6.1369, Sum: Creeping reflexives

[The Linguist List]

---------------------------------------------------------------------------
LINGUIST List:  Vol-6-1369. Fri Oct 6 1995. ISSN: 1068-4875. Lines:  439
 
Subject: 6.1369, Sum: Creeping reflexives
 
Moderators: Anthony Rodrigues Aristar: Texas A&M U. <aristar at tam2000.tamu.edu>
            Helen Dry: Eastern Michigan U. <hdry at emunix.emich.edu>
 
Associate Editor:  Ljuba Veselinova <lveselin at emunix.emich.edu>
Assistant Editors: Ron Reck <rreck at emunix.emich.edu>
                   Ann Dizdar <dizdar at tam2000.tamu.edu>
                   Annemarie Valdez <avaldez at emunix.emich.edu>
 
Software development: John H. Remmers <remmers at emunix.emich.edu>
 
Editor for this issue: dseely at emunix.emich.edu (T. Daniel Seely)
 
---------------------------------Directory-----------------------------------
1)
Date:  Fri, 06 Oct 1995 18:04:32 +0700
From:  t-markdr at microsoft.com (Mark Dras)
Subject:  Summary: creeping reflexives
 
---------------------------------Messages------------------------------------
1)
Date:  Fri, 06 Oct 1995 18:04:32 +0700
From:  t-markdr at microsoft.com (Mark Dras)
Subject:  Summary: creeping reflexives
 
 
This topic has generated a lot more interest than I expected, ranging
from "I'm surprised that you find this usage unusual" (a few
respondents) to "I consider it an abomination which should be stamped
out" (although, being Real Linguists, the prescriptivist tendencies
were generally more subtle, or only reluctantly admitted to).  So
apologies to those I haven't directly responded to by email.  Respondents were:
 
Steve Matthews		 Larry Trask
Sean Jensen		 Peter Blok
Ivan Uemlianin		 Kersti Bo"rjars
Francis Bond		 Matthew Whelpton
Robert Millar		 Bora
Dorothy Disterheft	 Stavros Macrakis
Catherine Rudin		 Steve Seegmiller
Joseph Davis		 Larry Rosenwald
Markell R West		 Jack Wiedrick
Dara Connolly		 Feride Erku
Michael McIlrath	 Heidi Quinn
Jeffrey L Lidz		 Claudia Brugman
Anthea F Gupta		 Michel Platt
Peter Daniels		 David Denison
Scott Herron		 Steven Blackwelder
Anthea Fallen-Bailey	 Cathy Ball
Hyeran Lee		 Laurence Kitching
Glenn Bingham		 Nobue Mori
Trey Jones		 Nancy Frishberg (at home)
 
The following summary is, of necessity, short - it isn't counted as
real work by my supervisor (Robert - if you're reading this, hi!) - but
it contains more references at the end if you're interested in
exploring the topic further; I only summarise those papers I've read,
and the speculations of Linguist respondents.
 
DEFINITION
 
It was pointed out by a couple of respondents that the name reflexive
is misleading; there's a distinction between its use for pronouns in a
purely morphological sense, and its use to denote an action where the
recipient is the agent.  I'll follow Baker (1995) in using it in the
former sense, so whenever I talk about reflexives I mean -self words,
whether or not they're being used reflexively.  I'll somewhat
inaccurately call the usage of reflexives described in my original post
Unbound Reflexives (URs).
 
The following sections describe how UR usage is seen now: the
Prescriptive View, Actual Usage, Proposed Theories, and Linguist
Speculations; these are followed by a section on Trends, and then Wild
and Groundless Speculation.
 
PRESCRIPTIVE VIEW
 
The use of reflexives in sentences like
 
	Paul and myself wrote the report.
	Gavin saw Paul and myself this morning.
 
is generally regarded as wrong by "How to Write" books (e.g. Harbrace
College Handbook, which says "Formal English does not accept 'myself'
as a substitute for 'I' or 'me'", giving the example "John and I [not
myself] work at the gas station").
 
Quirk et al (1985) comment that the use of reflexive pronouns in these
contexts "is felt by many to be a hyperurbanism, a genteel evasion of
the normal personal pronoun" (p359).  However, in some situations,
using a reflexive instead of the ordinary personal pronoun is
acceptable; Quirk lists 'as far as', 'like', and similar constructions.
 
