‘Greek Scripture’), since the dividing line between the so-called Old Greek translation and other early translations and revisions is often unclear, as are the exact limits of what may be considered Scripture.

• All early papyri that could be located in the libraries of Tübingen and Macquarie University (especially in the Ancient History Documentary Research Centre [AHDRC]) were examined, with the exclusion of very fragmentary texts. The table, which is rather exhaustive, lists in presumed chronological sequence, all the texts that were given dates up to and including the fourth century CE in their publications. Most texts examined were dated to the third and fourth centuries CE (see the statistics in Van Haelst, Catalogue, 419). The large codices A, B, S, and G are excluded from the analysis.

• The distinction between Jewish and Christian copies is relevant in as far as the former are more likely to preserve ancient Jewish scribal practices. Although this distinction is often very difficult, all texts antedating the middle of the first century CE are Jewish. According to K. Treu, “Die Bedeutung des Griechischen für die Juden im römischen Reich,” Kairos NF 15 (1973) 123–44, it is possible that several texts written after that period might also be recognized as Jewish; they are indicated in the first column of the table as ‘Jewish (Treu).’ A major though not exclusive criterion for the Jewish nature of a text is the writing in scrolls, indicated as ‘S’ in the second column of the table (see already C. H. Roberts, “The Christian Book and the Greek Papyri,” JTS 50 [1949] 155–68, especially 157–8). The Christian nature of Scripture texts can usually be detected by their inscription in codex form (indicated as ‘C’ in the second column), and their use of abbreviated forms of the divine names (indicated in the seventh column). See further Kraft, “Textual Mechanics.”

a. Small spaces without additional indications

The earliest sound evidence from the second century BCE onwards for the indication of small sense units in Scripture texts (verses) pertains not to Hebrew manuscripts, but to the Targumim and several early Greek sources. See ch. 5a2. The use of spacing for the indication of small sense units (verses) was a natural development in the tradition of Scripture-writing, both in Hebrew and in the translations, as larger units (sections) were likewise indicated with spacing. The size of the space indicating new sections was always larger than that indicating verses, which usually equaled a single letter-space, and sometimes slightly more.

A small group of early Greek texts indicated the ends of verses, and sometimes also groups of words, with small spaces without additional notations. In all these texts, with the exception of hand B of 8H≥evXIIgr in Zechariah (end of 1 BCE), word-division was not indicated with spaces. Since almost all these texts are early, they undoubtedly reflect early Jewish traditions. Some of these texts reflect early Jewish revisions (P.Fouad 266a–c, 8H≥evXIIgr), while others probably reflect the tradition of the Old Greek translation (4QpapLXXLevb, 4QLXXNum), different from the text contained in other witnesses. The nature of P.Oxy. 3522 is unclear.

P.Rylands Greek 458 of Deuteronomy (2 BCE): after Deut 24:1 (+ high dot); 25:2; 26:17, 18; as well as after
groups of words.362
P.Fouad 266a–c of Genesis and Deuteronomy (1 BCE) consisting of three different scrolls (Aly–Koenen, Three
Rolls): after verses, and sometimes also after groups of words: P.Fouad 266a of Genesis (942; middle of 1 BCE);
P.Fouad 266b of Deuteronomy (848 [middle of 1 BCE]), in the latter case, e.g. after Deut 22:8; 28:67; 32:19, 25, as
well as after 28:9a, 65a, but not after Deut 28:65; 31:25; 32:46; P.Fouad 266c of Deuteronomy 10–11, 31–33 (847
[second half of 1 BCE]). See Dunand, Papyrus grecs, Texte et planches; Aly–Koenen, Three Rolls, p. 5, n. 24 and p.
7, n. 32; Oesch, Petucha und Setuma, 297–8.
4QpapLXXLevb of Leviticus 2–5 (1 BCE): after Lev 3:11; 4:26.


