mesolect

In the introduction to his Urban Jamaican Creole: Variation in the Mesolect (Benjamins 1999), Peter L. Patrick writes (emphasis and annotation mine):

Accordingly, creolists have often focused on ‘folk speakers’: people living traditional lives in rural areas, especially the elderly, who are usually the best source for archaic speech (Kurath 1949: 7ff). Their conservative varieties, called basilectal, are idealized and opposed conceptually to the most standard local speech, characteristically used by people at the top of the social scale, which is known as the acrolect. Varieties intermediate between the two are referred to as the mesolect; and the scale along which all these are ranged, as the creole continuum. Many creolists have until recently been primarily concerned with describing and analyzing the basilect, which is typically idealized as “the creole” in opposition to everything else — for example, Jamaican Creole as opposed to Jamaican English (the acrolect conceived as a dialect of international English). (Patrick 1999:5) 

I was somewhat stunned to read these words, not because I learned anything new, quite the opposite: these 11 lines (in the physical book) perfectly summarize a problem I have been trying to wrap my head around for the last two or three years. Except where Patrick speaks to and about creolists, my problem involves Arabic varieties and Arabic dialectologists. Disregarding the issue of the creole continuum, Patrick's diagnosis fully applies to a large portion of research in Arabic dialectology:

• focus on elderly rural speakers
• focus on archaic varieties/features
• the idealization of said archaic varieties as "the dialect"

Don't believe me? Then consider the following passage from a recent description of an Arabic variety of ... Well, I don't want to put the authors on the spot and also I am writing a commissioned review of the book, so names shall be changed to protect the not-entirely-innocent. Let's call the rural town Ayn Wadi and the country Watan, the book would then be titled The Dialect of Ayn Wadi (Watan) (emphasis mine):

The major data collected during the field research are free speech audio recordings of improvised narratives about customs and activities in the town. Many informants I tried to work with were unable to narrate facts and stories in front of a recorder without mixing the local dialect with standard Watani. With these informants, I could only check lexemes without recording them to verify the recorded data with a broader spectrum of speakers. All the speakers were over 50 and many recalled the dialect spoken by their parents. The younger generation represented by youths and children were only briefly interviewed and they seemed to preserve only a few of the archaic features..."

It's all there: elderly speakers, archaic features, and the idealization of varieties spoken by elders as THE dialect of Ain Wadi. Only data from speakers over 50 is taken into account and on top of that, all the texts in the book were produced by people in their late 70s and early 80s. And then they are exactly what you would expect if you have ever read any description of an Arabic variety: memories of the old world, religious traditions and holidays, and traditional crafts.

Now there is nothing wrong with collecting spoken data from elders, far from it; there should be thousands upon thousands of field researchers armed with RØDE Wireless GO III doing exactly that all over the world. The problem here is that if you discount all the speakers younger than 50 - many of whom, nota bene, report not on their language, but on the language of their parents - is the variety you describe really the dialect of Ayn Wadi? No, of course not. Such research is not only misleading at best and disingenous at worst, but also - and this is far more important - constitutes a missed opportunity. Variation within the dialect could have been studied in multiple dimensions,  the contact between the local dialect and standard Watani or even between the local dialect and the other varieties spoken in the immediate area could have been examined, and of course we could have so much data on the speech of young people in a complex linguistic landscape, not to mention data on standard Watani (if there even is such a thing). But alas, speakers were apparently turned away (or, worse, their data was discarded) because they did not "preserve... the archaic features". That is less than ideal for the field, not to mention a questionable use of the money provided by whatever funding mechanism financed the research.

Don't get me wrong, this is not how things are done everywhere in Arabic dialectology and The Dialect of Ayn Wadi (Watan) is not a bad book. It is, however, a perfect example of one way of doing Arabic dialectology, a way that we should leave behind for good.