Johanna,

You wrote:

"The "a-" examples (at least some of them) may originate in old
progressives -- these used to be formed with "on", e.g. "the house is on
blazing". The "on" was reduced in pronunciation to "a-" (which survives
in, e.g., Appalachian English, "I'm a-comin' "). The absence of "-ing"
needs to be explained -- maybe Herb has this in his vast stock of
knowledge of the history of English. Most speakers of English have
dropped the "a-" in the progressive construction, but it remains.  In
fact, it seems to have given us a prefix for deriving adjectives from
verbs: "a-twitter" from "twitter"; "alight" from "to light (a fire)";
"aglow", "askew". But I don't think it's productive -- we don't use it
to make new verbs as a matter of course. This would render words like
"afall" (falling); "atwist", "asink" (cf. "afloat")."

As some wise person once said, "The only thing you can do when you open
a can of worms is get a bigger can."  I checked the a- prefix in the
OED.  There are 14 of them, including those derived from Latin ad- and
ab- and those from Greek a- (negative), which we can safely ignore here.
There are also 22 divisions of the "a" preposition, derived from "on".
Some forms that look today like a- prefixes became that by analogy.  The
a- of "afraid" is such a case.  This one is a fossilized past participle
of the 14th c. Anglo-Norman verb "affray", where the "af-" probably
indicates Latin "ex-".  The a-, which came into English as an
unanalyzable part of the root "affray" was then reanalyzed as the a-
participle prefix that descended from OE ge-, which, in another usage,
we still preserve in Modern English in "enough".

The OED also has an entry for "a" as a preposition that, in Middle
English, lost its final /n/ before consonants just as the indefinite
article did.  Unlike a/an, however, the a/on alternation didn't survive
as such into Modern English and survives only in fossilized forms as a
prefix, so that the a- prefix, from multiple sources, and the a/on
preposition have fallen together in Modern English as a less
differentiated prefix.  One of these is the present participle
"ahunting".  Up into the 19th c. a participle like "abuilding" was
interpreted as a passive, as in "My new house is abuilding", meaning
that it's being built.  The progressive passive "is being built", which
starts developing in the 17th c., finally replaces the use of the
present participle with passive meaning in the mid 19th c.

As a/on settled in as a prefix, we developed words like aback, afore,
afar, ahead, aside.  Other nouns took this prefix, as in alive, asleep,
awork, ajar, athirst, ablaze, afright, afloat, astare.  The stem to
which a- was added was a noun, frequently deverbal, so that this a-
prefix took on an adverbial and later adjectival function.

In abide, ago, arise, awake, the prefix goes back to OE ar-, distinct
from the ge-derived and on-derived prefixes.  This one has fallen
together with the other two in Modern English.

Back to "afraid" for a moment.  The reason we can use it only
predicatively and not attributively is not that it is a past participle,
at least etymologically.  "Molten" is also an etymological past
participle, but "molten lead" is good.  In fact, it's hard to use
"molten" predicatively.  

?The lead was molten.

Rather, "afraid" behaves as it does because it has been reanalyzed as an
a- prefix word.

The whole historical background is rather messier, and the OED entries
for a- as a prefix and as a preposition are worth reading.

Herb

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/