a not irrelevant dissertation

[Peter T. Daniels]

(from LINGUIST List)

Institution: Cambridge University
Program: PhD in Linguistics
Dissertation Status: Completed
Degree Date: 2011

Author: Catherine Evans Davies

Dissertation Title: Over-Informativeness in Referential Communication

Dissertation URL: http://www.srcf.ucam.org/~cnd24/CDaviesCompleteThesisPaperback.pdf

Linguistic Field(s): Language Acquisition
Pragmatics
Psycholinguistics


Dissertation Director(s):
Napoleon Katsos

Dissertation Abstract:

This thesis investigates the pragmatic mechanisms responsible for
detecting
how much information is appropriate to include in referring
expressions by
adults and children. Theoretically, it rests on the Cooperative
Principle
(Grice, 1975/1989) which is comprised of four conversational maxims to
which all cooperative speakers are assumed to adhere. The experiments
herein test whether the fundamental processes behind these theoretical
ideas are psychologically real in adults and children, with a focus on
informativeness/quantity maxims. The novel contribution of the thesis
to
this domain is the verification of sensitivity to and use of
informativeness cues; an integral skill in parsing abilities.


Using referential communication tasks for production and
comprehension, the
experimental findings reveal penalties resulting from non-optimally
informative utterances. Under-informative referring expressions cause
delays for the hearer, and speakers are penalised on judgments of
expression quality. Over-informative utterances also elicit processing
delays and penalties. This is compatible with hypotheses predicting
that
interlocutors hold expectations of optimal amounts of information and
suggest that Grice's Quantity maxim is psychologically real in
comprehenders.


Experiments 1 to 4 respond to a previous study reporting that speakers
and
hearers are sensitive to under-informativeness but not to
over-informativeness (Engelhardt, Bailey & Ferreira, 2006). In
comprehension, experiment 1 replicates the original findings regarding
under-informativeness but also documents a tentative sensitivity to
over-informativeness; revealed more robustly by experiment 2.
Experiments 3
and 4 focus on production, finding that speakers do not under- or
over-inform in neutral contexts, but may over-inform when aspects of
the
referent are made salient. This constitutes evidence that speakers and
hearers are sensitive to both Quantity maxims in simple contexts,
suggesting that the effects obtained in previous literature should be
attributed to pragmatic factors rather than structural constraints.


Experiments 5 to 7 investigate the development of pragmatic
expectations of
informativeness. They document five-year-old children's off-line
ability to
detect non-optimal informativeness. In production, experiment 5 finds
that
whilst children are frequently under-informative, they produce very
low
rates of over-informative referring expressions overall. From the
comprehender's perspective, experiment 6 shows that using binary
judgments,
five-year-olds do not reliably reject over-informative utterances
(unlike
adults), although they show an adult-like sensitivity to under-
informative
expressions. Experiment 7 tests the same sensitivities but this time
allows
intermediate ratings by using a gradable scale, yielding sensitivity
to
both under- and over-informative expressions. Thus, a major finding is
that
children do have adult-like processing mechanisms regarding the
detection
of Quantity-based infelicities. This pattern of results is
accommodated
within a novel account, the pragmatic tolerance hypothesis (Katsos &
Bishop, 2011) and extends the hypothesis beyond accounting for
sensitivity
to under-informativeness by additionally encompassing sensitivity to
over-informativeness.


The thesis provides evidence for adult and child sensitivity to both
Quantity maxims, with implications for pragmatic theory, for
psycholinguistic theory, and for methods in experimental pragmatics.

[Marius Hancu]

Is it just me or Google Groups has become very nasty in terms of text
wrapping at the end of the line?
I was able to cut and paste easily, now I see lots of straggling
ends :-[

Marius Hancu

[Iain Archer]

Peter T. Daniels wrote on Thu, 15 Dec 2011

>Dissertation Title: Over-Informativeness in Referential Communication
>
>Dissertation URL:
>http://www.srcf.ucam.org/~cnd24/CDaviesCompleteThesisPaperback.pdf

[...]

>Dissertation Abstract:

The version of the abstract in the PDF copy is a little fuller and does
read better.
--
Iain Archer To email, please use Reply-To address

[Jack Campin]

> Dissertation Title: Over-Informativeness in Referential Communication

TL;DR

-----------------------------------------------------------------------------
e m a i l : j a c k @ c a m p i n . m e . u k
Jack Campin, 11 Third Street, Newtongrange, Midlothian EH22 4PU, Scotland
mobile 07800 739 557 <http://www.campin.me.uk> Twitter: JackCampin

[CT]

Jack Campin wrote:

> > Dissertation Title: Over-Informativeness in Referential
> > Communication
>
> TL;DR

heh, I was about to post exactly that!

--
Chris

[aruzinsky]

What kind of person says, "not irrelevant" instead of "relevant?"

[Peter T. Daniels]

On Dec 16, 12:05 pm, aruzinsky <aruzin...@general-cathexis.com> wrote:

> What kind of person says, "not irrelevant" instead of "relevant?"

That would seem to be _exactly_ the sort of thing treated in the
dissertation. Do you suppose the subject heading was assigned
randomly?

Also, if you've read *1984* (by George Orwell), you might recall
"Newspeak," a politically prescribed language in which all nuance was
eliminated. Is that the sort of thing you advocate?

It's fascinating that five people have commented on the initial
posting -- with not one showing the slightest interest in its content.

[António Marques]

Well, interest and something to say are separate things to have...

[R H Draney]

aruzinsky filted:

>
>What kind of person says, "not irrelevant" instead of "relevant?"

I'd be far from lying if I neglected to deny that I couldn't help but fail to
disagree less....r


--
Me? Sarcastic?
Yeah, right.

[Evan Kirshenbaum]

R H Draney <dado...@spamcop.net> writes:

> aruzinsky filted:
>>
>>What kind of person says, "not irrelevant" instead of "relevant?"
>
> I'd be far from lying if I neglected to deny that I couldn't help
> but fail to disagree less....r

Okay, that's going into my sig quote file.

--
Evan Kirshenbaum +------------------------------------
Still with HP Labs |I'd be far from lying if I neglected
SF Bay Area (1982-) |to deny that I couldn't help but
Chicago (1964-1982) |fail to disagree less.
|
evan.kir...@gmail.com | R H Draney

http://www.kirshenbaum.net/

[Nathan Sanders]

In article <ehw4p2...@gmail.com>,

Evan Kirshenbaum <evan.kir...@gmail.com> wrote:

> R H Draney <dado...@spamcop.net> writes:
>
> > aruzinsky filted:
> >>
> >>What kind of person says, "not irrelevant" instead of "relevant?"
> >
> > I'd be far from lying if I neglected to deny that I couldn't help
> > but fail to disagree less....r
>
> Okay, that's going into my sig quote file.

I had something similar used as a question on an exam in one of my
logic courses:

Put a checkmark in exactly one of the two boxes below,
Either the one marked "YES" or the one marked "NO".
If you don't want to avoid failing to get this incorrect,
Don't leave the box not marked "YES" unchecked.

Nathan

--
Department of Linguistics
Swarthmore College
http://sanders.phonologist.org/

[Robert Bannister]

On 17/12/11 1:05 AM, aruzinsky wrote:
> What kind of person says, "not irrelevant" instead of "relevant?"

They don't mean quite the same thing.

--
Robert Bannister

[Peter Brooks]

On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article <ehw4p26v....@gmail.com>,
>  Evan Kirshenbaum <evan.kirshenb...@gmail.com> wrote:

>
> > R H Draney <dadoc...@spamcop.net> writes:
>
> > > aruzinsky filted:
>
> > >>What kind of person says, "not irrelevant" instead of "relevant?"
>
> > > I'd be far from lying if I neglected to deny that I couldn't help
> > > but fail to disagree less....r
>
> > Okay, that's going into my sig quote file.
>
> I had something similar  used as a question on an exam in one of my
> logic courses:
>
> Put a checkmark in exactly one of the two boxes below,
>    Either the one marked "YES" or the one marked "NO".
> If you don't want to avoid failing to get this incorrect,
>    Don't leave the box not marked "YES" unchecked.
>

That's easy! They don't ask you to tick the boxes at all, so, once
you've checked them and satisfied yourself that they are, indeed,
square, you've got the right answer.

[Nathan Sanders]

In article
<c250cf1c-3432-4be4...@i6g2000vbe.googlegroups.com>,

Peter Brooks <peter.h....@gmail.com> wrote:

> On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article <ehw4p26v....@gmail.com>,
> >  Evan Kirshenbaum <evan.kirshenb...@gmail.com> wrote:
> >
> > > R H Draney <dadoc...@spamcop.net> writes:
> >
> > > > aruzinsky filted:
> >
> > > >>What kind of person says, "not irrelevant" instead of "relevant?"
> >
> > > > I'd be far from lying if I neglected to deny that I couldn't help
> > > > but fail to disagree less....r
> >
> > > Okay, that's going into my sig quote file.
> >
> > I had something similar  used as a question on an exam in one of my
> > logic courses:
> >
> > Put a checkmark in exactly one of the two boxes below,
> >    Either the one marked "YES" or the one marked "NO".
> > If you don't want to avoid failing to get this incorrect,
> >    Don't leave the box not marked "YES" unchecked.
> >
> That's easy! They don't ask you to tick the boxes at all,

"Put a checkmark in exactly one of the two boxes below"!

You have to mark one of them!

> so, once
> you've checked them and satisfied yourself that they are, indeed,
> square, you've got the right answer.

[Nathan Sanders]

In article <9l23cj...@mid.individual.net>,

Robert Bannister <rob...@bigpond.com> wrote:

> On 17/12/11 1:05 AM, aruzinsky wrote:
> > What kind of person says, "not irrelevant" instead of "relevant?"
>
> They don't mean quite the same thing.

There's certainly some extra pragmatic meaning, since any difference
in word choice automatically creates some pragmatic meaning, but I'm
not sure there's a semantic difference, because relevance is binary.
Everything is either relevant or irrelevant; there is no neutral
ground (as there is with, say, hot and cold, or tall and short). If
something has zero relevance, it is irrelevant, but otherwise, if it
has even 0.0000001 relevance, it is relevant, no matter how slight
that relevance may be.

[Jared]

Every quality is binary by that standard.

--
Jared

[Nathan Sanders]

In article <jchdce$44m$1...@speranza.aioe.org>,

> Every quality is binary by that standard.

But not every quality has a single word like "irrelevant" to cover the
entire "not X" half of the spectrum. We have "relevant/irrelevant",,
but there is no single word to cover "not X" for "hot" or "tall" or
other similar qualities, so there isn't a pair of one-word terms
creating a binary distinction the way there is for
"relevant/irrelevant" (and similar pairs like "possible/impossible"
and "legal/illegal").

"Not irrelevant" just literally means "not not relevant", which only
means "relevant" by simple cancelation of true negations (cf. "not
impossible" = "possible"). Contrast this with "not unhappy", which
doesn't mean "happy", because "unhappy" doesn't mean "not happy" (it
means "sad").

There is room for neutral emotions between "happy" and "unhappy", but
there is no room for "neutral" relevance (whatever that would be!)
between "relevant" and "irrelevant". What do you envision something
to be if it is neither relevant nor irrelevant?

And keep in mind, I'm *only* talking about semantics
(context-independent truth conditions), not pragmatics
(context-dependent implications and connotations).

You can answer "are you happy or unhappy?" with "neither" because the
question presents a false dichotomy, but you can't felicitously answer
"neither" to "is this book relevant or irrelevant?", because the
dichotomy isn't false.

(You might be able to get away with answering "both", if the book has
some parts are relevant and others aren't. But crucially, the whole
book itself *is* relevant, since some parts of it are, and you
certainly can't answer "neither" in that case.)