4QLXXNum of Numbers 3–4 (1 BCE): after 3:41 and 4:6, as well as after groups of words.
8H≥evXIIgr hand A (end of 1 BCE) containing substantial segments of the Minor Prophets: after almost all
verses (E. Tov, DJD VIII, 11–12), as well as after some groups of words.
8H≥evXIIgr hand B (end of 1 BCE): after Zech 9:5, but not after 9:1. This scribe also indicated divisions
between words with small spaces.
P.Oxy. 50.3522 of Job 42 (1 CE): after 42:11, as well as after groups of words (42:11a, 12a).
P.Baden 56b = P.Heidelberg gr. 8 of Exodus 8 (2 CE): some spaces, e.g. after 8:8, 9.
P.Mich. 22 of Psalms 8–9 (3 CE): spaces in different positions in the line after each hemistich.
P.Chester Beatty XIII of Psalms 72–88 LXX (4 CE): after hemistichs, including a dicolon in 82:9.
This group may be extended by the following group of texts that likewise indicated spaces.

362. Roberts, Two Biblical Papyri (see n. 355) with plates. A more complete photograph is found in E. Würthwein, The Text of the Old Testament (Grand Rapids, Mich. 1979) 177.

b. Graphic indicators

(usually high dots or dicola) added in spaces left by the original scribes Graphic indicators (usually high dots or dicola, rarely oblique strokes or obelus signs [P.Hamb. bil. 1 of Qohelet]) were often added in texts that already had spaces marking the ends of verses. With two exceptions, all these texts are from the third century CE onwards, which shows that the ancient tradition of marking verse-endings with spacing was supplemented with a Greek system of indicating small sense units with dots. Sometimes these dots were inserted by the original scribes, but often they were added after the completion of the writing (§ c below). In those instances, extant spaces could not be erased, and if no spaces had been left, new markings were inserted between existing letters. The use of either a dot or dicolon depended on the preference of the scribe.

