The following answer summarizes the exhaustive survey of information given by John J. Collins, in his Hermeneia commentary on Daniel.
Regarding the split of the Book of Daniel into two languages, Collins presents four theories that prevail in scholarship (pages 12-13):
- The author deliberately switched between Aramaic and Hebrew. This view, based on the footnotes, appears to have been common in the eighteenth, nineteenth, and early twentieth centuries.
- The whole book was written in Hebrew, but portions were lost, and were filled in from an Aramaic translation. This view enjoyed some popularity in the late nineteenth century, but has since been largely abandoned.
- The whole book was written in Aramaic, with portions supplanted, for whatever reason, by their Hebrew translation. This view, Collins notes, has received most attention by English-speaking scholarship.
- The plurality of languages is incidental, due to the book being a composite of older Aramaic material being added to newer Hebrew material.
The view of the original question is expressed in the third theory. Collins writes (13):
Unlike the theory of a Hebrew original, this position is based on philological arguments that have been presented and are discussed below.
In addressing the character of the Hebrew chapters, Collins outlines Aramaisms, vocabulary preferences, features of syntax and grammar, late verb forms, and late idioms (21-22). The survey shows an affinity with the Hebrew of Chronicles, Ezra, Nehemiah, Ecclesiastes, and Qumran scrolls.
Following S.R. Driver, Collins mentions that 'Late Biblical Hebrew is commonly regarded as a stylistically inferior and somewhat degenerate form of the language'. With that being the case, it is especially important that Collins then notes:
The prayer in chap. 9, a traditional composition made up largely of Deuteronomic phrases, stands out from its context because of its fluency.
At minimum, we would expect Daniel 9 to have been composed in Hebrew rather than Aramaic, since it borrows from the Hebrew text of Deuteronomy. Nevertheless, Collins immediately addresses the third theory, that the Hebrew of Daniel was translated from an Aramaic original.
Daniel 1:1-2:4a is taken separately from Daniel 8-12, since it is widely thought the opening chapter was written specifically as an introduction which foreshadows the individual court tales:
Although the presence of Aramaisms is not necessarily a mark of translation, it is noteworthy that about half the Aramaisms attested in the Hebrew sections of Daniel occur in chap. 1. Also, the Hebrew of chap. 1 can be retroverted into Aramaic without recourse to emendations or positing misunderstandings on the part of the translator. None of this proves that Aramaic 1:1-2:4a was originally composed in Aramaic, but the evidence is highly compatible with that theory.
Regarding Daniel 8-12, John J. Collins turns to the works of H.L. Ginsberg's Studies in Daniel, and L.F. Hartman and A.A DiLella's The Book of Daniel. According to these three scholars, Daniel 8-12 was originally written in Aramaic, and the Hebrew translator repeatedly misunderstood what he was reading. Their claims of alleged mistranslation, as provided by Collins, are exampled in Daniel 8.11[-12]:
- Hebrew והשלך מכון מקדשו וצבא came from Aramaic ותהפס מקדש וחסין
- והשלך corrupts ותרמוס, mistaking תהפס
- מכון is displaced
- The ו suffix on מקדש is a corruption
- צבא corrupts חילא, mistaking חסין
Collins blasts such attempts at restoring an original Aramaic as ignoring most the surrounding context, which resists attempts at correction. 'A theory that so drastically changes the meaning of the received text, without textual support, cannot be accepted' (23).
In several cases, Aramaic features that have been taken as evidence for translation [into] Hebrew are otherwise attested as features of Second Temple Hebrew [...] The author of the Hebrew chapters was presumably bilingual, since he retained the preceding chapters in Aramaic, and we may supposed that he was "more at home in Aramaic that in Hebrew." The Hebrew text is attested from an early date at Qumran, and no trace of the supposed Aramaic original has been found.
Ultimately, Collins — relying on J.A. Montgomery (A Critical and Exegetical Commentary on the Book of Daniel), K. Koch (Das Buch Daniel), and S. Niditch (The Symbolic Vision in Biblical Tradition) — states the position that Daniel 1.1-2.4 may have been translated from Aramaic (but need not be); chapters 2.5-6.28 originated in Aramaic; chapter 7 also originated in Aramaic, though comes from a later time than the previous chapters; and chapters 8-12 originated in Hebrew, though from an Aramaic-minded author who came from a bilingual community that experienced a surge of enthusiasm for Hebrew in the wake of the Maccabean Revolt.