On 8 September 2017 at 11:31, Raymond Brown <[log in to unmask]>

> It so happens I've been skimming through a paper on English dialect
> forms.

What is the paper? (I'd like to have a read of it.)

On 7 September 2017 at 23:38, Mike S. <[log in to unmask]> wrote:

> On Thu, Sep 7, 2017 at 9:14 AM, Raymond Brown <[log in to unmask]>
> wrote:
> > On 07/09/2017 05:15, Mike S. wrote:
> >
> > Even if we were to pinpoint more precisely the present theoretical
> >> objection to the term "case" being applied to English, we'd still be
> left
> >> with the fact that English pronouns have traditionally been described in
> >> terms of case, the term is universally understood, and there is no
> obvious
> >> substitute term for it.
> >>
> >
> > The problem is that the term is universally understood.  This is why,
> > when the term is applied to modern English pronouns, we find that they
> > are "weird".  Just look at Wikipedia:
> >
> >
> > It's so peppered with exceptions that slavishly sticking to the case
> > tradition IMHO makes description more complicated.
> >
> Actually, I would say that summarizing the case usage rules from that
> Wikipedia section is relatively uncomplicated.  Here is the gist, ignoring
> details:
> "As a general rule, the subjective form is used when the pronoun is the
> subject of a verb, as in 'he kicked the ball', whereas the objective form
> is used as the direct or indirect object of a verb, or the object
> (complement) of a preposition."

We can't expect better of a Wikipedia article that hopefully is
intelligible to laymen, but that 'rule' is tacitly circular, since it
requires 'subject' to be defined as 'environments where pronouns take
subjective form'. There is no need whatever for such a definition of
'subject' other than to define an environment where pronouns take
subjective form. I can't offer an atheoretical layman-intelligible
characterization of this environment, but we can at least say that for all
English lects the presence of Tense is a key ingredient of the environment
(especially if Tense is present in subjunctives like "that she be"). The
same environment triggers number and person agreement on tense-inflected
verbs, but a key consideration is that only the two inflectional
superficies of subjective forms and verb agreement render this environment
grammatically significant: if English lacked subjective forms and verb
agreement, there would be no independent need to define this environment;
it plays no further role in the rules that constitute English.

Even setting aside the enormous amount of geolectal variation and looking
at usages with some claim to standardness (perhaps of an incipient or
obsolescent or hypercorrect sort), we find other environments where some
lects have subjective forms:

* As subject in non-complement small clause:
"*She* a philosopher of some renown, we plied her with questions."
"We plied her with the crassest questions, and *she* a philosopher of some
renown too."

* As a conjunct:
"They spoke to she and he."

* As a final conjunct
"They spoke to her and he"

* Vocative
"O thou noble conlanger!"

* In comparatives
"than she", "as she"

* As a quasideterminer
"It means nothing to we conlangers."

* As predicative complement of _be_
"It is they", "This is she"

The sole generalization over these environments is that they trigger
subjective forms.

In summary, then, no grammatical property unites the environments that
trigger subjective forms other than the property of triggering subjective
forms, and the 'core' tense-propinquant triggering environment for
subjective forms has no grammatical significance other than superficial

> I'll try to be clear.  What I argue is:
> > 1. 'case' is a distinctive, marked form assumed by a Noun Phrase to
> > indicate that the NP bears some _identifiable *grammatical* or
> > *semantic* relation_ to the rest of the sentence.
> > 2. The allomorphs of the modern English personal pronouns do not do
> > this.  They are conditioned, as And observed, "by various heterogeneous
> > syntactic conditions -- with these conditions being, in contemporary
> > English, subject to quite extreme interlectal variation."
> >
> The form of English pronouns (in the big, standard dialects) is entirely
> predictable when standing alone and governed directly by a verb or
> preposition.  In the handful of situations indicated earlier, there is a
> muddling of what case forms to use, but to suggest that standard English
> pronouns bear no "_identifiable *grammatical* or *semantic* relation_ to
> the rest of the sentence" in the face of obvious, simple examples like "We
> saw them yesterday" is (IMHO) at best to make a gross oversimplification.

The suggestion is not that the pronouns bear no identifiable grammatical or
semantic relation to the rest of the sentence but rather that subjective
forms do not mark pronouns in a single identifiable grammatical or semantic
relation to the rest of the sentence (i.e. there is not a one-to-one
correspondence between subjective form and semanticosyntactic triggers but
rather there is an eclectic range of semanticosyntactic triggers).
Furthermore, although in the instance of the core tense-propinquant
trigger, there is an identifiable grammatical (but certainly not semantic)
relation between the pronoun and the rest of the sentence, this relation
figures in no rules but the superficial inflectional rules of pronoun forms
and verb agreement; I lack the requisite knowledge to know whether there
are languages that indubitably have case but have a nominative so empty of

Assuming for the sake of argument (and perhaps counterfactually), that
passives and perfects are not synchronically related, the fact that they
have the same inflectional form necessitates recognition of a form type --
'past participle'. This allows us to say that in both the passive and the
perfect constructions the verb has past participle form. Likewise we must
recognize that certain personal pronouns have a 'subjective form', which is
triggered by any of a range of syntactic environments. This is in contrast
to, say, plural inflection, where a noun's Plural inflection is triggered
by a single, semantically importful semanticosyntactic trigger

 > What I find weird is that while we are (all) happy enough at the notion

> > that allophones may be positionally conditioned and, indeed, be subject
> > to extreme interlectal variation (as the very numerous YAEPT and YAEDT
> > testify), the notion that allomorphs may be condition "by various
> > heterogeneous syntactic conditions -- with these conditions being, in
> > contemporary English, subject to quite extreme interlectal variation" is
> > so difficult to accept.  That IMO is the real weirdness.
> >
> > Ray
> >
> I respect your position, but I have to differ.  A case system with
> weirdness is still a case system.  And while yes there is some weird stuff
> going on, there is also a good deal of normal stuff going on, and the
> normal stuff is at the core of the system, in the most basic sorts of
> sentences in which bare pronouns serve as subjects and objects.  How
> exactly should we account for the behavior of pronouns in sentences like
> "We saw them" and "They saw us" if it's not really a case system, not
> really, in your view, even a _vestigial_ sort of a case system?   What are
> we going to suggest the world call this behavior if not "case"?

If we generalize across lects (even if only (para)standard ones), then the
system overall is not a case system. But perhaps there is still room for
argument over whether the core triggering environment could legitimately be
called 'nominative case' -- i.e. over whether it could be legitimate to say
that one of the functions of subjective forms is to mark 'nominative case'.
My answer to that is that looking at English in isolation, absolutely
nothing is gained by introducing the notion 'nominative case' -- it would
essentially be vacuous, an addition label for what is already labelled and
defined by other means -- but if you instead want to describe English using
an analytical apparatus that foregrounds similarities across languages then
it might indeed be legitimate to say that one of the functions of
subjective forms is to mark 'nominative case', where a family resemblance
holds between nominative cases across languages.