Pronouns and mind reading

[Jeff Barnett]

I have noticed that some people make quite specific mistakes in their
use of pronouns (and other constructs to reference items mentioned in a
dialogue). That type of mistake is an introduction of a particular sort
of ambiguity: the reference seems to be to an item just introduced by
the dialogue or forward to an item in the same utterance; however, the
reference is meant to be resolved in a conversation that was 5 minutes
ago, 5 days ago, etc. It seems that something pops into their mind and
they use language as if you had a simultaneous pop. Since that's
unlikely, the only mechanism left to understand what is meant is mind
reading - a skill I lack. I have noticed also that some people almost
never make this sort of mistake

I presume their has been some study of this and some conjectures. I'd be
interested in knowing what's going on and I have a few questions:

1. Is there a sexual "dimorphism" in who makes this class of errors?

2. Is the frequency of this class of errors about the same for speakers
of "modern" and ancient languages? Does frequency depend on language family?

3. Do native and non-native speakers have similar statistics?

4. Etc?

Thanks in advance for any conversation. Also please forgive my lack of
knowledge that has lead to probable misuse of terms about language.
--
Jeff Barnett

[Peter T. Daniels]

This doesn't seem to be a question about language (the pronouns
have antecedents, and all that), but about psychology. The people
think that who they're talking to consider the topic is as important
as they do.

[Jeff Barnett]

I agree that this phenomenon is psychological but so is a lot of
linguistic theory. For example, a speaker's choice between "the" and "a"
involves consideration of what is thought about the state of knowledge
or mind of the listener. (I think I've noticed a shift in this pattern
in the last few decades to where "the" is used more than in the past.)
The use of pronouns involves similar mechanisms.

I'm also assuming that this might be a sign of a lack of empathy. Here I
using "empathy" to mean an effort, either conscious or unconscious to
understand another's state of mind. I am distinguish empathy and
sympathy here.

Part of my reason for seeking other's views is that I have a
contradiction in my observations: on one hand, I believe that females
are more empathetic; on the other, I believe females are more likely to
make the error I described.
--
Jeff Barnett

[Peter T. Daniels]

I say, it's not similar. The choice of "a" vs. "the" is determined by the
syntax of the sentence (or the discourse) and the "context of situation."

Whether one is considerate of one's interlocutor isn't.

One might be assuming that the other recalls as much of an earlier
situation as oneself does.

> I'm also assuming that this might be a sign of a lack of empathy. Here I
> using "empathy" to mean an effort, either conscious or unconscious to
> understand another's state of mind. I am distinguish empathy and
> sympathy here.

Again, not a language thing.

> Part of my reason for seeking other's views is that I have a
> contradiction in my observations: on one hand, I believe that females
> are more empathetic; on the other, I believe females are more likely to
> make the error I described.

As if all "females" share something other than a specific anatomy.

[Jeff Barnett]

It seems that you and I differ in a basic premise: T believe that any
theory of language must consider not only context but the interlocutor's
assumptions about the other participants state of mind or the intended
context. I will give three examples, one new, and the other two
elaborations of above discussion.
----------------
Case I - a vs the

Case I.A John and I have often talked about a green car we've seen in
the neighborhood. He approaches me today and asks "Why the police and
all the excitement?" I respond "The green car was stolen!"

Case I.B John and I have never talked about a green car in the
neighborhood. He approaches me today and asks "Why the police and all
the excitement?" I respond "A green car was stolen!"

Case I.C I don't recall if John and I ever talked about a green car in
the neighborhood. He approaches me today and asks "Why the police and
all the excitement?" I respond "There's a green car kept around here and
it's been stolen!"

The difference in my response surely rests on what I believe John knows
about green cars in the neighbor. In fact. if I interchange the
responses to John's question, I claim that in a serious sense I'm not
speaking properly.
----------------
Case II - pronouns that skip current target and refer back in time

A few weeks ago John and I talk briefly about an acrobat who did well in
a gymnastics competition. Today I say "George stood on a chair and
changed a light bulb he could barely reach." John replies "He should
have stood on his tip toes; that would make it easier." Note that George
has balance problem. When John replied "he" referred to the acrobat but
there was no context to make that intelligible to me. John's statement
is nonsense unless he does something to refresh my memory and point me
to the proper referent.
----------------------------
Case III - a speech act example (though I'm told that's the wrong term)

Example III.A A man carrying a suit case runs up to the ticket counter
in a train station and asks the agent "What track does the Chicago train
leave from?" the agent answers "Track 122 in 7 minutes." This answer
makes no sense unless we assume that the agent believes the questioner
is trying to make the train and probably needs to know how much time he has.

Example III.B A man carrying a suit case runs up to the ticket counter
in a train station and asks the agent "What time does the Chicago train
leave?" the agent answers "In 7 minutes from track 122." This answer
makes no sense unless we assume that the agent believes the questioner
is trying to make the train and probably needs to know from where as
well as how much time he has.
---------------------------

In all these cases and examples, reasonable communication requires that
participants at least guess at what the other might know or believe as
well as where there attention is focused at the time.

The questions in my original message where trying to find out if the
degree of assumed "mind reading" depended on particular languages,
culture, or sex.
--
Jeff Barnett

[Ross Clark]

OK, now I can see what you are talking about. "Reference tracking" is a
term used by some linguists for this general phenomenon.

You are concerned with cases where the "mind reading" is less than what
you would take to be the norm. I recognized this in cases where I'm
reading a newspaper or magazine article and an individual is mentioned
(typically by name) who has been named and identified 17 paragraphs
earlier, but not mentioned since; it is annoying to have to search back
for this information. Sometimes it may result from careless editing.
I even have trouble with novels -- I think all novels should come with a
character index or dramatis personae table.
Or my wife, out of the blue, will say something for which I cannot find
a referential anchor. She has, I think, several detailed discourse
histories relating to different people, problems, etc., in her mind
which she is (silently) working on all the time. When she chooses to
involve me in one of these, she does not always provide enough
information to orient me to the particular discourse, something we may
have talked about yesterday, or last week.

This is not my area of linguistics, but I suspect that there are few if
any studies looking at differences between genders or cultures in the
incidence of this kind of problem. One difficulty is that in order to do
such a study you would have to collect a reasonable sample of recorded
real-life examples of it, and that is a much more difficult matter than,
say, collecting a few hundred examples of the pronunciation of a common
word. Still, this might be within the range of conversational analysis
(cf. Deborah Tannen's work on differing styles of conversation), and
perhaps somebody has addressed the matter.

Finally, let me point out that the "assuming" in your last sentence
above is in effect a kind of meta-"mind-reading".