Journal: Tyndale Bulletin
Volume: TYNBUL 47:1 (NA 1996)
Article: A New Method For Reconstructing Biblical Scrolls
Author: Edward D. Herbert

---

TynBul 47:1 (1996) p. 177

A New Method For
Reconstructing Biblical Scrolls1

Edward D. Herbert

The Qumran biblical scrolls are of crucial importance for developing our understanding of the textual diversity that existed around the turn of the era, and for their text–critical contributions. Biblical scholars have generally depended primarily upon common sense and general scholarly judgement in reconstructing such scrolls. Chapter 1 advances a range of scientifically–based tools for reconstructing biblical scrolls, which will be summarised below under six headings. In the remainder of the thesis, these tools are applied to the reconstruction of the 4QSam^a fragments of 2 Samuel.

Development 1: Expected Widths. Amounts of text are traditionally measured by counting characters. This constitutes an approximation of the space occupied that takes no account of the widely varying widths of the letters of the Hebrew alphabet. In 1QIs^a, for example, the largest letter (ç) is over 2.5 times the size of a z and over 3.5 times the size of a final ˆ. The proposed new measure takes full account of the differing letter widths. It entails calculating, for the scroll in question, the average width of each letter of the alphabet, and, based on this, an ‘expected width’ for each proposed reconstruction is calculated, which is the sum of the average letter widths of its constituent letters. Thus the expected width of la is the sum of the average widths of an a, a l, and the following space.

Development 2: Vertical Dividers. A closely linked development is the construction of vertical dividers, which are vertical lines which link points that are directly above each other on an extant fragment. Expected widths are therefore calculated, not just for

TynBul 47:1 (1996) p. 178

complete lines, but for sections, which represent a part of a line that is bounded on both sides by a vertical divider, or by a vertical divider and the right margin. The expectation here is that the amount of material between two given dividers would be similar for each line. This not only places the focus on shorter sections of a line, but also removes the effect of the ‘ragged’ left margin from all but the left–hand section.

Development 3: Table of Critical % Deviations. The question now arises as to how much the expected length of a section has to differ from the average of the corresponding sections of neighbouring lines before one has to conclude that the reconstruction is implausible. Where the difference is substantial (say, 50mm against a range of 39–42mm for corresponding sections of neighbouring lines), it is reasonably clear that the proposed...

visitor : : uid: ()