Craig,

There's no question that words like "that" have cognitive force.  We wouldn't use them otherwise.  Bolinger has a wonderful little book titled That's That in which he investigates the meaning of "that" with content clauses, looking both at cases where it's deletable and those where it's not.  In a carefully constructed argument he argues that "that" carries a sense of definiteness, not surprising given its historical relationship to the determiner system.  While he didn't look at relative clauses in that study, I'd argue that "that" carries pretty much the same sense there, which would align with the fact that it doesn't generally occur in non-restrictive relatives.

When I look for evidence of category membership, I have to look to morphosyntax.  I can't go on feeling because there's no way to test such a thing.  That does not mean we can't have fuzzy categories.  There's solid evidence that language uses such things.  The only evidence you've provided that relative "that" is a pronoun is a strong feeling that that is the case. In your excellent example

"The picture, you might be interested to know, that is hanging on the wall was painted by your cousin."

Illustrates a way in which hunches can lead us to interpret one thing as another.  "that," now that we've introduced the notion clitic, is a clitic.  It is, in fact, enclitic, that is, it attaches to the preceding stressed unit, and so normally it would cliticize to "picture."  However, since there is an interruption after "picture," it can't cliticize and so it takes slight stress; it's not completely unstressed as it would be if it could cliticize to "picture."  This slight stress gives it an unusual prominence, which, in terms of cognitive grammar, I would say informs us that was follows is subordinate.  It's not filling the place of subject, though.  We can have relative clauses without either a subject or "that," as sentences like "The guy 0 met me at the airport drove me home."  Granted these don't occur in formal written English, but they're not uncommon in speech and I wouldn't be surprised to find them in a corpus of informal letter writing.  I added the 0 to the sentence just as a convenience, because our first reaction would be to interpret "met" as the main verb, an interpretation that runs into trouble when we get to "drove."  "That," if used here, would signal clearly that we are to begin parsing another clause before we get the rest of the main clause.  The fact that both "that" and the subject are deleted, and at the same time, makes it pretty hard either to claim that "that" is the subject or that it's pronominal, since, if it's not the subject there's no other nominal role for it in the sentence.

I agree with you that relative "that" carries a cognitive load; I disagree that it's pronominal in any way.

Herb

-----Original Message-----
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Craig Hancock
Sent: 2009-01-21 13:44
To: [log in to unmask]
Subject: Re: Pedants that or who?

Herb,
   I hope we are not boring people with this. I would be happy to make
it a private talk if people on the list believe it is going on too long.
    The fact that "that" is a subordinating marker has a cognitive force
to it, changing or contributing to (building) the meaning. And when
"that" stands in pronominally, we understand that it is carrying over a
meaning we hope is recoverable from context. Those don't seem to be
naive observations. What it means and how it functions are dynamically
linked. I think our internal intuitions about that are deeply relevant.
Meaning happens in us, not in the forms. These meanings may float below
the radar, but they are nevertheless meaningful. You can make the case
that grammatical meanings are the most meaningful of all, since they are
so central to our cognitive processing. It may very well be that a
primarily formal approach to language distorts what it is trying to
describe, creating the impression that content is cast into form rather
than built through it. As you know, linguists are not in agreement over
these questions.
   But what it comes down to, I think, is that we are wondering whether
to classify "that" as a pronoun, and that has everything in the world to
do with how we construct the category. You have yet to tell us how you
define pronoun. That it doesn't have gender forms or singular/plural
forms are part of it, but certainly not all. "Who" is singular and
plural. "Which" has no gender forms or plural forms, though it tends to
be inanimate. So we may have central prototypes for the category (like
the personal pronouns), but other words that seem to fit the category
very marginally. This is, as you know, very common.
  Pronouns tend to fill a grammatical slot (like subject), but stand in
for a meaning established elsewhere. That's a functional definition
rather than a formal one, but I don't think it is unscientific.
   Let me give one more example of something that feels pronominal to
me.  "The picture, you might be interested to know, that is hanging on
the wall was painted by your cousin."  Here, the "that"  feels to me to
stand in for "the picture" in the same way "it" would in "It is hanging
on the wall."  And, quite frankly, in the same way "which" would if the
clause were non-restrictive. "That picture, you may be interested to
know, which hangs so nicely on the wall, was painted by your cousin." If
someone asks, what does "that" refer to, I think we would answer "the
picture" without batting an eye. I don't think anyone would say "It
doesn't refer to anything. It's just an empty marker." We could also say
it functions as relative clause subject and does so because relative
clauses seem to require an explicit subject. Otherwise, you would be
forced to say that it stands in as an empty marker to let us know no
real subject is there. I'm guessing at that result, but it could be
formally tested. I don't think anyone would be surprised by the question.
   Can we describe relative "that" as a word that seems very
pronoun-like in some of its manifestations? Can we say that it's a word
that seems to many people to be acting like a pronoun, even though it is
just a place holder? If so, then what is it about the word that fools us
in that way? Again, it seems to me that if we agree about what we are
observing, then disagreeing about categories is far less important.
   Sometimes, classifying just brings us away from the richness of the
language by oversimplifying. But it is also possible to say that it is a
unique pronoun, sharing enough with other pronouns to make us put it
reluctantly in that category, but differing in substantial ways.
   If it's not a pronoun, doesn't it have to be in a category of its
own? The "that" in content clauses subordinates, but doesn't have a
placeholder function and never seems to refer back to a previous
meaning, even in an illusory way.
   "I believe that the aliens have landed." If you asked "What does
"that" refer to," it would probably be what follows, not what precedes.
It has no role, like a pronoun would, within the clause.
    I hope that at least makes some sense.

