subjunctive in indirect questions
Adam
Thanks for the help with my question about subject-accusative, I think
I understand better now. It is difficult studying on one's own!
Encouraged by the previous response, I'd like now to ask you this:
From Cicero's De Divinatione:
Si sunt di neque ante declarant hominibus quae futura sint, aut non
diligunt homines, aut quid eventurum sit ignorant, aut existumant
nihil interesse hominum scire quid sit futurum, aut non censent esse
suae maiestatis praesignificare hominibus quae sunt futura, aut ea ne
ipsi quidem di significare possunt.
The phrase "quae futura sint" at the beginning takes the subjunctive
because it is an indirect question. But towards the end of the
sentence, the same indirect question takes the indicative. Is it not
an indirect question? And if so, is the use of the subjunctive a
matter of tone, as in a relative clause in indirect statement?
These questions got me thinking. The proper Latin that we know of and
study, how close was that to actual spoken Latin in classical times? I
imagine there were a variety of dialects and accents and that proper
Latin was a sort of common language that no one actually spoke at
home.
Many thanks,
Adam
Thomas Bushnell, BSG
> From Cicero's De Divinatione:
>
> Si sunt di neque ante declarant hominibus quae futura sint, aut non
> diligunt homines, aut quid eventurum sit ignorant, aut existumant
> nihil interesse hominum scire quid sit futurum, aut non censent esse
> suae maiestatis praesignificare hominibus quae sunt futura, aut ea ne
> ipsi quidem di significare possunt.
>
> The phrase "quae futura sint" at the beginning takes the subjunctive
> because it is an indirect question. But towards the end of the
> sentence, the same indirect question takes the indicative. Is it not
> an indirect question? And if so, is the use of the subjunctive a
> matter of tone, as in a relative clause in indirect statement?
Good eye! The last "quae sunt futura" is not an indirect question, it
is a relative clause.
So the first "quae futura sint", which is an indirect question, is:
"...they do not declare ahead of time to men what will happen..."
and the last is
"...they do not count it to be of their majesty to foretell to men
those things which are to be..."
In the former, the declaring is of the answer to the question "What
will be?" In the latter, the fortelling is of the things themselves,
which will be.
There is a slight difference in meaning, but not so much as to matter
all that much. It does add to the variation in the passage (note that
Cicero is carefully avoiding repetition; existumant vs. censent;
declarant vs. praesignificare, and so forth.)
Thomas
Rolleston
[...]
>There is a slight difference in meaning, but not so much as to matter
>all that much. It does add to the variation in the passage (note that
>Cicero is carefully avoiding repetition; existumant vs. censent;
>declarant vs. praesignificare, and so forth.)
This reminds me of an earlier discussion, from which:
[BA 176, edited]
(1) Quae vere sentio dicam
I will utter my real sentiments
(2) Quae vere sentiam dicam
I will tell you what are my real sentiments
R.
Adam
> is a relative clause.
>
> So the first "quae futura sint", which is an indirect question, is:
>
> "...they do not declare ahead of time to men what will happen..."
>
> and the last is
>
> "...they do not count it to be of their majesty to foretell to men
> those things which are to be..."
>
> In the former, the declaring is of the answer to the question "What
> will be?" In the latter, the fortelling is of the things themselves,
> which will be.
>
> There is a slight difference in meaning, but not so much as to matter
> all that much. It does add to the variation in the passage (note that
> Cicero is carefully avoiding repetition; existumant vs. censent;
> declarant vs. praesignificare, and so forth.)
>
> Thomas
Hi Thomas, thanks for pointing out the difference! I guess I was
looking for an antecedant to a possible relative pronoun such as 'ea',
but I take it that the antecedant can be left out in certain cases?
Thanks again,
Adam
Thomas Bushnell, BSG
> Hi Thomas, thanks for pointing out the difference! I guess I was
> looking for an antecedant to a possible relative pronoun such as 'ea',
> but I take it that the antecedant can be left out in certain cases?
Antecedents are often left out in cases like this.
Neeraj Mathur
"Adam" <xanad...@yahoo.com> wrote in message
news:bb13062b.04080...@posting.google.com...
> I guess I was
> looking for an antecedant to a possible relative pronoun such as 'ea',
> but I take it that the antecedant can be left out in certain cases?
They are often left out in Latin, as in fact they are in English. You will
also find some cases of 'relative attraction', where the relative pronoun is
put into the case that its antecedent has (rather than following the normal
rule, that each pronoun gets its case from its own clause) and then the
antecedent is left out.
Example:
'Expected / Normal' Sentence: "da pecuniam eis quos rex amat".
Above with Relative Attraction: "da pecuniam quibus rex amat".
Although this is taught as a strange 'deviant' usage, you will find that in
certain authors this is the rule rather than the exception; it is standard
and expected in Attic Greek (although try finding a book that warns you of
that) and is thus to be expected in those many authors who use Attic as a
model.
Caveat: all of the above is written at around 2 am, without recourse to any
reference books. It should be all correct more or less, but use it rather as
a springboard for your own reading on the grammar than as a rule for now.
Neeraj Mathur
Ed Cryer
"Neeraj Mathur" <neeraj...@chch.ox.ac.uk> wrote in message
news:cf96ku$b4k$1...@news.ox.ac.uk...
I'd like to add another consideration about the Latin subjunctive. It is
very common in good classical writers to use it instead of "amat" in your
example.
"Da pecuniam quibus rex amet".
I'm not sure what grammarians call this, but I phrase it to myself as "of
the type that"; something like "whomsoever the king may love"; "all those
whom the king loves".
Ed
Rolleston
[...]
>I'd like to add another consideration about the Latin subjunctive. It is
>very common in good classical writers to use it instead of "amat" in your
>example.
>"Da pecuniam quibus rex amet".
>
>I'm not sure what grammarians call this, but I phrase it to myself as "of
>the type that"; something like "whomsoever the king may love"; "all those
>whom the king loves".
A "characteristic relative clause"?
R.
Thomas Bushnell, BSG
> I'd like to add another consideration about the Latin subjunctive. It is
> very common in good classical writers to use it instead of "amat" in your
> example.
> "Da pecuniam quibus rex amet".
>
> I'm not sure what grammarians call this, but I phrase it to myself as "of
> the type that"; something like "whomsoever the king may love"; "all those
> whom the king loves".
I just phrase it with an English subjunctive:
"Give mone to those whom the king would love."
Rolleston
[...]
Would you care to give us an example? :)
>"Give mone to those whom the king would love."
R.
Thomas Bushnell, BSG
Um, there's the example. "Give money to those whom the king would
love."
Rolleston
Are you saying that your "would" is an English subjunctive?
I don't think it is.
R.
Thomas Bushnell, BSG
> >Um, there's the example. "Give money to those whom the king would
> >love."
>
> Are you saying that your "would" is an English subjunctive?
>
> I don't think it is.
The auxiliaries would, could, and should normally carry the sense of
the subjunctive. Think of it as an analytic tense. If "locutus sum"
can be a perfect (despite actually being a present-tense verb and a
participle), then "would love" can be a subjunctive, despite actually
being an auxiliary and a complementary infinitive lacking "to".
Thomas
ken quirici
> "Da pecuniam quibus rex amet".
>
> I'm not sure what grammarians call this, but I phrase it to myself as "of
> the type that"; something like "whomsoever the king may love"; "all those
> whom the king loves".
>
> Ed
There seem to be several possible English translations with differing
meanings that count on the notion of subjunctive as associated
with possibility or wish (I'm just getting to subjunctives in my own
study so bear with me if this is old hat or off the mark)
Give money to anyone the king might love (i.e. it is conceivable that the
king loves - i.e. cover all your bases)
Give money to anyone the king would love (i.e. it is likely that the king
loves, upon due consideration)
Give money to anyone it happens that the king in fact loves - where what
is 'subjunctive' is that we don't know ourselves whom the king loves, but
whoever it might be, etc. - this is actually what Ed refers to.
The 'characteristic relative clause' which didn't seem to turn up in
Allen & Bacon (Perseus) doesn't seem to me to be a 'true' subjunctive in
the might/wish class, but I haven't actually got far enuf to know for sure.
FWIW.
Ken
Rolleston
>Rolleston <roll...@tiscali.co.uk> writes:
>
>> >Um, there's the example. "Give money to those whom the king would
>> >love."
>>
>> Are you saying that your "would" is an English subjunctive?
>>
>> I don't think it is.
>
>The auxiliaries would, could, and should normally carry the sense of
>the subjunctive.
But not the name. Never mind. The world will survive this.
R.
Rolleston
[...]
>The 'characteristic relative clause' which didn't seem to turn up in
>Allen & Bacon (Perseus) doesn't seem to me to be a 'true' subjunctive in
>the might/wish class, but I haven't actually got far enuf to know for sure.
Consult Allen & Greenough at Perseus (Sec. 534 and onwards):
http://perseus.uchicago.edu/cgi-bin/ptext?doc=
Perseus%3Atext%3A1999.04.0001&layout=&loc=534
My apologies for the broken link - TinyURL is not responding.
R.
Ed Cryer
"Rolleston" <roll...@tiscali.co.uk> wrote in message
news:ga1lh0d5lbtmm2aq4...@4ax.com...
Here's one for a dedicated grammarian, R.
A passage from Cicero with so many subjunctives that it's almost as if he
set out to try and put into one sentence as many different usages as he
could.
Ac non vereor, ne assentatiuncula quadam aucupari tuam gratiam videar, cum
hoc demonstrem, me a te potissimum ornari celebrarique velle; neque enim tu
is es, qui, qui sis, nescias et qui non eos magis, qui te non admirentur,
invidos quam eos, qui laudent, assentatores arbitrere, neque autem ego sum
ita demens, ut me sempiternae gloriae per eum commendari velim, qui non ipse
quoque in me commendando propriam ingenii gloriam consequatur.
Cicero Letters.
Ed
Thomas Bushnell, BSG
> >The auxiliaries would, could, and should normally carry the sense of
> >the subjunctive.
>
> But not the name. Never mind. The world will survive this.
There is not only one assignment of names to grammatical features.
Rolleston
Yes, indeed. And some of them are reasonably sensible.
Do give us your criteria for identifying the English subjunctive.
R.
Thomas Bushnell, BSG
I already did: it's an analytic construction, just like saying that
"the subject is completed" is a passive. (And not every sentence with
"to be" and a past participle is necessarily passive. For example,
"the door is closed" is not really passive in most uses.)
For example, "the pedant has exceeded reason" is a perfect. The fact
that in English we use compound constructions for the passive or the
perfect is an excellent clue to what's going on with an optative
subjunctive like "Oh, would that the pedant realize language
changes!".
This used to be called a "subjunctive-equivalent", but these days, it
doesn't get called that; it gets called a subjunctive--which nicely
captures the semantic similarity between most "would/could/should"
sentences and, for example, the Latin subjunctive.
Now all English passives and perfects are analytic, whereas only some
subjunctives are. Maybe this is your stumbling block. The situation
is hardly strange; perfect passives are analytic in Latin, but no
other perfects or passives are. Are you going to insist that "locutus
sum" is not a real passive? Or not a real perfect? Or is it just
that the grammatical terms you were taught years ago might no longer
be the only ones in use that gets your goat?
You may not like the fact that this gets called a subjunctive, but
it's hardly a usage I invented. Maybe the British schools aren't as
good at grammar anymore, or maybe they are still pretending that the
grammatical rules of Latin should govern English. Pray tell, can we
end sentences with prepositions in your school? How about split
infinitives?
"If the pedant were not so tiresome, the subject would be
interesting." Now isn't it interesting how we can see that both
clauses here are forms of the English subjunctive, and once you see
*that* it can make the similar constructions in Latin or Greek a
little easier to fathom?