[Dr Nick]

Nathan Sanders <san...@alum.mit.edu> writes:

> In article
> <c250cf1c-3432-4be4...@i6g2000vbe.googlegroups.com>,
> Peter Brooks <peter.h....@gmail.com> wrote:
>
>> On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
>> > In article <ehw4p26v....@gmail.com>,
>> >  Evan Kirshenbaum <evan.kirshenb...@gmail.com> wrote:
>> >
>> > > R H Draney <dadoc...@spamcop.net> writes:
>> >
>> > > > aruzinsky filted:
>> >
>> > > >>What kind of person says, "not irrelevant" instead of "relevant?"
>> >
>> > > > I'd be far from lying if I neglected to deny that I couldn't help
>> > > > but fail to disagree less....r
>> >
>> > > Okay, that's going into my sig quote file.
>> >
>> > I had something similar  used as a question on an exam in one of my
>> > logic courses:
>> >
>> > Put a checkmark in exactly one of the two boxes below,
>> >    Either the one marked "YES" or the one marked "NO".
>> > If you don't want to avoid failing to get this incorrect,
>> >    Don't leave the box not marked "YES" unchecked.
>> >
>> That's easy! They don't ask you to tick the boxes at all,
>
> "Put a checkmark in exactly one of the two boxes below"!
>
> You have to mark one of them!

Yes you do. You have to check a particular box and then put a mark in
one of the two.

[he's playing a BrE game - in our version of the language "check"
doesn't mean "tick" and we have no single word for "put a cross in"]
--
Online waterways route planner | http://canalplan.eu
Plan trips, see photos, check facilities | http://canalplan.org.uk

[Dr Nick]

"Peter T. Daniels" <gram...@verizon.net> writes:

> On Dec 16, 12:05 pm, aruzinsky <aruzin...@general-cathexis.com> wrote:
>
>> What kind of person says, "not irrelevant" instead of "relevant?"
>
> That would seem to be _exactly_ the sort of thing treated in the
> dissertation. Do you suppose the subject heading was assigned
> randomly?
>
> Also, if you've read *1984* (by George Orwell), you might recall
> "Newspeak," a politically prescribed language in which all nuance was
> eliminated. Is that the sort of thing you advocate?

I thought you were thinking of "Politics and the English Language" in
which he particularly attacked the "not un-" formation:

"One can cure oneself of the not un- formation by memorizing this
sentence: A not unblack dog was chasing a not unsmall rabbit across a
not ungreen field."

[Evan Kirshenbaum]

Dr Nick <3-no...@temporary-address.org.uk> writes:

> [he's playing a BrE game - in our version of the language "check"
> doesn't mean "tick"

It doesn't in my dialect, either. A tick mark is like a little
backwards "N", while a check mark is like a "v" with a a long right
diagonal.

> and we have no single word for "put a cross in"]

--

Evan Kirshenbaum +------------------------------------
Still with HP Labs |Yesterday I washed a single sock.
SF Bay Area (1982-) |When I opened the door, the machine
Chicago (1964-1982) |was empty.

evan.kir...@gmail.com

http://www.kirshenbaum.net/

[Nathan Sanders]

In article <87ehw3m...@temporary-address.org.uk>,

I assumed as much, but the very first sentence says to "put a
checkmark in", not to "check". Surely "put" means the same here in
BrE as it does in AmE! :-)

I grant that the second sentence won't work as intended in BrE, so no
particular box is specified to have a checkmark, but *some* box must,
according to the first sentence.

[J. J. Lodder]

Dr Nick <3-no...@temporary-address.org.uk> wrote:

> "Peter T. Daniels" <gram...@verizon.net> writes:
>
> > On Dec 16, 12:05 pm, aruzinsky <aruzin...@general-cathexis.com> wrote:
> >
> >> What kind of person says, "not irrelevant" instead of "relevant?"
> >
> > That would seem to be _exactly_ the sort of thing treated in the
> > dissertation. Do you suppose the subject heading was assigned
> > randomly?
> >
> > Also, if you've read *1984* (by George Orwell), you might recall
> > "Newspeak," a politically prescribed language in which all nuance was
> > eliminated. Is that the sort of thing you advocate?
>
> I thought you were thinking of "Politics and the English Language" in
> which he particularly attacked the "not un-" formation:
>
> "One can cure oneself of the not un- formation by memorizing this
> sentence: A not unblack dog was chasing a not unsmall rabbit across a
> not ungreen field."

Ah, an example of an unblack unraven,

Jan

[Peter T. Daniels]

On Dec 17, 2:56 am, Dr Nick <3-nos...@temporary-address.org.uk> wrote:

> "Peter T. Daniels" <gramma...@verizon.net> writes:
>
> > On Dec 16, 12:05 pm, aruzinsky <aruzin...@general-cathexis.com> wrote:
>
> >> What kind of person says, "not irrelevant" instead of "relevant?"
>
> > That would seem to be _exactly_ the sort of thing treated in the
> > dissertation. Do you suppose the subject heading was assigned
> > randomly?
>
> > Also, if you've read *1984* (by George Orwell), you might recall
> > "Newspeak," a politically prescribed language in which all nuance was
> > eliminated. Is that the sort of thing you advocate?
>
> I thought you were thinking of "Politics and the English Language" in
> which he particularly attacked the "not un-" formation:

Then I'm glad to have disabused you of that thought in the first half
of the relevant sentence: "doubleplusungood" does not mean 'very bad'
in English.

> "One can cure oneself of the not un- formation by memorizing this
> sentence: A not unblack dog was chasing a not unsmall rabbit across a
> not ungreen field."

As Nathan points out in bizarrely mathematicized terms, "not un-" is
not meaningless or pointless.

[CDB]

Evan Kirshenbaum wrote:
> R H Draney <dado...@spamcop.net> writes:
>> aruzinsky filted:
>>>
>>> What kind of person says, "not irrelevant" instead of "relevant?"
>>
>> I'd be far from lying if I neglected to deny that I couldn't help
>> but fail to disagree less....r
>
> Okay, that's going into my sig quote file.
>

Sig, nothing. I'm'a use it on somebody.

[Peter Brooks]

On Dec 17, 6:51 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <c250cf1c-3432-4be4-b24a-e6dcba4d4...@i6g2000vbe.googlegroups.com>,

>  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article <ehw4p26v....@gmail.com>,
> > >  Evan Kirshenbaum <evan.kirshenb...@gmail.com> wrote:
>
> > > > R H Draney <dadoc...@spamcop.net> writes:
>
> > > > > aruzinsky filted:
>
> > > > >>What kind of person says, "not irrelevant" instead of "relevant?"
>
> > > > > I'd be far from lying if I neglected to deny that I couldn't help
> > > > > but fail to disagree less....r
>
> > > > Okay, that's going into my sig quote file.
>
> > > I had something similar  used as a question on an exam in one of my
> > > logic courses:
>
> > > Put a checkmark in exactly one of the two boxes below,
> > >    Either the one marked "YES" or the one marked "NO".
> > > If you don't want to avoid failing to get this incorrect,
> > >    Don't leave the box not marked "YES" unchecked.
>
> > That's easy! They don't ask you to tick the boxes at all,
>
> "Put a checkmark in exactly one of the two boxes below"!
>
> You have to mark one of them!
>

Do you? I've no idea what a 'checkmark' might be - nor does the OED.
I'd read it as meaning that I check the box ( as it says below ) and
mark ( as in notice ) the inside of one of them. It isn't at all
clear, though, what the meaning is. Which is odd since it's supposed
to be a logic paper - maybe it's intended to emphasise how poor humans
actually are a logic - including logic teachers, it seems.

I suppose that you could interpret a tick as being a mark showing that
you'd check the box - it'd be one way of doing it. Then you could put
it in each of the inside of the boxes 'exactly'. I'm not very good at
being so exact myself, but, with with two boxes, I might be able to
put one in exactly. I'd not be sure which one to do it to, though, so
I'd probably choose one a random. Being a logic paper, though, I think
I'd go further and put a tick in the one box. to show that I'd checked
it, and a tick as a surrogate mark of my having checked the box in the
other. Thus complying with the instruction as fully as possible.

So the most correct solution seems to be a tick in both boxes to show
that you've checked them.

[Peter Brooks]

On Dec 17, 9:54 am, Dr Nick <3-nos...@temporary-address.org.uk> wrote:

> Nathan Sanders <sand...@alum.mit.edu> writes:
> > In article
> > <c250cf1c-3432-4be4-b24a-e6dcba4d4...@i6g2000vbe.googlegroups.com>,

> >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
> >> On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> >> > In article <ehw4p26v....@gmail.com>,
> >> >  Evan Kirshenbaum <evan.kirshenb...@gmail.com> wrote:
>
> >> > > R H Draney <dadoc...@spamcop.net> writes:
>
> >> > > > aruzinsky filted:
>
> >> > > >>What kind of person says, "not irrelevant" instead of "relevant?"
>
> >> > > > I'd be far from lying if I neglected to deny that I couldn't help
> >> > > > but fail to disagree less....r
>
> >> > > Okay, that's going into my sig quote file.
>
> >> > I had something similar  used as a question on an exam in one of my
> >> > logic courses:
>
> >> > Put a checkmark in exactly one of the two boxes below,
> >> >    Either the one marked "YES" or the one marked "NO".
> >> > If you don't want to avoid failing to get this incorrect,
> >> >    Don't leave the box not marked "YES" unchecked.
>
> >> That's easy! They don't ask you to tick the boxes at all,
>
> > "Put a checkmark in exactly one of the two boxes below"!
>
> > You have to mark one of them!
>
> Yes you do.  You have to check a particular box and then put a mark in
> one of the two.
>

Not necessarily, as I've said. To 'mark' something is to notice it -
when you 'mark my words', you don't have to put any physical label on
them. Similarly when you check something, you don't have to leave any
indication that you've done it. I often check thing are correct on the
screen of my computer, for example, but I don't leave indication that
I've done this.

In any event, you might check the two alleged 'boxes' below and find
that they are, in fact, not square, but oblique, so aren't boxes at
all - having been asked to check the boxes, the answer, if asked
verbally about it later, would be that the question is mistaken -
there aren't two boxes below. If one was really helpful, one could
write on the paper for the marker to know this; 'I've checked what
your question claims to be boxes and you can mark my words that they
are not, in fact, boxes at all'. On a flat piece of paper this would,
almost certainly, be true as they'd be squares, rather than cubes.

If the question had been better phrased, it'd ask the examinee to tick
the approximate representation of a box (in profile) [ or, better, to
tick the approximate representation of a square ] next to the correct
answer to the question. I suspect that that's what the examiner was
trying, but not managing, to say.

[Leslie Danks]

Have you never marked anyone's card?

> Similarly when you check something, you don't have to leave any
> indication that you've done it. I often check thing are correct on the
> screen of my computer, for example, but I don't leave indication that
> I've done this.
>
> In any event, you might check the two alleged 'boxes' below and find
> that they are, in fact, not square, but oblique, so aren't boxes at
> all

Is this no longer a box?

<http://t0.gstatic.com/images?q=tbn:ANd9GcTFUOL6gVzhygWnD_9WHPHG_36Qd
ZmtGgCtyuIQY_qKeh2irQEx>

<http://tinyurl.com/cqas6qs>

> - having been asked to check the boxes, the answer, if asked
> verbally about it later, would be that the question is mistaken -
> there aren't two boxes below. If one was really helpful, one could
> write on the paper for the marker to know this; 'I've checked what
> your question claims to be boxes and you can mark my words that they
> are not, in fact, boxes at all'. On a flat piece of paper this would,
> almost certainly, be true as they'd be squares, rather than cubes.
>
> If the question had been better phrased, it'd ask the examinee to tick
> the approximate representation of a box (in profile) [ or, better, to
> tick the approximate representation of a square ] next to the correct
> answer to the question. I suspect that that's what the examiner was
> trying, but not managing, to say.