Craig


STAHLKE, HERBERT F wrote:
> Craig,
>
> I'm surprised at your surprise.  I did not suggest that language reduces to form.  Rather there is an important part of language that is form, and this is particularly true of what we call "function words," as opposed to "content words."  Function words tend to be pretty bleached of content and rather have grammatical function.  "That" in relative clauses is such a word, and it has to be analyzed formally.  If it also has pronominal function, then I would expect to find evidence of that, and so far there is none.  If it were to develop a gender or number contrast, if, for example, we began using "these" if the head noun is plural, then there would be evidence that it's a pronoun.  It may become a pronoun, but if it does, the shift will leave structural traces.  I say that with confidence because that's how language change works.  The problem with relying on what we believe "that" to mean in a relative clause is that so far we've found no way to test that belief.  That's different from deciding that "disinterested" also means "uninsterested" or that "infer" means "imply."  For many speakers these meanings have changed, and we can see that in their usage.  We haven't seen anything comparable in the usage of "that" in relative clauses.
>
> Herb
>
> Herbert F. W. Stahlke, Ph.D.
> Emeritus Professor of English
> Ball State University
> Muncie, IN  47306
> [log in to unmask]
> ________________________________________
> From: Assembly for the Teaching of English Grammar [[log in to unmask]] On Behalf Of Craig Hancock [[log in to unmask]]
> Sent: January 21, 2009 9:19 AM
> To: [log in to unmask]
> Subject: Re: Pedants that or who?
>
> Herb,
>    I am deeply surprised by this. Do you mean to imply that language can be reduced to form? What a word means has nothing to do with what we believe it means? I tend to look at grammar as  form/meaning pairing, and cutting out the meaning side is highly problematic. Am I misreading you on this one?
>   I try to be honest with my students as much as I can. Part of that means telling them that my views are sometimes idiosyncratic, at other times in harmony with the mainstream. If views differ or some aspect of grammar is deliberately simplified, I try to make that clear.
>
> Craig
>
> STAHLKE, HERBERT F wrote:
> Craig,
>
> What do we base science on if not formal observation?  That's how it works.  Now what we choose to teach and how we choose to teach it is another matter, that's pedagogy, as we all know, more art than science.  If it helps to call "that" a relative pronoun in teaching grammar to your writing students, then by all means do so.  But recognize at the same time that this is a useful lie, just like telling high school physics students that f=ma, or high school music student that a seventh chord has to resolve.  If they get to the point where they understand linguistic methods and how to use them, they'll have forgiven you for simplifying earlier.
>
> Herb
>
> From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Craig Hancock
> Sent: 2009-01-20 11:23
> To: [log in to unmask]<mailto:[log in to unmask]>
> Subject: Re: Pedants that or who?
>
> Herb,
>    I'm a little uncomfortable with the notion that we have to base our "science" totally on formal observations and that observations about what something seems to mean are "naive." That may be the writing teacher speaking. It may also be from a growing interest in cognitive lenses. But I'm not ready for a full defense yet.
>    I guess what I'm holding out for, not so much from you but from my fellow public grammarians,  is a much clearer differentiation between "that" in a content clause and "that" in a relative clause. You and I seem close on that, but then you draw back from saying there's a possibility of a third "that" and I'm still ruminating.
>    I don't mean to replace your thoughtful phrasing for all this, but here's what I would tell my grammar class students. In a relative clause, the nominal group being "modified" has a role within the modifying clause. The marker (relative pronoun?) helps us establish that role. This doesn't happen in a content clause, where the subordinator (I call it a complementizer) simply stands outside the clause. This complementizing even happens in some clauses that seem relative by position, but turn out to be structurally different.
>    "He believes that the aliens have landed."  "His belief that the aliens have landed is absurd." In neither of these is "belief" or a pronoun stand-in functioning within the subordinate clause. to me, these are complements rather than modifiers.
