tersest proscription of the week (was PSAT Glitch)

[Arnold Zwicky]

but wait!  i have a new entry in the terseness competition.  this
is from X. J. Kennedy & Dorothy M. Kennedy, The Bedford Guide for
College Writers, 2nd ed.  Boston: Bedford Books of St. Martin's
Press, 1990.  on p. 639, they give the rule, and this is the whole
thing:

  Watch out for possessive nouns.  They won't work as antecedents.

that wins over the Barron's 1001 Pitfalls proscription, hands down, on
number of words.  *however*, context supplies a bit more content.  to
start with, the proscription comes in a section headed (oh dear):

  Be sure your antecedents are named, not implied.

and then there's a comment on the example (so that maybe the Bedford
Guide exceeds Barron's 1001 Pitfalls, after all):

  VAGUE  On William's canoe _he_ painted a skull and bones.
   (For all we know, _he_ might be some joker named Fred.)

  CLEAR  On his canoe William painted a skull and bones.

note that this is yet another example with the possessive antecedent
inside an adverbial.

note also that the explanation for why the first canoe example is
VAGUE is strikingly lame.  after all, "his" in the CLEAR rewriting
could equally refer to some joker named fred; that's the way of
pronouns.

there is one, and only one, other example illustrating the rule that
antecedents must be "named, not implied":

  VAGUE  Ted wanted a Norwegian canoe because he'd heard that _they_
    produce the lightest canoes afloat.

i agree that this is a terrible english sentence, but it's an awfully
subtle example (involving phonological identity of an adjective and a
noun), and the actual claim, that "the antecedent of a pronoun has to
be a noun or pronoun" is not quite right, since it treats "Norway"
and "Norwegian" differently in examples like

  Ted is making a Norway/Norwegian stop on this trip,
    just because it's in between Scotland and Finland.

    (or: ...because he heard summers are beautiful there.)

these strike me as roughly equivalent in acceptability, and as
much more acceptable than the canoe example.

what's odd here is that this manual gives no meat-and-potatoes
examples of sentences that are bad because the antecedent of a pronoun
is in fact only implied.  lots of handbooks do, and most of the
handbooks give *only* such examples (and not cases of possessive
antecedents or anaphoric-island-like violations).

so here's John C. Hodges & Mary E. Whitten, Harbrace College Handbook,
8th ed. (NY: Harcourt Brace Jovanovich, 1977), p. 272, cautioning you,
"As a rule, do not refer to a word or an idea not expressed but merely
implied" ("as a rule" here means 'for the most part, usually'), and
providing the (truly wretched) example:

  He wanted his teachers to think he was above average, as he could
    have been if he had used it [= his ability] to advantage.

this handbook does address the possessive antecedent issue, but in a
different section (pp. 270-1), having to do with "remote or obscure
reference".  we are told not to place a pronoun "too far from its
antecedent" and also to "avoid an obscure reference to an antecedent
in the possessive case."  (alas, it's not entirely clear, to me at any
rate, whether "obscure" is to be understood here as a restrictive or
non-restrictive modifier.  my first reading was that it was
restrictive, so that it's only obscure reference to a possessive
antecedent that we're to avoid, and that strikes me as pretty good
advice.  if "obscure" is non-restrictive, then we're back in
absolute-prohibition territory.)  the one example the handbook gives
has, yes, once again, a possessive antecedent inside an adverbial:

  When Johnson's club was organized, _he_ asked Goldsmith to
    become a member.

(this would of course be fine if johnson was already salient in the
preceding discourse.  or even in some situations where johnson hasn't
even been mentioned before:  "Seven students were chosen to head
clubs.  When Johnson's club was organized, he asked Goldsmith to
become a member.")

many handbooks say quite sensible things about (straightforward cases
of) implied reference, as in this comment from George S. Wykoff &
Harry Shaw, The Harper Handbook of College Composition, 2nd ed/ (NY:
Harper & Brothers, 1957), p. 522:

  Frequent occurrence of implied reference is found in the work
  of many reputable writers, and when there is no possibility of
  confusion, the use is effective, but remember that there _should_
  be no possibility of confusion.

arnold (zwicky at csli.stanford.edu)