Thomas
Rolleston
>> >The auxiliaries would, could, and should normally carry the sense of
>> >the subjunctive.
>> Do give us your criteria for identifying the English subjunctive.
>
>I already did: it's an analytic construction
So, the English subjunctive is an analytic construction...
>Now all English passives and perfects are analytic, whereas only some
>subjunctives are.
...unless it isn't. I see. I fear you have misunderstood my words.
R.
Thomas Bushnell, BSG
As if you just didn't bother to read my post.
Like the Latin passive, the English subjunctive is an analytic
construction in some contexts, and not in others.
Of course, I made exactly that point in my post, which makes me wonder
if your education was as deficient in reading skills as it seems to be
in other areas.
If you can't bother to read what I write, then don't bother to reply.
Thomas
ken quirici
>
> http://perseus.uchicago.edu/cgi-bin/ptext?doc=
> Perseus%3Atext%3A1999.04.0001&layout=&loc=534
>
> My apologies for the broken link - TinyURL is not responding.
>
> R.
I found it as you said at section 534. Thanks for the reference.
An English subjunctive might be used in this construction, as in
a sentence vaguely like the ones (don't have it front of me right now)
in Allen & Greenough (and possibly in this thread, although I also
don't have it in front of me right now either):
He's a king who would help an ally in need.
So it ties in to the subjunctive as generally the tense of what is not
known to be but is possible. Except of course in result clauses, and
the like. Is there any sense in which a result clause has the element
of potentiality otherwise associated with the subjunctive?
How did the subjunctive evolve into such a partially hybrid animal?
I vaguely also remember that it's used as a matter of course in
sub-sub-clauses whatever it's meaning or signification (having done
Moreland & FLeisher many years ago).
Thanks.
Ken
Gordon
says...
> Are you going to insist that "locutus
> sum" is not a real passive?
I certainly will. (Hint: you need to use a different verb as source for
your example)
--
Gordon
"I have just as much authority as the Pope.
I just don't have as many people who believe it."
Rolleston
[...]
>Here's one for a dedicated grammarian, R.
>A passage from Cicero with so many subjunctives that it's almost as if he
>set out to try and put into one sentence as many different usages as he
>could.
>
>Ac non vereor, ne assentatiuncula quadam aucupari tuam gratiam videar, cum
>hoc demonstrem, me a te potissimum ornari celebrarique velle; neque enim tu
>is es, qui, qui sis, nescias et qui non eos magis, qui te non admirentur,
>invidos quam eos, qui laudent, assentatores arbitrere, neque autem ego sum
>ita demens, ut me sempiternae gloriae per eum commendari velim, qui non ipse
>quoque in me commendando propriam ingenii gloriam consequatur.
Thank you. It's beautifully balanced.
http://www.thelatinlibrary.com/cicero/fam5.shtml
R.
Ed Cryer
"Thomas Bushnell, BSG" <tb+u...@becket.net> wrote in message
news:87n011w...@becket.becket.net...
What the dickens is this "analytic" thing? Am I just wrong in thinking it
goes back to Kant's 1781 book in which he distinguishes between
analytic/synthetic propositions?
Please explain to this man, who swears that he will give you his attention.
So eager is he for the new learning.
Ed
Rolleston
[...]
>What the dickens is this "analytic" thing?
It would appear that TB's analytic constructions are those, such as
"he has eaten" and "locutus est", in which independent function
words are used to express what might otherwise be expressed by
inflexion. Compare with "he ate" (no auxiliary), "loquitur" and Greek
perfect forms such as "keklęka".
R.
Rolleston
>Are you going to insist that "locutus sum" is not a real passive?
It has a passive _form_. Not that you'd necessarily think that:
Thomas Bushnell wrote in http://tinyurl.com/57eew:
>rexc...@aol.com (Rex Caeser) writes:
>> There are several words in Latin that are deponent that I think *should* have
>> passive form (i.e. to be aquired). How?
>
>There is none.
Toodle pip!
R.
Thomas Bushnell, BSG
> In article <87n011w...@becket.becket.net>, tb+u...@becket.net
> says...
>
> > Are you going to insist that "locutus
> > sum" is not a real passive?
>
> I certainly will. (Hint: you need to use a different verb as source for
> your example)
Blech, of course, that's foolish of me. :) I meant, "Are you going
to insist that 'locutus sum' is not a real perfect?"
Thomas Bushnell, BSG
> I absorbed every feeble word of your flocculent drivel, in
> which we find remarks and questions touching on analytic
> constructions, the Latin passive perfect, past participles,
> English passives and perfects, subjunctive-equivalence,
> the British school system, Latin rules governing English,
> split infinitives, sentences ending with prepositions, Greek,
> and goats getting got. None of which provided any criteria
> for identifying the English subjunctive.
What are the criteria for identifying the English passive voice?
Thomas Bushnell, BSG
> What the dickens is this "analytic" thing? Am I just wrong in thinking it
> goes back to Kant's 1781 book in which he distinguishes between
> analytic/synthetic propositions?
Nope. An analytic construction is one in which multiple words are
used to represent a tense. The metaphor is that the tense is being
"analyzed" or "broken up" into different parts.
Thomas Bushnell, BSG
> Thomas Bushnell wrote:
> >Are you going to insist that "locutus sum" is not a real passive?
>
> It has a passive _form_. Not that you'd necessarily think that:
I meant, of course, are you going to insist that it is not a real
perfect?
What are the criteria for identifying the English passive voice?
Thomas Bushnell, BSG
> It would appear that TB's analytic constructions are those, such as
> "he has eaten" and "locutus est", in which independent function
> words are used to express what might otherwise be expressed by
> inflexion. Compare with "he ate" (no auxiliary), "loquitur" and Greek
> perfect forms such as "keklęka".
And in this, I follow the Oxford English Dictionary (perhaps you've
heard of it?) which gives the following as definition 1b for the word
"analytical":
Expressing the various notions and relations into which a
proposition or complex notion may be analyzed, by distinct words,
instead of combining several into one word; as, they shall be sent
out for {emac}-mitt-{emac}-nt-ur; with a sword for gladio; plus fort
for fortior; of man for man's.
In the United States, the word "analytical" is rather uncommon, and
"analytic" is used instead.
Thomas
Thomas Bushnell, BSG
> Do give us your criteria for identifying the English subjunctive.
What are your critera for identifying the Latin subjunctive,
incidentally? Is "legam" a subjunctive or not? And if you say "it
depends on context", why is the same not satisfactory for "would read"
in English?
Rolleston
>Rolleston <roll...@tiscali.co.uk> writes:
>
>> Do give us your criteria for identifying the English subjunctive.
>
>What are your critera for identifying the Latin subjunctive,
>incidentally?
Ah, the old respond-with-a-question gambit. You could save
us all a lot of time by saying "I don't have any" if you don't.
Let me put this question to you: what is your definition of
the English subjunctive?
Be as formal as you wish. You can use lambda calculus for all
I care. The important thing is that you make clear what you
mean by the word "subjunctive", because that word is used
in very different ways by different people at different times.
R.
Neeraj Mathur
> But not the name. Never mind. The world will survive this.
With all due respect, wouldn't this have been the best place to end this
pointless argument? My goodness, I would have expected more from a group of
Classicists, but maybe I was wrong.
You two are squabbling over a rather trivial point, which neither has been
willing to grasp: one of you wants to treat 'subjunctive' as a purely
morphological category, while one of you wants it to be a grammatical
category. Obviously, 'would love' is not a subjunctive, morphologically,
because there are no morphosyntactic features that are recognised as marking
mood present in it. On the other hand, it belongs to a grammatical category,
defined by compounding with 'modal verbs', that can be used in certain
contexts to translate a Latin subjunctive.
I do happen to have a personal opinion on the matter, related to my general
view that linguists should only produce grammars that are descriptive, and
that each language should be approached on its own terms. I include my
opinion below. I feel it is well reasoned and considered, as all people feel
about their opinions; if you find a fault in what I say or have a relevant
counter-argument, I am open to scholarly persuasion.
I do not believe that the term 'subjunctive' is appropriate to the current
English language other than as a name for the morphology of the verb found
in clauses such as 'whether this be subjunctive or not is hardly worthy of
petty arguments'. I believe this because there are distinctions in meaning
of each modal auxiliary verb which are not accurately conveyed by the single
category 'subjunctive'. Thus, I feel that, while the 'modal group' as a
whole usually offers at least one appropriate translation of subjunctives in
certain languages, it is inappropriate to call it 'subjunctive'. I worry
that trying to use 'subjunctive' as a grammatical category stems from a
prescriptive dedication to Latin grammar, which would want to call
'subjunctive' anything that translates a Latin subjunctive (which is defined
morphologically, not grammatically). The trouble is that by this approach,
many Greek optatives would have to be reckoned subjunctives (North & Hillard
and other older grammarians mistake Greek grammar for Latin grammar with a
rather disturbing frequency). I think that in both Latin and Greek,
'subjunctive' (and 'optative' in the case of Greek) are morphological
categories; these categories are independent of semantics. In English, there
is a morphological contrast between the third persons 'he love' and 'he
loves', which have different semantic values; the first of these is often
called 'subjunctive' and that is fair enough. However, this form is not
equivalent in meaning to analytical compounds with modal auxiliaries, other
than in the irrelevant fact that both can translate, in certain contexts, a
Latin subjunctive.
Neeraj Mathur
Thomas Bushnell, BSG
> You two are squabbling over a rather trivial point, which neither has been
> willing to grasp: one of you wants to treat 'subjunctive' as a purely
> morphological category, while one of you wants it to be a grammatical
> category.
The problem of course is that if one wants such terms to be only
morphological categories, then "locutus sum" is not a perfect, and
"legam" is what exactly?
> I do not believe that the term 'subjunctive' is appropriate to the current
> English language other than as a name for the morphology of the verb found
> in clauses such as 'whether this be subjunctive or not is hardly worthy of
> petty arguments'. I believe this because there are distinctions in meaning
> of each modal auxiliary verb which are not accurately conveyed by the single
> category 'subjunctive'.
This is surely true, in so far as it goes. The problem is that the
Latin subjunctive, for example, also has many meanings for different
uses of the subjunctive. Now that's not strictly parallel: in Latin
we have one construction with different uses, but in English we have
different constructions with different uses. So this would seem to
argue for a strictly morphological usage of such terms.
But that fails, as I've already explained, when you consider that the
morphology is ambiguous. "Legam" is either a subjunctive or not,
depending on its use and meaning in context.
> Thus, I feel that, while the 'modal group' as a whole usually offers
> at least one appropriate translation of subjunctives in certain
> languages, it is inappropriate to call it 'subjunctive'. I worry
> that trying to use 'subjunctive' as a grammatical category stems
> from a prescriptive dedication to Latin grammar, which would want to
> call 'subjunctive' anything that translates a Latin subjunctive
> (which is defined morphologically, not grammatically).
The Latin subjunctive cannot be defined morphologically. It almost
can, but such a definition fails. We agree that "amem" is a
subjunctive, but what about "legam"? Likewise, the Latin perfect
cannot be defined morphologically; is "amatus sum" perfect or not? It
simply depends on context and meaning. There is no escaping this.
So that means that such terms *cannot* have such rigid bounds as being
strictly morphological, but instead must have a more fluid application
which takes into account morphology, syntax, and semantics.
> The trouble is that by this approach, many Greek optatives would
> have to be reckoned subjunctives (North & Hillard and other older
> grammarians mistake Greek grammar for Latin grammar with a rather
> disturbing frequency). I think that in both Latin and Greek,
> 'subjunctive' (and 'optative' in the case of Greek) are
> morphological categories; these categories are independent of
> semantics.
