PIE *gn- > know/ken

[Dennis Philps]

Thanks to Jens & Miguel for their responses to parts of my question
concerning the reflexes in English
of PIE *gen- 'to know'. I'd still be most grateful for any opinions as to :

a) at what stage in linguistic history the initial voiced consonant in PIE
roots such as *gno- "know/ken/can" or *gen-  "knife", etc. became devoiced
(esp. in respect of Grimm's Law), and

b) whether it is correct to state that the form 'know' is derived from the
zero grade of *gen-  (*gno-), whilst the forms 'ken (dial.)/can' are
derived from the full (e) grade (*gen-)?

Best Wishes,
Dennis.

[ Moderator's comments:
  1.  Please be aware that the root is *gneH3-, with the o-coloring laryngeal.
      The form 'know' is derived from the full grade *gneH3-w-; forms such as
      'can' are derived from the zero grade *gn.H3-.

  2.  The unrelated *gen- is not "knife", but rather the hypothetical basis for
      a large number of Germanic stems referring to knobs, lumps, sharp blows,
      and the like.

  3.  It may be that the initial *g was never voiced, but rather a glottalic
      /k{^?}/, which lost its glottalic articulation.  However, that aside, it
      appears that Germanic was a standard IE language in the first half of the
      1st millenium BCE, so not before 1000 BCE and not (much) later than 500
      BCE.

  --rma ]