Some Notes on Lexicographic Criticism and Biblical Hebrew Studies. Revisiting David J. A. Clines’s Method

1. Introduction

⌅

The late Prof. David J.A. Clines (1938-2022) devoted his life to the enthusiastic advancement of Hebrew philology. Clines’s scholarly legacy is distinguished by his three-volume commentary on Job (1989-2011)Clines, David. 1989-2011. Job. 3 vols. Word Biblical Commentary. Dallas: Word Books.
and the Dictionary of Classical Hebrew (hereafter DCH), of which he served as an editor in chief throughout its duration (1993-2011). In spite of its shortcomings and acknowledged ideological underpinnings, the DCH represents a significant milestone in Hebrew lexicography. In addition to his academic accomplishments and prolific writings, Clines should be remembered as a Hebrew scholar deeply committed to ensuring the open accessibility of his research ¹ His academia.edu website gathers early drafts of his publications and conference papers, including bibliographical lists of yearly contributions. . Between the years 2014-2021, Clines explored a new direction ―previously inaugurated in New Testament Studies by Lee (2003)Lee, John A. 2003. A History of New Testament Lexicography. Studies in Biblical Greek 8. New York: Peter Lang.
― which he named “Comparative (Classical) Hebrew Lexicography” (hereafter, CHL). During the years 2018-2022, thus partly coinciding with this timeframe, I dedicated myself to crafting my doctoral dissertation on the same subject as Clines’s, albeit from a different perspective and under an alternate name (see below, §2). Regrettably, I remained unaware of Clines’s contributions until 2021, when his final article on the subject was published in Vetus Testamentum. On the one hand, coming across his research confirmed my intuition about the future of Hebrew lexicographic studies ―namely, that the examination of existing dictionaries would slowly take precedence over the writing of new ones. On the second hand, and most importantly, my late discovery of Clines’s work in CHL enabled me to develop an independent method of analysis. The aim of this article is to counterbalance Clines’s method and to attempt to fix some of the cracks I found in the way he paved.

Before discussing CHL, I shall first provide a brief bibliographical overview. In 2015, Clines wrote a book chapter on the intriguing topic of Asian lions and their portrayal in the Hebrew Bible. The core of the analysis focused on the alleged anthropomorphised characterisation of their predatory behaviour in biblical accounts. However, Clines also decided to include an appendix where he undertook his first exercise in CHL, entitled “The Gender of לָבִיא.” There, he argued that the meaning “lioness” is a bogus gloss rooted in 17^th-century lexicography whose popularity has not faded away in contemporary dictionaries despite being morphologically suspicious and textually untenable (2015, 70-76Clines, David. 2015. “Misapprehensions, Ancient and Modern, about Lions (Nahum 2:13).” In Poets, Prophets, and Texts in Play. Studies in Biblical Poetry and Prophecy in Honour of Francis Landy, edited by Ehud Ben Zvi, Claudia V. Camp, David M. Gunn, and Aaron W. Hughes, 58-76. London / New York: T&T Clark.
). In this study, the survey of lexica went back as early as Buxtorf’s Lexicon hebraicum et chaldaicum (1615). The chronological aspect of Clines’s CHL constitutes its primary drawback (and peril), and it will be duly discussed later (see §3.1).

In 2017, Clines systematised his approach and presented it to Hebrew scholars under its official name ² An early draft of this contribution had been available online since 2014. . The aims of CHL are specified in the first place, followed by a pithy depiction of the history of these dictionaries, where the information dealing with the mediaeval period is unfortunately misleading (see §4 and §5.1). In order to show the functioning of his method, Clines analysed the words חֵיל, לָבִיא, גלה I and II, and שׁקע I and II. This time, the earliest item in his lexicographic corpus is Reuchlin’s De rudimentis hebraice (1506)Reuchlin, Johannes. 1506. De rudimentis hebraicis. Pforzheim: Thomas Anschelm.
.

In 2019, Clines read a paper on the emendation of אֳנִיּו̇ת in Psa 104:26 for אֵימו̇ת. He traced its presence in modern lexica from the beginning of the 20^th century onwards and managed to locate its origins in a proposal made by Gunkel in the late 19^th century ³ An earlier draft was available online since 2015 under the title “Revisiting the Hebrew Dictionary. 3. There Go the Ships,” which belongs to a series of articles regarding the preparation of the DCH Revised. .

Finally, in 2021 and as the corrected version of a conference paper read six years before, Clines published an exercise in CHL on the basic meanings of the verb קָדַשׁ in the history of Hebrew dictionaries ⁴ An earlier draft was available online since 2015 under the title “The Holy and the Clean: An Excursion in Comparative Lexicography.” It was read at the session of the International Syriac Language Project, Society of Biblical Literature Annual Meeting, in Atlanta. . Like in previous publications, dictionaries prior to Reuchlin’s were ignored.

Clines pioneered this approach in Biblical Hebrew Studies and his contributions are valuable and ground-breaking. Nevertheless, he built a chronologically mutilated corpus that, given the history and nature of lexicography, is bound to produce inconclusive results. At the end of this article, this problem is illustrated through the reappraisal of the position of Aramaic in these dictionaries (see §5.1) and the long-lived interpretation חֵיל = “outer wall” (see §5.2).

2. The Name of a Recent Discipline

⌅

In the 1960s, lexicography experienced what may be described, without fear of exaggeration, as a Copernican shift. The status of dictionaries switched. So far, they were viewed as tools that describe the reality of words but, suddenly, their definitions became a collection of words to be examined. This metamorphosis into cultural artifacts that could be close read is a remarkable change in the history of lexicographic studies. The doctoral dissertation of Quemada (1967)Quemada, Bernard. 1967. Les dictionnaires du français moderne. 1539-1863. Paris: Didier.
set the beginning of this revolution.

His disciple Pruvost (2003)Pruvost, Jean. 2003. “Some Lexicographic Concepts Stemming from French Training in Lexicology (Part One).” Kernerman News 11: 10-15.
underscored the terminological differentiation between the triplet lexicography-dictionaric-metalexicography. Whereas the meaning of the first word is clear, even if qualified or not with the adjectives “practical” and “theoretical,” the distinction between the pairs practical lexicography-dictionaric and theoretical lexicography-metalexicography is problematic. The terms overlap and interchange in the jargons of different lexicographic schools. For example, the Spanish school equates metalexicography with theoretical lexicography, provided that it reflects upon and creates terminology about itself (Seco 2003Seco, Manuel. 2003. Estudios de lexicografía española. 2^nd ed. Madrid: Editorial Gredos.
). Simultaneously, lexicography is understood by some as inherently practical and becomes, thus, synonymous to practical lexicology (Coseriu 1986Coseriu, Eugenio. 1986. Introducción a la lingüística. 2^nd ed. Madrid: Editorial Gredos.
). In the French school to which Quemada and Pruvost belong, lexicography entails the study of a set of words that is not necessarily intended for publication in the format of a dictionary. Meanwhile, dictionaric refers to the management of all the commercial, editorial, and typographical aspects involved in dictionary-writing (Pruvost 2010-2011Pruvost, Jean. 2010-2011. “La traque lexicographique et dictionnairique: du loup au chat en passant par le vin, le mariage et le citoyen.” ÉLA. Études de linguistic appliquée 157: 103-110. https://doi.org/10.3917/ela.157.0103 .
). Pruvost provided a succinct and accurate definition of metalexicography as “l’analyse des dictionnaires, forme et fond, d’hier et d’ajourd’hui” (Pruvost 2010-2011, 103Pruvost, Jean. 2010-2011. “La traque lexicographique et dictionnairique: du loup au chat en passant par le vin, le mariage et le citoyen.” ÉLA. Études de linguistic appliquée 157: 103-110. https://doi.org/10.3917/ela.157.0103 .
). A wider concept of metalexicography is envisioned by Burkhanov and described as the theoretical study of lexicographic activity and lexicographic products (1998, 154), which brings harmony to the Spanish and French definitions mentioned above. “Metalexicography” was the term I selected for my doctoral dissertation (2022), but the terminological clutter surrounding this word challenges its suitability. For clarity, other options will be explored below.

In his dictionary of lexicographic terminology, Burkhanov included dictionary criticism as a subfield within metalexicography and said that it “deals with dictionary reviews and general models for the assessment of existing dictionaries, encyclopedias, usage guides, and other kinds of lexicographic product” (Burkhanov 1998, 66Burkhanov, Igor. 1998. Lexicography. A Dictionary of Basic Terminology. Rzeszow: Wydawnictwo Wyzszej szkoly Pedagogicznej.
). Hartmann and James qualified this by stating that dictionary criticism “may involve studies of the historical background of the work and/or its compiler(s), detailed comparisons of its contents with other products of its type […] and result in a review in a periodical publication” (Hartmann and James 2002, 32Hartmann, Reinhard R.K. and Gregory James. 2002. Dictionary of Lexicography. 2^nd ed. London / New York: Routledge.
). Both excerpts make a connection between dictionary criticism and reviews. In fact, dictionary criticism (calque of German Wörterbuchkritik) is widely used as a synonym for the writing of reviews in specialised journals (Ripfel 1989Ripfel, Martha. 1989. Wörterbuchkritik: Eine empirische Analyse von Wörterbuchrezensiones, Lexicographica Series Maior 29. Tübingen: Max Niemeyer Verlag.
). Dictionary criticism is market-oriented, aiming to enhance the effectiveness of the dictionary together with the user’s satisfaction and comfort. Additionally, dictionary criticism is overly preoccupied with today’s lexicography and today’s spoken languages, which inevitably diminishes the relevance of certain discussions for Biblical Hebrew Studies.