Perhaps it was a disguised intelligence test. Or an intelligence test in
disguise.

--
Les
(BrE)

[Nathan Sanders]

In article
<e642f7dd-2cae-4b4a...@f1g2000yqi.googlegroups.com>,

Peter Brooks <peter.h....@gmail.com> wrote:

> On Dec 17, 6:51 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <c250cf1c-3432-4be4-b24a-e6dcba4d4...@i6g2000vbe.googlegroups.com>,
> >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> > > On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > I had something similar  used as a question on an exam in one of my
> > > > logic courses:
> >
> > > > Put a checkmark in exactly one of the two boxes below,
> > > >    Either the one marked "YES" or the one marked "NO".
> > > > If you don't want to avoid failing to get this incorrect,
> > > >    Don't leave the box not marked "YES" unchecked.
> >
> > > That's easy! They don't ask you to tick the boxes at all,
> >
> > "Put a checkmark in exactly one of the two boxes below"!
> >
> > You have to mark one of them!
> >
> Do you? I've no idea what a 'checkmark' might be -

Really? *No* idea at all? You have *never* been exposed to American
English usage, or to Unicode, or to OED's "check" (v.1) definitions
16b or 16f?

For what it's worth, the original wording on the exam also included a
sample checkmark inside parentheses after the word "checkmark", using
the second symbol (U+2714, HEAVY CHECK MARK) at:

http://en.wikipedia.org/wiki/Tick_(check_mark)

Note the name of the mark.

> - nor does the OED.

The OED has "check" (definition 16f: "To note with, or indicate by,
some mark.") and "check off" ("check" 16b: "to mark as examined and
found correct"), so what do you call the mark you have made when you
"check" something ("off")? Is "checkmark" not transparently
compositional enough?

> I'd read it as meaning that I check the box

Why? It says to "put a checkmark", not to "check".

> ( as it says below )

Nowhere in the problem does it direct you to "check" anything.

> and mark ( as in notice )

Nowhere in the problem does it direct you to "mark" anything.

The only imperative verbs are "put" and "don't leave", neither of
which is "check" or "mark".

> the inside of one of them. It isn't at all
> clear, though, what the meaning is.

Since overwhelmingly more than 50% of my students regularly get this
question correct (I don't think more than 2 in a single class have
gotten it wrong), it obviously is clear enough.

> I suppose that you could interpret a tick as being a mark showing that
> you'd check the box - it'd be one way of doing it.

It's certainly strongly implied by how the OED defines "check off", so
the connection between "check", "mark", and "tick" is already
established, even in BrE.

> Then you could put
> it in each of the inside of the boxes 'exactly'. I'm not very good at
> being so exact myself, but, with with two boxes, I might be able to
> put one in exactly.

The problem says "put a checkmark in exactly one", not "exactly put a
checkmark in one" or "put a checkmark in one...exactly". Adverbial
scope matters in English.

> I'd not be sure which one to do it to, though, so
> I'd probably choose one a random.

My students apparently don't need to resort to random selection, based
on how many of them get it correct.

(I also change "YES" to "NO" in the final line of the problem for half
of the exams, and alternate the two versions when handing them out, so
that copying one's neighbor is not a viable strategy.)

> Being a logic paper, though, I think
> I'd go further and put a tick in the one box. to show that I'd checked
> it, and a tick as a surrogate mark of my having checked the box in the
> other. Thus complying with the instruction as fully as possible.
>
> So the most correct solution seems to be a tick in both boxes to show
> that you've checked them.

That contradicts "exactly one of the two", so you'd get it wrong. (It
also contradicts your original claim that the problem doesn't ask you
to tick them at all...)

In eight years of teaching logic to hundreds of students, I've never
had a single one put a mark in both boxes (or leave both boxes blank).
Every single has put a mark in exactly one box. Not two, and not zero.

And that includes the British and Australian students.

Besides, "unticked" doesn't rhyme as well with "correct"!

And any student who is confused is always free to ask for
clarification at anytime in the exam (though I've never had a single
student ask for clarification on that problem).

[António Marques]

Yeah, but nobody told you either to check, or to mark, or even to checkmark
anything. They told you to _put_ a checkmark _in_ it. Now, you may not know
what a checkmark is, and just putting a mark in something may (may it?) not
entail putting physical labels on it in your patois, but seriously doubt
that's the case when the object of put is a noun with the structure of
'checkmark' - which structure must be transparent to you even if the meaning
is not.

[Nathan Sanders]

In article <jcisv1$e6l$1...@dont-email.me>,

I *think* he's just being difficult for the sake of argument, but it's
hard to tell.

[Frank S]

"Nathan Sanders" <san...@alum.mit.edu> wrote in message
news:sanders-92003A.15301617122011@[74.209.136.88.rev.sfr.net]...

He's regressing to his mean.

--
Frank ess

[Peter T. Daniels]

On Dec 17, 1:39 pm, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> On Dec 17, 9:54 am, Dr Nick <3-nos...@temporary-address.org.uk> wrote:
> > Nathan Sanders <sand...@alum.mit.edu> writes:

> > >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:

> > >> That's easy! They don't ask you to tick the boxes at all,
>
> > > "Put a checkmark in exactly one of the two boxes below"!
>
> > > You have to mark one of them!
>
> > Yes you do.  You have to check a particular box and then put a mark in
> > one of the two.
>
> Not necessarily, as I've said. To 'mark' something is to notice it -
> when you 'mark my words', you don't have to put any physical label on
> them. Similarly when you check something, you don't have to leave any
> indication that you've done it. I often check thing are correct on the
> screen of my computer, for example, but I don't leave indication that
> I've done this.
>
> In any event, you might check the two alleged 'boxes' below and find
> that they are, in fact, not square, but oblique, so aren't boxes at
> all - having been asked to check the boxes, the answer, if asked
> verbally about it later, would be that the question is mistaken -
> there aren't two boxes below. If one was really helpful, one could
> write on the paper for the marker to know this; 'I've checked what
> your question claims to be boxes and you can mark my words that they
> are not, in fact, boxes at all'. On a flat piece of paper this would,
> almost certainly, be true as they'd be squares, rather than cubes.

When did "box" cease to be the word for 'a usu. rectangular space that
is frequently outlined or demarcated on a surface' or 'a space on a
page for printed matter or in which to make a mark' (MW11C 5, 5b [it
also gives as exclusively Brit. 'a gift in a box' and 'BOX STALL']) or
'a square or rectangle' (AHD5 2 [with several other "Chiefly British"
senses]) ?

BTW MW11C's definition of "checkmark" is 'CHECK 10'.

[António Marques]

:) That usually works better when the pretense misunderstanding has
something to base itself on (my money is on 'put' having passed him by, they
say the profitability is high).

[Robert Bannister]

I'm not joining in this nitpicking, but I note that the majority of
official forms that I have to fill in say specifically something like
"by placing a cross in the box provided" or they say "tick" or they say
"by drawing a horizontal line through the box like this [picture]".

This stops any stupid confusion, but of course, because nobody ever
reads that stuff until later, near the end you have to find some way of
altering all your ticks to crosses or vice versa.


--
Robert Bannister

[Robert Bannister]

On 17/12/11 1:39 PM, Nathan Sanders wrote:
> In article<9l23cj...@mid.individual.net>,
> Robert Bannister<rob...@bigpond.com> wrote:
>
>> On 17/12/11 1:05 AM, aruzinsky wrote:
>>> What kind of person says, "not irrelevant" instead of "relevant?"
>>
>> They don't mean quite the same thing.
>
> There's certainly some extra pragmatic meaning, since any difference
> in word choice automatically creates some pragmatic meaning, but I'm
> not sure there's a semantic difference, because relevance is binary.
> Everything is either relevant or irrelevant; there is no neutral
> ground (as there is with, say, hot and cold, or tall and short). If
> something has zero relevance, it is irrelevant, but otherwise, if it
> has even 0.0000001 relevance, it is relevant, no matter how slight
> that relevance may be.

Is a person who is not ugly necessarily beautiful?


--
Robert Bannister

[Leslie Danks]

Robert Bannister wrote:

[...]

> I'm not joining in this nitpicking, but I note that the majority of
> official forms that I have to fill in say specifically something like
> "by placing a cross in the box provided" or they say "tick" or they say
> "by drawing a horizontal line through the box like this [picture]".
>
> This stops any stupid confusion, but of course, because nobody ever
> reads that stuff until later, near the end you have to find some way of
> altering all your ticks to crosses or vice versa.
>

What happens to you if you don't bother?

--
Les
(BrE)

[Robert Bannister]

On 17/12/11 3:11 PM, Nathan Sanders wrote:
> In article<jchdce$44m$1...@speranza.aioe.org>,

OK, I see now what you were driving at in your reply to me. However,
whatever the etymology of "irrelevant", I don't understand "not
irrelevant" to mean merely "not not relevant" so much as "not totally
irrelevant". I am not sure whether I am alone in this, but I don't think so.

The in-between states you ask about in your last sentence are where
something "has some relevance, but is not important enough to be
bothered with". A subjective view, but then without evidence so are
"relevant/irrelevant".


--
Robert Bannister

[António Marques]

Around here ballots used to be considered null if you didn't write a cross
closely corresponding to the full length of the diagonals of the box, and
nothing more. As law-abiding, state-respecting citizens, we considered that
perfectly just - and couldn't help but laugh at the US with their recounts
of votes where complex criteria to determine 'voter intention' were put
forth. Our idea was that if a voter wants to count, they'd better fill their
ballot with care - conversely, if they can't be bothered to fill their
ballot carefully, then we can't be bothered to guess what their intention was.

[António Marques]

It certainly depends on context. Its meaning _is_ 'not totally irrelevant',
but not when used in the way it was here - just as in 'you're forgetting a
detail', 'detail' means some thing of crucial importance.

> The in-between states you ask about in your last sentence are where
> something "has some relevance, but is not important enough to be bothered
> with". A subjective view, but then without evidence so are
> "relevant/irrelevant".

As to your point ('Every quality is binary by that standard', i.e. this is
no special case and qualities aren't necessarily binary), I'm with you.

[DKleinecke]

On Dec 17, 4:01 pm, Robert Bannister <robb...@bigpond.com> wrote:
> On 17/12/11 1:39 PM, Nathan Sanders wrote:
>
>
>

> > In article<9l23cjFm4...@mid.individual.net>,
> >   Robert Bannister<robb...@bigpond.com>  wrote:

>
> >> On 17/12/11 1:05 AM, aruzinsky wrote:
> >>> What kind of person says, "not irrelevant" instead of "relevant?"
>
> >> They don't mean quite the same thing.
>
> > There's certainly some extra pragmatic meaning, since any difference
> > in word choice automatically creates some pragmatic meaning, but I'm
> > not sure there's a semantic difference, because relevance is binary.
> > Everything is either relevant or irrelevant; there is no neutral
> > ground (as there is with, say, hot and cold, or tall and short).  If
> > something has zero relevance, it is irrelevant, but otherwise, if it
> > has even 0.0000001 relevance, it is relevant, no matter how slight
> > that relevance may be.
>
> Is a person who is not ugly necessarily beautiful?
>
> --
> Robert Bannister

Ugly is NOT the opposite of beautiful not is beautiful the opposite of
ugly. They are the tails of a distribution, say, one sigma out and
beyond. In my idiolect the space in between is filled with ordinary.

[DKleinecke]

On Dec 16, 11:11 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:

> And keep in mind, I'm *only* talking about semantics
> (context-independent truth conditions), not pragmatics
> (context-dependent implications and connotations).

Is this a widely made distinction ? I would use semantics to cover
both.

