Analyzing הֶז Grammar and Reading הֶז Texts of Ps 68 : 9 and Judg 5 : 5

Articles in JHS are being indexed in the ATLA Religion Database, RAMBI, and BiBIL. Their abstracts appear in Religious and Theological Abstracts. The journal is archived by Library and Archives Canada and is accessible for consultation and research at the Electronic Collection site maintained by Library and Archives Canada. ISSN 1203–1542 http://www.jhsonline.org and http://purl.org/jhs Volume 14, Article 8 DOI:10.5508/jhs.2014.v14.a8


I. INTRODUCTION
It is an almost unavoidable problem of the specialization in modern scholarship that with all the insights achieved, just as many broad-and sometimes deeper-insights and connections are lost.For the study of the ancient Hebrew language, the advantages of specialization are clear and numerous.Advances in many areas of general linguistics, from historical-comparative to typological to a variety of theoretical frameworks, have been and continue to be applied with success to the description of ancient Hebrew phenomena.And yet, exegetical competence is often the unfortunate casualty of achieving competence in historical-comparative Semitics or a linguistic theory.Grammatical descriptions typically present data as if they exist in a syntactic and semantic vacuum, disconnected from the communicative, discourse pragmatic context of the texts out of which the data come.
The typical disconnect between grammar and text may work for many examples, perhaps even a majority of the data, but never for all.For example, in a clause like ‫‬ ‫‬ ִ ‫י‬ ‫ים‬ ִ֑ ‫ִיָמ‬ ‫יף‬ ֣ ‫ִתֹּוס‬ ‫ה‬ ‫ְ֭הוָ‬ ‫ִיְ‬ ‫֣ת‬ ‫אַ‬ ְ ‫ר‬ (Prov 10:27), the 3 rd f. sg.features of the verb unambiguously identify the f.sg.‫ת‬ ‫אַ‬ ְ ‫ר‬ ‫י‬ as the syntactic subject.But, of course, there are many examples in which both the subject and verbal complement match the features of the verb, and only the context allows reasonable interpretation.Consider ‫את‬ ָֽ ָ ‫טּ‬ ַ ‫ִח‬ ‫ֵ֥ף‬ ‫לּ‬ ‫סַ‬ ְ ‫ִתּ‬ ‫ה‬ ָ֗ ‫עָ‬ ְ ‫שׁ‬ ‫ר‬ (Prov 13:6), in which both ‫ה‬ ‫עָ‬ ְ ‫שׁ‬ ‫ר‬ and ‫את‬ ‫טָּ‬ ַ ‫ח‬ match the 3 rd f. sg.features of the verb ‫ף‬ ‫לּ‬ ‫סַ‬ ְ ‫,תּ‬ allowing either to be the syntactic subject.The discourse context, though-in this case, the poetic parallelism-strongly points to ‫ה‬ ‫עָ‬ ְ ‫שׁ‬ ‫ר‬ as the subject since the clear subject of the preced-

II. DEMONSTRATIVE SYNTAX IN ANCIENT HEBREW: AN OVERVIEW
The most recent thorough study of demonstratives from a typological perspective is Diessel 1999. 1 In his introduction, he helpfully previews the nature and distribution of demonstratives.First, he defines demonstratives as "deictic expressions serving specific syntactic functions" (2) and notes a common discourse-pragmatic function: "they are primarily used to focus the hearer's attention on objects or locations in the speech situation (often in combination with a pointing gesture), but they may also function to organize the information flow in the ongoing discourse" (2).Second, Diessel observes that all languages have two sets of demonstratives used for proximal deixis and distal deixis (2).And finally, he organizes the typology of demonstratives into four categories: pronominal, in which a demonstrative serves as the argument (e.g., subject) of a verb or adposition (e.g., the object of a preposition); adnominal, in which a demonstrative accompanies a co-occurring noun; adverbial, in which a demonstrative modifies a verb (e.g., a locative deictic); and identificational, in which a demonstrative serves as the expletive subject of a non-verbal or copular clause, often with a presentative nuance (4-5).Ancient Hebrew ‫ה‬ ‫זֶ‬ fits Diessel's typological profile in straightforward terms.It is often clearly deictic, pointing to entities or events in the preceding discourse, and it is also used in pronominal and adnominal functions.The basic deictic nature (1) as well as the first two of Diessel's four functions-pronominal (2) and adnominal (3)-are illustrated in the examples below.
(2) Pronominal And God said, "This is the sign of the covenant that I am setting between me and you and every living creature that is with you, for everlasting generations" (Gen 9:12).
(3) Adnominal And Yhwh appeared to Abram and said, "To your seed I will give this land."And he built there an altar to Yhwh, who appeared to him" (Gen 12:7).