Another current label is critical lexicography. Scholars in this field claim that dictionaries are subject to the (postmodern) triad of ideology-power-politics, and their analytical tools mostly derive from a combination of Critical Theory, Critical Discourse Studies, and Postcolonial Studies, which eventually means that they target the lexicographers’ social accountability. For example, concerning critical lexicographic discourse studies, Chen has proudly asserted that “it is seen as a social movement” (Chen 2019, 16Chen, Weng. 2019. “Towards a Discourse Approach to Critical Lexicography.” International Journal of Lexicography 39 (3): 1-27. https://doi.org/10.1093/ijl/ecz003 .
) and, as such, that it is “oriented to critiquing and improving society (e.g. contributing to social equality)” (Chen 2019, 7Chen, Weng. 2019. “Towards a Discourse Approach to Critical Lexicography.” International Journal of Lexicography 39 (3): 1-27. https://doi.org/10.1093/ijl/ecz003 .
). Although the presence of ideology in dictionaries is hardly refutable, this approach is embedded in a Foucauldian academic culture. It is often driven by politics to such an extent that the boundaries between scholarly engagement and activism become blurred, sometimes even merging in the role of the activist scholar (see §3.2.2).

Clines decided to employ the label comparative lexicography, which he found in Hartmann and James as

[a] branch of general lexicography which contrasts the dictionary traditions of various cultures, languages and countries with a view to distilling from them common principles, by considering the external factors that have led to divergent practices. Examples include issues such as how different scripts influence the format of reference works, which different genres predominate, and what constitutes good practice in dictionary-making and dictionary use (Hartman and James 2002, 228Hartmann, Reinhard R.K. and Gregory James. 2002. Dictionary of Lexicography. 2^nd ed. London / New York: Routledge.
).

According to Clines, CHL should be understood as “the systematic study of Hebrew lexica in comparison with one another” (Clines 2017, 228Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). I would argue that the problem with the label “comparative lexicography” is that the juxtaposition of the meanings of each word only makes sense if we assume that lexicography here stands for theoretical lexicography and if theoretical lexicography means something beyond the technical debates on dictionary-making. The semantic duality of the term “lexicography” (theoretical vs. practical) makes the label “comparative lexicography” sound odd.

The syntagm dictionary criticism is, I anticipate, very appealing to biblical scholars. It echoes traditional interests and methods regarding the scientific study of the Bible, such as textual criticism, form criticism, source criticism… By virtue of constant use, biblical scholars have become immune to the negative connotations of the word “criticism” / “Kritik” that cause so much unease amongst scholars in the field of Western lexicographic studies (Ripfel 1989, 1Ripfel, Martha. 1989. Wörterbuchkritik: Eine empirische Analyse von Wörterbuchrezensiones, Lexicographica Series Maior 29. Tübingen: Max Niemeyer Verlag.
; Swanepoel 2017, 12Swanepoel, Piet. 2017. “The contribution of dictionary criticism to dictionary research.” In Wörterbuchkritik ― Dictionary Criticism, edited by Monika Bielinska, and Stefan J. Schierholz, 11-32. Berlin / Boston: De Gruyter.
; Nuccorini 2017, 222Nuccorini, Stefania. 2017. “English dictionary criticism in two lexicographical specialist journals.” In Wörterbuchkritik ― Dictionary Criticism, edited by Monika Bielinska and Stefan J. Schierholz, 221-243. Berlin / Boston: De Gruyter.
). However, “dictionary criticism” has meant “writing reviews of dictionaries” for decades and the method of analysis that I propose trespasses the boundaries of reviewing. Indeed, I consider it counterproductive to limit this type of research to one academic genre in particular. Therefore, and unwilling to further complicate the terminological landscape with resignifications or overlapping terms, I shall henceforth employ a label whose components are semantically intuitive to the biblical scholar: lexicographic criticism (hereafter LC).

Before concluding this discussion, one last detail about Clines’s terminology should be addressed. In the presentation of his method, he wrote the following:

If you search for the term “Comparative Hebrew Lexicography” or “Comparative Classical Hebrew Lexicography” on Google, you will find ― zero hits (apart, that is, from references to the present article, which has been on the Web since July 2014) (Clines 2017, 227-228Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

Although the lack of results for “Comparative Hebrew Lexicography” is symptomatic of an academic gap to be filled, the absence of results for “Comparative Classical Hebrew Lexicography” is fallacious. Note that Clines did not try the combination “Comparative Biblical Hebrew Lexicography,” which was motivated by his well-known reluctance to trap the language within the contours of the canon (the Hebrew Bible) ⁵ See, for example, the first volume of the DCH (Clines 1993, 14 and 27-29). . The problem of these search terms is that they force research to agree on a recent redefinition of the linguistic corpus (classical Hebrew; all linguistic strata prior to the Mishnah) to the detriment of a traditional and standardised one (biblical Hebrew)⁶ For an historical survey of the expressions “classical Hebrew” and “ancient Hebrew,” see Carbonell Ortiz (2022b, 20-23). See especially p. 23 n. 57 for a critique of the use of the adjective “classical” in the context of Hebrew. . If we push the argument to its reductio ad absurdum, it would mean that, stricto sensu, Comparative Classical Hebrew Lexicography should limit itself to the comparison between the data found at the DCH and the SAHD, since these two are the only existing dictionaries whose vocabulary is classical rather than biblical⁷ For a presentation of the theoretical framework of the SAHD (Semantics of Ancient Hebrew Database) and a first sample of entries, see Muraoka (1998). The history of this online dictionary has been addressed in Carbonell Ortiz (2022b, 74-76). .

3. Clines’s Method vis-à-vis an Alternative Proposal

⌅

The following pages contain a detailed examination of Clines’s proposal of CHL regarding four different methodological aspects. This reappraisal is complemented with the presentation of an alternative approach as I developed it over the last years.

3.1. A Mutilated Corpus: The Absence of Jewish Dictionaries

⌅

The main limitation in Clines’s method lies in the chronological range of his lexicographic corpus. In his articles, 16^th-century Christian Hebraism is taken as the starting point of the corpus and dictionaries written in European languages are exclusively used. The dismissal of mediaeval Jewish lexicography is systematic. The first explicit acknowledgement of this reads as follows:

My scope here is strictly the ancient Hebrew language, which I call “Classical Hebrew,” and lexica in European languages from the 16th century onwards (leaving aside the mediaeval and later dictionaries in Hebrew or in Hebrew and in Arabic) (Clines 2017, 228Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

This statement is problematic in a number of ways. Let us break the argument into three points, starting from the end of the quotation:

The label “dictionaries in Hebrew and in Arabic” is a misnomer. Such a thing would mean either bilingual dictionaries (Hebrew-Arabic) or dictionaries containing bilingual definientes in Hebrew and Arabic ⁸ There are indeed some dictionaries with bilingual definientes in the history of Hebrew lexicography. A recent example of this phenomenon is the KBL (1967). As part of a witty commercial strategy, this dictionary included definitions in German and, immediately afterwards, in English. Another case is Fünn’s biblical and Mishnaic dictionary, entitled האוצר (1884-1903), in which Hebrew definitions were accompanied with Russian equivalents. Ben Yehudah's מילון הלשון העברית הישנה והחדשה (1908-1958) encompassed a wide array of languages (Hebrew, German, French, and English), which met the pedagogical requirements of such ambitious an enterprise. However, and to my knowledge, a Hebrew dictionary providing double definitions in Hebrew and Arabic is still unwritten. . Despite the unfortunate wording, Clines was obviously referring to dictionaries written in Judaeo-Arabic. Works on exegesis, grammar, and lexicography were often composed in Judaeo-Arabic in the Mediterranean Basin during the Middle Ages. In fact, the landmarks of Hebrew philology in this period are works written in this language, which makes dispensing with this material in the Middle Ages equivalent to studying modern Hebrew philology while ignoring Gesenius.
“Leaving aside post-mediaeval dictionaries written in Hebrew” means wiping out all modern Israeli lexicography, like Eliezer ben Yehuda’s מילון הלשון העברית הישנה והחדשה (1908-1958), Abraham Even-Shoshan’s מילון חדש (1960), Yaakov Choueka’s רב מלים (1997), and, most importantly given the linguistic stratum targeted by Clines, Menahem-Zvi Kaddari’s מילון העברית המקראית (2013). Dispensing with monolingual Hebrew material is as ludicrous as suggesting English philologists do away with sources penned by Anglophones.
“Leaving aside mediaeval dictionaries” at large is a methodological mistake and a dangerous bet in practice. Firstly, it entails the disregard of the foundation of Hebrew philology, when fruitful grammatical debates took place and splendid dictionaries were written. Moreover, the exclusive selection of dictionaries written in Latin and vernacular languages betrays a Europe-centred bias that is unjustifiable on strictly academic grounds. Secondly, starting with Renaissance Hebraists is like beginning to read a story in medias res with no prospective flashbacks. The first Christian Hebraists always bore Qimḥī in mind, either implicitly or explicitly. This ultimately means that ignoring at least the first three centuries of mediaeval lexicography ―that is, Qimḥī and his sources― hinders a proper understanding of the Christian Hebraists themselves.

In Clines’s CHL, primary sources are discriminated against if they meet one of the following requirements: 1) belonging to a certain historical period (Middle Ages); or 2) being written in certain languages (Hebrew and Judaeo-Arabic). In practice, however, these conditions are inextricably intertwined. Provided that mediaeval lexicography was authored by Jews and that post-mediaeval Jewish lexicography was composed mostly in Hebrew in the modern period (19^th century-present), Clines’s criteria expel the vast majority of Jewish dictionaries out of the lexicographic corpus ⁹ This is the general situation, but there are a few dictionaries that escape these criteria, for example: in the pre-modern period (16^th-19^th centuries) in English, David Levi’s Lingua Sancta in Three Parts (1785-1788); in the modern period (19^th century-present) in German, Julius Fürst’s Hebräisches und Chaldäisches Handwörterbuch über das Alte Testament (1857-1861). .