I remain unconvinced that there is such a thing as "context-
independent truth conditions"

[pauljk]

"António Marques" <anton...@sapo.pt> wrote in message
news:jcjbip$3ne$1...@dont-email.me...

I bet the US has significantly more lawyers per capita than PT.
They all need to be engaged in doing something "useful" to earn
good living their families have become accustomed to. :-)

pjk

[Peter Brooks]

> <http://tinyurl.com/cqas6qs>
>
No, it's not a box, it's a representation of a box - as pointed out by
the famous painting '"Ceci n'est pas une pipe".

>
> > - having been asked to check the boxes, the answer, if asked
> > verbally about it later, would be that the question is mistaken -
> > there aren't two boxes below. If one was really helpful, one could
> > write on the paper for the marker to know this; 'I've checked what
> > your question claims to be boxes and you can mark my words that they
> > are not, in fact, boxes at all'. On a flat piece of paper this would,
> > almost certainly, be true as they'd be squares, rather than cubes.
>
> > If the question had been better phrased, it'd ask the examinee to tick
> > the approximate representation of a box (in profile) [ or, better, to
> > tick the approximate representation of a square ] next to the correct
> > answer to the question. I suspect that that's what the examiner was
> > trying, but not managing, to say.
>
> Perhaps it was a disguised intelligence test. Or an intelligence test in
> disguise.
>

Maybe. One in which the examiner failed.

The point is that, if this were a question in any exam other than
logic, pretty well, one would accept the ambiguity and looseness of
language in favour of going for what the examiner was probably trying
to say. The whole point of logic, though, is to be precise and to
achieve unambiguous clarity - the question is thus inexcusable and a
good student should find a way to make this clear.

[António Marques]

DKleinecke wrote (18-12-2011 01:40):
> On Dec 17, 4:01 pm, Robert Bannister<robb...@bigpond.com> wrote:
>> On 17/12/11 1:39 PM, Nathan Sanders wrote:
>>
>>
>>
>>> In article<9l23cjFm4...@mid.individual.net>,
>>> Robert Bannister<robb...@bigpond.com> wrote:
>>
>>>> On 17/12/11 1:05 AM, aruzinsky wrote:
>>>>> What kind of person says, "not irrelevant" instead of "relevant?"
>>
>>>> They don't mean quite the same thing.
>>
>>> There's certainly some extra pragmatic meaning, since any difference
>>> in word choice automatically creates some pragmatic meaning, but I'm
>>> not sure there's a semantic difference, because relevance is binary.
>>> Everything is either relevant or irrelevant; there is no neutral
>>> ground (as there is with, say, hot and cold, or tall and short). If
>>> something has zero relevance, it is irrelevant, but otherwise, if it
>>> has even 0.0000001 relevance, it is relevant, no matter how slight
>>> that relevance may be.
>>
>> Is a person who is not ugly necessarily beautiful?
>

> Ugly is NOT the opposite of beautiful not is beautiful the opposite of
> ugly. They are the tails of a distribution, say, one sigma out and
> beyond. In my idiolect the space in between is filled with ordinary.

Well, isn't one tail of a distribution the opposite of the other tail?

In my language, the space between irrelevant and relevant is filled with
almost irrelevant, largely irrelevant, acceptably relevant.

[António Marques]

[Not that I agree with what I've snipped, but there is one point I'd like to
address...]

Peter Brooks wrote (18-12-2011 03:54):
> [T]he question is thus inexcusable and a good student should find a way
> to make this clear.

...that way lies misfortune. A good student is one who knows what teachers
are worth the student's wit, and a teacher posing inexcusable questions
certainly doesn't belong in that category.

[Peter Brooks]

On Dec 17, 9:37 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <e642f7dd-2cae-4b4a-beb9-be78e2796...@f1g2000yqi.googlegroups.com>,

>  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Dec 17, 6:51 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article
> > > <c250cf1c-3432-4be4-b24a-e6dcba4d4...@i6g2000vbe.googlegroups.com>,
> > >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> > > > On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > > I had something similar  used as a question on an exam in one of my
> > > > > logic courses:
>
> > > > > Put a checkmark in exactly one of the two boxes below,
> > > > >    Either the one marked "YES" or the one marked "NO".
> > > > > If you don't want to avoid failing to get this incorrect,
> > > > >    Don't leave the box not marked "YES" unchecked.
>
> > > > That's easy! They don't ask you to tick the boxes at all,
>
> > > "Put a checkmark in exactly one of the two boxes below"!
>
> > > You have to mark one of them!
>
> > Do you? I've no idea what a 'checkmark' might be -
>
> Really?  *No* idea at all?  You have *never* been exposed to American
> English usage, or to Unicode, or to OED's "check" (v.1) definitions
> 16b or 16f?
>

I haven't seen the Unicode names before, not, or, at any rate, I'd not
noticed them.

>
> For what it's worth, the original wording on the exam also included a
> sample checkmark inside parentheses after the word "checkmark", using
> the second symbol (U+2714, HEAVY CHECK MARK) at:
>

Good. At least the examiner was aware of his lack of clarity. That
would make it a much better question.

>
> > - nor does the OED.
>
> The OED has "check" (definition 16f: "To note with, or indicate by,
> some mark.") and "check off" ("check" 16b: "to mark as examined and
> found correct"), so what do you call the mark you have made when you
> "check" something ("off")?  Is "checkmark" not transparently
> compositional enough?
>

Not in a logic examination, no.

>
> > I'd read it as meaning that I check the box
>
> Why?  It says to "put a checkmark", not to "check".
>

Without the clarifying Unicode symbol ( which, otherwise, would have
been otiose ), it's necessary to try to interpret the sentence giving
allowance for the chance that there are other errors in it.

>
> > ( as it says below )
>
> Nowhere in the problem does it direct you to "check" anything.
>
> > and mark ( as in notice )
>
> Nowhere in the problem does it direct you to "mark" anything.
>
> The only imperative verbs are "put" and "don't leave", neither of
> which is "check" or "mark".
>

Maybe, but this isn't clear because of the peculiar positioning of the
words 'check', and 'mark' and the lack of the simple, clear, English
word, 'tick'.

>
> >  the inside of one of them. It isn't at all
> > clear, though, what the meaning is.
>
> Since overwhelmingly more than 50% of my students regularly get this
> question correct (I don't think more than 2 in a single class have
> gotten it wrong), it obviously is clear enough.
>

Or the students don't give enough weight to clarity - which would be
unfortunate if they'd just been exposed to a logic course.

>
> > I suppose that you could interpret a tick as being a mark showing that
> > you'd check the box - it'd be one way of doing it.
>
> It's certainly strongly implied by how the OED defines "check off", so
> the connection between "check", "mark", and "tick" is already
> established, even in BrE.
>

No, it isn't. The examples I find for 'check it off' in the OED are
marked as 'U.S.' and refer to agricultural practices there.

Dictionaries, by the way, particularly the OED, are not usually
consulted with an aim to discover what they appear to 'strongly imply'
- it's definitions that are usually sought.

>
> > Then you could put
> > it in each of the inside of the boxes 'exactly'. I'm not very good at
> > being so exact myself, but, with with two boxes, I might be able to
> > put one in exactly.
>
> The problem says "put a checkmark in exactly one", not "exactly put a
> checkmark in one" or "put a checkmark in one...exactly".  Adverbial
> scope matters in English.
>

Yes, it does. However, as I've pointed out, this is supposed to be a
question in a logic exam. Its lack of clarity requires one to cast
about for possible meanings as the error may not only be in the
apparently strange coinage ( though, as you've pointed out, in the
actual exam, realising your lack of clarity, you include a Unicode
symbol to indicate what you mean - rendering much of our discussion
moot ).

>
> > I'd not be sure which one to do it to, though, so
> > I'd probably choose one a random.
>
> My students apparently don't need to resort to random selection, based
> on how many of them get it correct.
>

That's a false conclusion. You haven't, until now, told us what the
actual question was - now we know you provided a graphical example of
what you intended to mean, it's not surprising that the students got
there.

>
> (I also change "YES" to "NO" in the final line of the problem for half
> of the exams, and alternate the two versions when handing them out, so
> that copying one's neighbor is not a viable strategy.)
>

How very suspicious. Why, though, if you are suspicious, not set a
different exam for each student - that'd make copying impossible? A
cheating neighbour could, surely, notice that the words were switched.

>
> > Being a logic paper, though, I think
> > I'd go further and put a tick in the one box. to show that I'd checked
> > it, and a tick as a surrogate mark of my having checked the box in the
> > other.  Thus complying with the instruction as fully as possible.
>
> > So the most correct solution seems to be a tick in both boxes to show
> > that you've checked them.
>
> That contradicts "exactly one of the two", so you'd get it wrong.  (It
> also contradicts your original claim that the problem doesn't ask you
> to tick them at all...)
>

No, it doesn't. As I say, given the error, one can adopt a number of
strategies - I've listed a few. Some involve making the interpretation
that you guide them to by supplying a graphical example of what you
meant by 'checkmark'. If you'd said 'tick', then you'd not have needed
to add that clarification.

>
> In eight years of teaching logic to hundreds of students, I've never
> had a single one put a mark in both boxes (or leave both boxes blank).
> Every single has put a mark in exactly one box.  Not two, and not zero.
>

Clearly the students haven't been precise, or observant, enough -
possibly English wasn't their first language or, in taking a logic
course, they were demonstrating a preferential leaning away from
language.

>
> And that includes the British and Australian students.
>

They were either being indulgent with the lack of precision, or guided
by the Unicode you neglected, originally, to mention. I'm not sure
what Scotch, Irish, Welsh or Cornish students might have made of it,
my comments were related to English students.

>
> Besides, "unticked" doesn't rhyme as well with "correct"!
>

You encourage your logic students to interpret things according to
rhyme?

>
> And any student who is confused is always free to ask for
> clarification at anytime in the exam (though I've never had a single
> student ask for clarification on that problem).
>

The ones that notice are guided by the Unicode clarification, I'd
think. Why not leave it out one year ( unless you improve the question
by asking people to 'tick' ) and see if the error is spotted more
commonly?

[Peter Brooks]

On Dec 17, 10:30 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
>
>
> I *think* he's just being difficult for the sake of argument, but it's
> hard to tell.
>

If you find that hard to tell, then it's no wonder that you have this
problem - the one that requires clarifying Unicode graphics to make
intelligible.

[Peter Brooks]

Something like those bears that consume little boys who step on the
cracks in the pavement.

[Peter Brooks]

On Dec 18, 1:57 am, Robert Bannister <robb...@bigpond.com> wrote:
>
>
> I'm not joining in this nitpicking, but I note that the majority of
> official forms that I have to fill in say specifically something like
> "by placing a cross in the box provided" or they say "tick" or they say
> "by drawing a horizontal line through the box like this [picture]".
>

Indeed they do. I think that one could use picked nits to obscure the
white-space that one is intended to use to indicate the answer, this
would save ink. The only problem being that, these days, most logic
students, in most of the world, are unlikely to be lousy.

I have never had a form returned for correction because I've used
ticks rather than crosses - as I generally would, ticks being quicker,
easier, and less wasteful of ink. Crosses would only be required by
primitive OCR equipment that might miss a tick.

[Peter Brooks]

On Dec 18, 2:19 am, António Marques <antonio...@sapo.pt> wrote:
>
>
> Around here ballots used to be considered null if you didn't write a cross
> closely corresponding to the full length of the diagonals of the box, and
> nothing more. As law-abiding, state-respecting citizens, we considered that
> perfectly just - and couldn't help but laugh at the US with their recounts
> of votes where complex criteria to determine 'voter intention' were put
> forth. Our idea was that if a voter wants to count, they'd better fill their
> ballot with care - conversely, if they can't be bothered to fill their
> ballot carefully, then we can't be bothered to guess what their intention was.
>

It seems highly undemocratic to disenfranchise people for imprecision.
Particularly as it could be caused by shaky hands, poor eyesight,
dodgy motor control or a buggered pencil.