Whether Hebrew ‫ה‬ ‫זֶ‬ behaves as a demonstrative adverbial, that is, Diessel's third functional category, is unclear due to lack of sufficient data.Diessel restricts the term demonstrative adverbial to "locational deictics such as here and there in English … which are primarily used to indicate the location of the event or situation that is expressed by a cooccuring verb; that is, locational deictics function as some sort of verb modifiers " (1999: 74).In all but a handful of cases, the presence of ‫ה‬ ‫זֶ‬ within an adverbial modifier is complicated by the fact that the ‫ה‬ ‫זֶ‬ is not itself the adverbial, but rather the complement of a preposition (i.e., ‫ה‬ ‫זֶ‬ is a pronominal demonstrative), such as ‫ה‬ ‫זֶּ‬ ‫מ‬ "from this [place]" > "from here" or following a temporal or spatial noun that is used adverbially (i.e,.‫ה‬ ‫זֶ‬ is an adnominal demonstrative), such as ‫ה‬ ‫זֶּ‬ ‫יֹּוםִהַ‬ ‫הַ‬ "this day" > "today."Only in the six cases of ‫ה‬ ‫ה-זֶ‬ ‫נּ‬ ‫ה‬ "hey this!" > "look here"2 does ‫ה‬ ‫זֶ‬ itself approach an adverbial sense, like English here.
Finally, with regard to the fourth functional category, it is doubtful that ‫ה‬ ‫זֶ‬ functions as a "demonstrative identifier," at least as it is described by Diessel (1999: 78-88).There is no distinct class of demonstrative identifiers in Hebrew and ‫ה‬ ‫זֶ‬ when used as the subject of a copula (overt or null) always retains its deictic demonstrative characteristics (Diessel's comment on English is apropos;1999: 79).
Beyond the four core demonstrative functions (two of which are not attested for ‫ה‬ ‫,)זֶ‬ Hebrew exhibits two additional, derivative uses of ‫ה‬ ֶ‫-ז‬as a non-verbal copula (4) and as a relative marker (5).
The copular use of ‫ה‬ ‫זֶ‬ is similar to the grammaticalization of the independent pronoun ‫ה‬ ‫וּ‬ ‫א‬ into a pronominal copula, though the precise path is slightly different and the copular ‫ה‬ ‫זֶ‬ is rare (see Holmstedt and Jones 2014).The relative use of ‫ה‬ ‫זֶ‬ (and ‫זֹו‬ and ‫)זוּ‬ reflects a grammaticalization of the Semitic demonstrative, perhaps as early as proto-West Semitic (PWS); moreover, the ‫ה‬ ‫זֶ‬ relative represents a type of relative element-the relative marker-that is unlike other early Semitic relative elements: the ‫ה‬ ‫זֶ‬ (PWS ) relative marker is inflected (contrary to, e.g., ‫ר‬ ֶ ‫שׁ‬ ‫)אֲ‬ though not for the relativized noun's position within the relative but in the matrix clause, resulting in an inflectional redundancy (see Deutscher 2001Deutscher , 2002;;Holmstedt 2013).

III. THE SYNTAX OF ‫ה‬ ֶ‫-ז‬NP
Grammarians have long noted a small number of cases in which the demonstrative precedes its associated NP, often a quantified NP or number phrase.In Gesenius-Kautzsch-Cowley, the syntax of these examples is explained as apposition, in which the NP is appositive to ‫ה‬ ‫,זֶ‬ or a null copula clause in which the ‫ה‬ ‫זֶ‬ is the subject and the NP is the copular complement (1910: §126aa, 136d Few previous studies have focused on the syntax of ‫ה‬ ‫,זֶ‬ especially the non-conforming examples.Those that have attempted explanations, such as Pennachietti 1980, Joosten 1991, and Pat-El 2007, have contributed important pieces to the grammatical puzzle, though none provides an overall adequate and thus convincing analysis. Pennachiatti argues that the temporal adverbial use of ‫ה‬ ‫זֶ‬ in the ‫ה‬ ֶ‫-ז‬Number phrases (e.g., ‫ים‬ ‫מ‬ ‫עָ‬ ‫ֹלשִׁפְ‬ ָ ‫הִשׁ‬ ‫זֶ‬ Judg 16:15) developed out of a cleft-construction.That is, for discourse structure reasons, the cleft construction ‫י‬ ‫ִבּ‬ ָ ‫תּ‬ ‫לְ‬ ַ ‫ת‬ ‫ִה‬ ‫ים‬ ‫מ‬ ‫עָ‬ ‫ִפְ‬ ‫ֹלשׁ‬ ָ ‫ִשׁ‬ ‫ה‬ ‫זֶ‬ "this is three times (that) you have deceived me," serves both to focus the clefted constituent (the ‫ה‬ ֶ‫-ז‬NP) and "to establish a relationship of identification between the first and second section." 5 From this original cleft structure, Pennachiatti argues that the ‫ה‬ ֶ‫-ז‬Number examples were reanalyzed as adverbial phrases rather than full predications, e.g., "già tre volte ti sei burlato di me!" for Judg 16:15 (226).