>    "The aliens that have landed are green." In this case, aliens have done the landing and it's these specific aliens that are green. This is what feels pronominal to many of us, but could be explained thoughtfully as a subordinating stand-in for an absent subject.
>    I think we are both in full agreement to this point.  I'm still leaning toward calling the clause itself a "relative clause" even if it turns out "that" isn't acting pronominally in the full sense of the word. For pedagogical reasons, certainly, that makes the most sense. For pedagogical reasons, it might also be less confusing to say that the "relative pronouns" that help us out in these clauses all act a bit differently. (I would include "where" and "when". "The place where I was born." "The time when I'm most alert.")
>    Certainly the fact that we can say "the dresser in which I keep my socks" but not "the dresser in that I keep my socks" gives me deep pause.
>    I have to admit that the more we discuss this, the clearer your position becomes. Could it be that "that", because it is both subordinator and pronoun in other instances, can be acting in ways here that are very unique?
>   If it feels like a subject, can it become one over time?
>   Ultimately, it is more useful to agree on how something acts than it is to agree on classification, since classification categories can change with a change in definition. I think I'm arguing for a more fluid definition for pronoun, in part because it still feels to me that the "that" that shows up in these clauses is different from the other two, the subordinator and the demonstrative.
>    Again, though, I thank you for leading us patiently toward the light. That I agree with you more and more as time goes on should tell me something.
>
> Craig
>
>
>
> STAHLKE, HERBERT F wrote:
>
> Craig,
>
>
>
> My point was that while we do still use "that" after "except," "now," and a few other adverbs it used to have a much broader distribution.  It was used regularly with another word (since, if, when, while, which, etc.) to form show subordination, and the use of "that" in such cases was quite consistent.  In LME, on the other hand, that use in combination with a content word to mark subordination has become restricted to just a few holdouts.  As to your other "now that" sequence, your example has "that" as a demonstrative, and it is clearly pronominal.  I don't know of any grammar that identifies relative that with demonstrative that.  (I also don't find the comma necessary, but that's another matter.)  You're right, of course, that the "that" in these modern cases does mark a subordinate, non-relative clause, and, consequently, the clause will contain no nominal gaps, since those occur only in relatives (to avoid undue redundancy).
>
>
>
> As to your feeling that the arguments for "that" as a subordinator don't consider the possibility of a third "that," it's worse than that.  Those arguments explicitly reject that possibility.  On morphosyntactic grounds, there is no evidence that relative "that" is in any way pronominal.
>
>
>
> Your argument that "the music that moves her" and "the people who move her" are parallel is based solely on intuition, not on evidence.  Intuition is, of course, a double-edged term in grammar.  Within a large body of linguistic literature the term is carefully and narrowly defined to mean native speaker judgments of well-formedness of an utterance.  You may or may not accept that definition, but it is at least reasonable rigorous.  In your usage below, "intuition" is more like "gut feeling," something on the basis of which I might choose whom to hire out of several otherwise equally qualified applicants, but we can't base science on gut feeling.  We can create hypotheses in part that way, but we can't test them that way.  That sort of intuition is not evidence.
>
>
>
> I agree, however, that the two are parallel, and they are because reference can be represented by a pronoun or by zero anaphora, and in that-relatives we have the latter.  The gap occurs in just the place where the co-referential NP would be if it were a main clause.  If I remember right, Halliday and Hassan dealt at some length with deletion as a cohesive device.  In both sentences, the head noun represents what's doing the moving.  We interpret the subject of "move" either from the pronoun "who," which co-indexes with "the people," or from the zero subject after "that."  As I argued earlier, that subject is zero precisely because a lot of speakers can say, "The man 0 met me at the airport dropped me off at my office."  It is partly intonation that helps us parse the utterance.  People differ as to whether they would use this construction, but it represents a simple asyndetic relative clause where the zero subject co-indexes with th