It occurs to me that proper transparency would involve telling any
voter who was disenfranchised by a spoiled vote that this had
happened. The subsequent riots might be instructive.

[Peter Brooks]

On Dec 18, 2:01 am, Robert Bannister <robb...@bigpond.com> wrote:
> On 17/12/11 1:39 PM, Nathan Sanders wrote:
>
>
>
>
>
>
>
>
>

> > In article<9l23cjFm4...@mid.individual.net>,
> >   Robert Bannister<robb...@bigpond.com>  wrote:

>
> >> On 17/12/11 1:05 AM, aruzinsky wrote:
> >>> What kind of person says, "not irrelevant" instead of "relevant?"
>
> >> They don't mean quite the same thing.
>
> > There's certainly some extra pragmatic meaning, since any difference
> > in word choice automatically creates some pragmatic meaning, but I'm
> > not sure there's a semantic difference, because relevance is binary.
> > Everything is either relevant or irrelevant; there is no neutral
> > ground (as there is with, say, hot and cold, or tall and short).  If
> > something has zero relevance, it is irrelevant, but otherwise, if it
> > has even 0.0000001 relevance, it is relevant, no matter how slight
> > that relevance may be.
>
> Is a person who is not ugly necessarily beautiful?
>

Good point! I'd go further, though. Something that has nought ( or, as
the man said 'zero' ) relevance would be, for precisely that reason,
relevant. It'd be quite difficult, I think, to prove that a fact X has
absolutely no relevance to a fact Y - and, by creating the relevance
by your proof itself, your effort would, in any event, be doomed to
failure.

So I'd consider all claims for no relevance to be exaggerated.

[Peter Brooks]

When it's a logic examination where clarity of thought is being
sought.

The above definitions are imprecise as a 'box', in that sense, being a
representative of a cube, that represents an actual container, should
be 'parallelogram' rather than 'square' or 'rectangle' to allow for
printing faults and for representations of squashed boxes.

[Peter Brooks]

On Dec 18, 3:44 am, DKleinecke <dkleine...@gmail.com> wrote:
>
>
> I remain unconvinced that there is such a thing as "context-
> independent truth conditions"
>

What context is required for:
"
When P is true, ~P is false; and when P is false, ~P is true. ~~P
always has the same truth-value as P.
"
?

I mean, of course, the truth condition itself, as understood by a
conscious agent, not the sentence in English as above.

[Christopher Ingham]

According to Burchfield (1996, 529), repeated negatives (to be
distinguished from double negatives), common in meiosis and
periphrasis, cancel one another, and are therefore equivalent to an
affirmative. Fowler attributed the popularity of the idiom to “a
stubborn national dislike of putting things too strongly.” The SE
construction is attested as early as 1657 (OED, s.v. “not” 10.c.).

Peter’s use of repeated negatives was not irrelevant to the
dissertation’s theme of redundancy.

Christopher Ingham
– in modum robot

[António Marques]

On Dec 18, 4:39 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> On Dec 18, 2:19 am, António Marques <antonio...@sapo.pt> wrote:
>
> > Around here ballots used to be considered null if you didn't write a cross
> > closely corresponding to the full length of the diagonals of the box, and
> > nothing more. As law-abiding, state-respecting citizens, we considered that
> > perfectly just - and couldn't help but laugh at the US with their recounts
> > of votes where complex criteria to determine 'voter intention' were put
> > forth. Our idea was that if a voter wants to count, they'd better fill their
> > ballot with care - conversely, if they can't be bothered to fill their
> > ballot carefully, then we can't be bothered to guess what their intention was.
>
> It seems highly undemocratic to disenfranchise people for imprecision.
> Particularly as it could be caused by shaky hands, poor eyesight,
> dodgy motor control or a buggered pencil.

Not really, it's done with pens.

> It occurs to me that proper transparency would involve telling any
> voter who was disenfranchised by a spoiled vote that this had
> happened. The subsequent riots might be instructive.

How much trouble could a mob of shaking people with poor eyesight
armed with pointless pencils cause?

[António Marques]

You do know abstractions don't occur in the wild and as such 'aren't
there', don't you?

[pauljk]

"António Marques" <ent...@gmail.com> wrote in message
news:6d668f48-85ae-4451...@t38g2000yqe.googlegroups.com...

LOL.
Perhaps they still might be able to stage a successful sit-down
or lie-down demonstrations.

pjk

[Peter Brooks]

On Dec 18, 8:32 am, António Marques <ento...@gmail.com> wrote:
> On Dec 18, 4:39 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
> > On Dec 18, 2:19 am, António Marques <antonio...@sapo.pt> wrote:
>
> > > Around here ballots used to be considered null if you didn't write a cross
> > > closely corresponding to the full length of the diagonals of the box, and
> > > nothing more. As law-abiding, state-respecting citizens, we considered that
> > > perfectly just - and couldn't help but laugh at the US with their recounts
> > > of votes where complex criteria to determine 'voter intention' were put
> > > forth. Our idea was that if a voter wants to count, they'd better fill their
> > > ballot with care - conversely, if they can't be bothered to fill their
> > > ballot carefully, then we can't be bothered to guess what their intention was.
>
> > It seems highly undemocratic to disenfranchise people for imprecision.
> > Particularly as it could be caused by shaky hands, poor eyesight,
> > dodgy motor control or a buggered pencil.
>
> Not really, it's done with pens.
>

Buggered pen, then.

>
> > It occurs to me that proper transparency would involve telling any
> > voter who was disenfranchised by a spoiled vote that this had
> > happened. The subsequent riots might be instructive.
>
> How much trouble could a mob of shaking people with poor eyesight
> armed with pointless pencils cause?
>

Ah, there's a sterling silver response from a true democrat! Might is
right. As long as the disenfranchised can't cause too much shit; who
gives a toss?

[Peter Brooks]

What has the wild got to do with anything? If something occurs, then
it exists, and is, in the proper sense, natural.

That bears shit in zoos has no serious impact on the assertion that
bears tend to shit in the woods. Nor should it.

[John Holmes]

They tick you off.

--
Regards
John
for mail: my initials plus a u e
at tpg dot com dot au

[benl...@ihug.co.nz]

You seem to be under the misapprehension that "logic" refers to the
type of performance you are presently engaged in.

[António Marques]

On Dec 18, 7:25 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> On Dec 18, 8:32 am, António Marques <ento...@gmail.com> wrote:
>
>
>
>
>
>
>
> > On Dec 18, 4:39 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
> > > On Dec 18, 2:19 am, António Marques <antonio...@sapo.pt> wrote:
>
> > > > Around here ballots used to be considered null if you didn't write a cross
> > > > closely corresponding to the full length of the diagonals of the box, and
> > > > nothing more. As law-abiding, state-respecting citizens, we considered that
> > > > perfectly just - and couldn't help but laugh at the US with their recounts
> > > > of votes where complex criteria to determine 'voter intention' were put
> > > > forth. Our idea was that if a voter wants to count, they'd better fill their
> > > > ballot with care - conversely, if they can't be bothered to fill their
> > > > ballot carefully, then we can't be bothered to guess what their intention was.
>
> > > It seems highly undemocratic to disenfranchise people for imprecision.
> > > Particularly as it could be caused by shaky hands, poor eyesight,
> > > dodgy motor control or a buggered pencil.
>
> > Not really, it's done with pens.
>
> Buggered pen, then.

We'll condone no such buggery.

> > > It occurs to me that proper transparency would involve telling any
> > > voter who was disenfranchised by a spoiled vote that this had
> > > happened. The subsequent riots might be instructive.
>
> > How much trouble could a mob of shaking people with poor eyesight
> > armed with pointless pencils cause?
>
> Ah, there's a sterling silver response from a true democrat!

We're all republicans around here, even the royalists.

> Might is
> right. As long as the disenfranchised can't cause too much shit; who
> gives a toss?

Is there any place where it's done differently?

The campaign seems to be producing good results, since we actually
currently manage a lower percentage of null ballots than the countries
that obsess over them.

[António Marques]

On Dec 18, 7:28 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> On Dec 18, 8:34 am, António Marques <ento...@gmail.com> wrote:
>
>
>
>
>
>
>
> > On Dec 18, 4:45 am, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
>
> > > On Dec 18, 3:44 am, DKleinecke <dkleine...@gmail.com> wrote:
>
> > > > I remain unconvinced that there is such a thing as "context-
> > > > independent truth conditions"
>
> > > What context is required for:
> > > "
> > > When P is true, ~P  is false; and when P is false, ~P  is true. ~~P
> > > always has the same truth-value as P.
> > > "
> > > ?
>
> > > I mean, of course, the truth condition itself, as understood by a
> > > conscious agent, not the sentence in English as above.
>
> > You do know abstractions don't occur in the wild and as such 'aren't
> > there', don't you?
>
> What has the wild got to do with anything? If something occurs,

'Occurs' where, in a sentence? I've got some pure-bred pink purple
unicorns you might be interested in.

[Peter T. Daniels]

> printing faults and for representations of squashed boxes.-

It is simply not the case that "a 'box', in that sense,[is] a
representative of a cube, that represents an actual container." A box
(in that sense) is a rectangular outline for making a mark in, not a
"representation" of anything else whatsoever.

[Peter T. Daniels]

On Dec 17, 7:19 pm, António Marques <antonio...@sapo.pt> wrote:
> Leslie Danks wrote (18-12-2011 00:03):
>

> > Robert Bannister wrote:
>
> > [...]
>
> >> I'm not joining in this nitpicking, but I note that the majority of
> >> official forms that I have to fill in say specifically something like
> >> "by placing a cross in the box provided" or they say "tick" or they say
> >> "by drawing a horizontal line through the box like this [picture]".
>
> >> This stops any stupid confusion, but of course, because nobody ever
> >> reads that stuff until later, near the end you have to find some way of
> >> altering all your ticks to crosses or vice versa.
>
> > What happens to you if you don't bother?
>

> Around here ballots used to be considered null if you didn't write a cross
> closely corresponding to the full length of the diagonals of the box, and
> nothing more. As law-abiding, state-respecting citizens, we considered that
> perfectly just - and couldn't help but laugh at the US with their recounts
> of votes where complex criteria to determine 'voter intention' were put
> forth. Our idea was that if a voter wants to count, they'd better fill their
> ballot with care - conversely, if they can't be bothered to fill their
> ballot carefully, then we can't be bothered to guess what their intention was.

The votes that were difficult to count did not involve marks on paper.
Never in my lifetime have I encountered a paper ballot in a government
election.

From my earliest days (when I went with my mother into the voting
booth) until after I left New York City in 2004, voting was done on
electromechanical machines, where physical interlocks prevented voting
for more than one candidate for the same office (or both "yes" and
"no" on questions).

Chicago changed from that sort of machine to punchcards early in my
time there, and there was never a "hanging chad" or "butterfly ballot"
problem as there was in Florida in 2000.

Now in Hudson County, New Jersey, the voting machines are electronic-
mechanical, with buttons that give tactile feedback and light up when
pushed in -- but still there is no paper receipt, as I thought became
required by Federal law after the 2000 fiasco.

Electronic voting machines -- operated and maintained by private
enterprises rather than government employees -- are by far the easiest
to tamper with, and there is reason to believe that vote tampering has
occurred on a widespread basis in certain key states.

(NY State changed a couple of elections ago to marked paper ballots
that are scanned and electronically counted, but it turns out the
software doesn't prevent the voter from "overvoting," i.e. marking two
candidates for the same office; it simply provides an uninterpretable
message and accepts the ballot anyway, invalidating the vote for the
office in question.)