	For someone like {me/myself}, this is a big surprise.
 
Picture sentences are an often-quotes example of this:
 
	They thought that pictures of {them/themselves}
		would be on sale.
 
[Quirk believes the use of the reflexive to be obligatory here,
although all other sources I've read describe it as optional.]
 
Anecdotally, it appears that many technical writing teachers are
teaching their students some version of this rule discouraging reflexives.
 
ACTUAL USAGE
 
Despite this view of the usage I described being a bad thing, there are
many attested examples of its acceptability.
 
One respondent, Cathy Ball, notes that URs were acceptable in Middle
English, a fact which is discussed in Mustanoja (1960).  She gives as
an example from Malory's Morte d'Arthur
 
	Now, sayd Syr Gawayne, it is yourself I loue so wel ...
 
Later usage is also attested in Baker (1995) and Zribi-Hertz (1989),
who quotes authors from Jane Austen to Margaret Atwood; an example,
from Edith Wharton, is
 
	Miss Stepney's heart was a precise register of facts as
	manifested in their relation to herself.
 
(Many) Irish and Scottish speakers of English consider URs totally
acceptable, even with 3rd person pronouns (although almost all other
respondents found this 3rd person usage impossible or very marked).
 
	Himself is coming down the stairs.
 
Apparently, this is an intensive form of the pronoun, used to emphasise
the referent of the pronoun, particularly when the referent is a person
of high status and/or importance.
 
PROPOSED THEORIES ABOUT URs
 
The URs I've described cause problems for the Binding theory, as well
as being unacceptable to prescriptivists; there have been a number of
explanations about how they fit in.
 
[I confess, at this point, that I'm not a linguist, and know very
little about GB; so I won't try to gloss particularly linguistic papers.]
 
Ross (1970) proposed that all declarative sentences are embedded within
performative ones of the form "I tell you that ...", using URs as one
of his arguments.  So
 
	This paper was written by Ann and myself.
 
is OK because the deep structure is really
 
	I tell you that this paper was written by Ann and myself.
 
This accords with the intuition that 1st and 2nd person URs without
explicit antecedents are more acceptable than 3rd person URs, and also
perhaps why 1st person is better than 2nd.
 
Kuno (1987) and Zribi-Hertz (1989) suggest that the use of URs is a
discourse phenomenon indicating a Subject of Consciousness; these are
called logophoric occurrences.  Compare the two sentences (taken from
Zribi-Hertz)
 
	The women were standing in the background, with the
	children behind them.
 
	The women were standing in the background, with the
	children behind themselves.
 
Then imagine that these sentences describe a photograph where the
women's backs are towards the camera.  The first sentence more
naturally suits a situation where the children are more distant from
the camera than the women; this is reversed for the second.  This is
caused, it is suggested, by 'themselves' reflecting the viewpoint of
the women, whereas 'them' reflecting an observer's viewpoint.
 
[I tried this on a couple of unsuspecting non-linguists, asking them
what positions they thought camera, women and children occupied
relative to each other; it accorded with the intuition expressed.]
 
A similar example (also from Z-H) suggesting perspective is
 
	Zapp sat down at the desk and opened the drawers.  In
	the top right-hand one was an envelope addressed to
	{him/himself}.
 
Reinhart and Reuland (1993) contend that the phenomena are, in fact,
generally syntactic in nature, and revise Conditions A and B of the
Binding theory; however, they do agree that, when the option of UR use
exists, the choice can be influenced by discourse considerations.
 
Baker (1995) describes the use of URs as an intensification phenomenon,
along the lines of those noted earlier by Ross (1970) and Quirk (1985).
 They suggest that the known emphatic use of the reflexive, as in
 
	Anyone but you yourself would have noticed the change.
 
has, in some cases, a deletable personal pronoun, giving the semi-emphatic
 
	Anyone but yourself would have noticed the change.
 
Baker also note that URs are used for contrastive purposes.
 
	Unless you believe that you have a reasonable chance of
	CONVICTING John (as opposed to merely HARASSING
	him), you shouldn't waste time deposing the board members.
 
	Unless you believe that you have a reasonable chance of
	convicting JOHN HIMSELF (as well as John's subordinates),
	you shouldn't waste time deposing the board members.
 
	?? Unless you believe that you have a reasonable chance
	of CONVICTING John himself (as opposed to merely
	HARASSING him), you shouldn't waste time deposing
	the board members.
 