Joosten specifically addresses the very rare cases of "a proper noun with an adjectival demonstrative" that precedes the noun (1991: 413) ‫שׂ‬ ֵ֥ ֶ ‫מּ‬ ַ ‫.)דּ‬6He argues that the ‫ה‬ ֶ ‫ֹשׁ‬ ‫הִמ‬ ‫זֶ‬ pattern "implies that the speakers take their distance" from the named person or entity, thus assigning a plausible pragmatic function to the demonstrative-noun sequence, but providing little insight into the syntactic structure.Since he does not find the "distancing" function in Isa 25:9 and Ps 48:15, he suggests that they are better understood as null copula clauses, i.e., "this is Yhwh" and "this is God?" The most recent attempt is in Pat-El 2007, and it is with her study that I will end my survey.Pat-El concentrates on the cases of ‫ה‬ ‫זֶ‬ preceding a number phrase or quantified NP.Though her argument is not always perspicuous, she appears to make the following claims: 1. ‫ה‬ ‫זֶ‬ in these constructions is adverbial (151); 2. The ‫ה‬ ‫נָ‬ ָ ‫יםִשׁ‬ ‫ע‬ ‫בָּ‬ ְ ‫ר‬ ‫הִאַ‬ ‫זֶ‬ construction is just as often at the end of a clause and so cannot be, or be derived from, a cleft construction, contra Pennachietti 1980 (153-54); 3.There are no comparable syntactic patterns in other Semitic languages (154); 4. The ‫ה‬ ‫נָ‬ ָ ‫יםִשׁ‬ ‫ע‬ ‫בָּ‬ ְ ‫ר‬ ‫הִאַ‬ ‫זֶ‬ pattern developed from an "adverbial circumstantial sentence," such as in Gen 31.41 [see appendix, below], "lost the possessive marker l-, perhaps first in colloquial parlance, and became more and more dependent on the main sentence until it was understood to be a circumstantial adverbial phrase, rather than a clause with full nexus" (156).
Though the recent proposals, such as Pat-El's have been creative, I suggest that when it comes to the syntax of ‫ה‬ ‫זֶ‬ we should admit that Gesenius-Kautzsch-Cowley articulated a more accurate description over a century ago.First, nowhere do they make the illadvised claim that ‫ה‬ ‫זֶ‬ is adverbial, either in origin or use.Hebrew ‫ה‬ ‫זֶ‬ transparently patterns according to the pronominal and adnominal functions of demonstrative in all but a very few cases (see above, section II).Even when ‫ה‬ ‫זֶ‬ is used within adverbial phrases, its own function remains clearly demonstrative.Thus, it is confusing at best and simply inaccurate at worst to compare ‫ה‬ ‫זֶ‬ with demonstrative adverbials like English here and there (contra Joüon and Muraoka 2006: 143a).
Second, Gesenius-Kautzsch-Cowley were correct that two syntactic patterns explain all the examples in which ‫ה‬ ‫זֶ‬ precedes an NP or number phrase: either the NP following ‫ה‬ ‫זֶ‬ is in apposition to it or the NP is the complement of a null copula clause, of which ‫ה‬ ‫זֶ‬ is the subject.Consider again Exod 32:1 from (6), repeated in (8).
(8) Exod 32:1 And the people saw that Moses was delaying to come down from the mountain.And the people gathered to Aaron and they said to him, "Get up.Make for us gods that walk before us, because this one, Moses, the man who brought us up out of the land of Egypt-we do not know what has happened to him." In cases like Exod 32.1, the ‫ה‬ ‫זֶ‬ cannot modify ‫ה‬ ֶ ‫ֹשׁ‬ ‫,מ‬ since that implies the existence of at least two men named Moses, "this one (that led us)" versus "that one (that did not lead us)" (see also 1 Sam 21:12; Isa 23:13; Song 7:8).Rather, the deictic pronominal ‫ה‬ ‫זֶ‬ is the people's way of dismissively referring to Moses, who has not met their expectations (similarly, Joosten 1991; see also Brockelmann 1956: 20).The proper noun ‫ה‬ ֶ ‫ֹשׁ‬ ‫מ‬ follows the ‫ה‬ ‫זֶ‬ to clarify it appositively, as does a second appositive, the more complex ‫ישׁ‬ ָ֗ ‫א‬ ‫הָ‬ with its relative clause.Thus, "because this one, Moses, that man who took us out of the land of Egypt-we don't know what has happened to him." In the second pattern, where ‫ה‬ ‫זֶ‬ is followed by a temporal phrase, the ‫ה‬ ‫זֶ‬ points backward or forward to an event that can be quantified, mostly in years.Consider the example in (9).
(9) Deut 8:2 The exclamative force of ( 10) and ( 11) are due to the implicatures arising from the nature of the discourse.Yet the syntax in the two examples is simply a null copula clause with the ‫ה‬ ‫זֶ‬ as the subject and the number phrase as the copular complement. 7hese two patterns, that ‫ה‬ ‫זֶ‬ is followed by an appositive NP or that ‫ה‬ ‫זֶ‬ is the subject of a null copula clause that is completed by a number phrase, build on known, widely attested features of Hebrew grammar.On these points alone, the principle of parsimony suggests that the more complex descriptions, no matter how creative, are dispreferred.If the well-established patterns also ade-quately explain the data within their literary context, then the novel proposals should be firmly rejected.