[Peter T. Daniels]

On Dec 17, 9:27 pm, "pauljk" <paul.kr...@clear.net.nz> wrote:
> "António Marques" <antonio...@sapo.pt> wrote in message

It has very few "public interest" lawyers, because that is not a
lucrative practice.

[Peter T. Daniels]

On Dec 17, 11:18 pm, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> On Dec 17, 9:37 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <e642f7dd-2cae-4b4a-beb9-be78e2796...@f1g2000yqi.googlegroups.com>,
> >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:

> > The OED has "check" (definition 16f: "To note with, or indicate by,
> > some mark.") and "check off" ("check" 16b: "to mark as examined and
> > found correct"), so what do you call the mark you have made when you
> > "check" something ("off")?  Is "checkmark" not transparently
> > compositional enough?
>
> Not in a logic examination, no.
>
> > > I'd read it as meaning that I check the box
>
> > Why?  It says to "put a checkmark", not to "check".
>
> Without the clarifying Unicode symbol ( which, otherwise, would have
> been otiose ), it's necessary to try to interpret the sentence giving
> allowance for the chance that there are other errors in it.

No American-English speaker is unfamiliar with the word "checkmark."

> > The only imperative verbs are "put" and "don't leave", neither of
> > which is "check" or "mark".
>
> Maybe, but this isn't clear because of the peculiar positioning of the
> words 'check', and 'mark' and the lack of the simple, clear, English
> word, 'tick'.

"Tick" does not mean 'checkmark' in American English.

[Peter T. Daniels]

On Dec 17, 7:08 pm, Robert Bannister <robb...@bigpond.com> wrote:
> On 17/12/11 3:11 PM, Nathan Sanders wrote:
> > In article<jchdce$44...@speranza.aioe.org>,
> >   Jared<jared4...@gmail.com>  wrote:

> >> On 12/17/2011 12:39 AM, Nathan Sanders wrote:
> >>> In article<9l23cjFm4...@mid.individual.net>,
> >>>    Robert Bannister<robb...@bigpond.com>   wrote:
> >>>> On 17/12/11 1:05 AM, aruzinsky wrote:

> >>>>> What kind of person says, "not irrelevant" instead of "relevant?"
>
> >>>> They don't mean quite the same thing.
>
> >>> There's certainly some extra pragmatic meaning, since any difference
> >>> in word choice automatically creates some pragmatic meaning, but I'm
> >>> not sure there's a semantic difference, because relevance is binary.
> >>> Everything is either relevant or irrelevant; there is no neutral
> >>> ground (as there is with, say, hot and cold, or tall and short).  If
> >>> something has zero relevance, it is irrelevant, but otherwise, if it
> >>> has even 0.0000001 relevance, it is relevant, no matter how slight
> >>> that relevance may be.
>

> >> Every quality is binary by that standard.
>
> > But not every quality has a single word like "irrelevant" to cover the
> > entire "not X" half of the spectrum.  We have "relevant/irrelevant",,
> > but there is no single word to cover "not X" for "hot" or "tall" or
> > other similar qualities, so there isn't a pair of one-word terms
> > creating a binary distinction the way there is for
> > "relevant/irrelevant" (and similar pairs like "possible/impossible"
> > and "legal/illegal").
>
> > "Not irrelevant" just literally means "not not relevant", which only
> > means "relevant" by simple cancelation of true negations (cf. "not
> > impossible" = "possible").  Contrast this with "not unhappy", which
> > doesn't mean "happy", because "unhappy" doesn't mean "not happy" (it
> > means "sad").
>
> > There is room for neutral emotions between "happy" and "unhappy", but
> > there is no room for "neutral" relevance (whatever that would be!)
> > between "relevant" and "irrelevant".  What do you envision something
> > to be if it is neither relevant nor irrelevant?
>
> OK, I see now what you were driving at in your reply to me. However,
> whatever the etymology of "irrelevant", I don't understand "not
> irrelevant" to mean merely "not not relevant" so much as "not totally
> irrelevant". I am not sure whether I am alone in this, but I don't think so.
>
> The in-between states you ask about in your last sentence are where
> something "has some relevance, but is not important enough to be
> bothered with". A subjective view, but then without evidence so are
> "relevant/irrelevant".

Nathan apparently continues to operate under the misconception that
language is or should be "logical."

[Andrew B]

On 18/12/2011 05:14, Christopher Ingham wrote:

> According to Burchfield (1996, 529), repeated negatives (to be
> distinguished from double negatives), common in meiosis and
> periphrasis, cancel one another, and are therefore equivalent to an
> affirmative. Fowler attributed the popularity of the idiom to “a
> stubborn national dislike of putting things too strongly.”

As opposed, presumably, to "a stubborn national liking for putting
things just strongly enough".

[Christopher Ingham]

In some cases that could also involve putting things too strongly.

Christopher Ingham

[Nathan Sanders]

In article
<9e04411b-772e-45c9...@d17g2000yql.googlegroups.com>,

DKleinecke <dklei...@gmail.com> wrote:

> On Dec 16, 11:11 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
>
> > And keep in mind, I'm *only* talking about semantics
> > (context-independent truth conditions), not pragmatics
> > (context-dependent implications and connotations).
>
> Is this a widely made distinction ?

What's the scope?

In linguistics, yes, this distinction is widely used. Outside of
linguistics, not so much.

> I would use semantics to cover both.

Many people do, even in linguistics.

> I remain unconvinced that there is such a thing as "context-
> independent truth conditions"

(1) I bought some milk and eggs.
(2) I bought some milk.

If (1) is true, then (2) is necessarily true. No possible
manipulation of the context can change that fact.

Nathan

--
Department of Linguistics
Swarthmore College
http://sanders.phonologist.org/

[Peter T. Daniels]

On Dec 18, 12:48 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <9e04411b-772e-45c9-b0ac-226a34804...@d17g2000yql.googlegroups.com>,

>
>  DKleinecke <dkleine...@gmail.com> wrote:
> > On Dec 16, 11:11 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
>
> > > And keep in mind, I'm *only* talking about semantics
> > > (context-independent truth conditions), not pragmatics
> > > (context-dependent implications and connotations).
>
> > Is this a widely made distinction ?
>
> What's the scope?
>
> In linguistics, yes, this distinction is widely used.  Outside of
> linguistics, not so much.
>
> > I would use semantics to cover both.
>
> Many people do, even in linguistics.
>
> > I remain unconvinced that there is such a thing as "context-
> > independent truth conditions"
>
>      (1) I bought some milk and eggs.
>      (2) I bought some milk.
>
> If (1) is true, then (2) is necessarily true.  No possible
> manipulation of the context can change that fact.

It is entirely possible that a convenience store, or a breakfast
diner, might sell only the complete package, so that buying only milk
was not a possibility.

The other day you threw "Spiro conjectures Ex-Lax" at me. Did you not
learn the lesson of that example?

[Nathan Sanders]

In article <9l4p8l...@mid.individual.net>,

Robert Bannister <rob...@bigpond.com> wrote:

> On 17/12/11 3:11 PM, Nathan Sanders wrote:

> > In article<jchdce$44m$1...@speranza.aioe.org>,

> > Jared<jare...@gmail.com> wrote:
> >
> >> On 12/17/2011 12:39 AM, Nathan Sanders wrote:

> >>> In article<9l23cj...@mid.individual.net>,

The 'totally' part of the meaning you are interpreting is not part of
the semantics of either "not" or "irrelevant". It comes from the
pragmatics of a speaker uttering "not irrelevant" instead of uttering
"relevant".

Semantically, "not irrelevant" is exactly equivalent to "relevant",
for the reasons described above.

Pragmatically, there is a difference between the two when uttered.
Speakers are assumed to be cooperative, and one aspect of cooperative
speech is brevity, part of Grice's Maxim of Manner. Like any Maxim,
the Maxim of Manner can be flouted (i.e., intentionally and obviously
violated) to impart additional meaning to the listener beyond the
literal semantic content.

In this case, Manner is flouted by using the larger "not irrelevant"
instead of its shorter, semantically equivalent counterpart
"relevant". The additional pragmatic meaning is something along the
lines of your 'totally'.

One of the classic examples from Grice of floutingthe Maxim of Manner
is uttering (1) instead of (2) to impart the additional pragmatic
meaning that the singing was particularly bad.

(1) Miss X produced a series of sounds corresponding closely
with the score of "Home Sweet Home".
(2) Miss X sang "Home Sweet Home".

> The in-between states you ask about in your last sentence are where
> something "has some relevance, but is not important enough to be
> bothered with".

But if it has any relevance at all, then it is relevant, by definition.

[Nathan Sanders]

In article
<2787b0b1-0a1a-4a35...@d10g2000vbh.googlegroups.com>,

Semantics is logical, but of course, language is larger than
semantics, which I'm obviously aware of by drawing a distinction
between semantics and pragmatics.

I don't know where you get "should be".

[Nathan Sanders]

In article <9l4orf...@mid.individual.net>,
Robert Bannister <rob...@bigpond.com> wrote:

> On 17/12/11 1:39 PM, Nathan Sanders wrote:
> > In article<9l23cj...@mid.individual.net>,

> > Robert Bannister<rob...@bigpond.com> wrote:
> >
> >> On 17/12/11 1:05 AM, aruzinsky wrote:
> >>> What kind of person says, "not irrelevant" instead of "relevant?"
> >>
> >> They don't mean quite the same thing.
> >
> > There's certainly some extra pragmatic meaning, since any difference
> > in word choice automatically creates some pragmatic meaning, but I'm
> > not sure there's a semantic difference, because relevance is binary.
> > Everything is either relevant or irrelevant; there is no neutral
> > ground (as there is with, say, hot and cold, or tall and short). If
> > something has zero relevance, it is irrelevant, but otherwise, if it
> > has even 0.0000001 relevance, it is relevant, no matter how slight
> > that relevance may be.
>

> Is a person who is not ugly necessarily beautiful?

Of course not. "Ugly/beautiful" is like "tall/short" and "hot/cold"
and "happy/unhappy"; there is plenty of middle ground between them.

There is no such middle ground in "relevant/irrelevant" or
"pregnant/not pregnant" or "unique/not unique" or "perfect/imperfect",
etc.

[Nathan Sanders]

In article
<6b693cf0-1918-4ebd...@p20g2000vbm.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

So? If I bought the package, I still bought milk.

(2) doesn't contain the word "only".

[António Marques]

What a surprise.
I was, however, mostly thinking of the 2008 Minnesota race. I distinctly
recall paper ballots filled out in ink.

What I find amazing is the closeness of electoral results. Quite sincirely,
I don't think a 0.01% difference between candidates gives one more
democratic legitimacy to fill the post than the other. It's a problem for
which I have no solution, but what's surprising is that it happens at all. I
suppose people have already written lengthy theses on the subject.

[Trond Engen]

António Marques:

> [...]

>
> What I find amazing is the closeness of electoral results. Quite
> sincirely, I don't think a 0.01% difference between candidates gives
> one more democratic legitimacy to fill the post than the other. It's
> a problem for which I have no solution,

Multi-seat constituencies solve that problem (even if it introduces
some other).

> but what's surprising is that it happens at all. I suppose people
> have already written lengthy theses on the subject.

A trivial answer is that it's the nature of campaigning that both (or
all) candidates try to move both local voter opinion in their direction
and their campaigns close to the political center (as defined by local
voter opinion). If both (or all) campaigns are effective whoever wins
will have been elected by the closest possible margin. Between systems
with single constituencies, this will be less obvious in a tradition
where the campaigns are less centered around the candidates' persons and
more around the national parties.