Clines did not explain the reason for discarding centuries of dictionary-making by Jews until 2021Clines, David. 2021. “Alleged Basic Meanings of the Hebrew Verb qdš “Be Holy”: An Exercise in Comparative Hebrew Lexicography.” Vetus Testamentum 71 (4/5): 481-502.
, when he included a footnote noting that he had confined himself to European bilingual lexica because he had “not been able for lack of resources to trace the history of the subject in mediaeval and modern monolingual Hebrew lexica” (Clines 2021, 482Clines, David. 2021. “Alleged Basic Meanings of the Hebrew Verb qdš “Be Holy”: An Exercise in Comparative Hebrew Lexicography.” Vetus Testamentum 71 (4/5): 481-502.
). It is true that some Israeli dictionaries are extraordinarily difficult to find outside of Israel, but unavailability is not a real issue for mediaeval dictionaries. Fortunately, the central mediaeval dictionaries were edited in the 19^th century for the first time ¹⁰ Šĕlomo ibn Parḥōn’s dictionary was edited in 1844; David Qimḥī’s, in 1847; Mĕnaḥem ben Sarūq’s, in 1854; Jonah ibn Ŷanāḥ’s, in 1875 (in Arabic) and in 1896 (Hebrew translation). Full bibliographical references are provided in §4. . This means that today they are digitalised and accessible at varied online repositories just as Reuchlin’s De rudimentis hebraice (1506)Reuchlin, Johannes. 1506. De rudimentis hebraicis. Pforzheim: Thomas Anschelm.
is.

Clines collected “over 600 Hebrew dictionaries in European languages from the 16^th century onwards” (Clines 2017, 227Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
) and concluded his article with a list of the lexica mentioned in the paper, which amounts to 50 (including some mediaeval dictionaries; see §4). This disparity in figures (50 out of 600) reminds us that not all dictionaries of biblical Hebrew are important when tracing the origins and developments of a word. However, the first step must be the creation of a database of dictionaries and other books belonging to cognate fields (mainly, exegesis) that is as comprehensive as possible. This requires us to rely thus on historical studies of Hebrew lexicography (such as Brisman 2000Brisman, Shimeon. 2000. A History and Guide to Judaic Dictionaries and Concordances. Jewish Research Literature 1/3. Hoboken: KTAV Publishing House.
). The second step is to break the corpus not only into periods (mediaeval, pre-modern, and modern; see §4) but also, and more importantly, into chains of dependence in order to provide it with coherence and cohesiveness. Only then can the researcher seek, or at least expect, some degree of intertextuality that will help them recount the journey of a word. That is, the corpus begins as a totum revolutum that becomes functional after careful examination and classification of its components.

3.2 Scope of Interests and Perspectives

⌅

Clines envisioned CHL as consisting

in its simplest form of identifying similarities and differences among lexica, such as their size and organization and treatment of cognates, and in a more critical form of evaluating the differences among lexica, making judgments about one lexicon over against another, or about commonalities among lexica that are open to criticism (Clines 2017, 228Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

This concise statement reveals Clines’s diverse interests and speaks of the expected fruitfulness of his method. Clines only addresses the issue of “similarities and differences” focusing on the analysis of some definitions that reached our current dictionaries and “makes judgements” about their in/appropriateness. His lamented death left other possibilities unexplored.

I deem desirable to point at some possible applications of this kind of philological research: 1) source LC; 2) ideological LC; and 3) formal LC. I shall outline them below in the hope that it will contribute to further research and the advancement of prospective models of analysis.

3.2.1. Source LC

⌅

Source LC aims to trace the history of lexicographic statements and attempts to locate their genesis. Clines’s research falls into this category.

Above, I have intentionally referred to “statements” instead of “meanings” because the information given in the definiens is not the only aspect that can be analysed from this perspective. The elements within the microstructure of biblical Hebrew dictionaries are varied and the application of source LC is pertinent in all cases:

The etymological section. Example: the analysis of the relation between some instances of biblical בָּעַל and בָּחַל and Arabic بعل as portrayed (or refuted) in dictionaries since the Middle Ages (Carbonell Ortiz 2022a, 258-269Carbonell Ortiz, Clara. 2022a. “On the Alleged Meaning of the Verb שׁגל: A Philological Analysis from a Gender Perspective.” In (Re)Gained in Translation I: Bibles, Theologies, and the Politics of Empowerment, edited by Sabine Dievenkorn and Shaul Levin, 207-262. Berlin: Frank & Timme.
).
The grammatical section and 3) the semantic section or, strictly speaking, the definiens. Example of the crossroads between the two: the examination of the non-sexual interpretation of זו̇נָה in some verses, which flourished in mediaeval lexicography and started to wither in the late pre-modern period. The proposed meanings were partly derived from the morphological patterns of different weak roots (זנ׳׳ה and זו׳׳ן) (Carbonell Ortiz 2024Carbonell Ortiz, Clara. 2024. “¿Prostituta o mesonera? Las metamorfosis de Rajab en la lexicografía de hebreo bíblico.” In Al margen de la norma. Manifestaciones y representaciones siniestras de lo femenino en el Próximo Oriente antiguo, edited by Claudia D’Amico, 169-192. Sevilla: Editorial Universidad de Sevilla.
).
Extra-lexicographic comments, like encyclopaedic, exegetical, or even moralising annotations. Although repudiated in modern lexicography, this sort of information is not uncommon in mediaeval and pre-modern dictionaries. Example: the digression on sexual postures that Ibn Parḥōn (1160/1) included in the entry ידע (Carbonell Ortiz 2022a, 216-222Carbonell Ortiz, Clara. 2022a. “On the Alleged Meaning of the Verb שׁגל: A Philological Analysis from a Gender Perspective.” In (Re)Gained in Translation I: Bibles, Theologies, and the Politics of Empowerment, edited by Sabine Dievenkorn and Shaul Levin, 207-262. Berlin: Frank & Timme.
). It must be noted that this case can be addressed from a source LC perspective as well as from an ideological LC perspective (see below). Source LC seeks to disclose the connection with previous materials (Ibn Ǧanāḥ’s previous treatment of ידע [ca. 1050], in this case) and its survival in future writings (in Běḵōr Šōr’s commentary [12^th c.]). Unlike ideological LC, it does not delve into the ideological dimensions of the matter (sexual morals and women’s subjection). The theoretical difference between source and ideological LC is substantial. Empirically, however, a detailed discussion of some passages often necessitates the intertwining of both approaches.
Textual emendations. Example: the enquiry into the lexicographic endorsement of the textual emendation אֵימו̇ת for אֳנִיּו̇ת in Psa 104:26 (Clines 2019Clines, David. 2019. “There Go the Monsters (Psalm 104:26): An Emendation and Its History.” Paper read in the Biblical Poetry Hebrew Section at the SBL Annual Meeting, San Diego. https://www.academia.edu/40949823/There_Go_the_Monsters_Psa_104_26_An_Emendation_and_its_History (Last access: 23/05/2024).
).

In addition to these applications, there are two key concepts that should be clarified. One is “lexicographic tradition” ¹¹ In my doctoral dissertation, this was named in Spanish as “discurso lexicográfico,” but the calque into English is inadequate. Furthermore, the expression resonates with a postmodern academic terminology that, as previously stated, I intend to avoid. , meaning any piece of information in the microstructure that proves to have been well-established. It is noteworthy that traditions can be interrupted, disrupted, and also forgotten (see זו̇נָה above), but their nature as “traditions” rests upon the fact that they were authoritative and popular at some point in history. For example, the interpretation חֵיל = “outer wall” is a lexicographic tradition not only in pre-modern lexicography, as Clines claimed, but also (and causally) in mediaeval dictionaries (see §5.2).

On the other side of the coin, the label “lexicographic innovation” stands for a new interpretative path that strays off a lexicographic tradition and might become, with the passage of time, traditional. Curiously enough, some innovations can be independent rediscoveries that occur when the author ignores previous literature. For example, in the eighth volume of the DCH the noun שְׁכ̇בֶת is defined as “pouring out […] emission of semen, perh. penis” (p. 348). In the bibliographical list that endorses the content of this entry, the meaning “penis” is attributed to Orlinsky (1944, 40)Orlinsky, Harry M. 1940. “The Hebrew Root ŠKB.” Journal of Biblical Literature 63 (1): 19-44.
. Orlinsky could not have known that he was agreeing with the Andalusi lexicographer Saadya ibn Danān, who already proposed this meaning in his Ḍarūrī fī al-luġa al-ʿibrāniyya more than five centuries ago, in 1468 ¹² This dictionary has been apparently preserved in a unicum and was edited by Jiménez Sánchez (1994), whilst Orlinsky’s paper was published fifty years before. Orlinsky made careful research on the interpretations of the root שׁכ''בby mediaeval lexicographers and grammarians and even consulted Skoss on the matter (1940, 38 n. 25). For a comprehensive analysis of this case, see Carbonell Ortiz (2022b, 477-478). .

The second key concept is “lexicographic construct,” that is, a statement concerning a word that is incompatible with the textual data found in the linguistic corpus ¹³ Etymology and homonymy are, of course, excluded, provided their inherently hypothetical nature. . There is no overlap or interconnectedness between these categories: a lexicographic construct is about textual verifiability, whereas a lexicographic tradition is about historical reception. Likewise, lexicographic constructs are not necessarily motivated by ideology. The semantic interpretation of the verb שָׁגַל as ‘to rape’ is a prominent example of a lexicographic construct with a long history in dictionaries of biblical Hebrew (Carbonell Ortiz, 2022aCarbonell Ortiz, Clara. 2022a. “On the Alleged Meaning of the Verb שׁגל: A Philological Analysis from a Gender Perspective.” In (Re)Gained in Translation I: Bibles, Theologies, and the Politics of Empowerment, edited by Sabine Dievenkorn and Shaul Levin, 207-262. Berlin: Frank & Timme.
).