P.Rylands Greek 458 of Deuteronomy (2 BCE): a space after Deut 24:1 (+ high dot).
P.Yale 1 of Genesis 14 (end of 1 CE ?): spaces with median dots.
P.Chester Beatty VIII of Jeremiah 4–5 (2–3 CE): high and median dots, also after groups of words; dicolon
before speech in 4:31.
P.Chester Beatty X (967) of Daniel and Esther (early 3 CE): oblique strokes both in spaces and when no space
was left (see group c below).
P.Scheide + P.Chester Beatty IX (967) of Ezekiel (early 3 CE): spaces with high dots, median dots, and dots on
the line (no spaces left), also after groups of words.
P.Oxy. 9.1166 of Genesis 16 (3 CE): spaces with median dots, also after some groups of words.
P.Berlin 17213 of Genesis 19 (3 CE): a space with a high dot after 19:17.
P.Oxy. 8.1075 of Exodus 40 (3 CE): a space with a high dot after 40:28; otherwise the text is continuous.
P.Fir. 8 of Isaiah (3 CE): high dots, mainly in spaces.
P.Merton 2 of Isaiah 8–60 = P.Chester Beatty VII (965) (3 CE): high dots + dicola, sometimes in spaces.
P.Berlin 17212 of Jeremiah 2–3 (3 CE): spaces with high dots.
P.Bodmer XXIV of Psalms 17–53, 55–118, hand A: (3 CE) spaces + high dots and some dicola, also after
hemistichs.
P.Bodmer XXIV of Psalms 17–53, 55–118, hand B: (3 CE) spaces + dicola and some high dots, also after
hemistichs.
P.Antinoopolis 9 of Proverbs 2–3 (3 CE): spaces with a dicolon or high dot.
P.Berlin 11778 (BKT 8.17) of Job 33–34 (3 CE): spaces + median dot, also after parts of verses.
P.Mil. 13 of Qohelet 3 + (?) P.Mich. 135 of Qohelet 6 (3 CE ): spaces with some dicola after hemistichs.
P.Egerton 4 (B.M.) of 2 Chronicles 24 (3 CE): high dots, usually in spaces.
P.Berlin 14039 of Exodus 34–35 (3–4 CE): spaces with high dots, also after groups of words.
P.Genova P.U.G. 1 of Psalms 21–23 LXX (3–4 CE): spaces with dicola after hemistichs.
P.Flor. B.L. 980 of Psalms 143–48 LXX (3–4 CE): dicola, mainly in spaces.
P.Hamb. bil. 1 of Qohelet (3–4 CE): spaces + 2 oblique strokes or obelus signs after hemistichs.
P.Lit. London 202 of Genesis 46–47 (c. 300 CE): spaces with high dots.
P.Lit. London 209 of Canticles 5–6 (early 4 CE): spaces, sometimes with high dots.
P.Vindob. Gr. 39786 of Psalm 9 (early 4 CE): spaces with oblique strokes after hemistichs.
Louvre MND 552 H–L of Psalm 146 LXX (early 4 CE): spaces with 2 oblique strokes after hemistichs.
P.Oxy. 11.1352 (leather) of Psalms 82–83 LXX (early 4 CE): spaces with dicola after groups of words.
Codex St. Cath. of Genesis 27–28 (4 CE): spaces with high dots, also after groups of words.
P.Hamb. Ibscher 5 of Genesis 41 (4 CE): some spaces with high dots, also after groups of words.
P.Berlin 11766 of Exodus 5–7 (4 CE): a space with dicolon after 6:24 and a space in the middle of 6:24;
otherwise the text is continuous.
P.Bodl. Ms. Gr. Bibl. f. 4 of Exodus 9–10 (4 CE): spaces with low and high dots.
P.Oxy. 10.1225 of Leviticus 16 (4 CE): spaces with high and low dots.
P.Rylands 1 of Deuteronomy 23 (4 CE): spaces with median dots.
P.Oxy. 9.1168 of Joshua 4–5 (4 CE): spaces with high dots.
P.Oslo 11, frg. 1 of Isaiah 42, 52, 53 (4 CE): spaces with high dots.
P.Genève Gr. 252 of Jeremiah 5–6 (4 CE): spaces with high dots.
P.Antinoopolis 10 of Ezekiel 33–34 (4 CE): inconsistently placed high dots, sometimes in spaces, but also
when no space was left.
P.Oxy. 15.1779 of Psalm 1 (4 CE): spaces with high dots, also after groups of words.
P.Chester Beatty XIV of Psalms 31, 26, 2 LXX (4 CE): spaces with dicola, also after groups of words.
P.Vindob. Gr. 26205 of Psalm 34 LXX (4 CE): hemistichs with commas.
P.Erlangen 2 of Genesis 41 (4–5 CE): spaces with high and low dots.
P.Yale Beinecke 544 of 1 Samuel 24–2 Samuel 1 (4–5 CE): minute spaces with high dots and dicola, also after
groups of words.
P.Vindob. Gr. 29274 of Psalm 32 LXX (4–5 CE): spaces + low dots with an apostrophe.
P.Oxy. 24.2386 of Psalms 83–84 LXX (4–5 CE): spaces with two oblique strokes after each stich.
Cod. Cambridge of 2 Kings 21–23 (Aquila; 5 CE): spaces with high dots, also after groups of words.

c. High dots and/or dicola superimposed on texts written without spaces

Many of the high dots and dicola were superimposed, often inelegantly, on texts that were initially written continuously. The secondary status of these interpunction signs is evident from the lack of space left and often also from the color of ink of the dots.

P.Scheide and P.Chester Beatty IX (967) of Ezekiel (early 3 CE), which originally formed one scroll,
systematically indicated the end of each verse, as well as segments of verses, with dots in different positions (high
dots, median dots, and dots on the line), apparently without assigning a different meaning to each one (Johnson,
Scheide, 16–17).
P.Chester Beatty VII of Ezekiel 11–17 (3 CE): high dots and strokes superimposed on a running text without
spaces.
P.Merton 2 = P.Chester Beatty VII (965) of Isaiah 8–60 (3 CE): usually high dots and dicola super-imposed on
a running text.
Pap. W (Freer) of the Minor Prophets (3 CE) indicates dots at the ends of verses as well as after groups of words.