--
Trond Engen

[Peter T. Daniels]

On Dec 18, 1:17 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <2787b0b1-0a1a-4a35-ab52-bbb6ac5fb...@d10g2000vbh.googlegroups.com>,

Whence the bald assertion "semantics is logical"?

That would only be the case given the prior assumption that ...
semantics is logical.

[Christopher Ingham]

On Dec 18, 1:15 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article <9l4p8lF8i...@mid.individual.net>,

>  Robert Bannister <robb...@bigpond.com> wrote:
>
>
>
>
>
> > On 17/12/11 3:11 PM, Nathan Sanders wrote:

> > > In article<jchdce$44...@speranza.aioe.org>,
> > >   Jared<jared4...@gmail.com>  wrote:

>
> > >> On 12/17/2011 12:39 AM, Nathan Sanders wrote:

> > >>> In article<9l23cjFm4...@mid.individual.net>,

In cases of understatement by use of the repeated negatives
construction, though, it seems more that an alternate way of
expressing something has been opted for without actually conveying
additional pragmatic meaning, at least from examples that I can come
up with.

For example, how can “the argument was unconvincing, but not
illogical” be construed to mean anything on any level other than “the
argument was unconvincing, but logical”? All I derive from the
repeated negatives in the message, once I dismiss the initial false
impression that I am at least subliminally aware of an implied (but
actually illusory) subcontext, is that the speaker has used a felicity
of speech.

Christopher Ingham

>
> > The in-between states you ask about in your last sentence are where
> > something "has some relevance, but is not important enough to be
> > bothered with".
>
> But if it has any relevance at all, then it is relevant, by definition.
>
> Nathan
>
> --
> Department of Linguistics

> Swarthmore Collegehttp://sanders.phonologist.org/- Hide quoted text -
>
> - Show quoted text -

[Nathan Sanders]

In article
<7cdee080-ccb1-4fbe...@y7g2000vbe.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

Truth.

[Peter T. Daniels]

On Dec 18, 1:21 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article <9l4orfF5q...@mid.individual.net>,

>  Robert Bannister <robb...@bigpond.com> wrote:
>
>
>
>
>
> > On 17/12/11 1:39 PM, Nathan Sanders wrote:

> > > In article<9l23cjFm4...@mid.individual.net>,
> > >   Robert Bannister<robb...@bigpond.com>  wrote:

>
> > >> On 17/12/11 1:05 AM, aruzinsky wrote:
> > >>> What kind of person says, "not irrelevant" instead of "relevant?"
>
> > >> They don't mean quite the same thing.
>
> > > There's certainly some extra pragmatic meaning, since any difference
> > > in word choice automatically creates some pragmatic meaning, but I'm
> > > not sure there's a semantic difference, because relevance is binary.
> > > Everything is either relevant or irrelevant; there is no neutral
> > > ground (as there is with, say, hot and cold, or tall and short).  If
> > > something has zero relevance, it is irrelevant, but otherwise, if it
> > > has even 0.0000001 relevance, it is relevant, no matter how slight
> > > that relevance may be.
>
> > Is a person who is not ugly necessarily beautiful?
>
> Of course not.  "Ugly/beautiful" is like "tall/short" and "hot/cold"
> and "happy/unhappy"; there is plenty of middle ground between them.
>
> There is no such middle ground in "relevant/irrelevant" or
> "pregnant/not pregnant" or "unique/not unique" or "perfect/imperfect",
> etc.

Granted, things cannot be somewhat pregnant, somewhat unique
(according to prescriptivists), or somewhat perfect; but why can
something not be somewhat relevant?

And even "pregnant" and "perfect" don't work the same, since something
can be almost perfect (or almost unique, for prescriptivists), but not
almost pregnant.

Moreover, "very perfect" and "very unique" don't make much sense, but
"very pregnant" does.

[Nathan Sanders]

In article
<1499a469-d811-4c5c...@f1g2000yqi.googlegroups.com>,

The short answer is that context matters to pragmatics. A given
utterance in one context won't necessarily have the same pragmatic
effect as the same utterance in a different context.

The "but" is also crucial to your example.

[Nathan Sanders]

In article
<601a5a7b-4755-4223...@z19g2000vbe.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

I never said it couldn't be.

What I have said is that is something is somewhat relevant, it is
relevant.

> And even "pregnant" and "perfect" don't work the same, since something
> can be almost perfect (or almost unique, for prescriptivists), but not
> almost pregnant.
>
> Moreover, "very perfect" and "very unique" don't make much sense, but
> "very pregnant" does.

I never claimed that these adjectives all work the same way in every
case, only in this one specific case.

(I guess you've just completely given up any pretense of ignoring me
"as much as possible".)

[Nathan Sanders]

In article
<a503f688-6c26-4581...@z19g2000vbe.googlegroups.com>,
Peter Brooks <peter.h....@gmail.com> wrote:

> On Dec 17, 9:37 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article

> > <e642f7dd-2cae-4b4a-beb9-be78e2796...@f1g2000yqi.googlegroups.com>,
> >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> >

> > > On Dec 17, 6:51 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > In article

> > > > <c250cf1c-3432-4be4-b24a-e6dcba4d4...@i6g2000vbe.googlegroups.com>,
> > > >  Peter Brooks <peter.h.m.bro...@gmail.com> wrote:
> > > > > On Dec 17, 12:55 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > > > I had something similar  used as a question on an exam in one of my
> > > > > > logic courses:
> >
> > > > > > Put a checkmark in exactly one of the two boxes below,
> > > > > >    Either the one marked "YES" or the one marked "NO".
> > > > > > If you don't want to avoid failing to get this incorrect,
> > > > > >    Don't leave the box not marked "YES" unchecked.

> >
> > > > > That's easy! They don't ask you to tick the boxes at all,
> >
> > > > "Put a checkmark in exactly one of the two boxes below"!
> >
> > > > You have to mark one of them!
> >

> > > Do you? I've no idea what a 'checkmark' might be -
> >
> > Really?  *No* idea at all?  You have *never* been exposed to American
> > English usage, or to Unicode, or to OED's "check" (v.1) definitions
> > 16b or 16f?
> >
> I haven't seen the Unicode names before, not, or, at any rate, I'd not
> noticed them.

And why should I change my teaching methods to accommodate your lack
of knowledge? Enroll in my classes, and then maybe I'll make
allowances for your particular knowledge gaps.

> > For what it's worth, the original wording on the exam also included a
> > sample checkmark inside parentheses after the word "checkmark", using
> > the second symbol (U+2714, HEAVY CHECK MARK) at:
> >
> Good. At least the examiner was aware of his lack of clarity. That
> would make it a much better question.

Your arguments have almost convinced me that I should *remove* the
checkmark symbol from future versions of the problem, precisely so
that I can count off points for people who aren't familiar with the
ordinary vocabulary of the language used around them every day. Such
people need to be punished for not paying enough attention, and should
likely take remedial English courses to improve their vocabulary.

> > > - nor does the OED.

> >
> > The OED has "check" (definition 16f: "To note with, or indicate by,
> > some mark.") and "check off" ("check" 16b: "to mark as examined and
> > found correct"), so what do you call the mark you have made when you
> > "check" something ("off")?  Is "checkmark" not transparently
> > compositional enough?
>
> Not in a logic examination, no.

I don't see why it matters. The problem wasn't written in logical
formulas; it was written in English.

> > > I'd read it as meaning that I check the box
> >
> > Why?  It says to "put a checkmark", not to "check".
> >
> Without the clarifying Unicode symbol ( which, otherwise, would have
> been otiose ), it's necessary to try to interpret the sentence giving
> allowance for the chance that there are other errors in it.

Why would you assume that an exam question has errors in it?

> > > ( as it says below )
> >
> > Nowhere in the problem does it direct you to "check" anything.
> >
> > > and mark ( as in notice )
> >
> > Nowhere in the problem does it direct you to "mark" anything.

> >
> > The only imperative verbs are "put" and "don't leave", neither of
> > which is "check" or "mark".
> >
> Maybe,

No, not "maybe". This is a fact. Look at the sentences. The only
imperatives are "put" and "don't leave".

> but this isn't clear

It's perfectly clear to anyone who knows English syntax and verb
morphology.

> because of the peculiar positioning of the
> words 'check', and 'mark'

The words "check" and "mark" do not occur in the problem.

> and the lack of the simple, clear, English
> word, 'tick'.

What do clock noises, arachnids, or mattress casings have to do with
putting checkmarks in boxes?

> > >  the inside of one of them. It isn't at all
> > > clear, though, what the meaning is.
> >
> > Since overwhelmingly more than 50% of my students regularly get this
> > question correct (I don't think more than 2 in a single class have
> > gotten it wrong), it obviously is clear enough.
> >
> Or the students don't give enough weight to clarity - which would be
> unfortunate if they'd just been exposed to a logic course.

Since the problem is written in English rather than logical formulas,
they are already aware of the limitations of clarity (it's a
significant topic of discussion in the course).

> > > I suppose that you could interpret a tick as being a mark showing that
> > > you'd check the box - it'd be one way of doing it.
> >
> > It's certainly strongly implied by how the OED defines "check off", so
> > the connection between "check", "mark", and "tick" is already
> > established, even in BrE.
> >
> No, it isn't. The examples I find for 'check it off' in the OED are
> marked as 'U.S.' and refer to agricultural practices there.

And what about the one I actually mentioned, "to check off" (not
"check it off")? It's listed under "check" (v.1), definition 16b. It
is not marked "U.S", and one of the given examples is from Charles
Dickens (who is certainly not American).

> > > Then you could put
> > > it in each of the inside of the boxes 'exactly'. I'm not very good at
> > > being so exact myself, but, with with two boxes, I might be able to
> > > put one in exactly.
> >
> > The problem says "put a checkmark in exactly one", not "exactly put a
> > checkmark in one" or "put a checkmark in one...exactly".  Adverbial
> > scope matters in English.
> >
> Yes, it does. However, as I've pointed out, this is supposed to be a
> question in a logic exam.

Again, the problem is written *in English*, not in logical formulas.
The students know this. They know that no human language is perfectly
clear, so they know to interpret English as English.

> Its lack of clarity requires one to cast
> about for possible meanings as the error may not only be in the
> apparently strange coinage

I still see no reason to leap to the assumption that an exam problem
has an error.

> > > I'd not be sure which one to do it to, though, so
> > > I'd probably choose one a random.
> >
> > My students apparently don't need to resort to random selection, based
> > on how many of them get it correct.
> >
> That's a false conclusion.

No, it's not. If they chose randomly, then in the long run, roughly
half of them would get the problem wrong. The statistics don't
support the hypothesis that students answer that problem randomly.

> You haven't, until now, told us what the
> actual question was

I didn't realize I needed to.

> > (I also change "YES" to "NO" in the final line of the problem for half
> > of the exams, and alternate the two versions when handing them out, so
> > that copying one's neighbor is not a viable strategy.)
> >
> How very suspicious.

Indeed! I catch about 1-2 cheaters per year. Lord knows how many I
don't catch.

> Why, though, if you are suspicious, not set a
> different exam for each student -

Practicality, of course.

Besides, since this particular problem only has two answers, it
doesn't do any good to reword it 50 different ways. Two different
versions is sufficient.

> that'd make copying impossible? A
> cheating neighbour could, surely, notice that the words were switched.

You clearly have not dealt with cheaters. I have. They cheat for a
reason, and most of the time, it's because they're stupid or lazy,
neither of which is going to help them notice that one single word has
been changed in a simple binary choice problem.

> > > Being a logic paper, though, I think
> > > I'd go further and put a tick in the one box. to show that I'd checked
> > > it, and a tick as a surrogate mark of my having checked the box in the
> > > other.  Thus complying with the instruction as fully as possible.
> >
> > > So the most correct solution seems to be a tick in both boxes to show
> > > that you've checked them.
> >
> > That contradicts "exactly one of the two", so you'd get it wrong.  (It
> > also contradicts your original claim that the problem doesn't ask you
> > to tick them at all...)
> >
> No, it doesn't. As I say, given the error,

There is no error.