The interests of source LC may be manifold and, to a great extent, they are determined by the nature and content of each entry as well as by the needs of the researcher. Two are the main applications of source LC:

To reconstruct the journey of a word and witness its changes and vicissitudes across the centuries. Source LC enables us to: a) date when a lexicographic tradition appeared; b) spot the authors that inherited it and handed it down; c) establish when and under what circumstances it was discontinued; d) assess if it was superseded; and, if applicable, e) examine the new interpretation that replaced it and its degree of acceptance in subsequent works, which leads us back to a).
To measure the influence of previous authors on the content of a given dictionary, even when the lexicographer anonymised or did not disclose his sources. If done systematically, it enhances our understanding of the inner history of Hebrew lexicography. This application can be seen in a recent article of Basal (2021a)Basal, Nasir. 2021a. “The Glossary Šarḥ al-ʾalfāẓ by Abū al-Farāj Hārūn: David ben Abraham Alfāsī’s Dictionary Jāmiʿ al-ʾalfāẓ as a Resource.” Jerusalem Studies in Arabic and Islam 51: 337-376.
where he proved that the Karaite Abū al-Farāǧ Hārūn, in his glossary Šarḥ al-ʾalfāẓ al-ṣaʿba fī al-Miqraʾ (1026), relied extensively on his Karaite predecessor David ben Abraham al-Fāsī’s dictionary (see §4).

3.2.2. Ideological LC

⌅

Ideological LC deals with the extra-linguistic content of dictionaries as it stems from the system of beliefs shared by the lexicographer and their community. A wide definition of “ideology” is of pivotal importance, as the term directs the modern reader’s attention to issues such as sexism (Feminist Theory), racism (Postcolonial Theory), and anything related to political agendas (Critical Theory). However, this range of themes is too narrow: it heavily relies on recent events and the methodologies involved frequently focus on hodiernal struggles. From a diachronic perspective, it is clear that religious polemics should be taken into consideration. They are common ―albeit not always overt― in mediaeval and pre-modern lexicography: be it interdenominational, such as Karaite versus Rabbanite or Protestant versus Catholic, or interfaith, such as Christianity versus Judaism, paradigmatically expounded by Hutchinsonian dictionaries in the second half of the 18^th c. (Craig 2014, 39-150Craig, Marie-Louise. 2014. “Hebrew-English Lexicons of the British Isles: From Parkhurst (1762) to Benjamin Davies (1872).” Doctoral dissertation. Carles Sturt University. https://researchoutput.csu.edu.au/en/publications/hebrew-english-lexicons-of-the-british-isles-from-john-parkhurst--3 (Last access: 23/05/2024).
). According to a broad understanding of what ideology means, I conceive ideological LC as the process of digging out “fragments of thought,” as we may call them. These are interspersed in the dictionary and echo the biases of the author. Their gathering gives us valuable insights into the psychological traits of the lexicographer and the dictionary’s Sitz im Leben.

It should be noted that ideological LC branches off into two parallel courses of analysis: macrostructure and microstructure. The former focuses on the examination of the vocabulary included and, consequently, prompts us to wonder why some other lexical items were put aside. The selective elimination of some words, which is characteristic of modern languages lexicography, is nearly absent in corpus lexicography. In these dictionaries, the cases of omission are due to obviousness, irrelevance ¹⁴ This remark seems contradictory to our modern concept of historical or corpus lexicography, but it requires consideration in dictionaries of previous centuries. For example, such was the critique of Elias Levita, expressed in his Meturgeman, to Nathan ben Yěḥiʾel’s Aramaic dictionary, known as ʿArūḵ. See Griño (1971, 354). , and also to unfortunate oversight. Ideological biases at the macrostructural level primarily affect dictionaries of spoken languages.

Regarding the microstructure, exercises in ideological LC can scrutinise the following:

The ideological biases found in those parts of the entry intended to explain the word, like the definiens and extra-linguistic annotations.
How the selection of examples illustrating the usage of a word (and, conversely, the discarding of other textually suitable examples) may reflect an ideological position. In the case of biblical lexicography, it is interesting to note that the set of verses is sometimes a portent of a certain theology that has been privileged by the author or, also, of a partial interpretation that is privileged over the whole picture.

Let us illustrate the second possibility by bringing a case of semantic and vocalic reinterpretation contra textum that is backed by a clever, deceitful selection of biblical verses. The treatment of בעלה in Bate’s dictionary (1767)Bate, Julius. 1767. Critica Hebraea or a Hebrew-English Dictionary without points. London: M. Folingsby.
betrays an intentional misrepresentation of the data with a clear theological purpose in mind. This lexicographer defended the vocalisation בַּעֲלַת בַּעַל as a refutation of the passive reading of Masoretic בְּעֻלַת בַּעַל. However, the only way to convince his readers thereof was to conceal the verses that contain the word with scriptio plena (בְּעוּלָה, in Isa 54:1; 62:4). These verses are accordingly absent in the entry (Bate 1767, 77Bate, Julius. 1767. Critica Hebraea or a Hebrew-English Dictionary without points. London: M. Folingsby.
) ¹⁵ Ironically, Bate knew that this revocalisation created a semantic incompatibility between the roles of the wife and the husband (“female owner of a male owner,” instead of “owned by a male owner”) and did his best in order to sort it out. . Bate was fond of Hutchinsonian theology, which aimed to vilify the Rabbis as corrupters of the Scriptures. The case of בעלה was taken as one of the many instances in his dictionary that proved the unreliability of the Jewish tradition and, thus, of the textus masoreticus.

It is paramount to underscore that exercises in ideological LC deal with the data as recorded in the dictionary and not as recorded in the texts. Differently put, its role is to thoroughly examine mismatches between what the dictionary claims about a word and how that word is actually used in the texts, when this divergence is rooted in ideology. Elsewhere, I have catalogued the array of lexicographic biases that tend to occur in the semantic domain “sexual intercourse” according to procedures and themes (Carbonell Ortiz 2022b, 135-146Carbonell Ortiz, Clara. 2022b. “Metalexicografía diacrónica aplicada a diccionarios de hebreo bíblico. Aproximación al campo semántico ‘sexo’.” Tesis doctoral. Universidad Complutense de Madrid. https://docta.ucm.es/entities/publication/32eca4cd-d53f-4c72-be2d-080ab7933122 (Last access: 23/05/2024).
). Here, I shall only refer to the former group and say that:

Some biases “centre” one context in which the word occurs (out of others of equal importance) and elevate it into a core semantic trait.
Contrarily, other biases “decentre” a core semantic trait and provide an inclusive definition that aligns with a post/modern Weltanschauung, despite its anachronistic value regarding the textual data.

Taking the issue of sexism as an example, exercises in ideological LC may focus on how the definition of a word is elaborated in a sexist fashion when the meaning of this word is not sexist. Reversely, these exercises may also focus on the neutral definition of a word that misses its sexist nature and explore the reasons behind it. Nevertheless, the definition of a sexist meaning in sexist terms is as irrelevant to ideological CL as the non-sexist definition of a non-sexist word. In both cases, the lexicographic treatment is correct.

The distinction between the ideologisation of the language and the ideologisation of the dictionary is key (Calero Fernández 1999, 149Calero Fernández, María Ángeles. 1999. “Diccionario, pensamiento colectivo e ideología (o los peligros del definir).” In Así son los diccionarios, edited by María Nieves Vila Rubio, María Ángeles Calero Fernández, M. Rosa Mateu, Monserrat Casanovas Catalá, and José Luis Orduña López, 149-201. Lleida: Edicions de la Universitat de Lleida.
). The former stems from multiple societal factors, whereas the latter emanates from the lexicographer’s cosmovision and becomes a layer on top of the linguistic reality. This difference is crucial to our subject because the lines between the two types of ideologisation are not only blurred in daily-life debates about dictionaries but also in some scholarly literature ¹⁶ A blatant example of this confusion is represented by the DCH. For example, Clines acknowledged in the prologue to the dictionary that the meanings of some words had been adjusted to be gender inclusive regardless of their historical accuracy. He also attempted to standardise the employment of gender suffixes in English by lecturing users on the inappropriateness of feminine endings. See volume 1 of the DCH (1993, 356) and volume 3 (1996, 12), respectively. . If lexicography is committed to recording the usage of language as it is (a fortiori, corpus lexicography), lexicographers cannot be blamed for the political incorrectness of our linguistic habits. However, they should be blamed for the inaccurate definition of awkward or sensitive meanings.

3.2.3. Formal LC

⌅

Exercises in formal LC deal with strictly technical data of dictionaries. There are multiple possibilities, but all of them revolve around the macrostructure/microstructure, the orthotypography or the like, and their evolution. For example, an intriguing challenge for formal LC would be the historical development of the macrostructure and the criteria for fixing the headwords ¹⁷ Arabic philologists already have some studies of this sort at their disposal. For a comprehensive study of classical Arabic lexicography, see Baalbaki (2014). A summary of the possibilities of macrostructure in mediaeval Arabic dictionaries can be found in Arias (1996, 45-47). . In modern Hebrew studies, Schwarzwald (2004)Schwarzwald, Ora. 2004. “Modern Hebrew Dictionaries.” Kernerman Dictionary News 12: 18-22.
has made valuable progress in this direction. Regarding dictionaries of biblical Hebrew, I have sketched a draft typology of the macrostructure elsewhere (Carbonell Ortiz 2022b, 5-8Carbonell Ortiz, Clara. 2022b. “Metalexicografía diacrónica aplicada a diccionarios de hebreo bíblico. Aproximación al campo semántico ‘sexo’.” Tesis doctoral. Universidad Complutense de Madrid. https://docta.ucm.es/entities/publication/32eca4cd-d53f-4c72-be2d-080ab7933122 (Last access: 23/05/2024).
) and intend to elaborate on it before long. A case study of formal LC is included below (see §5.1).

