ABSTRACT
A problem which has vexed d ia le c to lo g is ts from the e a r l ie s t days o f the sub ject is th a t o f the im p o s s ib ility o f making c le a r -c u t d iv is io n s between d ia le c t areas, e s p e c ia lly in reg ions w ith an uninterrupted geograph ica l spread o f popula t io n . The German neogrammarians o f the 19th century noticed th a t in such areas one speech v a r ie ty blends im percep tib ly in to the next through interm ediary v a r ie t ie s so th a t ra ther than d is c re te e n t i t ie s we are confronted w ith a continuum which is on ly d iv id a b le in to c le a r ly demarcated reg ions i f we choose ve ry simple c r i t e r ia fo r d i f fe r e n t ia t io n . As an example o f th is we may take the important iso g lo ss which d iv id e s areas w ith /j/ frctn areas w ith /y/ as a r e f le x o f OA j* in such forms as /rayyal, rajjal/ "man". The pronunci a tion /rayyal/ is regarded in Iraq as "southern" as con tra s ted w ith /rajjal/ which is the pronunciation o f the cen t r a l area and north. D isregarding fo r the moment the pronun c ia t io n /razzal/ which is c h a ra c te r is t ic o f parts o f the southern marshlands, th is g iv e s a f a i r l y c le a r d iv is io n in to a /J/ area north o f Batha on the Euphrates and Kut on the T ig r is , and a /y/ area south o f i t . Th is is an important " c h a ra c te r is t ic " is o g lo s s since the pronunciation /rayyal/ is w ide ly recognised as a "southern" c h a ra c te r is t ic and i t would be tempting to take th is l in e as the demarcating one between "southern" and "non-southern" types in Ira q . However when we examine other iso g lo sses which are equ a lly regarded as d iv id in g "southern" and "non-southern" types , we fin d th a t they do not correspond to t h is . For instance the s u ff ix /-an/ c h a ra c te r is t ic o f f i r s t person s in gu lar ve rb a l forms, such as /aridan/ " I want" /aruhan/ " I g o " , is used as fa r north as N a ja f on the Euphrates, w h ile the occurrence o f the s u ff ix /-man/ "who" or "which" p o s t-v e rb a lly in such forms as /sifit-man/ "who or which d id s e e " , /saf-man/ "who or which d id he see" is used as fa r north as H i l la on the Euphrates. Thus we are faced w ith the d ec is ion as to whether to regard as "southern" on ly the core d ia le c ts which have a l l three c h a ra c te r is t ic s or to include a l l d ia le c ts up to the outer
r in g a t H i l la . Furthermore there are other con trasts which cannot be s t r i c t l y c a lle d " is o g lo s s e s " since th e ir geograph ic a l d is tr ib u t io n is more fragmented, which cut across these boundaries e n t ir e ly , d iv id in g "nomadic" from "sedentary" speech t y p e s . (1) Th ere fo re , i f we p e r s is t in try in g to d i v id e "southern" from "non-southern" types we are fo rced to consider a "co re " area o f "southern" speech w ith a band o f "border d ia le c t s " , w h ile d isregard in g the fa c t th a t nomadic "southern" and nomadic "non-southern" are united by some fea tu res which separate them fran th e ir non-nomadic counter p a rts .