IV. THE SYNTAX OF ‫ה‬ ‫זֶ‬
ִ

‫י‬ ‫ינַ‬ ‫ס‬ EXPLAINED
How the syntactic analysis of ‫ה‬ ‫זֶ‬ relates to textual interpretation brings us back to the general principle of philology I asserted in the introduction, which I shall formulate a slightly different way heregrammatical analysis must make good sense of a text.To illustrate how this principle ought to work, as well as how it has been ignored, resulting in interpretive unlikelihoods, I will conclude this study by examining the syntax of ‫י‬ ‫ינַ‬ ‫ִס‬ ‫ה‬ ‫זֶ‬ in Ps 68:9 and Judg 5:5, provided in ( 12) and ( 13): (12) Ps 68:9 The earth quaked, the heavens poured down rain at the presence of God, the God of Sinai, at the presence of God, the God of Israel (NRSV; alternatively, "before God, the One of Sinai," e.g., Tate 1990: 160, 163). 8 (13) Judg 5:5 The mountains quaked before the LORD, the One of Sinai, before the LORD, the God of Israel (NRSV; alternatively, "Yhwh of Sinai," e.g., Pat-El 2007: 44). 9 Both examples are from notoriously difficult poems, sometimes considered to be among the earliest texts in the Bible (see, e.g., Sáenz-Badillos 1993: 56-57 as well as the references listed below in note 11). 10 It has become standard to analyze ‫ה‬ ‫זֶ‬ ‫‬ ‫י‬ ‫ינַ‬ ‫ס‬ in both 8 NJPS: "the earth trembled, the sky rained because of God, yon Sinai, because of God, the God of Israel"; NIV: "the earth shook, the heavens poured down rain, before God, the One of Sinai, before God, the God of Israel"; KJV: "The earth shook, the heavens also dropped at the presence of God: even Sinai itself was moved at the presence of God, the God of Israel." 9 NJPS: "The mountains quaked Before the LORD, Him of Sinai, Before the LORD, God of Israel"; NIV: "The mountains quaked before the LORD, the One of Sinai, before the LORD, the God of Israel"; KJV: "The mountains melted from before the LORD, even that Sinai from before the LORD God of Israel." 10 See Frolov 2011 for arguments placing the Hebrew of Judges 5 between c. 700 and c. 450 B.C.E.Though there is certainly not the space in this study to address Frolov's arguments, I find it compelling that the language of Judges 5 mirrors the early monarchic period, but I disagree with the implication that a monarchic setting suggests that there is no distinction between "archaic" and "standard" biblical Hebrew.Also, Frolov uses the same problematic argument typified by Young and examples as remnants of the proto-West Semitic -series "determinative-relative" and to take the ‫ה‬ ‫זֶ‬ as a marker of the genitive, resulting in "God of Sinai" (Ps 68:9) and "Yhwh of Sinai" or "Yhwh, the one of Sinai" (Judg 5:5). 11The sample translations in ( 12) and ( 13) and notes 8 and 9 reflect this analysis.
There are two good reasons that the genitive analysis is very unlikely for Ps 68:9 and Judg 5:5-1) it is an ad hoc grammatical analysis and 2) it makes little sense in their poetic contexts.I will spell out each of these issues below, but first a comment on the "genitive" analysis of ‫ה‬ ‫זֶ‬ in comparative Semitic perspective is in order.
Based on this comparative evidence, the Hebrew ‫ינַי‬ ‫הִס‬ ‫,זֶ‬ or more fully, the sequences ‫י‬ ‫ינַ‬ ‫הִס‬ ‫יהוהִזֶ‬ or‫י‬ ‫ינַ‬ ‫הִס‬ ‫יםִזֶ‬ ‫ֹלה‬ ‫,אֱ‬ are understood as "genitival" in that the ‫ה‬ ‫זֶ‬ marks a genitive relationship between ‫יהוה‬ or ‫ים‬ ‫ֹלה‬ ‫אֱ‬ (as the "possessed," broadly understood) and ‫י‬ ‫ינַ‬ ‫ס‬ (as the "possessor").While the comparative evidence for ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ is salient, the flaw lies in how the "genitive" origin of this Semitic construction has been understood and applied to the Hebrew phrase.
In an insightful argument concerning the origin of subordination, Deutscher (2009) points to this very structure in Old Akkadian, such as, TU.RA -"he should take those.of/(who) were delayed in illness" (209, with slight modification), in which the demonstrative is the East Semitic reflex of the West Semitic -demonstrative-cum-relative.In this example, the demonstrative is the head of an unmarked relative clause; moreover, it is bound to the unmarked relative that modifies it (a common Semitic structure; see, e.g., Holmstedt 2008).Deutscher then argues that relative clauses formed using the -demonstratives, such as --u "the judgment that he rendered," reflect the integration of an appositional structure, in which a noun (e.g., -um) is modified by demonstratively-headed relative clause (e.g., -u), that is, "the judgment, that one.of(that) he rendered." From this apposition structure, the "demonstrative pronoun was degraded from an independent head of a relative clause to a mere marker of the onset of a relative" (209).