The contrast idea also occurs where emphasis is needed to contradict an
implicit assumption: you carry a book with you, but you buy a present
for yourself.
 
[Thanks to Heidi Quinn, Jeff Lidz, Claudia Brugman and Nobue Mori for
these references.]
 
 
LINGUIST SPECULATION
 
Many respondents noticed that URs were more acceptable as part of
compound NPs, or where they were part of a complex syntactic structure.
 Ross (1970) noted this also, giving two different levels of acceptability to
 
	This paper was written by Ann and myself.
	?? This paper was written by myself.
 
It was suggested that the occurrence of 'myself' is a hypercorrection
phenomenon: there is a hesitancy about which case to choose ("How many
times do I have to tell you: 'Tim and I met Ian Roberts', but 'Ian
Roberts met Tim and me'!").  The use of 'me' sounds coarse, but the 'X
and I' as object has been branded incorrect.  (There are similar
occurrences of hypercorrectness in other languages: Robert Millar noted
that in Norwegian there is confusion over the use of 'de' (a polite
pronoun) and 'du' (the common one), which leads to unusual usage of
'de'.)  This would accord with the respondents' views that the 1st
person URs are much more acceptable than 2nd person ones, where there's
no confusion with case.
 
Peter Blok notes that similar binding violations occur in Dutch,
depending on whether the NP is compound:
 
	De generaal(i) schiep een ontsnappingsweg voor zichzelf(i)
	'The general made  a    escape route     for himself '
 
	* De generaal(i) schiep een ontsnappingsweg voor hem(i)
	* The general(i) made   an escape route    for  him(i)
 
	De generaal(i) schiep een ontsnappingsweg
		voor hem(i) en zijn(i) manschappen
	The general made an escape route
		for him and his men
 
	* De generaal(i) schiep een ontsnappingsweg
		voor zijn(i) manschappen en hem(i)
	*The general(i) made an escape route
		for his(i) men and him(i)
 
Another common remark is that URs are used to indicate politeness
and/or confer status on the referent.  Stan Dubinsky (of, I think,
South Carolina Uni - I haven't actually heard from him about this) has
apparently drawn a parallel between Japanese honorifics and 2nd person
URs; the conferring of status explanation would accord with the
Irish/Scottish usage mentioned earlier.
 
Most of the respondents who actually used URs themselves appeared to
come from business fields; other noted that that was where they found
the usage most common.  Jack Wiedrick noted that it crops up in
waiters' speech:
 
	Will there be just yourself for this evening?
 
	As far as myself is concerned, I would recommend the marlin.
 
Kersti Bo"rjars (and others) noted its use by salespeople and bank
managers, with examples like:
 
	If yourselves had to give up work for some reason,
	how would the mortgage get paid?
 
	In the unfortunate event that either of yourselves
	were to die....
 
In these cases, again, it was seen to be a politeness feature (although
to some people, it appeared to be more ingratiating than polite) - the
example immediately above also contains a subjunctive, another marker
of politeness or formality (unless the speaker wasn't following the
prescriptive rule of singular verb with an 'either' construct); the
usage confers status on the boss / customer / whomever.
 
This seems to tie in reasonably well with the logophoric,
centre-of-discourse ideas described above: by focussing the discourse
on a particular entity (e.g. your boss) you are conveying respect.  By
saying "Paul and myself", you are putting more emphasis on yourself.
The irritation felt by some respondents could be due to a feeling that
this emphasis is inappropriate - it is an attempt to be ingratiating
(for 2nd person URs), or that this emphasis is unnecessary when
attempting to be objective (for the 1st person usage in business reports).
 
That this politeness can be taken as pomposity is used by comedians:
apparently there is a Monty Python skit where Michael Palin and/or
Terry Jones use URs when imitating pompous policemen; and an informant
of one respondent remembers W.C. Fields using URs when he was "acting pompous".
 
Korean, however, is slightly different.  It has the reflexive 'caki'
(according to Hyeran Lee); standard reflexive usage would be
 
	John-i caki-lul miwehayssta
	John-NOM self-ACC hated
	John hated himself
 
But in the 2nd person, this is used as a UR informally, in more
intimate situations, particularly by young people.  Instead of
indicating distance or politeness, it indicates familiarity.  This UR
usage occurs, for example, in both subject and object cases:
 
	na-n caki-lul saranghay
	I-TOP self-ACC love
	I love you(rself)
 
	caki-ka na-eykey malhaysse
	self-NOM I-DAT told
	You(rself) told me
 
This fits to some extent with the idea of a UR as an emphasis or
discourse focus - however, in Korean, the emphasis is to convey
familiarity, not politeness.
 
