I. The Term ‘Judeo-Arabic’
With the Arabic/Islamic expansion in the seventh century, Arabic became the dominant language and beyond. Jews living under Islamic rulership adopted Arabic as a language for their daily life as well as for scholarly discourses. The common language enabled cultural entanglement attested in the various sources that are written by Jews in Arabic language and ranging from documentary to literary sources. This literature is often referred to as Judeo-Arabic. But what exactly is ‘Judeo-Arabic’? And what is the problem with this designation?
First, let’s break down the word: only one part of this expression is linguistic (-‘Arabic’), the other ‘Judeo-‘ expresses a religious or cultural affiliation. Now, if we were to assume the existence of a specific Jewish form of Arabic, then ‘Judeo-Arabic’ would represent a ‘sociolect’ or a ‘religiolect’, a certain type of language, used by a specific (religious) group within a (larger) language community, rather than a language (or a dialect) on its own. Yet, we then would still have to suppose a particular ‘Jewish’ element in this form of Arabic that makes it distinct from other Arabic varieties – and this claim is, obviously, problematic: If we choose a social component for the definition of a language, we concede him a decisive influence; but we can choose such social criteria quite arbitrarily: We could also define ‘politicians’ English’ or 'Pop song English’ and we would hardly create meaningful categories. Back to Judeo-Arabic: We have to ask whether it is suitable as a linguistic category at all.
One of the most illustrative and prominent features of Judeo-Arabic sources is that they are usually written in Hebrew letters. So, one might consider any Arabic text written in Hebrew letters as Judeo-Arabic. This definition poses a serious problem: If we were to claim that Judeo-Arabic were linguistically distinct from ‘non-Jewish’ Arabic because it is written in another alphabet), we would make a categorial mistake: We would confuse language with script. Since we will discuss this again in the chapter on orthography, we will content ourselves here with the reference that even if we transcribe Arabic into Latin letters (which is a common practice in Western research), we will still maintain that it is Arabic and not ‘Latin-Arabic’. Additionally, there are texts written by Jews in Hebrew but with Arabic letters (for example Karaite Bible transcriptions) and we know also from drafts for official letters that were written in Hebrew characters although the final letter is finally sent to Muslim officials in Arabic letters. In short, we fall short if we try to define a linguistic category simply by its graphic representation. The choice of script certainly tells us something, but it is rather a social practice it refers to than linguistics.
Another attempt at defining Judeo-Arabic could be regarding some deviances from the standard of Classical Arabic. We find forms that could be considered, from the perspective of Classical Arabic, wrong. But apparently, there is method in’t: So-called pseudo-corrections are explained as interferences between spoken vernaculars and the ideal of Classical Arabic, they characterise so-called Middle Arabic, a standardised form of Arabic somewhere between these two ends of the linguistic spectre. This Middle Arabic, however, is not used by Jews only, we find it also in Christian or even Muslim sources. So even certain structural divergences do not help us to define Judeo-Arabic properly.
Apart from Hebrew letters and pseudo-corrections, what then would make Judeo-Arabic Judeo-Arabic? “The main feature […] is the Jewish subject-matter” (Khan 2016, 30). It is its “social and communicative function in that they are addressed to Jews on Jewish subjects, with contents that are likely to be largely incomprehensible to non-Jews” (Khan 2016, 30). According to this definition, Judeo-Arabic is distinguished from the Arabic of non-Jews by its context, i. e. sender and recipient of a message), except for some minor features. A strong indicator in this regard are Hebrew and Aramaic loanwords, particularly in the religious field. Indeed, it seems that especially after Medieval times, Judeo-Arabic varieties have become, due to a stronger spatial and intellectual separation, linguistically more distinguishable from Arabic of non-Jews compared to Classical Judeo-Arabic: The Jewish dialect of Baghdad, for instance, belongs to the so-called qəltu-dialects, whereas the Muslim dialect of Baghdad is a gələt-dialect (according to the respective form for ‘I said’, classical Arabic qultu). Now, as clear-cut as this may seem, we must be cautious: Is the 'religious' affiliation really the cause for this divergence? In the case of Baghdad, the dialectal difference has rather to do with the migration history of different groups dwelling in the city. Other factors apart from religious affiliation (such as city/countryside) have been proven as equally or even more influential and the divergence between Jewish speakers from different regions may be greater than between Jews and their non-Jewish environment; the lesson from this is: It is vital to be very careful.