>
> e head noun.
>
>
>
> To demonstrate that relative "that" is in some way nominal, you'll need to show that it has clearly nominal behavior, and that's a morphosyntactic question, not one that can be answered from intuition.
>
>
>
> On the other hand, Edith raises the interesting question of whether "that" could be changing its function from subordinator to pronoun.  I certainly don't reject that possibility.  Such grammatical change is not unusual, and more radical cases abound.  My problem with the claim, though, is again an absence of evidence beyond, present company excepted, naïve assumptions about grammar.  A long time ago, 1973, I think, I published a paper that included an internal reconstruction of the Yoruba preverbal morphemes, including the subject pronouns. This is a fairly complex problem, and internal reconstruction is a historical linguistic methodology for extrapolating earlier stages of a language from synchronic alternations and irregularities.  On historical grounds, what every grammar and every Yoruba teacher I had called a third person singular pronoun was nothing of the sort.  Historically I could explain every phonological and morphologic
>
> al property of the form, and none of
>
>
>
> it had any historical source in the pronoun systems.  Rather, the third singular in the present affirmative indicative was a zero form, just as it was in all the other subject pronoun sets in the language, and there were different paradigms depending on tense, modality, and negation.  In spite of a total lack of morphosyntactic evidence that the form was a third singular subject pronoun and in the face of overwhelming evidence to the contrary, grammar writers, Yoruba language teachers, and speakers of the language who also spoke English insisted that the word did in fact translate as English he/she/it (the language is totally without gender marking).  On the basis of that, I can accept that for modern speakers that form has changed from what it was historically, a combination of two different morphemes neither of them pronominal, to a subject pronoun.
>
>
>
> Edith suggests that something similar has happened in English with relative that, but in the English case the grammars are far from in agreement, and the best of them, Jespersen and Huddleston&Pullum, reject the idea.  Grammar teachers are rather more in agreement, but I suspect that's because of what they've been taught, which tends not to come from the best sources.  So the change may be in progress, but there's no way of detecting it yet.
>
>
>
> Herb
>
>
>
>
>
> Herb,
>
>
>
>
>
>     In structures like "except that" and "now that", "that" is clearly
>
> part of a subordinating (or complementing) process. Both require an
>
> explicit subject for the clause that follows. "Now that Obama has been
>
> elected..." If we said "Now that is a good thing," "that" is clearly
>
> acting like a pronoun. I would also expect a comma after "now." "Now,
>
> that is a good thing."
>
>    I can't help feeling that the arguments for "that" as a subordinator
>
> rather than a pronoun don't consider the possibility of different
>
> "that's." I'm positing three: the subordinator, the demonstrative, and
>
> the relative.)
>
>    To me, "The music that moves her" is very parallel to "The people who
>
> move her."  Neither that or who can be deleted from these because (as I
>
> see it) the clause needs an explicit subject. Either can be deleted if
>
> another subject is present. ("The music she loves... The people she
>
> loves.") You keep saying there isn't any evidence for "that" as a
>
> relative, but the evidence seems strong to me. The feeling sense that
>
> the music is doing the moving and "that" stands in as the clause
>
> subject seems very compelling. None of the arguments against it seem to
>
> outwiegh that compelling intuition.
>
>    Is it just a matter of classifying "that" differently because it
>
> doesn't have as much flexibility as "which" ("with which") or have
>
> separate forms (like "whose" and "whom")? Can we have a relative
>
> pronoun with more constraints than other pronouns?
>
>    We call these clauses "relative" in part because they are adjectival
>
> and the pronoun stands in for what the whole clause modifies. Is there
>
> an example for "that" in which that doesn't at least seem to happen? Is