3.3. Aims and Consequences

⌅

In his 2017 article, Clines explained the procedure of this method of CHL but was a somewhat secretive about his objectives. He mentioned a pragmatic motivation ―namely, that CHL is useful “to know what Hebrew dictionaries have been like when one is considering what is valuable in a Hebrew dictionary” (Clines 2017, 234Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). Contrastingly, Clines was more open about the consequences of CHL. The first one relates to the perceived reliability of dictionaries:

[S]tudying Hebrew dictionaries in relation to one another reveals inescapably how antiquated much of the content of our most recent dictionaries is, how derivative they often are, and how often the mistakes of the past are perpetuated uncritically (Clines 2017, 234Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

All this is true, but only the third objection is harmful to the advancement of Hebrew lexicography. The fact that dictionaries “are often derivative” is natural and expectable, since they always rely on previous lexicography, plus other genres of linguistic literature (grammars, exegetical commentaries, academic papers…). The extension of vocabulary together with the wearing meticulousness of the lexicographic task makes plagiarism unavoidable. It is the degree of plagiarism what is troublesome. Finally, the first concern brought up by Clines is worrisome, I believe, as it implies that antiquated material deserves our distrust. Notwithstanding this, a definition is valuable not on grounds of its novelty but of its linguistic precision.

The second consequence of CHL stems directly from this:

A consequence of bringing back older lexica into the conversation, as well as of registering the numerous proposals that have been made in the last half-century […] is bound to be a destabilizing of the lexicon, with more options, more need of the term “perhaps,” more acknowledgement that we don’t really know the meaning of many of our words (Clines 2017, 239Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

The general overtone of this statement is hardly objectionable: having humility to acknowledge ignorance is certainly a virtuous rara avis in lexicography. However, when read in the context of Clines’s scholarly production at large, these words mean that dictionaries exert power and are cultural artifacts that can be “destabilized,” that is, deconstructed. Clines’s CHL was postmodern, much like his DCH, as recognised in the prologue of the first volume (Clines 1993, 26Clines, David. 1993-2013. Dictionary of Classical Hebrew. 8 vols. Sheffield: Sheffield Academic Press.
). Andersen (1995)Andersen, Francis. 1995. “Review Article and responses. The Dictionary of Classical Hebrew. Vol. 1 א.” Australian Biblical Review 43: 50-75.
wrote a lengthy and detailed critique of the problems of this approach in lexicography, which I entirely endorse.

3.4.The Chronological Directionality of the Analysis

⌅

The surveys of lexicographic content can be conducted in two ways, depending on what chronological directionality is preferred: 1) backwards: commencing with a recent dictionary and retracing the steps to the first occurrence of the examined matter; or 2) forward: beginning with the first dictionary that displays it and tracing its development in subsequent dictionaries.

Lee and Clines opted for the first possibility. Both directions are functional, but it seems to me that the goal of the analysis conditions the directionality. If the objective is to destabilise the lexicon, the optimal direction is counter-chronological, for it is best to highlight how blindly our current dictionaries rely on older sources. However, proceeding chronologically is more suitable if we seek to understand lexicographic phenomena in their historical dimension, without making evaluative statements.

4. Misconceptions about Mediaeval Lexicography in Clines’s Method

⌅

In his 2017 article, Clines devoted one section to outline a concise history of Hebrew dictionaries. This endeavour enables the researcher to “locate a particular lexicon in its appropriate historical context” (Clines 2017, 229Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). The location of dictionaries in their historical (and sociocultural) setting is not the end of the research but rather the means by which the researcher is led to a better understanding of the dictionary form and content. Being acquainted with the immediate bibliographical context of a work trains the researcher’s skill to recognise points of departure and tell innovative materials apart from traditional opinions.

In Clines’s short history of Hebrew lexicography, some misconceptions regarding the mediaeval period went unnoticed ¹⁸ For a periodisation, see Martínez Delgado (2008; 2011), Craig (2011), and Holtz (2011). . The dates are addressed here, whereas the problem of content will be discussed in section §5.1.

Clines addressed the mediaeval period (10^th-15^th c.) in five lines, the pre-modern period (16^th-18^th c.) in five lines, and the modern period (19^th c.-present) in seventeen lines (Clines 2017, 229-230Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). Although quality precedes quantity, the number of lines dedicated to each phase is a portent of the importance they have in Clines’s method. The length of the modern period is a corollary of (what I believe is) a generalised tendency amongst non-Israeli/non-Jewish and non-medievalist Hebrew scholars to overemphasise the relevance of Christian, post-Renaissance Hebraists ―and to the detriment of their predecessors― in the history of Hebrew linguistic literature. This scholarly short-sightedness might stem either from a Europe-centred bias or from a poor/lacking acquaintance with the mediaeval Jewish sages. Both causes are in reality a vicious circle where the latter contributes to the former and the former reinforces the latter.

Clines began the mediaeval section by appointing the Kitāb ǧāmiʿ al-ʾalfāẓ, written by the Karaite David ben Abraham al-Fāsī, as “the first true Hebrew lexicon” (Clines 2017, 229Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
) ¹⁹ It was edited by Skoss (1936-1945). . This elusive reference to which work merits the title of the firstborn is seemingly a confrontation with Saadya Gaon’s ʾEgrōn, although the latter is not mentioned²⁰ It was edited by Allony (1969). . If this excerpt is read against the grain, it implies that Mĕnaḥem ben Sarūq’s Maḥberet (ca. 960)²¹ The editio princeps belongs to Filipowski (1854). A critical edition was later carried out by Sáenz-Badillos (1986). A new commented critical edition has been recently prepared by Maman and Mirsky (2024). , the first Hebrew-Hebrew dictionary to ever be composed, is either 1) an “untrue lexicon,” in spite of its strictly lexicographic structure and content, far from the poetry-oriented ʾEgrōn (Sáenz-Badillos 1986, 14Sáenz-Badillos, Ángel. 1986. Menaḥem ben Saruq. Maḥberet. Granada: Universidad de Granada.
), or 2) that it is later than Al-Fāsī’s. There is some degree of disagreement amongst experts in mediaeval Hebrew lexicography concerning this. The dating of both works in the second half of the 10^th c. is widely accepted (Polliack 2010Polliack, Meira. 2010. “David Ben Abraham al-Fāsī.” In Encyclopedia of Jews in the Islamic World, vol. 1, edited by Norman A. Stillman, 34-36. Leiden-Boston: Brill.
; Martínez Delgado, 2010bMartínez Delgado, José. 2010b. “Ibn Sarūq, Menahem ben Jacob.” In Encyclopedia of Jews in the Islamic World, vol. 1, edited by Norman A. Stillman, 541-544. Leiden / Boston: Brill.
), but some speak of Al-Fāsī’s as the first comprehensive dictionary (Dotan 1988, 115Dotan, Aaron. 1988. “מראשית המילונאות העברית ― קטע ממילון קדום.” Te’udah 6: 115-126.
; Eldar 1992, 369Eldar, Ilan. 1992. “ראשית המילונאות העברית במזרח.” Meḥkarim ba-lašon 5/6: 355-382.
), while others have just suggested the existence of a common source used by Al-Fāsī and Ben Sarūq which is currently unknown to us (Skoss 1936, lxiiSkoss, Solomon L. 1936-1945. The Hebrew-Arabic Dictionary of the Bible Known as Kitāb Jāmi’ al-Alfaẓ (Agrōn) of David ben Abraham al-Fāsī the Karaite (Tenth Cent.). 2 vols. Yale Oriental Series 20. New Haven: Yale University Press.
; Sáenz-Badillos 1986, 14 and 21Sáenz-Badillos, Ángel. 1986. Menaḥem ben Saruq. Maḥberet. Granada: Universidad de Granada.
). Most recently, Basal (2021b)Basal, Nasir. 2021b. “דוד בן אברהם אלפאסי הקראי וזמן פעילותו.” Lĕšonénu 83 (4): 380-400.
has located Al-Fāsī’s dictionary in the last quarter of the 10^th c. In summary, the chronological relation between the Kitāb ǧāmiʿ al-ʾalfāẓ and the Maḥberet is uncertain.

At the end of his 2017 article, Clines included a list of Hebrew lexica mentioned in the paper (pp. 240-244). The Maḥberet was not referred to in the article and is accordingly absent in this section. This list comprises a set of bibliographical entries that begin with the date of composition of each dictionary. Two core mediaeval dictionaries are misdated or their dating is reconsidered without due justification. Firstly, Ibn Ǧanāḥ’s Kitāb al-ʾuṣūl is dated in “c. 990-c. 1050” (Clines 2017, 240Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). This span actually corresponds to the estimated dates of birth and death of the lexicographer and not to the composition of his work (Martínez Delgado 2010aMartínez Delgado, José. 2010a. “Ibn Janāḥ, Jonah (Abū ’l-Walīd Marwān).” In Encyclopedia of Jews in the Islamic World, vol. 1, edited by Norman A. Stillman, 500-503. Leiden / Boston: Brill.
). The Kitāb at-tanqīḥ, of which the Kitāb al-ʾuṣūl is the second part, was written at some point in the middle of the 11^th century, shortly before the author’s death in Saragossa, where he arrived after fleeing from turbulent Cordoba in the beginning of the century. The year 1039 serves as the terminus post quem for the writing of the Kitāb at-tanqīḥ, since Ibn Ǧanāḥ mentioned R. Haī Gaon (939-1038) as having already passed away (Sáenz-Badillos and Targarona Borrás 1988, 112Sáenz-Badillos, Ángel, and Judith Targarona Borrás. 1988. Gramáticos hebreos de Al-Ándalus (Siglos X.XII), Filología y Biblia. Córdoba: Ediciones El Almendro.
). The second wrong date is that of Qimḥī’s Sefer ha-šorašīm, which is allotted to the “12^th cent.” (Clines 2017, 241Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
). Nevertheless, it is conventionally thought to have been composed in the early 13^th century, probably around 1210 (Kogel 2016a, 232Kogel, Judith. 2016a. “Qimḥī’s Sefer ha-Shorashim: A Didactic Tool.” Sefarad 76 (2): 231-250: 232. https://doi.org/10.3989/sefarad.016.008.
).

