Curious Name: al-Wādī
By Basil Dragonstrike
Greetings, and welcome to another "Curious Names" article---quite possibly, the longest I will ever write.

In this article, I am going to explore an Arabic byname that has, in the SCA, recently become relatively popular----al-Wādī. First, though, there are a few terms I must define, so you will be able to follow what I'm saying.
I start with "generic toponym". This is a type of locative; a locative is a name based on a the name of a place, while a generic toponym is based on a kind of place. In modern English, this is the difference between John Winchester and John Rivers; the first comes from an ancester who came from Winchester, while the second from an ancester who lived by a river.
Next is "romanization"; we tend to speak of transliteration when describing writing something originally in one alphabet when given in a different alphabet. Properly, though, transliteration refers to taking one letter in alphabet X and turning it into one letter in alphabet Y. In Arabic, some of the letters do not correspond well to just one letter in the Roman alphabet (the one being used right here). A scholar who wishes to transliterate Arabic, therefore, must use unusual letters, such as š to transliterate ش, and so on. What is more often used is two-letter combinations; for instance, sh to represent ش. This method, which changes alphabet X into Roman letters is known as romanization.
Then there's "editing". In terms of Arabic texts, this means taking a manuscript of a text (or more than one, if there is more than one), checking to see if there are probable mis-spellings and making probable corrections, filling in as well as possible gaps and lost pages/lines, and in general neatening up, and then publishing one's results, in Arabic (with notes as to the changes, guesses, etc. made). This is useful if one reads Arabic, somewhat if one can merely romanize.

Now to the point of this article - - - -
When al-Wādī was first used in a submission, it was stated to be a generic toponym, with the assumption that generic toponyms were used in Arabic names. I thought it would be interesting to find other generic toponyms, and write an article on them. So, what does wādī mean? Well, the Arabic word وَادِي means an intermittent stream, a water course that flows irregularly, particularly one that flows only in certain seasons. I will herafter use "intermittent stream" to refer to the object, and use wādī when referring to the word as such. So I figured I'd look for other words for types of features to see if they occurred in Arabic; first, though, I thought I should double-check al-Wādī. And that's when I ran into difficulties.

The first problem is with the romanization. In romanizing وَادِي there are a number of possible forms, depending on how dilligent the writer was, and what printing resources were available. To be blunt, if you see "wadi" in some book or article or web-page, and if you don't know how careful the romanization is, you can't be sure what the Arabic is. It could be وَادِي  or وَادِ or وَدِع or---well, lots of possibilities. Should the a be better romanized as ā? Or is it a short a? Is the d actually ḍ? The i, ī? Should or shouldn't there be a ʻ on the end? In short, there are four places which have two possibilities each, giving 16 possible original Arabic. Here's a table, using LOC romanization, for the 16 possibilites:

Wadi Wādi Waḍi Wāḍi
Wadī Wādī Wāḍī Waḍī
Wadiʻ Wādiʻ Waḍiʻ Wāḍiʻ
Wadīʻ Wādīʻ Waḍīʻ Wāḍīʻ

I have put two of these in boldface: that in the second line because it's the correct form of the name of an intermitent stream, the other for reasons I'll get to soon.

There are three possible sources for al-Wādī that people in the SCA might use as documentation. Only two of them have been used so far, but I'm going to start with the third. This is Cariadoc's Miscellany, a work by David Friedman and Elizabeth Cook, containing a number of articles regarding various subjects that may be of interest to SCAers (and other re-creators). Part of it is recipes, some from Islamic sources. One of them goes by the ponderous title:
Tharîda in the Style of the People of Bijaya
(Bougie, a city in Algeria) Which They Call
the Shâshiyya of Ibn al-Wadi'
In the text of the recipe, the same person is referred to as Ibn al-Wadi, without the apostrophe.. So, we are already dealing with the problem of inconsistent romanization.

The history of some of the recipes in Cariadoc's Miscellany is worth recounting briefly. One source is known in the original Arabic as Kitab al tabikh fi-l-Maghrib wa-l-Andalus fi `asr al-Muwahhidin, li-mu'allif majhul and dates to the 13th century CE. In the early 1960's, a translation into Spanish, under the title La Cocina Hispano-Magrebi en la Epoca Almohade, was made by Ambrosio Huici Miranda. Somewhat later, a group of four SCAers started translating the Spanish work into English; when a preliminary version started to be circulated, Charles Perry, food critic for the LA Times, took this translation, the Spanish work of Ambrosio Huici Miranda, and Ambrosio Huici Miranda's edited version of the original Arabic, and made a full translation into English.