This would be nice if it were true, but it reflects an idealized
vision of what the morphology of the languages should be, but not what
it actually is. In fact, there is no assignment of mood, tense, or
voice terms in Latin which is strictly morphological. Sometimes
"amatus sum" is a perfect passive; sometimes "amatus sum" is really
just a present-tense predication. Sometimes "legam" is future
indicative, sometimes "legam" is present subjunctive. And we can only
tell between these by attention to the semantics and indeed the actual
meaning of the sentences concerned.
The normal (and I think correct) way to understand "amatus sum" or
"legam" is to hold that the word is ambiguous, and that the ambiguity
is resolved by considering which of various meanings is more likely.
But that view requires saying that "in this sentence, 'legam' is
subjunctive," and that determination simply is not based purely on
morpohological considerations.
> In English, there is a morphological contrast between the
> third persons 'he love' and 'he loves', which have different
> semantic values; the first of these is often called 'subjunctive'
> and that is fair enough. However, this form is not equivalent in
> meaning to analytical compounds with modal auxiliaries, other than
> in the irrelevant fact that both can translate, in certain contexts,
> a Latin subjunctive.
I once explained Latin perfect verbs to a Spanish speaker by saying
that the Latin perfect has two reflexes in Spanish: the Spanish
analytic perfect tenses, and the preterit. (My use of the word
"reflex" there is probably not correct sensu stricto, but you get my
point.)
What about the *lack* of morphological contrast between "he loved" and
"he loved", the former subjunctive and the latter not? Surely we do
not want to say that there is no subjunctive past! So there must be
something other than morphological considerations going on.
Thomas
Peter T. Daniels
Whereas the Latin subjunctive is pretty easily identified
morphologically, no?
--
Peter T. Daniels gram...@att.net
Thomas Bushnell, BSG
> Be as formal as you wish. You can use lambda calculus for all
> I care. The important thing is that you make clear what you
> mean by the word "subjunctive", because that word is used
> in very different ways by different people at different times.
I mean a semantic category, reflected in (but not defined by)
morphology, which is expressed in English by a variety of forms, some
with auxiliaries and some by conjugational variations, and which are
similar in meaning to the constructions known as "subjunctive" in
closely related Indo-European languages (such as Romance, Latin, and
German).
My "answer with a question" is designed to make a specific point: the
giving of such rigid definitions is well-nigh impossible; if you
cannot even define the Latin subjunctive or the English passive, then
why do you expect me to define the English subjunctive?
I am happy to say that I am unable to give a list of necessary and
sufficient conditions, but that this is no deficiency in my part, but
rather part of the nature of the beast. Absent a comprehensive and
complete grammar and semantics for the language, it is essentially
impossible.
But you could prove me wrong, and demonstrate that it is my failing,
by providing such a definition of the Latin subjunctive or the English
passive, one which covers all the cases, leaving nothing out.
Thomas
Thomas Bushnell, BSG
> > Be as formal as you wish. You can use lambda calculus for all
> > I care. The important thing is that you make clear what you
> > mean by the word "subjunctive", because that word is used
> > in very different ways by different people at different times.
>
> Whereas the Latin subjunctive is pretty easily identified
> morphologically, no?
No. Is "legam" subjunctive or not?
Carmen L. Abruzzi
> Rolleston <roll...@tiscali.co.uk> writes:
>
>
>>Be as formal as you wish. You can use lambda calculus for all
>>I care. The important thing is that you make clear what you
>>mean by the word "subjunctive", because that word is used
>>in very different ways by different people at different times.
>
>
> I mean a semantic category, reflected in (but not defined by)
> morphology, which is expressed in English by a variety of forms, some
> with auxiliaries and some by conjugational variations, and which are
> similar in meaning to the constructions known as "subjunctive" in
> closely related Indo-European languages (such as Romance, Latin, and
> German).
But the subjunctive in those languages is not defined by
meaning, but by morphology.
>
> My "answer with a question" is designed to make a specific point: the
> giving of such rigid definitions is well-nigh impossible; if you
> cannot even define the Latin subjunctive or the English passive, then
> why do you expect me to define the English subjunctive?
I would think it's so that he knows what you're talking
about. As you state, your meaning of "subjunctive" is a
rather loose semantic one, and does not depend on morphology
as its primary defining characteristic. Thus, your
"subjunctive" for English, is quite a different category
from what "subjunctive" means in Latin, or Greek, or
Sanskrit, where it is easy to define the subjunctive, since
it's a morphological category.
>
> I am happy to say that I am unable to give a list of necessary and
> sufficient conditions, but that this is no deficiency in my part, but
> rather part of the nature of the beast. Absent a comprehensive and
> complete grammar and semantics for the language, it is essentially
> impossible.
>
> But you could prove me wrong, and demonstrate that it is my failing,
> by providing such a definition of the Latin subjunctive or the English
> passive, one which covers all the cases, leaving nothing out.
>
The Latin subjunctive is a morphological category, it is
easily defined.
A Latin verb is in the subjunctive according to its form,
whatver its semantics may be.
One can say that a verb is in the subjunctive without having
it connected to any specific semantics.
This page:
<http://www.facstaff.bucknell.edu/gretaham/Teaching/latin102/latin/subjunctive_main.htm>
lists numerous uses of the Latin subjunctive, but it isn't
any of these uses that determine whether a verb is in the
subjunctive or not, it's the form of the verb that
determines whether it's subjunctive or not.
The subjunctive has many uses, but it's always clear when a
verb is in the subjunctive and when it is not, as this has
nothing to do with the use, but with the form of the verb.
Peter T. Daniels
Is it not a member of an identifiable paradigm?
Zobby
It could be a present subjunctive or it could be a future indicative. I
suspect that it is why this example was chosen.
Z.
Thomas Bushnell, BSG
> Thomas Bushnell, BSG wrote:
> >
> > "Peter T. Daniels" <gram...@worldnet.att.net> writes:
> >
> > > > Be as formal as you wish. You can use lambda calculus for all
> > > > I care. The important thing is that you make clear what you
> > > > mean by the word "subjunctive", because that word is used
> > > > in very different ways by different people at different times.
> > >
> > > Whereas the Latin subjunctive is pretty easily identified
> > > morphologically, no?
> >
> > No. Is "legam" subjunctive or not?
>
> Is it not a member of an identifiable paradigm?
It is a member of two. That means you cannot determine on strictly
morphological grounds whether "legam" is subjunctive or not. You also
cannot tell on syntactic grounds; and in general, only a consideration
of which of two possible meanings is more likely will determine
whether "legam" in some particular use is subjunctive.
Carmen L. Abruzzi
person (which have distinct fut. ind. and pr. subj. forms)
would be used in a given construction will tell you whether
"legam" is subjunctive or future indicative in a that
construction. You don't need to consider the meaning as
such, as this does not determine whether it's subjunctive.
The meaning has little to do with determining the
subjunctivity of a given form; the same meaning might be
expressed using the subjunctive or a nonsubjunctive form
(e.g. English "I ask you to go" vs. "I ask that you go").
Thomas Bushnell, BSG
> No, a consideration of which form of the second or third person (which
> have distinct fut. ind. and pr. subj. forms) would be used in a given
> construction will tell you whether "legam" is subjunctive or future
> indicative in a that construction. You don't need to consider the
> meaning as such, as this does not determine whether it's subjunctive.
Alas, if only it were true! As it happens, some of the uses of the
subjunctive are dependent on person. (As in my example.) The
deliberative subjunctive is basically only a first-person
construction, so there is no consideration of what the second or third
person form "would be". But we can amend your test to the correct
one, which is to ask what the plural form would be.
So now, the test is we look and ask ourselves: If I were to recast
this sentence in the plural, which form would I use? We just ask that
nice little question, and we will know! Splendid.
How do you propose to go about deciding which form to use in the
absence of information about meaning?
Thomas
Thomas Bushnell, BSG
> But the subjunctive in those languages is not defined by meaning, but
> by morphology.
Really? Is "legam" subjunctive or not?
> I would think it's so that he knows what you're talking about. As you
> state, your meaning of "subjunctive" is a rather loose semantic one,
> and does not depend on morphology as its primary defining
> characteristic. Thus, your "subjunctive" for English, is quite a
> different category from what "subjunctive" means in Latin, or Greek,
> or Sanskrit, where it is easy to define the subjunctive, since it's a
> morphological category.
Really? Is "legam" subjunctive or not?
Do you have a recipe for reliably telling whether a given usage of
"legam" is subjunctive that doesn't depend on semantics and meanings?
> The Latin subjunctive is a morphological category, it is easily
> defined.
Really? Is "legam" subjunctive or not?
> A Latin verb is in the subjunctive according to its form, whatver its
> semantics may be.
Really? What about the Latin perfect. Is "amatus sum" perfect or not?
> lists numerous uses of the Latin subjunctive, but it isn't any of
> these uses that determine whether a verb is in the subjunctive or not,
> it's the form of the verb that determines whether it's subjunctive or
> not.
An excellent theory, but alas, it doesn't apply to the actual
language.
> The subjunctive has many uses, but it's always clear when a verb is in
> the subjunctive and when it is not, as this has nothing to do with the
> use, but with the form of the verb.
Fascinating. So in the sentence, "legam illum librum" is "legam"
subjunctive or not? As far as I can tell, there are (at least) two
different meanings, one with present subjunctive "legam", and one with
future indicative "legam", and it is hardly clear without more
information which is the correct reading. And that more information,
of course, would be the context of the utterance. For example:
"Quid cras agabis?" - "Legam illum librum" (future indicative)
"Quid nunc cogitas?" - "Legam illum librum" (present subjunctive)
Nor is this the only example where categories that are allegedly
purely morphological are in fact not, and clearly depend on
semantics. As I have said repeatedly, the perfect and passive also
have this potential ambiguity.
"Aliquisne portam aperuit?" - "Porta aperta est." (perfect passive)
"Estne porta quae est aperta?" - "Porta aperta est." (present with
predication)
Now these are uncommon, but even one example suffices to prove that
"subjunctive" is not a strictly morphological category in Latin (nor
is perfect, or passive), and that the term "subjunctive" therefore
includes not just information about morphology, but also about
semantics.
Thomas
Thomas Bushnell, BSG
> No, a consideration of which form of the second or third person (which
> have distinct fut. ind. and pr. subj. forms) would be used in a given
> construction will tell you whether "legam" is subjunctive or future
> indicative in a that construction. You don't need to consider the
> meaning as such, as this does not determine whether it's subjunctive.
I owe a fuller response, to explain better my view of exactly what is
going on here. First, amend the proposed test to one which considers
plural instead of singular, as I mentioned already.
So in the sentence, "legam illum librum," I must ask myself: "If I were
to make that plural, which sentence would I use? And there are of
course two possibilities:
"Legemus illum librum" (future indicative)
"Legamus illum librum" (present subjunctive)
Now, the test says: "If you would use 'legemus,' then 'legam' was
indicative; if you would use 'legamus,' then 'legam' was subjunctive."
And I agree completely: this is the right test. (There may be corner
cases that it misses on, but it seems basically right to me.)
But how do I know which plural form I would use?
The only test I know of is to ask myself "how was 'legam' being used
in the original sentence" and then use the appropriate plural.
That is, I say, "In the context, 'legam' looks like a deliberative
subjunctive, so I will use 'legamus.'" Or I say, "In the context,
'legam' looks like a future indicative, so I will use 'legemus.'"
You see, the only way to know which plural to use, is to *already*
have decided whether "legam" is subjunctive or indicative! That is, I
cannot determine the mood of "legam" by asking how I would pluralize
it; in fact, I do exactly the opposite: I determine the mood of
"legam" (on the basis of meaning and context), and then I pluralize it
by choosing the correct mood, "legemus" or "legamus."