Deutscher's cogent analysis can be directly applied to most of the Semitic parallels adduced in the study of ‫י‬ ‫ינַ‬ ‫הִס‬ ‫.זֶ‬ In each of these demonstrative-noun constructions, the analytical options are either that the demonstrative retains its demonstrative force (and is bound to and so modified by an unmarked relative clause, e.g., -"that one of ara" or "that one (who is) of ara") or has become a relative word modifying a null head, "(the one) who is of ara."In either case, the demonstrative pronoun itself is not a genitive marker (contra, e.g., Allegro 1955;Pat-El 2010); rather, the "genitive" nature of the relationship is a product of the bound/cliticization construction in Semitic.
As most grammatical references recognize, and as I have briefly described in this study, there are four well-established uses of the deictic demonstrative ‫ה‬ ‫זֶ‬ in ancient Hebrew: 1) pronominal, 2) adnominal, 3) relative, and 4) copular.A "genitive" use is not among these.Moreover, Diessel 1999 provides no comparable typological evidence for a demonstrative used as a genitive marker as suggested for ancient Hebrew. 13This strongly suggests that we should consider whether one of the four established functions for ‫ה‬ ‫זֶ‬ can adequately explain the data.
Perhaps more damning for the "genitive" analysis of ‫ה‬ ‫זֶ‬ in Ps 68:9 and Judg 5:5 than the misunderstood comparative Semitic data is the poetic structure of the two verses, which seems to have been ignored by those who have proposed the "genitive" analysis.It is worthwhile starting our poetic discussion by noting that the Masoretic tradition divides the verse in Ps 68:9, given again in ( 14), by the ‫ד‬ ‫יֹור‬ ‫יִוְ‬ ‫ֹל‬ ‫ע‬ on ‫ים‬ ‫ֹלה‬ ‫,אֱ‬ leaving ‫ה‬ ‫זֶ‬ ‫‬ ‫י‬ ‫ינַ‬ ‫ס‬ in the second half. 14Similarly, in Judg 5:5, given again in (15), the ‫ח‬ ‫נָ‬ ְ ‫ת‬ ‫אַ‬ after ‫ה‬ ‫הוָ‬ ‫יְ‬ once again places ‫ה‬ ‫זֶ‬ ‫‬ ‫י‬ ‫ינַ‬ ‫ס‬ in the second half of the verse (so also Frolov 2011: 166).( 14) Ps 68:9 13 The only potential demonstrative-genitive connection Diessel describes is one of the proposed origins for a genitive linker in Chadic, that it developed from a demonstrative used as an adnominal determiner or as the demonstrative used pronominal with a possessor NP in apposition.Importantly for the discussion of Hebrew, though, is that in the Chadic data, the NP following the demonstrative is the possessor NP, i.e., John's in the horse of John's (1999: 131-32).This structure has no parallel in the Hebrew data, in which the alleged "genitive" noun is not a possessor.Indeed, the variable semantics of the cliticization process of the bound (or "construct") structure, which may but do not necessarily signal possession, mislead many Hebraists and Semitics to use the Latinate "genitive" label inaccurately. 14Joüon and Muraoka also invoke the Masoretic ‫ים‬ ‫מ‬ ‫עָ‬ ‫,טְ‬ but their point is opaque (2006: 500, §143i, n. 10).An anonymous reviewer questioned whether the ‫ח‬ ‫נָ‬ ְ ‫ת‬ ‫אַ‬ in Ps 68:9, which is placed on ‫ִ֑י‬ ‫ינַ‬ ‫,ס‬ does not imply that the Masoretes, at least, understood ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ to be associated with the preceding ‫ים‬ ‫ֹלה‬ ‫.אֱ‬The simple answer to this is, absolutely not.In a poetic verse with an ‫ד‬ ‫יֹור‬ ‫יִוְ‬ ‫ֹל‬ ‫,ע‬ this ‫טעם‬ marks the primary division of the verse, which results in a downgrade of the ‫ח‬ ‫נָ‬ ְ ‫ת‬ ‫,אַ‬ which then marks the primary division of whatever half-verse in which it appears (see Yeivin 1981: 266-67).Thus, in Ps 68:9, the ‫ח‬ ‫נָ‬ ְ ‫ת‬ ‫אַ‬ separates ‫ה‬ ‫זֶ‬ ִ ‫ינִַ‬ ‫ס‬ ‫י‬ from the remainder of the half-verse, ִ ‫ֹלה‬ ‫ִאֱ‬ ‫ים‬ ‫ֹלה‬ ‫ִאֱ‬ ‫י‬ ‫נ‬ ‫פְ‬ ‫מ‬ ‫ל‬ ‫א‬ ָ ‫ר‬ ְ ‫שׂ‬ ‫ִי‬ ‫י‬ .That is, the ‫ח‬ ‫נָ‬ ְ ‫ת‬ ‫אַ‬ neatly separates the subject from the complement of the gapped verb, ‫שׁ‬ ‫עַ‬ ָ ‫.ר‬ The earth shook-even the heavens dripped-before God; this one, Sinai, (shook) before God the God of Israel. 15(15) Judg 5:5 The mountains streamed 16 before Yhwh; this one, Sinai, (streamed) before Yhwh, God of Israel. 