 
TRENDS
 
This was what I meant by my subject heading "creeping reflexives":
there seemed to me to be a change in usage going on, and I was
wondering if there was an explanation for it.
 
>>From the historical evidence mentioned above, it seems that URs were
considered acceptable at some stages of English history.  David Denison
(who is writing a section on these for "The Cambridge History of the
English Language", to be published in 1996) also notes that there was
no adverse comment on the use of URs by 18th century grammarians.
 
However, there is a record from 1929 of the dismissal of UR usage as
"mere shopkeepers' Cant".  Also as mentioned above, it is now
considered incorrect by many grammarians and writing advice books.
 
Many respondents agreed with my feeling that the usage seems to be
increasing - Baker (1995) notes that it is more common in Britain - and
there were additional comments that it was usually younger people who
used URs in that way.
 
So it seems that the usage was OK for a long time, then suffered a
decline in acceptability, but is now attempting to make a comeback.
 
 
WILD AND GROUNDLESS SPECULATION
 
These are just some thoughts about reasons for the trend.
 
Maybe as grammar has become less prescriptively enforced (e.g. in
Australia, prescriptive grammar wasn't taught for about 23 years after
my birth), and with only a rearguard action being fought by some
how-to-write books, people are re-adopting usages which better fit
non-syntactic purposes.
 
Politeness is one of these, and URs are being used for politeness with
fewer (and different) syntactic constraints.  Rhythm is another: "The
report was written by Paul and me" sounds a little blunt and
"end-light" (i.e. there's not enough emphasis at the end of the
construction), so in the interest of balance, the pronoun is changed to
'myself'.  I wondered about this after reading about Hawkins's (1994)
ideas on end-weight; he makes a quantitative analysis of why it sounds
better to have the heavier element at the end, e.g.
 
	I introduced to Mary some friends that John had brought to the party.
 
is better than
 
	I introduced some friends that John had brought to the party to Mary.
 
This change in balance between prescriptive constraints and
non-syntactic considerations would also fit with the "recorrection" I
mentioned in my original posting, where my partner's correction from
'myself' to 'I' was changed back to 'myself' by another person who
thought 'I' was wrong (and this "recorrection was also mentioned by 2
other respondents): there are no longer strong prescriptive rules
against the usage, so the polite / balanced form is preferred (by some).
 
It could also be related to a desire for formality and distance (also
noted by Scott Herron), in perhaps the same way that the passive
construction is used for distance.
 
	I negotiated the deal.
	The deal was negotiated by me.
	The deal was negotiated by myself.
 
'Myself', while being more emphatic or discourse-focussed, seems less
central to the essence of the speaker than 'me' (almost as if 'myself'
were a separate entity, like 'my dog' or 'my teddy bear').
 
So, who knows?  I've enjoyed this ramble; thanks to those who helped.
 
 
REFERENCES
 
Baker, C.L. (1995).  Contrast, prominence, and intensification, with
special reference to locally free reflexives in British English.
Language 71(1), 63-101.
 
Kuno, Susumu (1987).  Functional syntax: Anaphora, discourse and
empathy.  Chicago: Uni of Chicago Press.
 
Mustanoja (1960).  A Middle English Syntax.
 
Quirk,  Randolph, Sidney Greenbaum, Geoffrey Leech and Jan Svartnik
(1985).  A Comprehensive Grammar of English.  Longman Group Limited.
London, UK.
 
Reinhart, Tanya and Eric Reuland (1993).  Reflexivity.  Linguistic
Enquiry 18, 445-479.
 
Ross, John Robert (1970).  Declarative sentences.  In Readings in
Transformational Grammar, ed. by Roderick Jacobs and Peter Rosenbaum,
222-272.  Waltham, Ma: Ginn.
 
Zribi-Hertz, Anne (1989).  Anaphor binding and narrative point-of-view:
English reflexive pronouns in sentence and discourse.  Language 65, 695-727.
 
 
------------------------------------------------------------------------
LINGUIST List: Vol-6-1369.