Now if we are talking about querying native speakers who have an
intuitive grammar in their heads, the test proposed is exactly the
right one (and that's why I agree it's the right test). But if you
ask, "What's going on in their heads to produce that answer?" the
answer cannot be, "They have a mystical faculty for producing the
correct plural." It must be, "They consider (perhaps unconsciously)
the mood of 'legam' and then form the corresponding plural." And
their consideration of the mood of "legam" *cannot* be based upon
already knowing what the right plural is!
So yes, an excellent test is to ask how the construction behaves if we
change number (and sometimes, though not in this case, person); but
actually applying that test requires already having the ability to
determine the mood of "legam"; it cannot be that the mood is actually
determined based upon the correct plural, and the plural is in turn
based on the correct mood!
So morphology cannot do it (ever!) when we have the same morphological
form. Sometimes grammar will do, if say the subjunctive can only live
in one kind of construction, so that we couldn't have a minimal pair
of ambiguous identical sentences. This is why I presented not just the
word "legam", but the sentence, "legam illum librum", where you cannot
tell from the complete sentence what the mood is. Sometimes you might
have identical surface forms where the parse will tell you the correct
meaning (perhaps that's the way to understand "amatus sum", as having
two different parses); but in the case of "legam illum librum" both
readings have exactly the same parse tree.
The only way to tell the mood of "legam" in "legam illum librum" is to
consider the context--the meaning.
Thomas
Thomas Bushnell, BSG
> So morphology cannot do it (ever!) when we have the same morphological
> form. Sometimes grammar will do, if say the subjunctive can only live
> in one kind of construction, so that we couldn't have a minimal pair
> of ambiguous identical sentences. This is why I presented not just the
> word "legam", but the sentence, "legam illum librum", where you cannot
> tell from the complete sentence what the mood is. Sometimes you might
> have identical surface forms where the parse will tell you the correct
> meaning (perhaps that's the way to understand "amatus sum", as having
> two different parses); but in the case of "legam illum librum" both
> readings have exactly the same parse tree.
>
> The only way to tell the mood of "legam" in "legam illum librum" is to
> consider the context--the meaning.
And lest people think this is some weird feature of this one sort of
example, it happens all over.
I have already given the example of the Latin passive perfect tenses,
where I think you must know the parse tree to determine what tense the
verb is. I am unsure whether this should be called a semantic or a
syntactic test, since the only way to determine which of the two
possible parse trees is correct is the meaning.
There are many cases similar to that one, where you must know the
correct parse tree to know the way to construe the syntax of a word,
and in which you can only know the correct parse tree by considering
which meaning was more likely in context.
For example, most Latin vocatives are morphologically identical to
nominatives, and you need the parse tree to distinguish.
But there are also cases closer to my "legam" example. A very common
verbal one is verbs whose perfect stem is identical to their present
stem; in such cases (for example, abluo) the third person singular and
first person plural have identical present and perfect forms, which
can only be distinguished by context; the parse tree of both
construals is exactly the same.
Similarly, plural nouns always have identical dative and ablative
forms, and there are constructions which could have both meanings (and
have identical parse trees), and in which you can only tell "what is
the case of this noun" by considering the meaning of the sentence, in
context.
The same happens in English. I already gave the example of the
English passive voice, which is potentially ambiguous in just the same
way as the Latin perfect passive. There are also nouns in English
which have identical plural and singular forms, and except in the
present indicative, there is no way from the surface structure or the
parse tree to determine whether the noun is singular or plural.
All this shows that such categories are simply not strictly
morphological terms, but that labels like "subjunctive", "plural", and
"passive" depend on more than just morphological catogorizing for
their meaning and correct application.
Thomas
Rolleston
[...]
>> > No. Is "legam" subjunctive or not?
>>
>> Is it not a member of an identifiable paradigm?
>
>It is a member of two. That means you cannot determine on strictly
>morphological grounds whether "legam" is subjunctive or not. You also
>cannot tell on syntactic grounds;
Is "legam" unambiguously in the subjunctive in this:
utinam haec legam!
?
There may be a smallish proportion of instances of "legam"
that can be identified with high probability as subjunctives
on syntactic grounds, if we can come up with an adequate
description of the relevant Latin syntax.
Instances of "legam": http://tinyurl.com/45vez
I would guess, if you know your principal parts, most
subjunctive forms are morphologically identifiable.
As always, I'd be pleased to hear the views of others.
Thanks,
R.
Thomas Bushnell, BSG
> Is "legam" unambiguously in the subjunctive in this:
>
> utinam haec legam!
I'm pretty sure, yes.
> There may be a smallish proportion of instances of "legam"
> that can be identified with high probability as subjunctives
> on syntactic grounds, if we can come up with an adequate
> description of the relevant Latin syntax.
Undoubtedly true.
> I would guess, if you know your principal parts, most
> subjunctive forms are morphologically identifiable.
Most, yes.
The same is much less certain if you look at some of the other
morphological ambiguities in the language; I chose this one because of
its pertinence to the previous discussion. But the ambiguity of
dative and ablative plurals is endemic, and a source of frequent
confusion to beginning students.
Charles Riggs
BSG) wrote:
No-one cares, Tom baby. More interesting is why anyone would put BSG
after their name. What, by the way, does it indicate? In Latin does it
equate to 'Pompous Asshole', perchance?
--
Charles Riggs
Charles Riggs
BSG) wrote:
>"Carmen L. Abruzzi" <carmenl...@yahoo.com> writes:
>
>> But the subjunctive in those languages is not defined by meaning, but
>> by morphology.
>
>Really? Is "legam" subjunctive or not?
>
>> I would think it's so that he knows what you're talking about. As you
>> state, your meaning of "subjunctive" is a rather loose semantic one,
>> and does not depend on morphology as its primary defining
>> characteristic. Thus, your "subjunctive" for English, is quite a
>> different category from what "subjunctive" means in Latin, or Greek,
>> or Sanskrit, where it is easy to define the subjunctive, since it's a
>> morphological category.
>
>Really? Is "legam" subjunctive or not?
>
>Do you have a recipe for reliably telling whether a given usage of
>"legam" is subjunctive that doesn't depend on semantics and meanings?
>
>> The Latin subjunctive is a morphological category, it is easily
>> defined.
>
>Really? Is "legam" subjunctive or not?
You're repeating yourself, Tom old bean. Your question was out of
place and boring enough the first time around.
--
Charles Riggs
Neeraj Mathur
Right, first things first - thanks to all for getting this back into a
scholarly discussion, rather than a flame-war.
"Thomas Bushnell, BSG" <tb+u...@becket.net> wrote in message
news:873c2s2...@becket.becket.net...
> This is surely true, in so far as it goes. The problem is that the
> Latin subjunctive, for example, also has many meanings for different
> uses of the subjunctive. Now that's not strictly parallel: in Latin
> we have one construction with different uses, but in English we have
> different constructions with different uses. So this would seem to
> argue for a strictly morphological usage of such terms.
>
> But that fails, as I've already explained, when you consider that the
> morphology is ambiguous. "Legam" is either a subjunctive or not,
> depending on its use and meaning in context.
It suffers from no such failure. You seem to be insisting that, since Latin
morphology does not produce unique forms, it must be jettisoned as a means
of recognising linguistic features. The question to ask is not 'what is this
particular verbal form' but rather 'what does morphology allow this verbal
form to be'. With 'legam' you conclude it is either a future or a
subjunctive - to the historical linguist, this is splitting an incredibly
fine hair (since the development of these two categories from IE tells us
interesting things about their semantics). The virtual identity of these two
categories, when viewed diachronically backwards, is in fact proved by the
fact that the Latins replaced the form for one with the other: the fact that
they are identical is not an accident.
Anyway, I think we need to remember that to a native speaker, the process of
speaking is productive. The speaker decides to use 'the subjunctive' - how
does he do so? This is simple: the subjunctive is the name for a
morphologically defined set of verbal forms, so he applies the relevant
morphology and produces a morphologically-marked subjunctive. When he
thought of 'the subjunctive' in the first place, he was thinking of this
morphological set and nothing else.
The 'subjunctive' was so named by ancient grammarians on the basis of
morphology; even today the word remains the name for a morphological set.
The fact that morphology does not produce unique forms is perhaps a flaw of
the language, but does not mean that 'subjunctive' is the name for anything
other than a morphological set. Or put it this way: what allows you to call
'legam hunc librum' subjunctive but 'lego hunc librum' indicative? Nothing,
if not the morphology: these are morphological labels only. The fact that
the morphology of 'legam' can be analysed in two different ways cannot be
used to argue that a morphological category does not exist.
> The Latin subjunctive cannot be defined morphologically. It almost
> can, but such a definition fails. We agree that "amem" is a
> subjunctive, but what about "legam"?
Let me summarise what I wrote above: the morphology of 'legam' allows it to
be classified as a subjunctive. It also allows it to be classified as a
future indicative - for no reason other than that the Latins identified the
two categories and used one form for the other (ie, the future is modal in
origin). The subjunctive *is* defined morphologically, but the morphology
does not produce unique forms from a synchronic perspective. I do not
believe that this constitutes a failure of the morphological definition in
Latin.
> Likewise, the Latin perfect
> cannot be defined morphologically; is "amatus sum" perfect or not? It
> simply depends on context and meaning. There is no escaping this.
>
> So that means that such terms *cannot* have such rigid bounds as being
> strictly morphological, but instead must have a more fluid application
> which takes into account morphology, syntax, and semantics.
I object to your lumping of 'subjunctive', 'perfect', 'passive' etc. under
the heading 'such terms' above. The subjunctive is defined morphologically.
The perfect passive is not. This is because the 'perfect passive' category
is late (from an IE perspective), and was created by a reorganising of the
tense-aspect system of the language at a time when morphological forms were
not being created anymore. Consider: just as the two uses of 'legam' betray
their original identity, so the two interpretations of verbal adj in -to-
demonstrates that the 'perfect passive' category was created by a
grammatical reanalysis of an existing grammatical (not morphological)
construction. Proof of this is that, even when being used as a perfect
passive, the 'sum' is still considered copulative and is left out, as would
not be possible for a morphological category. (I do not have the data with
me, but I would venture to expect the frequencies for the presence of 'sum'
with the adj in -to- to exactly match its frequency with other adjectives).
In other words, the fact that Latin has an analytic perfect passive cannot
be used to argue that its subjunctive is not morphologically defined.
> The normal (and I think correct) way to understand "amatus sum" or
> "legam" is to hold that the word is ambiguous, and that the ambiguity
> is resolved by considering which of various meanings is more likely.
> But that view requires saying that "in this sentence, 'legam' is
> subjunctive," and that determination simply is not based purely on
> morpohological considerations.
Restating my view for legam only, amatus sum being dealt with above: it is
morphology which must justify mood labels in Latin, and nothing else. When
considering 'legam' we must conclude that it is either a future or a
subjunctive, because the morphology will permit only these two analyses.
That the morphology is not unique, that it allows two analyses forms a
diachronic question but not a synchronic defeat.
> I once explained Latin perfect verbs to a Spanish speaker by saying
> that the Latin perfect has two reflexes in Spanish: the Spanish
> analytic perfect tenses, and the preterit. (My use of the word
> "reflex" there is probably not correct sensu stricto, but you get my
> point.)
The Latin perfect can be translated in two ways into Spanish; the Latin
subjunctive can be translated in many ways into English. I am not sure what
you are arguing here, but I see no reason for saying that this constitutes a
single grammatical category of subjunctive in English any more than a single
category of perfect in Spanish.
> What about the *lack* of morphological contrast between "he loved" and
> "he loved", the former subjunctive and the latter not? Surely we do
> not want to say that there is no subjunctive past! So there must be
> something other than morphological considerations going on.
I am not convinced that we need a subjunctive past in English. 'Would that
he love her, that I need not marry her myself' - what is its 'past' tense?