17 15 Knohl (2012) proposes that ‫י‬ ‫ינַ‬ ‫ס‬ in Ps 68:9, 18 is not a geographic reference but an epithet for ‫,יהוה‬ thus "The earth quaked, the heavens poured down rain, before Elohim-that is, Sinai."With this supposed epithet, Knohl sees an early equation of ‫יהוה‬ with the Mesopotamian moon god Sin.But the textual argument for this in Psalm 68 is very weak.First, Knohl builds the equation on his analysis of 68:18, in which he finds the reference to Mount Sinai too close to the description of Mount Bashan as the eternal residence of ‫יהוה‬ in vv.16-17.Even Knohl acknowledges, however, that by a small (and common) emendation the reference to ‫י‬ ‫ינַ‬ ‫ס‬ in v. 18 can be understood to "depict God as coming from his abode at Mount Sinai in order to assist his people in the war (see Judg 5 and Hab 3), and moreover, as deciding to reside on Mount Bashan and to make this mountain into His eternal bode" (2012: 16).Second, Knohl's analysis is definitely not "based on the usual meaning of the word ‫זה‬ in Biblical Hebrew" (2012: 17), since his translation "that is, Sinai" does not transparently reflect any accepted understanding of ‫ה‬ ‫,זֶ‬ but suggests that he takes it (uniquely) as an appositive marker (and Isa 23:13, which he provides in n. 44 as a similar use of ‫ה‬ ‫,זֶ‬ is not a helpful comparison; see below, example (41) in the Appendix).In sum, there is no good literary or grammatical reason to take ‫י‬ ‫ינַ‬ ‫ס‬ as an epithet of ‫יהוה‬ in Psalm 68. 16The Qal ‫נזל‬ "flows" is often repointed, based on the putative evidence of ἐσαλεύθησαν in the LXX, as a Nifal from ‫,זלל‬ so ‫ֹלּוּ‬ ‫ז‬ ‫נָ‬ "quaked"; see NRSV; DCH, s.v.‫;נזל‬ HALOT, s.v. ‫;נזל‬ Keil 1874: 249;Boling 1975: 108;Butler 2009: 115-16.However, the Greek evidence is not as clear as the lexica and commentators suggest or imply and the sense of "stream" for mountains is neither illogical nor out of place in the literary context of both Judges 5 and Psalm 68.First, the LXX rendering ἐσαλεύθησαν for ‫לוּ‬ ‫זְ‬ ‫נָ‬ in Judg 5:5 does not clearly suggest that a verb from the root ‫זלל‬ II "to quake" (HALOT) was in its Vorlage.The root ‫זלל‬ II "to quake" (DCH ‫זלל‬ III) appears only in Isa 63:9 and 64:2.In both cases the LXX has the noun τρόμος 'trembling' (not a verb).Moreover, at the end of the same clause in Is 63:9 is τακήσονται (< τήκειν "to melt"), which reflects ‫נזל‬ "to flow" (Muraoka 2010: 275), not ‫זלל‬ II.Thus, the Greek evidence does not clearly support repointing the MT Hebrew ‫לוּ‬ ‫זְ‬ ‫נָ‬ to ‫ֹלוּ‬ ‫זּ‬ ‫.נָ‬ Second, the image of mountains melting and subsequently "flowing" (see, e.g,.Ps 147:18 for the sequence "melting-flowing") is neither illogical nor necessarily foreign to the biblical conception of theophanic effects.The language of the mountains "dripping" and "flowing" occur in images of agricultural bounty connected to restoration (see, e.g,.Joel 4:18, Amos 9:13) and is arguably extended to Yhwh's appearance in judgment in Mic 1. 4 14) and to ‫ים‬ ‫ר‬ ‫הָ‬ in (15).But given the difficulties these verses have posed for interpreters, I will briefly situate each verse within its literary context and describe the poetic features that I have represented in my translations.
Psalm 68 covers numerous themes, often dealing with Yhwh's kingship, though scholars have had difficulty determining a clear structure. 18However the psalm is structured overall, vv.8-11 are widely recognized as a unit in which most interpreters see a theophany that builds both on the imagery of the storm-god and God's advocacy for his people.The poetic structure of v. 9 consists of a well-balanced bicolon, with three prosodic word-pairs in each half, illustrated in ( 16) by the division of pairs using ||: (both to clarify who this ‫אלהים‬ is and to balance the lines), the first half includes its own complication-a parenthetical clause, beginning with ‫ף‬ ‫,אַ‬ that sits between the verb ‫ה‬ ָ ‫שׁ‬ ‫עָ‬ ָ ‫ר‬ and its adjunct ‫ים‬ ‫ֹלה‬ ‫יִאֱ‬ ‫נ‬ ‫פְ‬ ‫.מ‬In sum, in Ps 68:9, when Yhwh marched before his people, the earth ‫,רעשׁ‬ the heavens ‫,נטף‬ and Mount Sinai ‫רעשׁ‬ (by gapping from the preceding main clause), all of which reflects a progression of nature's responses to Yhwh's theophany during the Exodus. 19 Like Psalm 68, the complications surrounding the interpretation of Judges 5 are numerous, and run the gamut from genre to 17 Moore (followed by Seeligman 1964: 80, n. 1) suggests that ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ in Judg 5:5 is a gloss (1895: 142).Fishbane argues similarly that ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ in Ps 68:9 is a scribal gloss to connect the earthquake imagery with the Sinai theophany: "It is therefore readily conceivable that a later glossator-scribe read the particular imagery of Ps. 68:9 within its particular historicogeographical context as an allusion to the Sinaitic revelation-and so interrupted the a∼ab parallelism ('before Elohim … before Elohim, God of Israel') with the specifying comment: ‫סיניִזה‬ 'this [earthquake caused by Elohim] refers to [the theophany of] Sinai '" (1985: 55).Knohl allows that ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ in Judg 5:5 might be a gloss, but dismisses Fishbane's argument for Ps 69:9 (2012: 16, n. 43).I see no reason to take ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ as a gloss in either text; it fits the literary development and poses no grammatical problems, as I argue below.