Although I found the title of Perry's work, I have not, despite the help of a number of Inter-Library Loan (ILL) librarians, been able to find hide nor hair of that work; that is, no library taking part in the ILL network had a copy---nor a copy of the magazine issues with Perry's preliminary work on the translation. However, via ILL, I did get my hands on a copy of La Cocina Hispano-Magrebi en la Epoca Almohade. In that work, the recipe is titled
Torta al estilo de la gente Bugía
La llaman šāšiya de Ibn al-Waḍīʻ.
Note the use of š to romanize ش! in šāšiya. The name Ibn al-Waḍīʻ; is also used in the text of the recipe.

This means that Friedman & Cook's Wadi' and Wadi are very definitely non-scholarly (and may mis-copy Perry's work). It also means the scholar Ambrosio Huici Miranda used Waḍīʻ, which is the other form I boldfaced in the chart above. Thus, the Arabic would not be وَادِي , but وَضِيع . In conclusion: it's a good thing no-one has yet used Friedman & Cook's work, as it does not, in the slightest, support al-Wādī.

So, what of those two sources I mentioned that have been used? I start with the online work by Julia Smith, "Andalusian Names: Arabs in Spain" found at:
https://www.s-gabriel.org/names/juliana/andalusia/
where Smith states the names she lists are from Estudios Onomastico-Biograficos de al-Andalus. She includes a byname spelled
al-Wadi
In al-Andalus there was an intermittent stream called Wādī Āsh. On its banks was a town also called Wādī Āsh. The adjective form is WādīʼĀshī; a person using a byname based on the adjective would use the form al-WādīʼĀshī. I have limited access to Estudios Onomastico-Biograficos de al-Andalus, but I found there a person called Muḥammad Ibn Ŷābir al-Wādī Āšī; changing that to a romanization you are more likely to have seen, this would be Muḥammad ibn Jābir al-WādīʼĀshī. This person has an article in The Encyclopaeidia of Islam, 2nd version, and is not the only person with this byname (al-WādīʼĀshī) found there (and elsewhere). After a long e-mail conversation with Smith, she admitted she could not find her notes, that my knowledge of Arabic was notibly greater than hers, and that she must have, unfortunately, mis-transcribed this name. She agreed with my conclusion that the byname ought to be al-WādīʼĀshī. Hence, her work does not support al-Wadi as a byname.

The other source people have used for al-Wadi is Selections from The Art of Party-Crashing in Medieval Iraq by al-Khatib al-Baghdadi, translated and edited by Emily Selove. In Selove's book, she does not use either macrons nor under-dots, but, she does differentiate between ʻayn and hamza [ʼ], and does not show ʻayn in her version of "Wadi". Thus, any of the first 8 forms in the above table might be what the original was. Note that in her story, Hakam is said to come from "Wadi," not "a wadi"; this clearly is meant as a town name, not a generic toponymic. From the beginning, I was concerned this might be part of a name, as al-Wadi in WādīʼĀshī proved to be, though the fact Selove is a professor alleviated some of my concern.

I decided to try and find another translation. However, so far as I can find, there isn't one. So, I looked for the Arabic text Selove used. I'm not sure I found it, but I did find an edited text (remember my explanation of "edting," above). The title was not exactly the same as that given by Selove, but it's undoubtedly the same text. My ability to read Arabic is rather small, but between what I have, Google Translate, Lane's Lexicon, and Wehr's Dictionary, I was able to work out what I needed. It is worth pointing out that the numbers Selove gives to the stories do not exist in the edited text I had available. The long introduction, also not in Selove's book, was a further problem. And, to tell you the truth, I'm not sure the stories are in the same order, which is on top of the fact Selove skipped some (as her numbering shows). All this made finding the correct story difficult.

But find it I did! And there is no doubt that Hakam is said to have left Wādī, and no doubt he is called al-Wādī. Hence, Selections... is a valid source for al-Wadi, although a more scholarly romanization would be al-Wādī; I have seen an Arabic text and can assure one and all that al-Wādī is the correct romanization.

To conclude: Friedman & Cook's Cariadoc's Miscellany does not support al-Wādī at all, Smith's "Andalusian Names: Arabs in Spain" contains an unfortunate error, but Selove's Selections from The Art of Party-Crashing in Medieval Iraq does show al-Wādī was used as a name element in period, though it makes clear al-Wādī is not a generic toponym (and thus, does not support the idea that generic toponyms were used in Arabic).

And now you can take a deep breath and relax - - - I'm done.

Until my next article, that is.