'Would that he had loved her, that I need not have married her myself' does
not seem to me (if it is correct; I think you cannot say this) to have a
subjunctive. Why would we insist on more than one tense for the subjunctive
in English?
Neeraj
John W. Kennedy
necessary to point out the error on both sides, it will be done." But I
see no one has. Frankly, were it any more obvious, it would bite you.
--
John W. Kennedy
"Only an idiot fights a war on two fronts. Only the heir to the throne
of the kingdom of idiots would fight a war on twelve fronts"
-- J. Michael Straczynski. "Babylon 5", "Ceremonies of Light and Dark"
Rolleston
[...]
>It suffers from no such failure. You seem to be insisting that, since Latin
>morphology does not produce unique forms, it must be jettisoned as a means
>of recognising linguistic features. The question to ask is not 'what is this
>particular verbal form' but rather 'what does morphology allow this verbal
>form to be'. With 'legam' you conclude it is either a future or a
>subjunctive - to the historical linguist, this is splitting an incredibly
>fine hair (since the development of these two categories from IE tells us
>interesting things about their semantics). The virtual identity of these two
>categories, when viewed diachronically backwards, is in fact proved by the
>fact that the Latins replaced the form for one with the other: the fact that
>they are identical is not an accident.
I wonder whether the similarity of some of the perfect
subjunctive and future perfect forms is accidental or not.
R.
Rolleston
[...]
>> But that fails, as I've already explained, when you consider that the
>> morphology is ambiguous. "Legam" is either a subjunctive or not,
>> depending on its use and meaning in context.
>
>It suffers from no such failure. You seem to be insisting that, since Latin
>morphology does not produce unique forms, it must be jettisoned as a means
>of recognising linguistic features. The question to ask is not 'what is this
>particular verbal form' but rather 'what does morphology allow this verbal
>form to be'.
I hope I haven't caused all this trouble by suggesting that TB give
us criteria for identifying the English subjunctive. It was not my
intention to suggest that there are perfect identification criteria
that one can apply in every case. However, one needs to be able
to identify a sufficient number of examples of the blasted thing to
formulate rules that will define the class.
R.
Neeraj Mathur
"John W. Kennedy" <jwk...@attglobal.net> wrote in message
news:K%3Tc.23943$js3.8...@news4.srv.hcvlny.cv.net...
> As I read this thead, I found myself saying to myself, "If it be
> necessary to point out the error on both sides, it will be done." But I
> see no one has. Frankly, were it any more obvious, it would bite you.
My bad, being unbitten. I'm not sure if I've been fully with my wits, or if
I've let myself say things that don't really formulate an argument, or if
I'm just wrong - any or all are likely enough. But do go ahead, and give us
the error on both sides; I for one am most receptive.
Neeraj Mathur
Neeraj Mathur
> I wonder whether the similarity of some of the perfect
> subjunctive and future perfect forms is accidental or not.
I'm trying to find a hard copy of a tutorial essay I wrote last year on the
subject, which summarised a lot of the historical development (I don't want
to have to go back to the library and search through Sihler, Ernout and
Meillet once again), but the laptop with the stuff I need was stolen on New
Year's and I'm having difficulty going through the rubble from my last move
to find it.
Indo-European, amongst its many stem-forming suffixes, had ones which
provided the origin of what became the Greek and Sanskrit subjunctive and
optative. Sihler argues, on the basis of distribution in Vedic and certain
other observations, that these were not really 'moods' of the verb as such
but rather suffixes added to the root directly and that formed stems in the
same way that causatives, desideratives, inceptives etc. did.
The suffix *-yeH-/-iH- (ablaut alternations; the H is a laryngeal) is called
the optative suffix, because it produces the mood called optative in Greek
and Sanskrit (eg. with 'be': *Hs-yeH-m > Grk eien). The suffix *-e-/-o-
(which is identical to the thematic vowel; if the verbal root had a thematic
vowel it seems to have fused with it and produced a lengthened thematic
vowel) is called the subjunctive suffix, because it produces the mood called
subjunctive in Greek and Sanskrit (eg. with 'be': *Hes-o-H > Grk ew, found
in Ionic).
In Latin, these two suffixes are still found: with 'be', for instance, the
IE subjunctive gives *Hes-o-H > Lat. ero. The IE optative gives *Hs-yeH-m >
Lat. siem. Siem is the archaic form (attested); the Classical forms show
levelling to the grade of the plural, giving sim.
In other words, the IE 'subjunctive' has become, by the time of Classical
Latin, a future. The IE 'optative' is what we call the Latin subjunctive.
Both modal forms have been analogically extended far beyond their original
forms, to become a full-fledged tense in one case, and mood in the other.
But in origin, these two appear to have been fairly similar modal suffixes
which formed IE stems.
There is, however, another type of Latin subjunctive, which has nothing to
do with the IE optative marker *yeH: this is the subjunctive in long -a-.
This also appears in Celtic, and a similar morpheme forms a past tense in
Balto-Slavic: Sihler thinks the Italic and Celtic forms are independent of
the Balto-Slavic and go back to another IE stem-forming suffix in long -a-
(or *-eH2- in IE terms) which also had a modal, optative-like meaning;
Szemerenyi groups them all and says that they were a type of IE preterite
which acquired modal characteristics in Celtic and Italic, to become a
subjunctive in Latin.
Now, for the disputed 'legam'. Its morphology shows that it come from the
long -a suffix, which most scholars seem to believe had a modal function in
a pre-Latin or Italic stage. From this modal function, it continues as a
subjunctive in Latin (forming the long -a subjunctive paradigm: legam,
legas, legat etc). However, watch what happens to the IE subjunctive form.
The suffix for the thematic verb was, as seen above, a long -e or long -o;
before the laryngeal ending it is -o which is found. Thus, the expected
Latin form would be *leg-o-o, which would contract to *lego - but this is
identical to the present indicative! In other words, the Latin sound changes
made this form indistinguishable from the present. It is thus natural that,
at a time when the modal force was still felt, it would be replaced with a
more overtly modal form - the one with the long -a. So, 'legam' is
originally a modal in long -a; it then comes to replace the inherited form
of the IE subjunctive while this is still modal; later one function, in the
paradigm legam, leges, leget (of which the latter two forms are direct
continuation of the IE long -e/-o suffix, but are not ambiguous with
non-modal forms) became a simple future, while the other function, with the
long -a paradigm, retained its modality to end up as the Latin subjunctive.
The Latin perfect subjunctive is easy: it shows the IE optative ending -iH
(zero grade generalised), added to the perfect marker -is-: thus 2sg
*vid-is-i(long)-s > Lat videris (*i > e by vowel weakening rules; the i is
long in the last syllable). The future perfect I think I was too tired to
tackle at that time, but let's have a look at it. If we have a root, perfect
marker, IE subjunctive -e/-o, and personal endings, we get for the 2sg
*vid-is-e-s, which would give, post vowel-weakening, videris! Alright, so
that one works too!
Why the similarity? Both forms have a root and a perfect marker; the
difference is that the perfect subjunctive shows a long -i- suffix, from the
IE optative -iH, while the future perfect has a short i, from vowel
weakening of an earlier -e/o, the IE subjunctive suffix.
There we go; that should provide some base for the answer to your question.
For the best detail of this, look at Ernout's _Morphologie historique du
latin_ (1953) and Sihler's _New Comparative Grammar of Greek and Latin_; a
peek at Szemerenyi's _Introduction to Indo-European Linguistics_ is always
fruitful, if nothing else than for his extensive bibliographical references.
Neeraj Mathur
Skitt
> (Thomas Bushnell, BSG) wrote:
>>>> Be as formal as you wish. You can use lambda calculus for all
>>>> I care. The important thing is that you make clear what you
>>>> mean by the word "subjunctive", because that word is used
>>>> in very different ways by different people at different times.
>>>
>>> Whereas the Latin subjunctive is pretty easily identified
>>> morphologically, no?
>>
>> No. Is "legam" subjunctive or not?
>
> No-one cares, Tom baby. More interesting is why anyone would put BSG
> after their name. What, by the way, does it indicate? In Latin does it
> equate to 'Pompous Asshole', perchance?
Hmm. British Society of Gastroenterology? British Society of Gerontology?
Broadband Stakeholder Group? Bingo Sales Group? Bison Specialist Group?
Argh, there's just too many possibilities, many of them self-deprecating.
--
Skitt (in Hayward, California)
www.geocities.com/opus731/
Thomas Bushnell, BSG
> It suffers from no such failure. You seem to be insisting that, since Latin
> morphology does not produce unique forms, it must be jettisoned as a means
> of recognising linguistic features.
No, certainly not. My claim is that "subjunctive" is not a purely
morphological category, but a mixed morphological/semantic category.
Without doubt, morphology is very important in understanding it!
> The 'subjunctive' was so named by ancient grammarians on the basis of
> morphology; even today the word remains the name for a morphological set.
Huh? The subjunctive was so named because of syntax, not morpohology,
because it was felt particularly appropriate to subordinate clauses.
Thomas
Thomas Bushnell, BSG
> No-one cares, Tom baby. More interesting is why anyone would put BSG
> after their name. What, by the way, does it indicate? In Latin does it
> equate to 'Pompous Asshole', perchance?
It's a religious order of the Episcopal Church.
Thomas Bushnell, BSG
> You're repeating yourself, Tom old bean. Your question was out of
> place and boring enough the first time around.
My name is Thomas, not Tom. If you cannot do me the simple courtesy
of using my name, and not poking fun at it, then you are beneath
contempt.
Thomas Bushnell, BSG
> "Neeraj Mathur" <neeraj...@chch.ox.ac.uk> writes:
>
> > It suffers from no such failure. You seem to be insisting that, since Latin
> > morphology does not produce unique forms, it must be jettisoned as a means
> > of recognising linguistic features.
>
> No, certainly not. My claim is that "subjunctive" is not a purely
> morphological category, but a mixed morphological/semantic category.
> Without doubt, morphology is very important in understanding it!
Here's a more specific account:
I believe that "subjunctive" is a semantic category, and a
morphological category, both. (Actually, it might well be multiple
semantic categories; I'll speak as if it's just one for simplicity's
sake. I have no particular opinion about that, and it doesn't affect
my argument.)
A note about history: the prehistory of the language cannot answer
this question. The speakers of Latin (or any language) were generally
ignorant of the prehistory, and indeed, a common engine of linguistic
change is the reanalysis of one construct as another, which could only
really happen if the history were unknown. I am interested in how
Latin (and other languages) actually work, and how they came to be as
they are is a secondary and different question; their history only
affects how they work through the medium of the actual language as it
exists (existed), and if the history is not reflected in the present
sense of the speakers, it can have no other effect.
It therefore does not matter how recent the perfect passive
construction is; nor did I use "perfect passive" as an example; my
example was "perfect". So onward:
The purpose of the morphological subjunctive is to communicate the
semantic subjunctive. For this to work, it must be different from the
indicative sufficiently often to avoid confusion. What counts as
"sufficiently often" can be well less than half: for example, the
Latin dative and ablative are distinguished in no plurals, and in no
second declension nouns; so well under half the forms distinguish them
at all. But this is plenty to distinguish.
(By contrast, the distinction of English past subjunctive forms and
past indicative exists only in the one verb "to be", and this does not
seem to be enough, so that the subjunctive "were" is rapidly being
replaced by indicative "was" for the first and third person, and when
that process is complete, we can say that the past subjunctive in
English is gone.)
Some words are ambiguous, so that "legam" could be future indicative
or present subjunctive (and the history of the Latin subjunctive and
future tenses, while interesting in explaining the origin of the
ambiguity, does not explain in itself the way the language works in
situ). That means that legam maps to either of two possible meanings:
a future indicative meaning, and a present subjunctive meaning.