18 See, among myriad others, Gray 1977; Dahood 1968: 133-34; Tate  1990: 170-75; Hossfeld and Zenger 2005: 160-64; Knohl 2012.  1Although not all scholars connect the mention of ‫י‬ ‫ינַ‬ ‫ס‬ in Ps 68:9 to the Exodus traditions (see, e.g., Knohl 2012), many do take such a connection to make sense not only of the word ‫י‬ ‫ינַ‬ ‫ס‬ in the verse but also of the perceived literary progression (or concentricity) of the Psalm as a whole (see, e.g., Tate 1990: 177;Hossfeld and Zenger 2005: 165).structure, language to dating. 20This passage is the Song of Deborah about the Israelite victory over the army of Hazor.But more than a poetic recital of the victory that was narrated in Judges 4, the song in Judges 5 celebrates both those who ‫ה‬ ‫הוָ‬ ‫ִיְ‬ ‫ת‬ ַ ‫ר‬ ‫זְ‬ ‫עֶ‬ ‫ִלְ‬ ‫אוּ‬ ‫בָּ‬ (to paraphrase 5:23) and Yhwh himself as Israel's champion (Webb 2012: 205-6).After the narrative introduction (v. 1) and then brief preliminary elements of the song in vv.2-3, the song quickly moves to the powerful image of Yhwh marching forth to battle, beginning from the southern areas associated with the beginning of his relationship with Israel, ‫יר‬ ‫ע‬ ‫שׂ‬ ‫דֹום/‬ ‫אֱ‬ (v.4).This battle march affects creation, including ‫י‬ ‫ינַ‬ ‫ס‬ (v.5), by which the song draws Israel's constitutional tradition into the glorious victory.Just as the psalmist invokes Mount Sinai to connect the theophanic imagery to the Exodus memory (see above, n. 15), so Deborah draws on the Exodus event by invoking Sinai in order to describe the similar victory provided by Yhwh against Sisera: This powerful poetry draws on the historical memory of the exodus, when Yahweh as the divine warrior manifested his presence by making wind and water change their normal behavior for the sake of his people, and Sinai, where his presence was manifested on the mountain in cloud and fire and earthquake … Here in Judges 5 it gives powerful poetic expression to the belief that the storm which broke over Sisera and his army and threw them into panic and retreat was no merely natural event, or amazing stroke of luck; it was unleashed by Yahweh, as he had divided the Sea at the exodus and shaken Mount Sinai by descending on it (Webb 2012: 208).
The poetic structure of Judg 5:5 consists of a bicolon with three words or word pairs in each half, illustrated in (17) using || to divide the units: Though the six units of the bicolon are not quite as neatly isolated as they are in Ps 68:9, the separation of subject phrases ‫ים(‬ ‫ר‬ ‫הָ‬ and ‫י‬ ‫ינַ‬ ‫הִס‬ ‫)זֶ‬ from the other syntactic elements of the clause is clear.Moreover, just as in Ps 68:9, the gapping of the verb in the first half into the second half is compensated for by the appositive ‫ל‬ ‫א‬ ָ ‫ר‬ ְ ‫שׂ‬ ‫ִי‬ ‫י‬ ‫ֹלה‬ ‫אֱ‬ in the second half so that the basic balance between the two lines is preserved.There is also a similar thematic progression in Judg 5:5 compared to Ps 68:9-the earth ‫‪‬רעשׁ‬ (5:4a), the heavens and the clouds ‫נטף‬ water (5:4b), and the mountains ‫‪‬נזל‬ (5:5a) as does Mount Sinai (5:5b gapped from v. 5a).‫‬In summary, in both Ps 68:9 and Judg 5:5, ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ "this (mountain), Sinai" is experiencing the effects of Yhwh's power, not defining who Yhwh is (as a ‫ה‬ ‫זֶ‬ relative clause would do, and presumably as the supposed ‫ה‬ ‫זֶ‬ genitive phrase would do).Thus, the weight of Hebrew grammar as well as the literary features of the verses in question indicate that it unnecessary to propose a novel grammatical function for the demonstrative ‫ה‬ ‫.זֶ‬ Rather, in both cases the ‫ה‬ ‫זֶ‬ operates as a deictic demonstrative followed by an appositive proper noun, ‫י‬ ‫ינַ‬ ‫,ס‬ which clarifies the deictic reference.As such, these examples are no different than the ‫ה‬ ֶ ‫ֹשׁ‬ ‫ִמ‬ ‫ה‬ ‫זֶ‬ (Exod 32:1) or ‫ה‬ ‫זֶ‬ ‫ה‬ ‫נָ‬ ָ ‫יםִשׁ‬ ‫ע‬ ‫בָּ‬ ְ ‫ר‬ ‫אַ‬ (Deut 8:2) types explained above.

