📄 Alma Brodersen, The End of the Psalter. Psalms 146–150 in the Masoretic Text, the Dead Sea Scrolls, and the Septuagint (Beihefte zur Zeitschrift fürdie alttestamentliche Wissenschaft 505; Berlin: De Gruyter 2017). ISBN 978-3-11-053476-4

Summary

This is Wilfred G.E. Watson's short review, in Folia Orientalia vol. 55, of Alma Brodersen's monograph on the closing five psalms of the Hebrew Bible. The book asks whether Psalms 146–150 were composed as one connected unit or began as independent poems. To settle the question, Brodersen looks at every psalm as it is preserved in three witnesses: the Septuagint (LXX), the Masoretic Text (MT), and the Qumran manuscripts known as the Dead Sea Scrolls (DSS). She uses the same fixed sequence of steps every time: text and translation, form (outline, syntax, structure, poetic devices), inner-biblical references, content framed as guiding questions, then genre, date, unity and an overall interpretation. The psalms are treated in reverse order to limit subjective bias, and the three versions are compared at the end. Her conclusion is that the five psalms were separate compositions in origin. Watson praises the work as meticulous and objective and recommends the method for other texts that survive in several traditions. He notes a few minor gaps.

Key points

• The corpus is Pss. 146–150, compared across three witnesses (MT, DSS/Qumran, LXX). The book appeared in 2017 as BZAW 505 (De Gruyter, ISBN 978-3-11-053476-4).
• The psalms are analysed in reverse sequence, from Ps. 150 back to Ps. 146, to avoid subjectivity.
• Content analysis is driven by explicit questions, for example what the "new song" of Ps. 149:1 means (p. 99).
• The final verdict (p. 270) is that the five psalms were originally independent texts.
• Watson's criticisms: tables would have helped, a footnote misattributes a Watson citation (n. 157, p. 246, which refers to Rebecca Watson), and the bibliography (pp. 279–307) omits Jan Fokkelman's The Psalms in Form (2002).

Interesting ideas

• I1. Processing items in reverse of their natural order keeps the expected "story" from colouring each local judgment. — For Mova (learn): Evaluate or repair-rule-induce on shuffled or reversed document order as a negative control, checking that results do not depend on narrative position.
• I2. Every unit goes through one identical, fixed analysis template, so results stay directly comparable across units and versions. — For Mova (frames): Give each clause the same ordered frame-extraction checklist (roles, speech act, implicature) and log every step, so explanations line up across sentences and texts.
• I3. Content is interrogated through explicit questions rather than free description, which turns interpretation into checkable answers. — For Mova (answer): Generate probe questions per passage from its frames and test whether the reader's world model answers them consistently.
• I4. A claim about unity is tested by first analysing each witness separately and only then comparing, instead of merging the witnesses up front. — For Mova (learn): Keep witness-parser outputs fully separate until a final agreement stage, so disagreements are visible and attributable before silver data is accepted.
• I5. Whether adjacent texts form one composition is treated as a hypothesis to test with structural, lexical and referential evidence. — For Mova (world): Decide whether consecutive passages share one world (entities, event order) using explicit evidence, rather than assuming document boundaries equal world boundaries.