re: "signalling" vs "signaling"

[Kent M Pitman <KMP@STONY-BROOK.SCRC.Symbolics.COM>]

    Date: 26 Jul 89  1440 PDT
    From: Dick Gabriel <RPG@SAIL.Stanford.EDU>

    Well, every dictionary I have lists it first,

Did you check the key up front to see if order matters? See below.

    the OED says one L is the American spelling, 

Since when is the OED the recognized authority on American usage?

    the Chicago Manual of Style says 

I've worked with such `standard' style guides on newspapers and I have
to say that I think they have are far from "authoritative" on the issue
of what is generally permissible. They are clear on what "they" permit,
but that may be very different than what "English" permits.  In most
cases, their real value is in their resolution of arbitrary issues which
would otherwise have perfectly balanced arguments on both sides, so that
people who have agreed up front to abide by them can just get along with
their work.  Even most places I know of that have used these things have
private exception lists because they can't abide by all the rules, or
because some rules are so domain-specific as to be not useful or not
complete in other domains.  Anyway, I certainly did not agree up front
to abide by any particular such guide, so I don't buy any attempts to
sneak such a thing in now after the fact--certainly not as an unbiased
authority on correctness, which these things do not typically profess to
be.

    every writer I talk to says to use one L,

Aha! I always suspected you didn't listen to our writers. :-}
Seriously, though, I just wanted to underscore that the set of writers
you talk to isn't exactly a universal quantification of all writers 
everywhere.

    every article I have read in the last 3 years spells it with one L,

Funny. You reviewed an article by me on condition handling within the past
three L's. I question your memory.

    and GNUEmacs wants me to spell it with one L.

Hardly authoritative. Again, it's more important for a SPELL program to
permit either spelling just so it can catch cases where you use one spelling
in one place and another in another place than it is for a program to allow
all possible spellings. They really ought to write a SPELL program that
knew "foo" and "fu" were both legit and that noticed you used "fu" once
and then complained when you used "foo" later (or vice versa). If they did
have such featureful SPELL programs, and if GNU Emacs was such a one, then
you might well not get barfed at because it might well permit both spellings.

    This has nothing to do with personal respect - I'm simply trying to
    follow contemporary American usage as best as I can discover it.

Well, I'm a contemporary American. I spell it that way. My dictionary
says it's ok to do that. My dictionary says that people are pretty
evenly split on the issue. Why not follow my lead.  Why is that not
exactly an issue of personal respect?

    Note that I also include most punctuation (but not all) within 
    right double quotes ``sort of like this,'' which is both illogical
    and contrary to British usage but is American usage.

No argument here.

    I used to spell ``signalled'' with two L's until I was confronted with
    the very facts I'm listing here.

Good, then look on this as an opportunity to go back to all those people
who tried to change you and tell them they were confused.

Quoting now, from my dictionary:

 ``VARIANTS

 ``When a main entry is followed by the word <or> and another
   spelling, the two spellings are equal variants.  Both are
   standard, and either one may be used according to personal
   inclination:

    ``mea.ger <or> mea.gre

  ``If two variants joined by <or> are out of alphabetical order,
    they remain equal variants. The one printed first is, however,
    slightly more common than the second.

    ``judg.ment <or> judge.ment

  ``When another spelling is joined to the main entry by the 
    word <also>, the spelling after <also> is a secondary variatn and occurs
   less frequently than the first:

    ``quintet <also> quintette

  ``Secondary variants belong to standard usage and may be used according
    to personal inclination. ...''

Note that it's clear here that the mere fact that "or" is used at all
means that it is proper -American- English no matter what the order.
However, the fact that in all three instances below they appear 
alphabetically means that, according to the above rules, they are 
also equally common in American usage. If they had not been, the 
dictionary guys would have been forced to use <also> instead of <or>,
or to use <chiefly Brit> or some such.

 ``2signal <vb> signaled <or> signalled; signaling <or> signalling ...
    signaler <or> signaller <n>.''

It's not like I'm picking a no-name obscure little dictionary that no
one has ever heard of.  The fact is that a reputable American dictionary
backs me up.  If you found another that didn't, I wouldn't believe
that that overruled me.  At worst, I'd believe that it meant there was
dispute.  And the only way for us to resolve this kind of dispute is to
just plain decide something on the basis of what suits our needs.
So I really do think it comes down to those non-technical criteria I
outlined before.