> one can adopt a number of
> strategies - I've listed a few. Some involve making the interpretation
> that you guide them to by supplying a graphical example of what you
> meant by 'checkmark'. If you'd said 'tick', then you'd not have needed
> to add that clarification.

But I don't want my students to put arachnids on their exams. Lyme
disease is a serious problem in this part of the country, and I would
prefer not to contract it.

> > In eight years of teaching logic to hundreds of students, I've never
> > had a single one put a mark in both boxes (or leave both boxes blank).
> > Every single has put a mark in exactly one box.  Not two, and not zero.
> >
> Clearly the students haven't been precise, or observant, enough -

No, they're just generally quite knowledgeable about the vocabulary of
the language used around them on a daily basis.

> > Besides, "unticked" doesn't rhyme as well with "correct"!
> >
> You encourage your logic students to interpret things according to
> rhyme?

It's a poem consisting of two couplets.

[Brian M. Scott]

On Sun, 18 Dec 2011 16:15:46 -0500, Nathan Sanders
<san...@alum.mit.edu> wrote in
<news:sanders-0FBD84.16154618122011@[74.209.136.88.rev.sfr.net]>
in sci.lang,alt.usage.english:

> In article
> <a503f688-6c26-4581...@z19g2000vbe.googlegroups.com>,
> Peter Brooks <peter.h....@gmail.com> wrote:

[...]

>> A cheating neighbour could, surely, notice that the words
>> were switched.

> You clearly have not dealt with cheaters. I have. They
> cheat for a reason, and most of the time, it's because
> they're stupid or lazy, neither of which is going to help
> them notice that one single word has been changed in a
> simple binary choice problem.

Nor recognize what to do about the change if they *do*
happen to notice it. (It's also been my experience over the
years that people who copy often choose unreliable sources.)

[...]

Brian

[Brian M. Scott]

On Sun, 18 Dec 2011 12:13:58 -0800 (PST), Christopher Ingham
<christop...@comcast.net> wrote in
<news:1499a469-d811-4c5c...@f1g2000yqi.googlegroups.com>
in sci.lang,alt.usage.english:

[...]

> In cases of understatement by use of the repeated
> negatives construction, though, it seems more that an
> alternate way of expressing something has been opted for
> without actually conveying additional pragmatic meaning,
> at least from examples that I can come up with.

Such litotes is commonly used for pragmatic effect in Old
Norse saga literature, e.g., <ekki vartu þá ófóthvatr> 'you
were not then not-fast-of-foot', actually meaning 'then you
ran like a scared rabbit'.

In English we have 'she was not a little upset', which
generally has the emphatic sense 'she was very much upset',
not its literal meaning, 'she was at least moderately
upset'. This is clearly a matter of pragmatics, not
semantics.

[...]

Brian

[Peter T. Daniels]

On Dec 18, 3:26 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <601a5a7b-4755-4223-a4f1-56bcecbbd...@z19g2000vbe.googlegroups.com>,

That, however, is not how English works. If it were relevant, it would
not be qualified with "somewhat."

> > And even "pregnant" and "perfect" don't work the same, since something
> > can be almost perfect (or almost unique, for prescriptivists), but not
> > almost pregnant.
>
> > Moreover, "very perfect" and "very unique" don't make much sense, but
> > "very pregnant" does.
>
> I never claimed that these adjectives all work the same way in every
> case, only in this one specific case.
>
> (I guess you've just completely given up any pretense of ignoring me
> "as much as possible".)

You had been managing to post without your personality intruding, for
a few messages.

[Peter T. Daniels]

On Dec 18, 3:20 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <1499a469-d811-4c5c-9389-aa38a9682...@f1g2000yqi.googlegroups.com>,

And, the prefixes un-, in-, etc., are not automatically equivalent to
"not."

[Peter T. Daniels]

On Dec 18, 3:16 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <7cdee080-ccb1-4fbe-89f5-597a9e8fa...@y7g2000vbe.googlegroups.com>,

That would only be the case given the prior assumption that ...
semantics is logical.

> Truth.

As I thought, you have no justification for your assertion.

[DKleinecke]

On Dec 18, 4:54 pm, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:

> In English we have 'she was not a little upset', which
> generally has the emphatic sense 'she was very much upset',
> not its literal meaning, 'she was at least moderately
> upset'.  This is clearly a matter of pragmatics, not
> semantics.

This is a good example of where I would use the word "pragmatics". It
involves why the speaker chose one way of making an utterance rather
than another. Of course it involves context - but I do not accept the
converse. If context is involved the matter does not necessarily
belong to pragmatics.

Only one example of a truth value alleged to not involve context has
been offered and I have replied to it pointing out the context it
demands. I still feel all semantics requires context - but there is
something called pragmatics that matters - but does not concern truth
values.

[DKleinecke]

On Dec 17, 8:45 pm, Peter Brooks <peter.h.m.bro...@gmail.com> wrote:

> What context is required for:
> "
> When P is true, ~P  is false; and when P is false, ~P  is true. ~~P
> always has the same truth-value as P.
> "
> ?
>
> I mean, of course, the truth condition itself, as understood by a
> conscious agent, not the sentence in English as above.

The most noteworthy context required is the notion of proposition and
the unstated assumption that P is a proposition. Also required are
the notion of an operator (here ~) and that true and false are truth
values. There are probably more.

The "of course .. " that these observations might induce proves just
how hard it is to escape context.

[pauljk]

"John Holmes" <s...@sig.instead> wrote in message news:4eeda626$1...@dnews.tpgi.com.au...

> Leslie Danks wrote:
>> Robert Bannister wrote:
>>
>> [...]
>>
>>> I'm not joining in this nitpicking, but I note that the majority of
>>> official forms that I have to fill in say specifically something like
>>> "by placing a cross in the box provided" or they say "tick" or they
>>> say "by drawing a horizontal line through the box like this
>>> [picture]".
>>>
>>> This stops any stupid confusion, but of course, because nobody ever
>>> reads that stuff until later, near the end you have to find some way
>>> of altering all your ticks to crosses or vice versa.
>>>
>> What happens to you if you don't bother?
>

> They tick you off.

That would make me crossed!

pjk

[Nathan Sanders]

In article
<f93ef2aa-22c1-4e82...@o14g2000vbo.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

Yes it is.

> If it were relevant, it would
> not be qualified with "somewhat."

"Somewhat relevant" is a subset of "relevant", just as "somewhat
pregnant" is a subset of "pregnant".

[Nathan Sanders]

In article
<97caa661-587c-4f74...@n6g2000vbg.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

> > Truth.
>
> As I thought, you have no justification for your assertion.

Truth isn't enough justification for you?

[Nathan Sanders]

In article
<5b7572c2-2cad-4e96...@y12g2000vba.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

I never said they were "automatically" equivalent to "not", and I even
gave explicit examples for "un-", such as "unhappy" showing this.
There are also example for "in-", such as "infrequent" and
"improbable", which, like "unhappy", point to the opposite end of a
scale, leaving open middle ground not covered by either the positive
or negative term.

But in some cases, these prefixes *are* equivalent to "not", as in
"irrelevant", "infinite", "unequal", and "unable", where there is no
middle ground between the positive and the negative.

You originally claimed that the thread title was redundant. If the
redundancy does not like in "not irrelevant", where does it lie?

[Peter T. Daniels]

On Dec 18, 10:28 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <f93ef2aa-22c1-4e82-a51b-a2842680a...@o14g2000vbo.googlegroups.com>,

That is absurd, because there's no such thing as "somewhat pregnant,"
but there is such a thing as 'somewhat relevant."

[Peter T. Daniels]

On Dec 18, 10:38 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <5b7572c2-2cad-4e96-9d2c-a35d95db5...@y12g2000vba.googlegroups.com>,

You cannot simply continue to assert without evidence or argument that
"relevant" is equivalent to "equal" for this parameter. It clearly is
not.

There are of course many degrees of infinity and of ability. The world
is not so black-and-white as you seem to think.

> You originally claimed that the thread title was redundant.  If the
> redundancy does not like in "not irrelevant", where does it lie?

Why would I have "claimed" that my thread title was "redundant"? I did
no such thing.

[Nathan Sanders]

In article
<0701b64b-b21e-4d7f...@z25g2000vbs.googlegroups.com>,

"Peter T. Daniels" <gram...@verizon.net> wrote:

Of course there is:

"Minnie Driver is Somewhat Pregnant"
http://www.woofactor.com/82446/

"Somewhat pregnant Katherine"
http://picasaweb.google.com/lh/photo/IMQn28EY-l71efW4Ry8FjQ

"I guess little things like that remind me that I am somewhat pregnant"
http://www.babyandbump.com/pregnancy-first-trimester/498141-hasnt-hit-m
e-yet-2.html

"This is relevant because Miss Jane is somewhat pregnant right now
(i.e. more than a little bit pregnant)"
http://niallniallorangepeel.blogspot.com/2010/09/breakfast-brought-to-y
ou-by-skype-and.html

"now that I am somewhat pregnant"
http://www.babycenter.com/400_no-baby-shown-up-on-ultrasound-tested-pos
itive-but-hcg-level_9612083_620.bc

"She is somewhat pregnant."
http://www.paklinks.com/gs/life-and-relationships/348847-manipulating-f
emales.html

"Malcolm poses with his mum and dad and his elder sister Louise, who
is somewhat pregnant."
http://malcolmcarline.tripod.com/theweddingday/id10.html

"I finally look somewhat pregnant 17 weeks!"
http://www.myspace.com/krbeavers/photos/58320935

"This is the only photo i look somewhat pregnant in"
http://www.myspace.com/karaannethomas/photos/20684551

"Things over which to obsess when you are somewhat pregnant."
http://deep-freeze.typepad.com/frozen/2005/04/things_over_whi.html

"a provocative shilm (short film) charting 7 terrifying minutes in the
life of Consuela Hernandez Rodriguez, a brave, young and somewhat
pregnant Spanish woman"
http://www.youtube.com/watch?v=fIvPRrgr0UA

"I missed my 10-year reunion because I was somewhat pregnant"
http://tahlequahdailypress.com/columns/x947031418/It-started-as-a-get-t
ogether-and-it-got-bigger

"According to Celebslam.com, the latest info buzzing on The HillsŒ
related rumor mill is that Lauren Conrad is somewhat pregnantŠ"
http://www.starzlife.com/20080314/babys-the-new-botox/

"the fact that his sister-in-law was somewhat pregnant and
circumstantial evidence seemed to mark him as the father."
http://www.american-buddha.com/illuminatus!.app3.htm

"could you girls tell me if I'm somewhat pregnant or not"
http://www.babycenter.com/400_hi-ladies-could-you-girls-tell-me-if-im-s
omewhat-pregnant-or_7671603_86.bc

"The female in that picture is somewhat pregnant."
http://answers.yahoo.com/question/index?qid=20070303212412AALjj9L

"Feeling somewhat pregnant as of late."
http://ashleyebacon.tumblr.com/post/7316952083/feeling-somewhat-pregnan
t-as-of-late-my-favorite

And there are of course many examples of the opposite end, "very
pregnant".

But you'd only know that if you bothered to pay attention to how real
people actual use the language, rather than just asserting by fiat
that your own personal idiolect is all that matters.

[Peter T. Daniels]

On Dec 18, 10:29 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article

> <97caa661-587c-4f74-9045-55785f7e9...@n6g2000vbg.googlegroups.com>,

"Truth" is not something that can simply be asserted. You have offered
no evidence or argument that "semantics is logical."