Shaligram Shukla
 

            The presence of corresponding phonetic features and semantic values is the basis for establishing cognates and reconstructing proto sounds, forms and meanings:

            Sanskrit                        Greek                           Latin                             Gothic

1.         ajras ‘pasturage’          agros  ‘field’                ager ‘field’                   akrs ‘acre’
2.         asti ‘is’                         esti  ‘is’                        est ‘is’                          ist ‘is’
3.         avis ‘sheep’                  ovis  ‘sheep’                 ovis ‘sheep’                  awi- ‘sheep’

Reconstruction: (sound)

1.         a                                  a                                 a                                  a : PIE *a
2.         a                                  e                                 e                                 i :  PIE *e
3.         a                                  o                                  o                                  a : PIE *o        

            The reconstruction of sounds is strictly mechanistic and relatively easy, but the reconstruction of semantics is not.  It is true that some semantic correspondences (2 and 3) offer no difficulty.  But 1 is complicated.  Although the meanings in 1 (pasturage, cultivated field, and acre) have a common semantic core, it is not clear which meaning should be reconstructed for PIE.  The meaning ‘cultivated field,’ which occurs most widely cannot be chosen to represent the proto meaning, because a new meaning does not replace the old meaning everywhere with the same regularity as a new sound.  Moreover, unlike the new and old sounds, the new and old meanings may co-exist, e.g., Greek agraulos ‘spending the night on the agros,’ said of a shepherd.  Here the reference seems to be to ‘pasturage,’ not a ‘cultivated field.’  Likewise, the Latin agrestis ‘standing on an ager’ is a metaphor for ‘wild,’ as opposed to ‘domestic, or cultivated.’  Thus, it appears that the proto meaning may have been ‘uncultivated field, pasturage,’ and through metaphor and metonymy it shifted to ‘cultivated field’ and ‘acre.’ 
            In this paper I look at cognate sets from Indo-European and Algonquian languages, discover morphological constructions in which the cognates may occur with shifted meanings, reconstruct the proto meaning, and trace the metaphorical, metonymic, and indexical paths that many proto meanings must have taken to arrive at their new meanings.