5. Two Case Studies

⌅

5.1. Exercise in Formal LC: the Place of Aramaic in Mediaeval Dictionaries

⌅

We are used to lexica of Biblical Hebrew including words in Biblical Aramaic. The earliest, Jewish, lexica contained no Aramaic, and the first lexicon that did was that of Alfonsus Zamorensis (Alfonso de Zamora) in the Complutensian Polyglot of 1515, in which the Aramaic words were interspersed among the Hebrew (Clines 2017, 231Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

A quick look at the history of mediaeval Hebrew lexicography suffices to refute this assertion. Pace Clines, the inclusion of Aramaic goes back five centuries before the printing of the dictionary of the Complutensian Polyglot (sixth volume). In fact, interspersing Aramaic amongst Hebrew words was the norm in the first period of Jewish lexicography in the Middle Ages. In the recent publication of the retrieved fragments of Yěhudah ibn Qurayš’s dictionary, this hallmark of the earliest works has been underscored: הוא (אבן קורייש) משלב את ערכי הארמית המקראית בין ערכי העברית, משל היו לשון אחת, וכך נהגו אחריו בן סרוק ב'מחברת' ודוד בן אברהם אלפאסי ב'ג̇אמע אלאלפאט̇ (Maman and Ben-Porat 2023, 20Maman, Aharon, and Efraim Ben-Porat. 2023. “שרידים ממילון אל''ף בי''ת לרבי יהודה בן קורייש.” Ginzei Kedem 19: 15-131.
). Not only did Ibn Qurayš mix Hebrew and Aramaic items in his dictionary, but also did his two successors, Ben Sarūq in his Maḥberet and Al-Fāsī in his Ǧāmiʿ al-ʾalfāẓ.

In the Maḥberet, the entries ארק and ארע (for Aramaic ארקא and ארעא, respectively) illustrate this matter. They appear in the book (maḥberet) of the letter ʾaleph and do so next to Hebrew lemmas. These entries show that Ben Sarūq sometimes expected the user to know that a word was Aramaic (a) whereas, other times, he warned about it in the definiens (b):

(a)	ארק די שמיא וארקא. ארק וארץ ותבל והרום ואדמה כנויי ארץ. (Sáenz-Badillos 1986, 59Sáenz-Badillos, Ángel. 1986. Menaḥem ben Saruq. Maḥberet*. Granada: Universidad de Granada. )	ארק: [It appears in the verse] דִּי־שְׁמַיָּא וְאַרְקָא (Jer 10:11). [The words] ארק, ארץ, תבל, הרום, and אדמה are names for ‘earth’ [ארץ].
(b)	ארע ארעית גובא. ארץ בלשון ארמית ארעא. (Sáenz-Badillos 1986, 63Sáenz-Badillos, Ángel. 1986. Menaḥem ben Saruq. Maḥberet*. Granada: Universidad de Granada. )	ארע: [It appears in the verse] לְאַרְעִית גֻּבָּא (Dan 6:25). [The word for] ‘earth’ [ארץ] is ארעא in the Aramaic language.

Here, the consonantal combinations of the Aramaic (ארק, ארע) are clearly distinct from the Hebrew (ארץ). However, a single root may be shared by Hebrew and Aramaic. The entry שׁרק bis is an example of this (Sáenz-Badillos 1986, 390*Sáenz-Badillos, Ángel. 1986. Menaḥem ben Saruq. Maḥberet. Granada: Universidad de Granada.
). There, an Aramaic verse (Dan 3:5, containing מַשְׁרוֹקִיתָא) is interspersed along with Hebrew verses (Zech 10:8, אֶשְׁרְקָה; Isa 5:26, וְשָׁרַק; Judg 5:16, שְׁרִקוֹת).

These two cases not only prove the coexistence of Aramaic and Hebrew prior to the dictionary of the Complutensian Polyglot, but they also evince the lack of separation between the words of these languages at two levels: macroestructural (ארע and ארק: Aramaic lemmas next to Hebrew lemmas) and microstructural (שרק: Aramaic and Hebrew words within the same entry).

The disappearance of Aramaic in mediaeval lexicography is an innovation introduced by Ibn Ǧanāḥ’s Kitāb al-ʾusūl (ca. 1050) ²² The Arabic text was edited by Neubauer (1875). The Hebrew translation of Yehudah ibn Tibbon (1171) belongs to Bacher (1896). . Lexicographers that followed suit are Šĕlomoh ibn Parḥōn in his Maḥberet he-ʿArūḵ (1160/1)²³ It was fragmentarily edited by Di Rossi (1805). The complete edition belongs to Stern (1844). and an anonymous 13^th-century writer who composed a Hebrew dictionary in Provence²⁴ The ms. Vat. Hebr. 314 was edited by Sáenz-Badillos (1987). However, Kogel (2016b) has found more manuscripts and is currently preparing a critical edition. . For different reasons, I have not been able to check this issue in the following two mediaeval dictionaries: the Kitāb al-istiġnaʾ, written by Šĕmuʾel ibn Nagrella (993-1055/6), which is too fragmentarily preserved²⁵ It was edited by Kokovtsov (1916, 204-224). , and the lexicographic section of the Kitāb al-kāmil, written by Yaʿaqoḇ ben ʾElʿazar (12^th-13^th c.), which is still unpublished²⁶ The grammatical section was edited by Allony (1977). Some small fragments of the dictionary have been published by Martínez Delgado (2013; 2019). Maman has prepared the edition of the dictionary but it is currently in press (personal communication of José Martínez Delgado). .

Clines also introduced a list of pre-modern dictionaries whose title indicates that they contain Hebrew and Aramaic words, and said:

I have, however, not yet been able to identify when the practice began of collecting the Aramaic words into a section of their own at the end of the lexicon, which is what we are used to today. [In the footnote:] I see it in Gesenius’s 12^th edition (1895), but not in the 6^th (=5^th) edn (1863). (Clines 2017, 231Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
)

The genesis of this practice is found, again, in the Middle Ages. Some lexicographers created an independent section for the Aramaic at the end of the dictionary and did so according to the macrostructural arrangement characteristic of glossaries (canonical order and textual lemmatisation) ²⁷ See Olszowy-Schlanger (2011, 165). . This innovation in the lexicographic treatment of Aramaic is first found in Qimḥī’s Sefer ha-ʿinyanīm (ca. 1210, eventually known as Sefer ha-šorašīm)²⁸ It was edited by Biesenthal and Lebrecht (1847). A critical, digital edition is currently in progress under the direction of Judith Kogel. The “carnet de recherche” is available at www.shorashim.hypotheses.org [last access: 25/03/2024]. . After concluding the dictionary, Qimḥī included an Aramaic section preceded by this explanation:

אמר דוד בן יוסף בן קמחי הספרדי ראיתי לכתוב עם הספר הזה אשר חברתי באור המלים אשר באו בלשון ארמי בספרי הקדש למען לא יחסר הקורא בספר הזה דבר, ולא תהיה מלה בארבעה ועשרים ספרים שלא ימצא באורה בספר הזה, עם מה שאני מוסיף בו עתה. לכן כל מעתיק ספר זה יכתוב זה בתכלית הספר כמו שכתבתי אני הנה.

(Biesenthal and Lebrecht 1847, 416Biesenthal, Johann H. R., and Fürchtegott Lebrecht. 1847. Rabbi Davidis Kimchi Radicum liber; sive, Hebraeum Bibliorum lexicon. Berlin: G. Bethge.
)

‘David ben Yosef ben Qimḥī, the Sephardic, said: In this book that I composed, I saw [convenient] to write a clarification of the words that are in the Aramaic language in the Sacred Scriptures, so that the reader of this book will not miss anything and the clarification of all words of the Twenty-Four Books [i.e., the Hebrew Bible] will be found in this book, with what I am adding now. Therefore, may the copyists of this book write it at the end of the book as I have written it here ²⁹ Own translation. .’

This Aramaic glossary contains all the elements of an appendix: it is placed at the end (בתכלית), considered to be an addition to the book (אני מוסיף בו), and intended to offer the user a most complete tool (למען לא יחסר הקורא). Maman already noticed and compared the nature of the Aramaic section in this dictionary to that of the two previous options: its amalgamation with Hebrew lemmas and its absence (2004, 20Maman, Aharon. 2004. Comparative Semitic Philology in the Middle Ages. Studies in Semitic Languages and Linguistics 40. Leiden / Boston: Brill.
).

To judge from the wording of the excerpt, the inclusion and independent treatment of Aramaic at the end of the dictionary is a decision pioneered by Qimḥī. It will be successful in forthcoming dictionaries in the Middle Ages. In England, Moše ben Yiṣḥaq ben ha-Nĕsiyya wrote a grammatical and lexicographic work entitled Sefer ha-šoham (ca. 1260) where the glossary of biblical Aramaic is also placed at the end (still unpublished) ³⁰ The first two sections of the dictionary were edited by Klar (1940). Olszowy-Schlanger (2012) announced a new edition. . Also, at some point between the 13^th and 14^th centuries and likewise influenced by Qimḥī, the Karaite Šĕlomoh ben Mobaraḵ ben Saʿīr penned his Kitāb at-taysīr, whose Aramaic appendix title reads: והד̇ה ג̇מלה אלאלפאט̇ אלסיריאני אלמוג̇ודה פי אלמקרא ותפסירהא (Martínez Delgado 2010c, 498Martínez Delgado, José. 2010c. Šĕlomo ben Mobaraḵ ben Saˁīr. El libro de la facilitación. Kitāb al-taysīr. 2 vols. Granada: Universidad de Granada.
) (“this is the compendium of Aramaic words that appear in the Scriptures and their explanation”). In the year 1468, in the Nasrid Kingdom of Granada, Saadya ibn Danān composed his Ḍarūrī fī’l-luġah al-ˁibrāniyya, in which the lexicographic section contained also an appendix with the Aramaic (unpublished due to illegibility; Jiménez Sánchez 1996, 12Jiménez Sánchez, Milagros. 1996. Sěˁadyah ibn Danān. Sefer ha-šorašim. Granada: Universidad de Granada.
)³¹ The dictionary was translated into Spanish by Jiménez Sánchez (2004). . All these authors were greatly influenced by Qimḥī.

