In response to Rick McCallister's question, I have written one article on
Nostratic. It's the fourth chapter in the recent volume, Nostratic:
Examining a Linguistic Macrofamily, edited by Colin Renfrew and Daniel
Nettle. In it I consider what emerges if we compare one version of
Nostratic with reconstructions from two sub-Saharan African language
families. I use those findings not so much give my views on Nostratic per
se as to raise issues about what we are dealing with and what we need to
take account of in pursuing deep-time reconstruction.

On the Afroasiatic family and its origins, I actually have several pieces,
some available in print and some in the works still. My arguments are
essentially that Afroasiatic is a resoundingly African family in origin and
that the proto-Afroasiatic peoples were (arguing from reconstructed
subsistence vocabulary) distinctly preagricultural in economy. The
evidence is very strong that they used grasses or grains as food. But
diagnostic vocabulary of cultivation is entirely lacking before the
proto-periods of the major branches of the family. Wild grass/grain
collection is attested in the archaeology of northeastern Africa from
before 15,000 BP down to the beginnings of cultivation. I argue in favor
of the correlation of the early Afroasiatic peoples with this archaeology.

A recent rather full account of the evidence and my reading of it, but
difficult to get because it was published on CD, is somewhat misleadingly
titled, "Who were the rock artists? Linguistic evidence for the Holocene
populations of the Sahara." It appears in a volume published about a year
ago, edited by Alfred Muzzolini and Jean-Loic Le Quellec, forming part of
the proceedings of the 1995 international rock art congress held in Turin.
(An unusual forum, I'm afraid; I'll choose better in future.)