But that only makes *sense* if there *is* a "present subjunctive
meaning" (or perhaps more than one). If subjunctive is a purely
morphological category, with no semantic meaning at all, then there is
no sense at all to understanding the situation that way. (And I have
no clue how you are supposed to describe the ambiguity--perhaps you
have to say that there is none, or wax poetic about how the future
tense used to be subjunctive and what not--considerations which we
know played no role in the actual brains of Latin speakers, since they
were mostly ignorant of such facts.)
Now it's not a purely semantic category, of course! I would never
assert that morphology plays no role; obviously it does. Heck, if we
want to mark up a table of "uses of the Latin subjunctive" the only
responsible way to do that is to tote up a bunch of uses where there
is no ambiguous morphology and then to construct our table. (And
after we've done that, we can even begin including ambiguous
examples.) Of course that's the procedure, and its validity is
sufficient to make the case that "subjunctive" is primarily a
morphological category.
But "primarily" and "exclusively" are worlds apart, and what I am
saying is that there is such a thing, in Latin, as the (or perhaps
more than one) semantic subjunctive, which is the semantic meaning (or
meanings) conveyed by the use of the subjunctive, and which are
conveyed sometimes by ambiguous words like "legam".
Likewise, the systematic ambiguity between dative and ablative plural,
combined with the relative ease of disambiguating, tells us that
dative and ablative are not exclusively morphological categories, but
rather mixed morphological and semantic categories.
Now if in saying that the subjunctive is a strictly morphological
category you mean that you can pick out subjunctives on strictly
morphological grounds, you are wrong. If you admit that you must
attend to semantics--to some notion of subjunctive meaning or
meanings--to pick out at least some subjunctives, you would be
right--at which point, what is the sense of this strictly
morphological category talk?
Thomas
Thomas Bushnell, BSG
> Anyway, I think we need to remember that to a native speaker, the process of
> speaking is productive.
Of course; but what about the process of listening? How does the
hearer disambiguate moods? Does he do so purely on morphological
grounds, or does he do so with a combination of morphological and
semantic grounds?
And lest you say he doesn't have to, I'll return you to that test
about pluralizing. The listener might, say, have to restate the
speaker's sentence in the plural: and in that case, he certainly does
have to.
> The speaker decides to use 'the subjunctive' - how does he do so?
> This is simple: the subjunctive is the name for a morphologically
> defined set of verbal forms, so he applies the relevant morphology
> and produces a morphologically-marked subjunctive. When he thought
> of 'the subjunctive' in the first place, he was thinking of this
> morphological set and nothing else.
Why did the speaker use "the subjunctive" in the first place? Was it
because he likes the sound of such verbal forms, or was it because
those forms are associated with particular meanings that he wished to
communicate?
When he did the ordinary task of pluralizing a sentence, did he not
(implicitly, at least) consider: "This use of legam is a subjunctive"?
And how on earth did he know that if he were only attending to
morphological criteria? He was attending to *both* morphology and
semantics to do it.
He models the speaker's head, thinking (at least implicitly): "What is
the meaning this speaker is trying to convey?" recognizing that the
words the speaker used could admit of multiple meanings, and that
context and "what is the likeliest meaning" and such other
considerations will disambiguate for him.
So he thinks (at least implicitly): "The speaker is using a
deliberative subjunctive" or "The speaker is using the future
indicative", and his discrimination CANNOT be based in every case only
on morphology, for there are cases where morphology cannot tell the
difference.
And he reliably, without doubt, produces either "legamus" or "legemus"
when pluralizing, fitting the semantics exactly, indicating that this
is not a grammarians phantom, but a genuine feature of the language.
In other words, I don't disagree with anything you say, except what
you say does nothing to prove that "subjunctive" is a strictly
morphological category. It shows, instead, that it is primarily a
morphological category, something I have never denied.
Thomas
Neeraj Mathur
> tb+u...@becket.net (Thomas Bushnell, BSG) writes:
> >
> > No, certainly not. My claim is that "subjunctive" is not a purely
> > morphological category, but a mixed morphological/semantic category.
> > Without doubt, morphology is very important in understanding it!
>
> Here's a more specific account:
Okay, in this, you have convinced me of your conclusion, and I have doubts
of whether I understood your original point. Conceded.
There is one point you make, however, which I would challenge; I don't think
it affects your conclusion, but why not talk about it nevertheless?
First, indulge me in a case study (with all credit going to Anna
Morpurgo-Davies, who used it in lecture). English has three negating
prefixes: un-, im-, and a-. I am going to invent two adjectives; you tell me
what you, as a native English speaker, would intuitively suggest is the
negative form of it. Just tell me which you think feels most natural as its
negative.
Example:
elegant > inelegant, not unelegant or anelegant.
Okay, here are my two invented English adjectives:
1) bilitic
negative:
2) blick
negative:
Did you answer abilitic, or perhaps imbilitic, but unblick? Most people in
the lecture room did, and virtually everybody I've tried it out on, linguist
or not, has given unblick, with abilitic being most common for the first
one.
What conclusion do we draw from this?
The three prefixes are derived, of course, from Germanic (un-), Latin (in-),
and Greek (a-). What is amazing is that English speakers seem to have an
intuitive, subconscious ability to decide which of these three provides the
most likely etymology for any given word, and then choose the negating
prefix to match. At any rate, they are certainly able to decide whether or
not a word has a Germanic ancestry.
Another example of this ability can be found in accent patterns. How would
you pronounce my first and last names? Which syllables do you stress? I've
listened to people deal with my name quite attentively over the years; it is
most interesting to note that the vast majority of people (particularly in
North America, but the phenomenon is visible on both sides of the Atlantic)
stress the second syllable of each: niRAZH maTHUR (the last name also gets
said MAYther quite a bit). This is intriguing: it seems that English
speakers, realising that the name is not in any way a part of the inherited
Germanic tradition of their language, try to display this knowledge. The
most common non-Germanic source of words in English is French, and so they
accent the names as if they were French words. Associated with this is the
fact that they make the French 'zh' sound, instead of the 'normal' sound of
English 'j'. (By the way, the names are Hindi, of Sanskrit origin, and are
both accented on the first syllables.)
Why do I mention all this? Because this demonstrates that English speakers
are not completely oblivious to features of their language that relate to
diachronic linguistics. Therefore, your statements that "if the history is
not reflected in the present
sense of the speakers, it can have no other effect" and "the history of the
Latin subjunctive and future tenses, while interesting in explaining the
origin of the ambiguity, does not explain in itself the way the language
works in situ" are based on an assumption which is incorrect - namely, that
synchronic linguistics can be completely separated from diachronic
linguistics. Anna Morpurgo-Davies argued in her lecture, and I suggest from
my name experience, that this is not in fact a tenable position. Diachronic
arguments must be considered, even when discussing issues at a given point
in time.
Now, about 'legam'. How are you able to know in what degree the future tense
in Latin was devoid of modal associations in the minds of its speakers? The
origin of the future is important because we do not have any Romans around
who can tell us if they did in fact conceive of their tenses in the exact,
tabular format that modern grammar books print them as. The fact that the
future was once a modal form may have had a residual presence in the
subconscious of the Romans, and therefore we need to be aware that the
tense-tables that we use to discuss Latin are our constructions as we
attempt to describe their language the way we see it, not an inherent,
verifiable truth. On the other hand, the modal origin of the future is such
a truth, and therefore we must not refuse to consider it when considering
the mood system of Latin.
Okay, that's my piece. Like I said, I'm not sure that this affects your
conclusion; I'm quite happy to accept it after your last few posts, that
'subjunctive' marks a grammatical category as well as a morphological one. I
simply point out that the divorce of synchronic and diachronic linguistics
that you assume is a point of contention.
Neeraj Mathur (that's NEE-rudzh MAH-thur) :)
Peter T. Daniels
Welcome to alt.usage.english. I've no idea why someone crossposted you
thither and hither, but we'd be glad if you went away again.
Carmen L. Abruzzi
> "Carmen L. Abruzzi" <carmenl...@yahoo.com> writes:
>
>
>>No, a consideration of which form of the second or third person (which
>>have distinct fut. ind. and pr. subj. forms) would be used in a given
>>construction will tell you whether "legam" is subjunctive or future
>>indicative in a that construction. You don't need to consider the
>>meaning as such, as this does not determine whether it's subjunctive.
>
>
> I owe a fuller response, to explain better my view of exactly what is
> going on here. First, amend the proposed test to one which considers
> plural instead of singular, as I mentioned already.
>
> So in the sentence, "legam illum librum," I must ask myself: "If I were
> to make that plural, which sentence would I use? And there are of
> course two possibilities:
>
> "Legemus illum librum" (future indicative)
> "Legamus illum librum" (present subjunctive)
>
> Now, the test says: "If you would use 'legemus,' then 'legam' was
> indicative; if you would use 'legamus,' then 'legam' was subjunctive."
>
> And I agree completely: this is the right test. (There may be corner
> cases that it misses on, but it seems basically right to me.)
>
> But how do I know which plural form I would use?
>
> The only test I know of is to ask myself "how was 'legam' being used
> in the original sentence" and then use the appropriate plural.
Of course. This is mere ambiguity. Because you may need to
ask whether the intended meaning is one in which the
subjunctive or the future indicative form is used does not
mean that the meaning determines the subjunctiveness of the
verb.
If we take a somewahat different English example, "I
hit the ball" is
ambiguous as to whether the verb is in the present or the
past. If I write "Yesterday I found a ball lying in the
street. I hit the ball, and it went over the neighbor's
fence." we can say it is past tense. But
if I use the "historical present" and write "Yesterday I'm
walking along the street, and I find a ball. I hit the
ball, and it goes over the neighbor's fence.", we say "hit"
is in the present tense. That it is
referring to a past event does not make it past tense. In
the same way, just because a meaning is one in which a
subjunctive form is normally (or often) used in Latin does
not make that meaning itself "subjunctive".
>
> That is, I say, "In the context, 'legam' looks like a deliberative
> subjunctive, so I will use 'legamus.'"
This is, I believe, where the confusion between semantics
and morphology arises in the classical languages, Latin
especially. '"Legam" looks like a *deliberative*
subjunctive'. "Deliberative" refers to the semantics,
"subjunctive" refers to the morphology. The two have been
mixed to create a hybrid category. This is all well and
good, as its the interplay between morphology and semantics
that makes morphology useful, but it should not obscure the
fact that it's the morphology that determines the mood, not
the semantics. There might well be a way of expressing
deliberative semantics without using the subjunctive mood; I
don't know about Latin, but certainly there is in English:
"I'm thinking about reading this book". One would hardly
say that "reading" is here in the subjunctive mood, even
though it expresses the same meaning as "I'm considering
whether I read this book", wherein I can agree that "read"
could be meaningfully classed as subjunctive.
In Latin grammar, there is, for both verbal forms and
nominal cases, generally a long list of various uses of
these morphological forms, e.g. "ethical dative", "jussive
subjunctive" and of course "deliberative subjunctive". This
often leads to the mistaken idea that the semantics of these
various uses are themselves inherently "dative" or
"subjunctive" or whatever--when all that is really going on
is that dative or subjunctive *forms* are used *in Latin* to
express these semantics.
No, of course morphology cannot tell us whether the verb is
subjunctive present or indicative future when both have the
same form, but it is the *form* and not the meaning, that
determines that the verb is one or the other.
The meaning may disambiguate to the extent that one form is
used for this meaning and the other form for that meaning.
> Sometimes grammar will do, if say the subjunctive can only live
> in one kind of construction, so that we couldn't have a minimal pair
> of ambiguous identical sentences. This is why I presented not just the
> word "legam", but the sentence, "legam illum librum", where you cannot
> tell from the complete sentence what the mood is. Sometimes you might
> have identical surface forms where the parse will tell you the correct
> meaning (perhaps that's the way to understand "amatus sum", as having
> two different parses); but in the case of "legam illum librum" both
> readings have exactly the same parse tree.