V. CONCLUSION
In conclusion, ‫ה‬ ‫זֶ‬ grammar in ancient Hebrew straightforwardly accords to cross-linguistically attested patterns of demonstratives.‫ה‬ ‫זֶ‬ and its feminine and common plural counterparts function primarily as deictic pronouns or deictic nominal modifiers.A small set of examples indicate that some stage of Hebrew witnessed the grammaticalization of ‫ה‬ ‫זֶ‬ as a relative marker and a copular pronoun.But beyond this, there is no compelling evidence for the use of ‫ה‬ ‫זֶ‬ as a "genitive" marker.Indeed, as I have shown with the two examples most often adduced as ‫ה‬ ‫זֶ‬ genitives, the genitive analysis fails the tests of both grammar and literary sensitivity.Instead, both Judg 5:5 and Ps 68:9 should be understood as the Masoretes understood them, with ‫י‬ ‫ינַ‬ ‫הִס‬ ‫זֶ‬ beginning the second half-verse and the ‫ה‬ ‫זֶ‬ in each as a pronominal demonstrative.
More than simply clarifying the grammar of ‫ה‬ ֶ‫-ז‬NP or ‫ה‬ ‫-זֶ‬ Number phrases in the Hebrew Bible, this study also serves as a bifold warning.First, comparative Semitic evidence must be used cautiously, since there are many nuances of syntax yet to be fully understood.To wit, the demonstrative does not ever serve as a "genitive" marker in ‫ה‬ ‫זֶ‬ constructions (nor in the Semitic parallels).Second, ancient language grammarians, whether operating within an historical-comparative framework or theoretical linguistic framework, must not-indeed, cannot!-ignore the literary context of data examples.To put it more bluntly, linguists and grammarians risk significant error if they ignore (or simply cannot read) texts.

-[
Gk. Dusares] "the [god] of [mount] ara," Li y nite -"the [god] of the Thicket," Classical Arabic -qarnayn "the [man] of two horns," Colloquial Arabic "the [man] of learning," Sabaic -Rezetko 2008 and Rezetko and Young 2014 And you shall remember the whole road that Yhwh your God led you on.This was forty years in the wilderness in order to humble you, to test you, to know what was in your heart … Rather than interpret ‫ה‬ ‫נָ‬ ָ ‫ִשׁ‬ ‫ים‬ ‫ע‬ ‫בָּ‬ ְ ‫ר‬ ‫ִאַ‬ ‫ה‬ ‫זֶ‬ in (9) as some sort of questionably formed adverbial phrase (Pat-El 2007), it is much simpler and avoids imaginative grammatical creations to follow the Masoretes in understanding a small prosodic break before the ‫ה‬ ‫זֶ‬ and so take ‫ה‬ ‫זֶ‬ as the subject of a new clause: "And you shall remember the whole road that Yhwh your God led you on.This was forty years in the wilderness in order to humble you, to test you, to know what was in your heart."ExampleslikeGen27:36,given in (10), should be read similarly but with an unmarked relative following the ‫ה‬ ‫זֶ‬ clause (so also Pennachietti 1980: 238):The proposed syntax of (10) is precisely what Pennachietti argues for the examples he identifies as originally cleft constructions.He is, in fact, correct, though I see no support for his subsequent proposal that these cleft sentences were transformed into non-cleft adverbial phrases.Although Waltke and O'Connor may not accurately describe the syntax of the ‫ה‬ ‫נָ‬ ָ ‫ִשׁ‬ ‫ים‬ ‫ע‬ ‫בָּ‬ ְ ‫ר‬ ‫הִאַ‬ ‫זֶ‬ pattern, the notion that the pattern "emphasize[s] the time" (1990: 311)points in the correct direction.Most of the examples exhibit a clear exclamative function within their discourse contexts.For example, in (10), Esau expresses his distress and anger at being cheated by Jacob two times.Similarly, in (11), the messenger's use of the separate clause ִ ‫ה‬ ‫זֶ‬ ‫ה‬ ‫נָ‬ ָ ‫יםִשׁ‬ ‫ע‬ ‫בְ‬ ‫,שׁ‬ at the end of his plaintive question is intended to reinforce his perspective that it has been entirely too long since Yhwh has acted on behalf of his people (see, e.g.,Petersen 1984: 146-51).