All of them were probably inserted after the completion of the writing. Dots or dicola appear in many of the texts listed in the previous section, sometimes in existing spaces, and often when no space had been left (see the table). In this regard, Psalms manuscripts form a special group since the graphic indicators are found only at the ends of the lines when the texts are arranged stichographically, and the secondary nature of such indicators cannot be determined easily. A special case are a few papyri in which high dots were superimposed on the text after each syllable as scribal exercises: P.Rendel Harris 166 of Exodus 22–23 (3 CE); P.Laur. Inv. 140 (34) of Psalm 1 (3–4 CE); P.Lit. London 207 of Psalms 11–16 (3–4 CE). Similar exercises are also known for literary texts.

d. Continuous writing involving no notation for the indication of verse endings

The scribes of some texts did not indicate verse endings. Although in principle such texts could reflect the first stage of the Greek translation, it appears that they reflect a secondary development since the earliest available evidence (group a above) reflects spacing between the verses.

P.Oxy. 4.656 of Genesis 14–27 (2 or 3 CE)
P.Coll. Horsley (Deissmann) of Exodus 4 (2 or 3 CE)
P.Chester Beatty VI (963) of Numbers and Deuteronomy (2 or 3 CE)
P.Oxy. 7.1007 of Genesis 2–3 (3 CE) (leather)
P.Heidelberg 290 of Leviticus 19 (3 CE)
P.Lit. London 204 of Psalm 2 (3 CE)
P.Schøyen 2648 of Joshua 9–11 (early 3 CE)
P.Chester Beatty V (962) of Genesis (second half of 3 CE)
P.Vindob. Gr. 39777 of Psalms 68, 80 LXX (3–4 CE ; Symmachus)
P.Chester Beatty IV (961) of Genesis (4 CE) 
P.Amherst 1 of Genesis 1 (4 CE) (Aquila)
P.Oxy. 9.1167 of Genesis 31 (4 CE)
P.Grenfell 5 of Ezekiel 5–6 (4 CE)
P.Oxy. 15.1779 of Psalm 1 (4 CE)
P.Flor. B.L. 1163 of Job 1–2 (4 CE)
P.Vindob. Gr 26205 of Psalm 34 LXX (4 CE)
P.Oxy. 6.845 of Psalms 68–70 LXX (4 CE)

When the text is laid out in stichoi, the layout renders the need for explicit verse-indications superfluous:

P.Bodl. MS . bibl. Gr. 5 of Psalms 48–49 (late 1 or early 2
P.Antinoopolis 7 of Psalms LXX (2 CE)
P.Leipzig 170 of Psalm 118 LXX (2–3 CE)
P.Vindob. Gr. 26035B of Psalms 68–69 LXX (3 CE)
P.Alex. 240 (PSI 921) of Psalm 77 LXX (3 CE)
P.Berlin Inv. 21265 of Psalm 144 LXX (3 CE)
P.Antinoopolis 8 of Proverbs 5–20 (3 CE)
P.Laur. Inv. 140 (34) of Psalm 1 (3–4 CE)
P.Oxy. 10.1226 of Psalms 7–8 LXX (3–4 CE)
P.Lit. London 207 of Psalms 11–16 (3–4 CE)
P.Bonn Coll. P147v of Psalm 30 LXX (3–4 CE)
P.Flor. B.L. 1371 of Psalm 36 LXX (4 CE)
P.Leipzig 39 of Psalms 30–55 LXX (4 CE)
P.Vindob. Gr. 35781 of Psalm 77 LXX (4 CE)
P.Berlin 18196 of Canticles 5–6 (4 CE)
P.Oxy. 24.2386 of Psalms 83–84 LXX (4–5 CE)
CE)

e. Presumed development

As the earliest available Greek sources reflect spaces between verses (group a), this practice probably reflects the oldest form of the Greek translation in which verse division was indicated, possibly in the original translation itself. See ch. 5a2. According to this assumption, the continuous writing of the Greek texts as recorded in group d is secondary. Over the course of time, graphic signs were added in these spaces in accord with the Greek manuscript writing tradition (groups b and c). See further the summary of Greek scribal traditions in ch. 5a2. According to Revell, the spacing and high dots in the middle of a verse in early Greek manuscripts reflected, though not consistently, the Hebrew accent system known from the later Masoretic sources regarding disjunctive accents. ³⁶³ A similar claim was made throughout the study of Korpel–de Moor, Structure. However, while this claim can be made for such probable Jewish sources as:

P.Rylands Greek 458 of Deuteronomy (2 BCE)
P.Fouad 266a–c of Genesis and Deuteronomy (middle 1
4QLXXNum of Numbers 3–4 (1 BCE)
8H≥evXIIgr hand A (end of 1 BCE)
P.Oxy. 50.3522 of Job 42 (1 CE),
BCE)
it is unlikely for such non-Jewish sources as:
P.Scheide + P.Chester Beatty IX (967) of Ezekiel (early 3 CE )
Pap. W (Freer) of the Minor Prophets (3 CE)
P.Antinoopolis 9 of Proverbs 2–3 (3 CE)
P.Berlin 14039 of Exodus 34–35 (3–4 CE)
P.Oxy. 11.1352 of Psalms 82–83 LXX (early 4 CE) (leather)
P.Hamb. Ibscher 5 of Genesis 41 (4 CE)
Codex St. Cath. of Genesis 27–28 (4 CE)
P.Oxy. 15.1779 of Psalm 1 (4 CE)
P.Chester Beatty XIV of Psalms 31, 26, 2 LXX (4 CE)

363 E. J. Revell, “The Oldest Evidence for the Hebrew Accent System,” BJRL 54 (1971–2) 214–22; idem, “Biblical Punctuation and Chant in the Second Temple Period,” JSJ 7 (1976) 181–98.

The later manuscripts A, B, S, and Q, which form the basis of the study of Korpel–de Moor, Structure, it is even more difficult to invoke the antiquity of the Greek tradition, as the Christian scribes of these late manuscripts probably would not have had access to the earlier Hebrew traditions. In Greek inscriptions, the high dot ( ̇ ), dicolon ( : ), tricolon (three vertical dots), as well as various additional graphic signs (see especially Threatte, Attic Inscriptions, 73–94), were used regularly from the seventh century BCE onwards to indicate small or large sense divisions, while in papyri this system was developed further. Thus in the punctuation system devised by Aristophanes of Byzantium (c. 257–180 BCE), and recorded by Aristotle, Rhet. 3.8.1409a.20 , different values were assigned to the dot as it stood above the line (a full stop), in the middle of the line (a comma), or on the line (a semicolon). See Hall, Companion, 13; Gardthausen, Griechische Palaeographie, II.400; Schubart, Palaeography, 173; Kenyon, Palaeogra-phy, 28; Pfeiffer, History, 180; Turner, Greek Manuscripts, documents 20, 21, and index, with examples from manuscripts from the second century CE onwards; D. C. Parker, Codex Bezae: An Early Christian Manuscript and its Text (Cambridge 1992) 31–4. The high dots were often inserted by a later hand, as illustrated by a papyrus of Homer’s Iliad presented in Turner, Greek Papyri, pl. IV. B. Indication of sections See ch. 5a. C. Special writing of the divine names The limited evidence for special writing of the divine names points to differences between Jewish and Christian texts.

• Jewish:
texts display the Tetragrammaton in various ways, see ch. 5d. scribes employed k(uvrio)", together with other abbreviated nomina sacra.364 See the penultimate column in the table.

• Christian:
As expected, early Jewish copies of Greek Scripture reflect some scribal phenomena of the Hebrew manuscripts from which the Greek translation was made. However, with the transmission of this translation by Christian scribes, these features were contaminated and in many cases can no longer be recognized. This pertains to the following features: indication of small sense units (verses) and sections with spacing, and the writing of the divine names in Hebrew characters. Several new features were introduced that reflected the Greek writing tradition (graphic indicators for the indication of verses; paragraphos signs; ekthesis) or early Christian practices (abbreviated nomina sacra).

364. The isolated case of the unabbreviated kuvrio" in P.Oxy. 4.656 of Genesis 14–27 (2 or 3 CE) is unclear. In this manuscript the spaces left by the original scribe were filled in by a second scribe with kuvrio".