>
> seeming (cognition) unimportant? Wouldn't it make just as much sense to
>
> call it a relative pronoun at least in these instances when it seems to
>
> act like one?
>
>    As far as I can tell, I recognize the same observations about how it
>
> acts as you do, but am just comfortable placing it in a relative
>
> category because it seems similar enough to other relatives to warrant
>
> that.
>
>    Clearly, in other cases, "that" acts like a subordinator or different
>
> kind of (non-relative) pronoun.
>
>    Am I totally missing the point?
>
>
>
> Craig
>
>
>
>
>
> This thread has teased out some of the complexity surrounding "that."  No
>
>
>
> one has mentioned the demonstrative, which is the direct descendant of the
>
> OE form "thaette" that Edith notes, and I think we're all agreed that the
>
> demonstrative and the conjunction are distinct forms.
>
>
>
> The one formal argument for pronoun status of relative-that is the fact
>
> that children and some non-standard dialects do have the form written as
>
> "that's." appending the genitive marker to the form in question.  This is
>
> not, however, a particularly strong argument because, as Arnold Zwicky has
>
> demonstrated elsewhere, the -'s genitive is not an affix but a clitic.
>
> Clitics are forms that typically attach to phrases, not to word stems.
>
> Affixes attach to word stems.  The fact that we can say "the Queen of
>
> England's decision" demonstrates that -'s attaches, here, to a nominal
>
> construction, not to a noun stem.  The fact, then, that for some speakers
>
> it can attach to the subordinator "that" simply means that for them it's
>
> behaving like a clitic, not like an affix.
>
>
>
> Bill's concern over zero forms is well-founded.  We don't want to
>
> proliferate zero forms every time we run into an anomalous distribution,
>
> which we certainly have in that and wh- relative clauses.  However, I
>
> would dispute, or at least strongly question, the claim that the absence
>
> of "that" in an RC represents a deletion.  That certainly doesn't reflect
>
> the historical facts, although the contemporary behavior of something
>
> doesn't necessarily have to reflect closely its historical development.
>
> I'd argue rather that asyndetic (unconnected) and that-marked relatives
>
> are simply two options--no zero involved here.
>
>
>
> As to the feeling that rel-that is pronominal, I suspect this is
>
> influenced in part by the syncretism with the demonstrative, which is
>
> quite a different form.  It is, for one thing, nearly always stressed, and
>
> the subordinator is almost never stressed.  As far as Craig's example of a
>
> "that it" relative construction, where the "it" is a resumptive pronoun,
>
> resumptives in subject position are particularly problematical.  They tend
>
> to occur, when they do, only in those places where "that" can't occur.  In
>
> subject position we're more likely to delete where the subject of the
>
> relative is coreferential with the head noun.  And that makes it very much
>
> like other subject deletions in dependent structures in English.
>
>
>
> In response to Dick's query about complementarity, wh- and that
>
> historically were not complementary.  In fact, in late Middle English and
>
> Early Modern English the two typically occurred together.  In fact, "that"
>
> occurred regularly after what we now consider adverbial subordinating
>
> conjunctions, so that expressions like "which that," "who that,"when
>
> that," and "if that (see the Sydney sonnet I posted not too long back),"
>
> etc. were the rule.  We still have reflexes of this in "now that" and
>
> "except that."  Over time, as we get into Late Modern English, the sense
>
> that the "that" is needed to mark subordination diminishes and the pronoun
>
> or adverb takes on that function itself.
>
>
>
> I understand the feeling that relative that is pronominal; I just haven't
>
> seen any evidence for it.
>
>
>
> Herb
>
>
>
> Herbert F. W. Stahlke, Ph.D.
>
> Emeritus Professor of English
>
> Ball State University
>
> Muncie, IN  47306
>
> [log in to unmask]<mailto:[log in to unmask]>
>
> ________________________________________
>
> From: Assembly for the Teaching of English Grammar
>
> 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/
>
>
>
> 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/
>
> 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/
>
> 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/
>
>
>

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/

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/