This historical survey allows us to classify the varied approaches regarding the treatment of Aramaic in the mediaeval dictionaries of Hebrew. Its presence (either separate or intermingled) or absence is inextricably connected to the linguistic theory proposed by the lexicographer. As quoted above, Maman and Ben-Porat referred to the treatment of Aramaic by Ibn Qurayš, Ben Sarūq, and Al-Fāsī as if it were a single language together with Hebrew (משל היו לשון אחת). A clear distinction between Hebrew and Aramaic was of course acknowledged by these lexicographers, but Hebrew and Aramaic were sometimes considered a unit (Maman 2004, 63Maman, Aharon. 2004. Comparative Semitic Philology in the Middle Ages. Studies in Semitic Languages and Linguistics 40. Leiden / Boston: Brill.
). It is key to see these works as dictionaries of the languages of the Hebrew Bible (which happen to be Hebrew and, coincidentally and to a vestigial degree, Aramaic), rather than dictionaries of biblical Hebrew (which happen to include Aramaic items).

A final word of caution must be expressed regarding this matter. Let us briefly imagine that the subject of enquiry was not when this practice first appeared, but rather when it became customary. The latter was already discussed by Koehler and Baumgartner (1953, xxxiv)Koehler, Ludwig and Walter Baumgartner. 1953. Lexicon in Veteris Testamenti Libros. Leiden: Brill.
in their celebrated KBL. In the preface to this dictionary, they pointed to Delitzsch’s Prolegomena (published in 1886) as the theoretical seed and to Siegfried-Stade’s Hebräisches Wörterbuch zum Alten Testamente (in 1893) as the first materialisation. The distinction between the genesis of a practice and its standardisation in modern times is key. In the case of the Aramaic appendix, the first question leads us to Qimḥī’s dictionary, whereas the latter brings us to late 19^th-century scholars.

This exercise in formal LC also makes noticeable that there are two linguistic factions that have co-existed since the Middle Ages onwards. The first one is a Bible-centred Hebrew lexicography, which is embodied in dictionaries that include Aramaic, regardless of its place. The second faction is a language-centred Hebrew lexicography. In the 1980s, this approach was revitalised not as a result of the major importance of Hebrew vis-à-vis Aramaic in Biblical Studies, as may have happened in previous periods. It also did not aim at correctness, as was sometimes the case in the Middle Ages, nor was it driven by nationalistic fervour, as in the early 20^th century, with Ben Yehuda as its main proponent. Instead, it was due to the debiblicisation of the linguistic corpus. This reduction of the corpus was accompanied by a simultaneous broadening ―namely, the introduction of Qumran, Ben Sira, and epigraphic material, as I have discussed elsewhere (Carbonell Ortiz 2022b, 21-22Carbonell Ortiz, Clara. 2022b. “Metalexicografía diacrónica aplicada a diccionarios de hebreo bíblico. Aproximación al campo semántico ‘sexo’.” Tesis doctoral. Universidad Complutense de Madrid. https://docta.ucm.es/entities/publication/32eca4cd-d53f-4c72-be2d-080ab7933122 (Last access: 23/05/2024).
).

5.2. Exercise in Source LC: the Deep Roots of the Meaning of חֵיל “Outer Wall”

⌅

There are two words in Classical Hebrew for a city wall, חוֺמָה and חֵיל (sometimes spelled חֵל), and lexicographers can never resist trying to distinguish the meanings of two apparent synonyms. After at least since Reuchlin (1506)Reuchlin, Johannes. 1506. De rudimentis hebraicis. Pforzheim: Thomas Anschelm.
and Pagnini (1529)Pagnini, Sancte. 1529. אוצר לשון הקדש, Thesaurus Linguae Sacrae sive Lexicon Hebraicum. Lyon: Sebastian Gryphius.
, two of the earliest Latin dictionaries of Hebrew, our dictionaries say חֵיל means an outer wall, though some think it the inner wall or else the space between outer and inner walls (Clines 2017, 234Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

After referencing Gesenius’s Thesaurus (first issue published in 1829Gesenius Wilhelm. 1829. Thesaurus philologicus criticus linguae hebraeae et chaldaeae Veteris Testamenti. Vol. 1. Leipzig: Vogel.
), the second edition of his Handwörterbuch (in 1815), and the BDB (between 1892 and 1906), Clines concluded that the meaning “outer wall” is a long-lived fabrication that is rooted in the architecture of mediaeval and early modern fortifications.

In a footnote, Clines retrieved non-biblical data from the DCH and completed the linguistic scenario with a case in Qumran, according to which the word must have meant

ditch or open space or dry moat around the temple, 100 cubits wide, a sense that Kimchi already noted in reference to b. Sanh. 88b, and that the BDB also refers to, even though it is hardly relevant to Biblical Hebrew (Clines 2017, 234 n. 23Clines, David. 2017. “Towards a Science of Comparative Classical Hebrew Lexicography.” In From Ancient Manuscripts to Modern Dictionaries, edited by Tarsee Li, and Keith Dyer, 227-245. Piscataway: Georgias Press.
).

I am unsure of the source from which Clines derived Qimḥī’s interpretation of חֵיל in T. B. Sanh. 88b (ובשבתות ובימים טובים יושבים בחיל), but this passage is not referenced in his dictionary. The opinion of the Provençal grammarian is recounted in Gesenius’s Thesaurus (1829, 454-455)Gesenius Wilhelm. 1829. Thesaurus philologicus criticus linguae hebraeae et chaldaeae Veteris Testamenti. Vol. 1. Leipzig: Vogel.
, and that Talmudic excerpt is also unmentioned there. In his commentary on 2Sam 20:15, Qimḥī gives two explanations, besides echoing that of Targum Jonathan: 1) חֵיל means ‘moat’, according to the Targumic translation of בגי as בחלתא; 2) חֵיל is another ‘wall’, according to Pesaḥ 86a (שורא ובר שורא) ³² All the quoted mediaeval commentaries are available online at www.alhatorah.org (last access: 23/05/2024). . Qimḥī elaborates on these interpretations in his dictionary (see below).

Let us go back to Clines to examine Reuchlin and Pagnini afresh. Reuchlin defined חֵיל as Paruus murus, uel antemurale (1506, 170Reuchlin, Johannes. 1506. De rudimentis hebraicis. Pforzheim: Thomas Anschelm.
) (“small wall, or outer wall”) and only provided Lam 2:8 as textual support. Reuchlin’s sources for this definition are the Vulgate (חֵל וְחוֹמָה in Lam 2,8: antemurale et murus) and Qimḥī, from which he took the size, “small” (see below). Pagnini also used the wording of the Vulgate but, unlike Reuchlin, he disclosed his sources:

Thre. 2.v.8. Et luxit (uel desolatus est) חֵל וְחוֹמָה .i. antemurale, & murus, Vñ R.D. in li.ra. exponit חֵל .i. murus paruus exterior, q est ante murũ magnũ […] & põt inquit R. Da. exponi חֵל murus interior et חוֹמָה. murus exterior […]: & rectũ est inqt ut exponãt, quoniam fossa quae erat in circuitu muri vocẽt חֵל Nã Targhũ pro ualle & fossa habet חֵלְתָא […] ()

Lam 2:8: ‘And he mourned (or desolated) חֵל וְחוֹמָה,’ that is, “the outer wall and the wall,” whence Qimḥī, in his dictionary, explained חֵל as “small, outer wall, which is before the big wall” […] and afterwards Qimḥī explained that חֵל [is] “inner wall” and חוֹמָה “outer wall” […]. And it is correct to interpret that the moat that is around the wall is called חֵל since, for instance, the Targum has חֵלְתָא for ‘valley’ and ‘moat’ […].

The abbreviated phrase “R.D. in li.ra.,” constantly recurring in Pagnini’s Thesaurus, stands for “Rabbī David [Qimḥī] in [his] Liber Radicum” (the Latin verbatim translation of Sefer ha-šorašīm). In Clines’s exercise, the unrecognised Qimḥīan genesis of this interpretation in early Christian lexicography is incomprehensible given its explicit acknowledgement by Pagnini.

Pagnini followed Qimḥī very closely, as is evident from the entry חיל in his Sefer ha-šorašīm. For the sake of clarity and given its length, let us break Qimḥī’s lexicographic discussion into three sections:

אבל ויאבל חֵל וחומה (איכה ב, ח.), וישׁפכו סוללה אל העיר ותעמוד בַּחֵל (שמואל ב׳ כ, טו.), אשר חֵיל ים (נחום ג, ח.), שיתו לבכם לְֽחֵילָה (תהילים מח, יד.), יהי שלום בְּחֵילֵךְ (שם קכב, ז.), כל אילו פירשו בהם החומה. ורבותינו ז׳׳ל פירשו חיל החומה הקטנה החיצונית שלפני החומה הגדולה באמרם (פסחים פו, ע׳׳א.) מאי חיל וחומה? אמר רבי חנינא שורא ובר שורא. ()

But [i.e., different from the meanings recorded above]: וַיַּאֲבֶל־חֵל וְחוֹמָה (Lam 2:8), וַיִּשְׁפְּכוּ סֹלְלָה אֶל־הָעִיר וַתַּעֲמֹד בַּחֵל (2Sam 20:15), אֲשֶׁר־חֵיל יָם (Nah 3:8), שִׁיתוּ לִבְּכֶם לְחֵילָה (Psa 48:14), יְהִי־שָׁלוֹם בְּחֵילֵךְ (Psa 122:7). All these [instances of חֵיל] have been interpreted as “wall” [חוֺמָה]. Our sages, blessed be their memory, interpreted that חֵיל was the “outer, small wall” [חוֺמָה] before the big wall [חוֺמָה], when they said: ‘What is a חֵיל and [what is a] חוֺמָה? R. Ḥanina said: [It means] a wall and a low wall [respectively]’ [T.B. Pes. 86a].

