It has been established beyond reasonable doubt on linguistic, epigraphic, and textual grounds that the Sn, and in particular certain portions of it, date from a relatively early stratum of the Pali TheravAda canon. First of all, the archaic character of the langauge of many parts of the Sn has been clear insce the time of Fausboll, as shown in the introduction to his translation (1881: xi-xii) and as further developed by, among others, Jayawickrama (see Jayawickrama 1951 and on, on the Khvs-P in particular, 1949: 123-5).
The epigraphic evidence consists of the celebrated list of seven texts recommended by Ashoka in his Calcutta-BairAT edict, of which three or four (according to varying opinoins; see, e.g., Jayawickrama 1948b:229-32; Norman 1992a:xxix-xxx) are likely to refer to suttas whose Pali versions are preserved in the Sn. This shows that some texts which later became part of the Sn were already extant and pouplar at least by the time of Asoka, that is, around the middle of the third century B.C.
Finally, abundant internal textual evidence in the Pali canon and in other Buddhist texts affirms that certain portions of what came to be the Sn were extant from a relatively early time. Besides the fact taht the earliest surviving independent commentary, the Niddesa, is a commentary on parts of the Sn and hence may have been based on an earlier core of such a collection (Jayawickrama 1951: 123), numerous references to sections of the Sn, especailliy the ATThakavagga and PArAyanavagga, in Pali and othe rBuddhist exts (see e.g., Jayawickrama 1948b: 232-6; von Hinuber 1996: 49 n. 165) prove that these parts, at least, of the Sn were old and important in the Buddhist tradition.
—A G. V. of the R. S., p.16