This is where the criticism of the term ‘Judeo-Arabic’ comes in: First, ‘Judeo-Arabic’ is a term that has been coined by Western researchers, thus it is an etic (i. e., an ‘external’) category, not internal or emic. From the perspective of the Jews whose native language it was, it was referred to as Arabic (or sometimes as ‘Arabic written in Hebrew letters’). Second, the notion of Judeo-Arabic should not obscure the fact that Judeo-Arabic does not exist as a uniform phenomenon and that we cannot distinguish it per se from non-Judeo-Arabic. How Judeo-Arabic looks like, and how it might be differentiated from non-Jewish Arabic is a question depending on the regional, temporal, cultural, and even formal context (a private letter might display other features than a scholarly treatise on philosophy, even if the same author wrote them). Third, focusing only on the Jewish part of Judeo-Arabic culture would not adequately address their interconnectedness with their non-Jewish environment. Emphasising the ‘Jewish’ character reflects the institution of ‘national philologies’ in the 19th century, it tends to regard Judeo-Arabic as a branch of a whole family of ‘Jewish’ languages, such as Yiddish, Judeo-Spanish, Judeo-Persian, Judeo-Greek, Judeo-Provençal etc. and thus suggests an overarching common feature of all those languages (that belong linguistically, of course, to different language families). This cultural essentialisation, however, is highly problematic and it leads to Judeo-Arabic varieties being removed from its specific cultural and historical settings, its contexts, and its entanglements with the surrounding culture, regardless of religious or social affiliation. In practice, this influences where ‘Arabic written in Hebrew characters’ is located: There are universities in Europe or North America where it is located in the Hebrew (and not the Arabic) department.
Remember the letter mentioned above whose draft was written in Hebrew characters but its final version in Arabic letters: Has it ever been Judeo-Arabic (since its raison d’être was to become an official letter)? Would it suddenly stop being Judeo-Arabic when its content is transferred into another alphabet? And it would become what? ‘Normal’ Arabic? This example illustrates, again, how fast such a categorisation can become problematic. Starting from ‘Judeo-Arabic’ without problematising leads to distortions in the end: It is not the evidence that is wrong, but it’s our interpretation when we make Jewishness the decisive point. The answer to the complexity we only touched upon is differentiation: First, the expression ‘Arabic written in Hebrew letters’ – and we would like to specifically refer to Connolly 2023 in this context – is more precise regarding the evidence. We then can continue to ask what makes a person choose rather one alphabet over another and compare this to similar choices (e.g., Hebrew in Arabic letters or Arabic in Syriac letters and so on). Second, when we want to find out about the motivation for this choice (if people could choose between different options at all), we have to be very careful about what we can derive from the material evidence – and what not. Perhaps the ‘choice’ of a script depends more on a person’s education rather than the intended expression of cultural identity. And third, in every case, exactitude is vital, and generalisation is dangerous: Even when there are linguistic differences as in the case of Baghdad, one should be wary of declaring religion to be the cause for them, and one cannot take the specific situation of Baghdad as a model for other places.
Now the difficulty arises as to how to continue working with it: ‘Judeo-Arabic’ is a hitherto well-established term in research and its avoidance in this unit would contravene its usage in the research literature. The pragmatic, albeit theoretically unsatisfying approach for the next pages is to use ‘Judeo-Arabic’, but in quotation marks.