>
> The only way to tell the mood of "legam" in "legam illum librum" is to
> consider the context--the meaning.
>
Of course. But it's still the form that determines that the
verb is in the subjunctive, or the future indicative. If
the same meaning could be expressed using another verb form
that form would not therefore become "subjunctive" or
"future indicative".
The idea that the meaning is the primary determinant of
subjunctive mood leads to the classification of such diverse
English forms as "may do", "might do", "could do", "ought to
do" as subjunctive, when in fact there is only one
subjunctive form in English, which, as it happens, is
identical with the indicative in all instances except
third person singular present and the singular past of "be"
(Leave it for another day to discuss the usages of the
subjunctive past and present--they don't normally refer to
actual past and present events respectively). "I suggest
that he do this". "If I were you." So we
must always ask about the context to determine whether an
English verb is subjunctive; however this should not lead us
into thinking that it is the context that determines the
verb's mood. "I suggest that he do this" is subjunctive, "I
suggest that he could/might/ought to do this" are not.
Thomas Bushnell, BSG
> There is one point you make, however, which I would challenge; I don't think
> it affects your conclusion, but why not talk about it nevertheless?
Hey, anything interesting is worth talking about! And this is
certainly a good question.
> First, indulge me in a case study (with all credit going to Anna
> Morpurgo-Davies, who used it in lecture). English has three negating
> prefixes: un-, im-, and a-. I am going to invent two adjectives; you tell me
> what you, as a native English speaker, would intuitively suggest is the
> negative form of it. Just tell me which you think feels most natural as its
> negative.
>
> Example:
> elegant > inelegant, not unelegant or anelegant.
I think I would tolerate unelegant, but not anelegant. I would insist
that "inelegant" is a far more elegant word! :)
> Okay, here are my two invented English adjectives:
>
> 1) bilitic
>
> negative:
I think I would say "abilitic", but I would accept "unbilitic".
> 2) blick
>
> negative:
Only "unblick".
> Did you answer abilitic, or perhaps imbilitic, but unblick? Most people in
> the lecture room did, and virtually everybody I've tried it out on, linguist
> or not, has given unblick, with abilitic being most common for the first
> one.
Bing! I didn't look at all, and I'm delighted at the findings of
regularity.
> The three prefixes are derived, of course, from Germanic (un-), Latin (in-),
> and Greek (a-). What is amazing is that English speakers seem to have an
> intuitive, subconscious ability to decide which of these three provides the
> most likely etymology for any given word, and then choose the negating
> prefix to match. At any rate, they are certainly able to decide whether or
> not a word has a Germanic ancestry.
Well, not quite. After all, we all agree about bilitic, and it's not
because it has a Greek ancestry! Rather, we agree about "what words
sound Greek" and "what words sound Latin" and "what words sound
German", and that agreement is delightful (and this is a very splendid
test of it!).
I also think my toleration of unbilitic and unelegant comes (I'm
guessing) from un- being the "default" negation prefix, and a general
tendency of words to be regularized (witness plurals like
"concertos").
What about the interesting "unchivalrous", where chivalrous is a word
we borrowed from French, but is a German root?
> Another example of this ability can be found in accent patterns. How would
> you pronounce my first and last names? Which syllables do you stress? I've
> listened to people deal with my name quite attentively over the years; it is
> most interesting to note that the vast majority of people (particularly in
> North America, but the phenomenon is visible on both sides of the Atlantic)
> stress the second syllable of each: niRAZH maTHUR (the last name also gets
> said MAYther quite a bit).
Hrm, my guess was NEEraj maTHUR. But I'm pretty attentive to names,
so I'm not at all likely to assume than an unfamiliar name gets a
French-style stress (especially a name that looks about as ungallican
as yours!)
> Why do I mention all this? Because this demonstrates that English speakers
> are not completely oblivious to features of their language that relate to
> diachronic linguistics. Therefore, your statements that "if the history is
> not reflected in the present
> sense of the speakers, it can have no other effect" and "the history of the
> Latin subjunctive and future tenses, while interesting in explaining the
> origin of the ambiguity, does not explain in itself the way the language
> works in situ" are based on an assumption which is incorrect - namely, that
> synchronic linguistics can be completely separated from diachronic
> linguistics.
Ah, no, not quite. I was trying to say something more nuanced, but
your prodding is good. I will try to be more exact.
What I mean is that there is no mystical ability of English speakers
to guess where a word came from. They must be reacting to, for
example, very subtle sound-pattern guesses (in the case of your first
example). They know a whole bunch of words in -ic, for example, and
those tend to get a- negations. So my claim is not that the past is
irrelevant, but that any relevance it has must come through features
of the language in the present.
> Now, about 'legam'. How are you able to know in what degree the future tense
> in Latin was devoid of modal associations in the minds of its
> speakers?
It's awfully hard to tell, I grant you, and the tendency of
subjunctives and future indicatives to swap places is very telling;
indeed, the semantic reasons for this swapping seem pretty clear--the
meanings are just very close, and if you are looking around for a
word, you might well bend one to the other.
Still, there are two plurals.
But my point bears up even so, because it was a point about a more
general use of morphological terms as if they were purely exactly and
only morphological. The case of words like "abluit" (perfect or
present?) or "hominibus" (dative or ablative?) make the same point.
Now something I have heard people mean when they say that subjunctive
is a purely morphological category is that they think there is no
single semantic field for it, but that the subjunctive covers a
variety of disjoint semantic domains. So that if the question is "what
do all the uses of the subjunctive have in common," they can only say,
"morphology." That answer may well be correct, but it wasn't what I
was criticizing in arguing that the subjunctive is not a purely
morphological category.
Thomas
Thomas Bushnell, BSG
I do not disagree with anything in the substance of what you wrote!
My only quibble (and perhaps it is only that), is that it is a
decision that must be made in deciding how one will describe the
language which terms should be used.
The unanalytic subjunctive is of course as you say; but the question
is "why should we not say that 'would eat' is not also a subjunctive?"
I do not claim that it is a subjunctive merely because it falls into
this or that semantic domain.
Why do we call "amem" a subjunctive in Latin? Didn't it arise from
some other form, that was not a subjunctive? At some point, a form
that was not subjunctive got relabelled as one, because the language
had changed.
So it looking at a language, we see a bunch of different words (or
analytic constructions), occupying a certain semantic domain, and we
say "those are the subjunctive words" or "those are the verbs in the
passive mood". We do that by a general sense of the semantic domain
that "subjunctive" or "passive" is supposed to cover, and by an
expectation of morphological regularity (not absolute regularity,
however), and by comparison with other languages.
Doing that should yield, in my opinion, should result in the view that
constructions with the modal auxiliaries "would," "could," and
"should" ought to be regarded as subjunctives. Are they the same as
the unanlytic subjunctives? No.
Is there a rule that the subjunctive must never be an analytic
construction? Where did that rule come from? Is it the rule that we
can only call a subjunctive what comes from a subjunctive in the
parent language? (Why does the same rule not apply to perfects and
passives then?)
What we have is that English grammarians used to speak of only the
unanalytic constructions as subjunctive, and would ignore or say
wildly incorrect things about the modal auxiliaries (for example, that
would/could/should are just ordinary past tenses of will/can/shall).
Then people started referring to "would eat" as a
subjunctive-equivalent (telling phrase, no?) and then more recently,
simply as a subjunctive. You can even see a fascinating snapshot of
this in progress if you look at the OED entry for "subjunctive," and
especially the last quote listed under "subjunctive-equivalent."
I was criticized for following that usage, as if I had made some
foolish mistake, when in fact, it is simply a shift in usage of the
application of "subjunctive" to this or that construction. I am not
the only one who has decided that a clearer and more felicitous
description is to widen the application of the word's use.
Thomas
John W. Kennedy
(Sigh!)
I did. Whether it be the case or not that most English subjunctives are
analytical, it were an error to say that all are.
--
John W. Kennedy
"Compact is becoming contract,
Man only earns and pays."
-- Charles Williams. "Bors to Elayne: On the King's Coins"
Charles Riggs
BSG) wrote:
'Beneath contempt', oh my, how easily some people's feathers are
ruffled. If worse than contemptible, there must be a number of us, at
least three of them being quite esteemed `round here, since they
normally write Coop, T**y C**per, Pooper or dickhead, those for one or
two here, when writing in response to one his posts or talking about
him in another. I guess you're too new to this group, having only been
crossposted into it, to have learned about some of its conventions.
But should you become a regular fixture here in AUE, I'll be sure to
call you Thomas from then on, since you obviously feel so strongly
about it. I'm more used to and prefer Tom or Tommy, Thomas, no doubt,
being a name I've never run across in Ireland, other parts of Europe,
Mexico, Japan, Canada, or America.
--
Charles Riggs
Rolleston
>Neeraj Mathur wrote:
>> "John W. Kennedy" <jwk...@attglobal.net> wrote in message
>> news:K%3Tc.23943$js3.8...@news4.srv.hcvlny.cv.net...
>>>As I read this thead, I found myself saying to myself, "If it be
>>>necessary to point out the error on both sides, it will be done." But I
>>>see no one has. Frankly, were it any more obvious, it would bite you.
>
>> My bad, being unbitten. I'm not sure if I've been fully with my wits, or if
>> I've let myself say things that don't really formulate an argument, or if
>> I'm just wrong - any or all are likely enough. But do go ahead, and give us
>> the error on both sides; I for one am most receptive.
>
>(Sigh!)
>
>I did. Whether it be the case or not that most English subjunctives are
>analytical, it were an error to say that all are.
Ahem, cough: http://tinyurl.com/3rmxk
R.
Thomas Bushnell, BSG
> (Sigh!)
>
> I did. Whether it be the case or not that most English subjunctives
> are analytical, it were an error to say that all are.
Oh, and I certainly wouldn't say that.
Rolleston
[...]
>assert that morphology plays no role; obviously it does. Heck, if we
>want to mark up a table of "uses of the Latin subjunctive" the only
>responsible way to do that is to tote up a bunch of uses where there
>is no ambiguous morphology and then to construct our table. (And
>after we've done that, we can even begin including ambiguous
>examples.)
Is there going to be anything in your table that is not a traditional
subjunctive _form_?
R.
Thomas Bushnell, BSG
> But should you become a regular fixture here in AUE, I'll be sure to
> call you Thomas from then on, since you obviously feel so strongly
> about it. I'm more used to and prefer Tom or Tommy, Thomas, no doubt,
> being a name I've never run across in Ireland, other parts of Europe,
> Mexico, Japan, Canada, or America.
I'll give you the easy foolproof way to figuring out what name to call
a person: use the name they use. No need to guess, you can just
notice that I always use Thomas, and conclude that this is the safest
strategy.
I find it curious that the only people I have ever run in to who wish
to mangle my name and think it is some curiously peculiar request that
they not do so are from the British Isles. Maybe this is a cultural
thing. But people in North America very frequently care how you refer
to them, and don't care for making up nicknames.
It might be insignificant to you, but on this side of the pond, at
least, it is really a basic human courtesy.
Thomas Bushnell, BSG
I don't know what you mean by a "traditional subjunctive _form_"; I
can guess but rather than say that, I would put on the table words
where there is no ambiguous morphology at work (so "legam" or
"amaverit" would not be on the table).
I thought I said that.
Thomas
Dylan Nicholson
>
> I find it curious that the only people I have ever run in to who wish
> to mangle my name and think it is some curiously peculiar request that
> they not do so are from the British Isles. Maybe this is a cultural
> thing. But people in North America very frequently care how you refer
> to them, and don't care for making up nicknames.
