A lot! In many ways, the publication of A. Tal’s A Dictionary of Samaritan Aramaic (2 vols., Brill, 2000), based on the decades-long research and collection of lexical material by the late Z. Ben-Ḥayyim, may be said only to have muddied the waters of Aramaic lexicography. To be sure, it provides a long-missing guide to understanding Samaritan Aramaic texts of all periods. But the language there described, most of whose unique lexemes come from the very late “A” version of the Pentateuch, often bears little relation to normative Aramaic, being largely a combination of Hebrew and Arabic words and meanings, as well as the unique words of that text, many of which seem to have a distant relationship to Latin (q.v. e.g. אסולה p. 573!) but not Greek (i.e. those that are not based on Arabic, Hebrew, or a misunderstanding of the Hebrew)! Much of the vocabulary may with certainty be ascribed to the artifical late language first described by Ben-Ḥayyim as “Shomronit” and recently the subject of fine study by M. Florentin in Late Samaritan Hebrew: A Linguistic Analysis of its Different Types (Brill, 2005). In addition, the scholarship presented in Tal’s work is idiosyncratic to say the least: Each group of words is introduced by a presumed root, fully a quarter of which must be said to be imaginary. Infinitive forms of various structures are all given their own entry. Noun forms of the qātōl type, that serve as common participles in later Samaritan, are all given their own lemma. Some words are used in a distinctive way that is based on a misunderstanding of a Pentateuchal form. Definitions often leave much to be desired. Varied spellings of what are obviously all the same lemma are given separate entries. Many of the supposedly Aramaic cited passages are in fact in Hebrew. Scribal errors are included as headwords. There is no index of passages (or any real index at all for that matter). And the less said about etymologies the better.
To give an example of a typical problematic entry that is not included in the CAL, take the root משק, rendered “to rule” (p. 491), which occurs in the A targum. It is without parallel elsewhere in Aramaic, or in Semitic for that matter. It is, however, known from Florentin’s Shomronit Hebrew, where it has been derived from the poorly understood Hebrew text of Gen. 15:2: בן משק ביתי. We see no reason to include such a thing in a dictionary designed to reflect the Aramaic lexicon.
Thus it has been necessary to adopt a strategy for dealing with this material in the CAL in a way that truly reflects the contribution of Samaritan to the Aramaic lexicon: All the words in the clearly Aramaic portions of manifestly earliest texts are to be included; this includes Targum J, book one of Marqe, and the earliest liturgical poems. Infinitives and participles from those texts are listed under the verbal form. Multiple spellings of the same word are combined into a single lemma. Definitions are made to conform to standard Aramaic ones where the connection is obvious. Hebrew usages are included only where they also appear in Jewish Aramaic texts. Words that occur only in the late texts that are also in our database (e.g. Asatir and the later books of Marqe), will be included as they are encountered, for the sake of comprehensiveness and usefulness for students.