For the study of the evolution of Kurdish, these 400-odd years are not really relevant; one therefore has to try to reconstruct the earlier stages of Kurdish by comparing the available Kd. data to those from other Iranian languages and dialects, past and present. The synchronic relation of Kurdish to other modern Iranian languages may then be translated into a diachronic, genealogical model. Both the synchronic and diachronic levels of analysis, of course, go hand in hand, since the postulation of relevant (synchronic) isoglosses presupposes a certain knowledge about their historical (diachronic) significance.
The position of Kurdish among modern Iranian languages and dialects is not easy to define. Many dialects of Kurdish are still imperfectly known, as are many other modern Iranian dialects to this day. The picture that may be drawn of the evolution of Kurdish is further obscured as a result of the continuous migration of Iranian peoples, and accordingly by many contact, areal, and loan phenomena among Iranian languages, above all by the far-reaching influence that Persian has exerted on most Iranian languages throughout its history. One of the universally accepted (but indeed also questionable) findings of Iranian dialectology is that, since Middle Iranian (MIr.) times, the Iranian languages may be divided into an eastern and a western subgroup; to the latter belong Persian, Kurdish, Balōči (see BALUCHISTAN iii. Baluchi Language and Literature), and all other Iranian languages and dialects that are spoken in present-day Iran (e.g., Gilaki, Tāleši, etc.). For a further, more detailed sub-grouping of the West Iranian (WIr.) languages there exists no commonly accepted dialectal or genealogical model that would do justice to the variety and complexity of these languages.
History of studies. During the earlier phases of Iranian dialectology, i.e., until ca. 1900, the Iranian languages and dialects were mostly grouped on geographical terms. From the West Iranian languages, only Persian, Kurdish, and Balōči were known relatively well; from the other WIr. languages, many were spoken in central and northwest Iran, i.e., approximately in the area that corresponds to most of ancient Media, and were accordingly subsumed summarily under the heading of “Median” (e.g., Geiger und Kuhn, I/2, p. 413; Hübschmann, p. 115). This changed only at the beginning of the 20th century, when so many documents in various MIr. languages were found in Central Asia that put the study of Iranian linguistics on a totally new basis.
Among these documents, those written in Manichean Middle Persian (ManMP.) and Manichean Parthian (ManPth.) were of particular importance for the study of WIr. dialectology. Parthian proved to be clearly distinct from Middle Persian; while the latter underwent several characteristic phonetic innovations (as compared to Proto- and Old Iranian), Parthian was more conservative phonetically. Some of these conservatisms are also found in various modern non-Persian languages and dialects. Since Parthian was spoken in the north of the region where WIr. languages were spoken, it was labeled the “prototype” of “NW-Iranian (languages and dialects)” in the MIr. period (in contrast to MP. as a representative of “SW-Iranian”). The modern non-Persian languages and dialects that are spoken in Iran—except a small number from southern Iran—were accordingly called “NW-Iranian” (the first time possibly by Mann, p. XXI). The label of “NW-Iranian” was used summarily, and negatively, in the sense of “different from Persian,” in quite the same way as “Median” had been used earlier. Paul Tedesco, in his pioneering work on the dialectal distinctions between Middle Persian and Parthian (Tedesco, 1921), saw various connections between Persian and Kurdish, and occasionally also Balōči (e.g., on pp. 193, 255). His sub-groupings of West Iranian, however, are mostly made ad hoc, often based on one feature only, even subject to frequent change (e.g., p. 253), and already implicitly question the very concept of NW-Iranian.