>
You think the UK's bad, try Australia. You know that you haven't been
accepted by the locals until they make up a name for you.
In your case, most likely Thommo, but it could very well be something not
obviously related to your name, like say...Spud (T.B. = tuber-culosis,
tuber=potato=spud).
Rolleston
[...]
>> subjunctive _form_?
>
>I don't know what you mean by a "traditional subjunctive _form_";
Forms of the kind that occur in traditional verb conjugation tables
under "present subjunctive", "imperfect subjunctive", etc. etc. etc.
R.
Thomas Bushnell, BSG
Some of them: the ones where there is no ambiguous morphology at work.
But as I said, that does not mean that the subjunctive is defined by
those forms, or that only those forms could be subjunctive (after all,
ambiguous ones have to be left off).
The question is: how does one draw up that table of verb conjugations?
How do you decide that "amaverit" is ever any kind of subjunctive?
You do it by comparing the uses of amaverit (some of them only, as it
happens, for some are indicative) to uses of amet and amavisset and
amaret, and you conclude that amaverit must sometimes be a
subjunctive.
And that happens to be a semantic test...
Charles Riggs
BSG) wrote:
>It might be insignificant to you, but on this side of the pond, at
>least, it is really a basic human courtesy.
No, Thomas, it is not insignificant to me. I think it crass to mangle,
as you put it, a person's name. I didn't think I'd done much mangling
when converting the more formal Thomas to the friendlier, to my ears,
Tom. For whatever reason, you clearly prefer Thomas, so Thomas it is
for me.
--
Charles Riggs
david56
I can counter that. Living and working in the USA in the 70s, I
found it impossible to get any local to call me "David", and had to
settle for being "Dave" while over there.
--
David
=====
Matti Lamprhey
>
> I can counter that. Living and working in the USA in the 70s, I
> found it impossible to get any local to call me "David", and had to
> settle for being "Dave" while over there.
Even by Jews?
Matti
david56
Oh yes indeed. Since I was working in a Jewish summer camp, 90% of
the people I met were Jews.
There being many different Davids, each was given an identifier to
indicate their position in the local society. We had Dave Swim, Dave
Art, Dave Basketball, etc. For two complete summers I was Dave Sail.
--
David
=====
Harvey Van Sickle
> Thomas Bushnell, BSG typed thus:
>> I find it curious that the only people I have ever run in to who
>> wish to mangle my name and think it is some curiously peculiar
>> request that they not do so are from the British Isles. Maybe
>> this is a cultural thing. But people in North America very
>> frequently care how you refer to them, and don't care for making
>> up nicknames.
>>
>> It might be insignificant to you, but on this side of the pond,
>> at least, it is really a basic human courtesy.
>
> I can counter that. Living and working in the USA in the 70s, I
> found it impossible to get any local to call me "David", and had
> to settle for being "Dave" while over there.
For what it's worth, I've found the situation to be as Thomas has
experienced it.
Here in the UK (and in her native NZ), my wife's name is very often
shortened from "Susanne" to "Sue" -- although she doesn't introduce
herself with the short form, and uses it with just a handful of long-
time friends.
She's so used to it in the UK and NZ settings, though, that when she
was in Canada and people called her as she introduced herself, it came
as quite a surprise.
I can also offer another bit of circumstantial evidence in support of
presumptuous shortening being more common in the UK than in NA. I'm
quite certain I've heard people in the UK taking the mickey out of
Americans who introduce themselves by saying "I'm David Jones; call me
Dave": behind the guy's back, such people are sometimes referred to as
"David-call-me-Dave-Jones".
What such mickey-takers invariably miss is the corollary: if David
*hasn't* said "call me Dave", you've not been given permission to do so
-- and it is thus ill-mannered to call him anything other than "David".
--
Cheers, Harvey
Ottawa/Toronto/Edmonton for 30 years;
Southern England for the past 22 years.
(for e-mail, change harvey.news to harvey.van)
John Briggs
Has it not occurred to you (or your wife) that if you are associating with
people who do not address you as you would wish, that you are perhaps
associating with the wrong sort of people? Have you no choice in the
matter?
--
John Briggs
Harvey Van Sickle
I've found these tend to be people one meets through --- and must deal
with at -- work.
So, no: there's virtually no choice.
John Briggs
Your willingness to undertake such uncongenial employment is, perhaps,
commendable.
--
John Briggs
Keith Edgerley
"John Briggs" <john.b...@ntlworld.com> wrote in message
news:eqITc.108$Bk...@newsfe5-gui.ntli.net...
> Harvey Van Sickle wrote:
> > On 15 Aug 2004, david56 wrote
> >> Thomas Bushnell, BSG typed thus:
> >
> > -snip-
> >
> >>> I find it curious that the only people I have ever run in to who
> >>> wish to mangle my name and think it is some curiously peculiar
> >>> request that they not do so are from the British Isles. Maybe
> >>> this is a cultural thing. But people in North America very
> >>> frequently care how you refer to them, and don't care for making
> >>> up nicknames.
> >>>
> >>> It might be insignificant to you, but on this side of the pond,
> >>> at least, it is really a basic human courtesy.
> >>
> >> I can counter that. Living and working in the USA in the 70s, I
> >> found it impossible to get any local to call me "David", and had
> >> to settle for being "Dave" while over there.
> >
> > For what it's worth, I've found the situation to be as Thomas has
> > experienced it.
> >
> > Here in the UK (and in her native NZ), my wife's name is very often
> > shortened from "Susanne" to "Sue" -- although she doesn't introduce
> > herself with the short form, and uses it with just a handful of long-
> > time friends.
> >
> > She's so used to it in the UK and NZ settings, though, that when she
> > was in Canada and people called her as she introduced herself, it came
> > as quite a surprise.
> >
Call me old-fashioned, but I expect people I have only just met to address
me by my surname prefixed with Mr.
I am always astounded when I get phone-calls from English-speaking strangers
who address me as Keith, as is or in a familiar form.
Mostly I speak French in everyday life and there, of course, no-one would
dream of calling me by my forename at a first meeting, even new colleagues,
with one exception: jazz musicians traditionally say "tu" to each other and
use first names from the outset.
--
Keith Edgerley
owe war sint verswunden
alliu miniu jar
John Briggs
In that case, it would probably a good idea to avoid being awarded a
knighthood, because then people get to use your forename whether you like it
or not :-)
The distinguished American classicist M.I. Finley taught for many years at
the University of Cambridge, and became a British citizen. His career
culminated in a knighthood, and he became, possibly to his surprise, Sir
Moses Finley.
Sir Yehudi Menuhin had less of a problem in that regard, but he solved it by
becoming Lord Menuhin, OM. Although what his formidable mother, who had
saddled him with the name "Yehudi" in the first place (and nearly outlived
him), thought of that is not recorded.
--
John Briggs
Harvey Van Sickle
>> Call me old-fashioned, but I expect people I have only just met
>> to address me by my surname prefixed with Mr.
>> I am always astounded when I get phone-calls from
>> English-speaking strangers who address me as Keith, as is or in a
>> familiar form.
-snip-
> In that case, it would probably a good idea to avoid being awarded
> a knighthood, because then people get to use your forename whether
> you like it or not :-)
>
-snip-
> Sir Yehudi Menuhin had less of a problem in that regard, but he
> solved it by becoming Lord Menuhin, OM.
Um...*he* didn't solve it: it was solved for him when he received an
appointment to the Lords.
(As for Lordship titles, try pronouncing "Lord Lloyd-Webber" out loud a
few times in a row: it doesn't roll off the tongue.)
John Briggs
> On 15 Aug 2004, John Briggs wrote
>> Keith Edgerley wrote:
>
>>> Call me old-fashioned, but I expect people I have only just met
>>> to address me by my surname prefixed with Mr.
>>> I am always astounded when I get phone-calls from
>>> English-speaking strangers who address me as Keith, as is or in a
>>> familiar form.
>
> -snip-
>
>> In that case, it would probably a good idea to avoid being awarded
>> a knighthood, because then people get to use your forename whether
>> you like it or not :-)
>>
> -snip-
>
>> Sir Yehudi Menuhin had less of a problem in that regard, but he
>> solved it by becoming Lord Menuhin, OM.
>
> Um...*he* didn't solve it: it was solved for him when he received an
> appointment to the Lords.
>
He chose to accept a peerage. Those who don't like titles (such as the late
Francis Crick) just accept the OM.
--
John Briggs
Harvey Van Sickle
But but but -- and assuming there was a "problem" in the first place:
if it hadn't been offered, he couldn't have accepted it.
He was the passive side of the "problem solving": somebody else came
up with a solution, and his actions were limited to accepting or
rejecting that solution.
(But it undoubtedly wasn't a problem anyway...)
Peter T. Daniels
Jeez, you talk nonsense at a.u.e. Could you _please_ stop the
crossposting?
--
Peter T. Daniels gram...@att.net
Thomas Bushnell, BSG
> Call me old-fashioned, but I expect people I have only just met to
> address me by my surname prefixed with Mr. I am always astounded
> when I get phone-calls from English-speaking strangers who address
> me as Keith, as is or in a familiar form.
In my case, I sign all my posts with "Thomas" so I could hardly
complain about that. If you want a title, then I'm "Br. Thomas", but
just "Thomas" will do. The only time I get miffed if people use my
first name uninvited is when they are reading it from a credit card
receipt or the like, or when they insist on a title themselves while
refusing me one (which is distressingly common for American
physicians).
Thomas
Richard R. Hershberger
> Is there a rule that the subjunctive must never be an analytic
> construction? Where did that rule come from?
This is not a rule. It is a definition. Could we define
"subjunctive" semantically? Of course we could. But we don't. Or
least we don't don't usually. It is the usual trend to define such
terms morphologically. Consider how linguists define "tense". There
have been lively exchanges of views on how many tenses English has.
The answer is two, as linguists define "tense". Not everybody is
happy with this.
If you find a semantic definition more useful than a morphological
one, you are free to use it that way. But if you use a non-standard
definition, you are obliged to make your definition clear. Otherwise,
it is perfectly reasonable for others to call you on it.
Richard R. Hershberger
Thomas Bushnell, BSG
> tb+u...@becket.net (Thomas Bushnell, BSG) wrote in message news:<87n00yi...@becket.becket.net>...
>
> > Is there a rule that the subjunctive must never be an analytic
> > construction? Where did that rule come from?
>
> This is not a rule. It is a definition. Could we define
> "subjunctive" semantically? Of course we could. But we don't. Or
> least we don't don't usually.
Ah, so this is a question of how the word is in fact defined? Well,
two things to say about that. The OED describes such modal
constructions as "subjunctive-equivalent" (see the entry under
"subjunctive")--and, interestingly, the last quote of
"subjunctive-equivalent" is a forty-year-old comment that it's time to
retire the term "subjunctive-equivalent", in favor of simply calling
such things "subjunctive".
Moreover, there is nothing non-morphological about allowing analytic
uses to count.
> If you find a semantic definition more useful than a morphological
> one, you are free to use it that way. But if you use a non-standard
> definition, you are obliged to make your definition clear. Otherwise,
> it is perfectly reasonable for others to call you on it.
Sure; now are you willing to notice that definitions change, that the
usage of words change, and that the change in question is perfectly
reasonable and not one I invented?
Rolleston
[...]
>two things to say about that. The OED describes such modal
>constructions as "subjunctive-equivalent" (see the entry under
>"subjunctive")--and, interestingly, the last quote of
>"subjunctive-equivalent" is a forty-year-old comment that it's time to
>retire the term "subjunctive-equivalent", in favor of simply calling
>such things "subjunctive".
Is the OED really the book to rely on here?
They may even quote madmen.
R.
Thomas Bushnell, BSG
> Is the OED really the book to rely on here?
It's an excellent source to rely on for the actual usage of English
words.