Qimḥī will collect an array of interpretations of חֵיל and expose the exegetical implications taking 2Sam 20:15 as an example. According to this interpretation, חֵיל means “outer wall” in contrast to חוֺמָה, “inner wall,” and a Talmudic passage in used as support.

He then continues:

ואפשר שקראו הפנימית חיל והחיצונה הקטנה חומה ויהיה פירוש ותעמד בחל שהפילו החומה החיצונה ועמדה העיר בחל שהיא החומה הפנימית. ()

They may have [also] called the inner [wall] חֵיל and the outer, small [one] חוֺמָה, and [in that case] the interpretation of וַתַּעֲמֹד בַּחֵל (2Sam 20:15) would be that they brought down the outer wall and the people of the city “stood by the חֵיל,” which is the “inner wall” [חוֺמָה].

This interpretation conveys the exact opposite: חֵיל means “inner wall.” Qimḥī did not resort to the Talmud here but to some authors, who are unmentioned. Yosef Kara (before c. 1060/70) described this wall as “inner” systematically (on Lam 2:8; 2Sam 20:15; Nah 3:8). Rašī (1040-1105) knew about this opinion and anonymised it in his commentary on 2Sam 20:15: some interpreted (יש פותרים) that the חֵיל was the inner wall (הפנימית). In Lam 2:8, Rašī referred to the low size of the חֵיל (חומה נמוכה).

The third and last section of Qimḥī reads:

According to this interpretation, the third possibility is that חֵיל means “moat.” Qimḥī resorted to the equivalence between Hebrew גיא and Aramaic חלתא in the Targum. If we search in previous mediaeval Hebrew lexicography, we will find that this translation had already appeared in Ben Sarūq’s Maḥberet (ca. 960):

האחת: ישועה ישית חומות וחל; ויאבל חל וחומה; ובאו במערות צורים ובמחלות עפר; בחל יזראל. ענין גאיות ושוחי עמק העשויות סביבות חומה. ()

The first [section]: יְשׁוּעָה יָשִׁית חוֹמוֹת וָחֵל (Isa 26:1), וַיַּאֲבֶל־חֵל וְחוֹמָה (Lam 2:8),וּבָאוּ בִּמְעָרוֹת צֻרִים וּבִמְחִלּוֹת עָפָר (Isa 2:19), בְּחֵל יִזְרְעֶאל (1Ki 21:23). [It has the] meaning of “valleys” and the “moats that are built around a wall” [חוֺמָה].

Ben Saruq used the word “valley” as part of the definiens and Qimḥī is likely to have drawn it from here, or at least to have been inspired by it. The Maḥberet had great diffusion in Ashkenaz and Qimḥī mentioned Ben Sarūq’s name thrice in his dictionary ³⁴ Qimḥī mentioned him as Měnaḥem. See the entries כרה, פאר, according to the enumeration of Biesenthal and Lebrecht (1847, 18). In relation to the success of the Maḥberet in 12^th-century Ashkenazi Jewry, see Ibn Parḥōn’s testimony in his prologue (Stern 1844, xxii). . In Al-Andalus, ʾAbraham ibn ʿEzraʾ also held the interpretation חֵיל = “a place around the wall” (מקום סביב החומה) for Lam 2:8 and Psa 122:7.

Qimḥī’s first option (חֵיל = “outer wall”) is rooted in Ibn Ǧanāḥ’s Kitāb al-ʾuṣūl (ca. 1050). Ibn Ǧanāḥ considered that the following verses are related to the meaning “strength” (القوة):

ومما هو عندى مشتقّ من هذا المعنى وان لم يكتب بياء قوله ויאבל חל וחומה. ותעמד בחל. وهو بالفصال والفصال حائط قصير [...] والاوائل رضى الله عنهم يقولون لهذا الفصيل בר שורא [...] ومن هذا المعنى ايضا שיתו לבכם לחילה. وايضا יהי שלום בחילך. ()

Hence, in my opinion, [the following] is derived from this meaning [“strength”], even though it is not written with yod: וַיַּאֲבֶל־חֵל וְחוֹמָה [Lam 2:8], וַתַּעֲמֹד בַּחֵל [2Sam 20:15]. It means “in the outer wall,” and the outer wall is a small wall […]. The Sages, may God be pleased with them, said about this fortress בר שורא […] And from this meaning is also שִׁיתוּ לִבְּכֶם לְחֵילָה [Psa 48:14] and also יְהִי־שָׁלוֹם בְּחֵילֵךְ [Psa 122:7].

To Ibn Ǧanāḥ, חֵיל means “outer wall” in all these instances. If we compare Ibn Ŷanāḥ’s list of biblical verses to Qimḥī’s we notice that the Provençal grammarian added Nah 3:8. This opinion had already been expressed by Rašī (sub loco). Ibn ʿEzraʾ, however, disagreed and claimed that the meaning of חֵיל in this verse was homonymous (“strength”): וכח יש לה מהים, כי אין צורך לה לחומה וזה ים סוף או ים ספרד (“it has strength from the sea, for there is no need for ‘wall’, and it is the Red Sea or the Sea of Sefarad”; sub loco).

Ibn Parḥōn (1160/1) was faithful to Ibn Ǧanāḥ (notice the absence of Nah 3:8). The main difference in this entry is that he spared Ibn Ǧanāḥ’s linguistic reflections and Talmudic references, and wrote a didactic, to-the-point entry:

מ''א ויאבל חיל וחומה (איכה ב) פי' חומה קטנה בנין בצד הגדולה ונקרית כך. וכן ותעמוד בחיל (ש''ב כ) וכן שיתו לבכם לחילה (תהילים סח) יהי שלום בחילך (שם קב''ב). ()

Another meaning: וַיַּאֲבֶל־חֵל וְחוֹמָה (Lam 2[:8]). It means a small wall that is built next to the big one and is thus called. Also וַתַּעֲמֹד בַּחֵל (2Sam 20[:15]) and also שִׁיתוּ לִבְּכֶם לְחֵילָה (Psa 48[:14]), יְהִי־שָׁלוֹם בְּחֵילֵךְ (Psa 122[:7]).

Mediaeval dictionaries after Qimḥī’s are irrelevant to the understanding of Reuchlin and Pagnini. However, they all confirm the authoritative status of the interpretation חֵיל = “outer wall” ³⁵ I have not found the entry חיל in the Diccionario hebreo de Provenza (Sáenz-Badillos 1987). . Let us check Ben ha-Něsiyya’s dictionary (ca. 1260). In the section of nominal patterns, subsection פַיִל, entry חיל, he wrote:

מ''א ואבל חיל וחומה ' שיתו לבכם לחיה ' יהי שלום בחילך פי' חומה קטנה לצד גדולה מ''א וגלות החיל הזה ' בחיל יזרעאל ' מענין זה ' פיר' חומה קטנה בצד גדולה גובה העיר.

Another meaning: וַיַּאֲבֶל־חֵל וְחוֹמָה [Lam 2:8], שִׁיתוּ לִבְּכֶם לְחֵילָה [Psa 48:14], יְהִי־שָׁלוֹם בְּחֵילֵךְ [Psa 122:7]. The interpretation is “small wall next to the big one.” Another meaning: וְגָלֻת הַחֵל־הַזֶּה [Oba 1:20], בְּחֵל יִזְרְעֶאל [1Ki 21:23] have this meaning. The interpretation is: “small wall next to the big one, [with] the height of a city.”

The core biblical verses remain, but Ben ha-Něsiyya adds a couple of verses that contain the alleged noun with scriptio defectiva and spells the yod in each case. Notice that this lexicographer spoke in terms of size rather than position. Both Ibn Parḥōn and Qimḥī exerted great influence upon Ben ha-Něsiyya, but, in this case, the definiens was clearly borrowed from the former.

The Karaite Ben Mobarak (13^th-14^th c.) wrote in his dictionary: אלפציל חאיט קציר ואלאואיל יסמון הד̇א אלפציל בר שור לאנה סור צגיר (Martínez Delgado 2010c, 332Martínez Delgado, José. 2010c. Šĕlomo ben Mobaraḵ ben Saˁīr. El libro de la facilitación. Kitāb al-taysīr. 2 vols. Granada: Universidad de Granada.
) (“‘outer wall, small wall’, and the Sages called this outer wall בר שור [sic], because [it is] a small wall”). Ben Mobarak repeats and summarises Ibn Ǧanāḥ to the detriment of Qimḥī. The influence of Qimḥī is, however, transparent in Ibn Danān (1468), who mentions the three different meanings reported by the former and innovated in the distribution of the senses:

וח ואבל חל וחומה. אלפציל. והו אלסור אלת̇אני אלקציר אלד̇י יסמיה רז''ל בר שורא. וט ותעמד בחיל. אלסור אלדאכלי. וקיל אלחפיר אלד̇י יחלק אלסור. תרגו בגיא, בחילתא. והו אצח. ()

Eighth [meaning]: וַיַּאֲבֶל־חֵל וְחוֹמָה [Lam 2:8]. “Outer wall.” It is the small, second wall that our Sages, blessed be their memory, called בר שורא. Ninth [meaning]: וַתַּעֲמֹד בַּחֵל [2Sam 20:15]. “Inner wall.” It is said [by some to be] the moat that surrounds the wall. The Targum בגיא is בחילתא, and it is the most valid.

This exercise in source LC allows us to conclude that the interpretation חֵיל = “outer wall” is not only four centuries and a half older than Reuchlin, but also that it was a lexicographic tradition in the Middle Ages, regardless of some small variations and adjustments in the number of biblical verses involved. The authors that conditioned the debate in this period were Ibn Ǧanāḥ and Qimḥī, whereas only Qimḥī was pivotal to Christian lexicographers in pre-modern lexicography.