/azəbbuž/ : Tracking a Wiktionary Etymology

If on an autumn’s night a Wiktionary user, flitting in bewildered confusion from one impulsively chosen word to another…

Imagine an Oulipian conceit of sorts, where you land on Wiktionary’s page for “Foreign word of the day”, and based on the etymology of the day’s pick, you are to write a poem, where the word’s equivalent in your writing language takes center stage (provided it exists in the language). I know! I know! Yet why not? Let’s say, the day’s word is acebuche. You click on the word to look-up the etymology, and you read ‘inherited from Old Spanish [...], from Andalusian Arabic, from Arabic الزَّنْبُوج (az-zanbūj)’. The section “Further Reading” links to the 2014 online edition of the RAE, Diccionario de la lengua española. Wiktionary being a crowd-sourced dictionary, you click through to the Diccionario’s entry, and there it says: ‘Del ár. hisp. azzabbúǧ’, and stops there. Why not check out Wikcionario, since this is a Spanish word after all ! It too says ‘Del árabe andalusi azzabbúǧ’, with no reference to any source. Might as well try out the derived acebuchal in the search : nothing about Arabic, the entry simply repeats the etymology of acebuche, yet it gives a reference : the 1914 edition of the Academia’s Diccionario, p.10. You open the page and there you read, under acebuche: ‘Del berb. ازبوج, azebuch’?! How about checking out what Wiktionary says about الزَّنْبُوج : ‘A Berber borrowing’, with a reference to Dozy’s 1881 Supplément aux dictionnaires arabes. And, if you go as far as to look through the discussion notes for these entries, you’ll notice that only زنبوج has a couple remarks disagreeing about its etymology : non-Berber (quoting Souag’s 2018 Berber Etymologies in Maltese) vs. Berber (with little to substantiate the assertion other than a vague reference to كتاب عمدة الطبيب re: tāzammūrt). And you thought you’d get some clarity !

What struck me weaving this web path was not that there is disagreement about the possible origin of the word, or the activism of some of Wiktionary’s contributors, but the seemingly small changes in the Spanish dictionary corpus : switching from berber to arabic (sometime between 1914 and 2014), dropping the link with arabic, and changing andalusi to hispánico. The rabbit hole beckoned, and unawares, I dove in, relying mainly on what can be found on the open web, by a non-specialist in diachronic lexicography. As I collected references (by no means exhaustive) of who said what and whom they cited, I needed to give the data a form(s) to make the temporal dimension of the etymologies visible, along with the embedded citation network that should ultimately emerge. The results are two visualizations : a timeline, and a citation network graph. I should add that the network graph accounts for S. Chaker and G. Camps, but I kept them out of the timeline (busy as it is already).      

   Figure 1. Timeline of an Etymology

         

Figure 2. Who Cites Who: A Network

Citation Politics : A Few Disjointed Observations

The most cited author: Dozy (7)
The biggest citer : Simonet (23 references)
The most ascribed source language : Latin (4), Levantine Arabic (2), Berber (1)
Detected communities : 6

The topography of the communities that emerge from the network, overlap neatly with what can be called schools : the Dutch-Spanish School (Dozy, Simonet, Eguilaz, Corriente) ; the solitary German School (Fischer) ; the French Colonial School (Laoust, Colin, Marçais, Ricard, Camps,*Chaker). Of all the languages in the fray, in this etymological probing, only the Spanish and German schools consider Levantine Arabic sources (to argue against, in the case of Eguilaz & Simonet); these sources are absent from the Dutch and French Schools’ references.

It is notable that while Fischer is cited by Colin and Marçais, Laoust who cites Marçais sidesteps Fischer, whose work he knows and refers to regarding the introduction of the cultivation of the olive tree in Tripolitania and Cyrenaica. The cleaving away of Fischer seems to follow the fault line of the presence of Levantine Arabic in the carved out corpus of references. (In Corriente’s case, it would seem to be an oversight). One could, perhaps, read the lexicographical interlocution between the Spanish and French authors as gesturing toward an affinity rooted in that they are both animated by a kind of philological reconquista, and the desire to step over the arabo-islamic reference, deemed invasive and alien, to (re)activate a more “authentic” and local stratum : the hispano-latin of the Mozarabes “los naturales de la tierra” (in the case of Simonet & Eguilaz), and pre-Arab North Africa (for the French). For instance, when Eguilaz’s rhetorical argument in his entry for acebuche, reaches out for the toponyms az-zembuchár, acebuchár, and acebuchál, to anchor Simonet’s etymology, contra Dozy, and in course marshalls out the Geography of al-Idrīssī, the Libro de la Montería (Book of the Hunt), which he ascribes to Rey D Alonso (Alfonso XI, el justiciero, “He of Rio Salado”), the Kitāb al-Filāḥa of Ibn al-Ɛawwām, and the botanical pharmacopeia of Kitāb al-Jāmiɛ (Compendium) of Ibn al-Bayṭār, Eguilaz in a way is citing the land to testify, to bear witness ; he is invoking what he considers to be the ultimate veriloquium–which, in essence, is the intrinsic meaning of etymology in classical rhetoric, translated from Greek into Latin, and captured in Cicero and Quintilian’s phrase “words are the tokens [notae] of things” (Bloch 1983:54). Needless to say, through “botany” and the “hunt”, as encoded knowledge of the land, its topography, fauna, and flora Eguilaz’s citing strategy is a kind of grounding, a siting of language in a topology of a reimagined hispano-latin Iberia. Etymology as a speaking from a place (loqui/locus cf. Bloch 1983:55), a topolect, a language of place unsullied by voces arabigas. Laoust is more explicit in this citing of the land as philological witness ; in discussing the word azemmur, he writes : “Il apparaît d’une antiquité déjà respectable si on en juge par ce fait qu’il s’est fixé comme toponyme dans des régions où le berbère n’est plus parlé [...]” (Laoust 1920:447). As such, this deployment of etymology, by Eguilaz and Laoust, enacts through toponymy, a scenario of return that forecloses the arabo-islamic referent and its historical relevance.

To return to where I started, with the seemingly trivial variations in the Spanish corpus, the RAE’s (1933-1936) edition of the Diccionario Historico de la Lengua Española still says ‘Del berb. ازبوج, azebuch’, just as did the 1914 edition. The (1960-1996) edition, however changes it to an unqualified ‘Del ár. azzanbūǧ’ ; and the 2014 edition further changes it to ‘Del ár. hisp. azzabbúǧ’. These changes are part of a broader ideological move to hispanicize al-Andalus. They should also be put in resonance with the Diccionario’s change (between 1914 and 2014) of the flag ‘En el imperio marroquí…’ to ‘En Marruecos…’ (Escribano 2016) for certain words, as well as the mid-19th century change from ‘España árabe’ to ‘España musulmana’ (Dominguez 2021).


< />

___________

BLOCH, Howard R., Etymologies and Genealogies, University of Chicago Press, 1983