Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

The garden of Eden

105 views
Skip to first unread message

Quinn C

unread,
Jan 14, 2022, 9:54:15 AM1/14/22
to
If that isn't a garden path, I haven't seen one:

[I space it the way it appeared on my phone screen]

Scientists find oldest ever
confirmed human fossils are
even older than previously
thought

This is especially bad because in context with "fossils", "find" really
wants to have another meaning.

<https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
--
... while there are people who are consecrated, chronic
assholes--like Donald Trump for example, or General Patton--
it's a condition that all of us are liable to.
-- Geoffrey Nunberg, 2012 interview

lar3ryca

unread,
Jan 14, 2022, 12:33:43 PM1/14/22
to
On Friday, January 14, 2022 at 8:54:15 AM UTC-6, Quinn C wrote:
> If that isn't a garden path, I haven't seen one:
>
> [I space it the way it appeared on my phone screen]
>
> Scientists find oldest ever
> confirmed human fossils are
> even older than previously
> thought
>
> This is especially bad because in context with "fossils", "find" really
> wants to have another meaning.

It's not even 1/10 as bad as the crap I hear every day on our local 'news' channels.

Anders D. Nygaard

unread,
Jan 14, 2022, 12:42:50 PM1/14/22
to
Den 14-01-2022 kl. 15:54 skrev Quinn C:
> If that isn't a garden path, I haven't seen one:
>
> [I space it the way it appeared on my phone screen]
>
> Scientists find oldest ever
> confirmed human fossils are
> even older than previously
> thought
>
> This is especially bad because in context with "fossils", "find" really
> wants to have another meaning.

I don't find it too bad - I got it in one.

"He gave the girl the ring impressed the watch"

/Anders, Denmark

Quinn C

unread,
Jan 14, 2022, 1:26:38 PM1/14/22
to
* Anders D. Nygaard:

> Den 14-01-2022 kl. 15:54 skrev Quinn C:
>> If that isn't a garden path, I haven't seen one:
>>
>> [I space it the way it appeared on my phone screen]
>>
>> Scientists find oldest ever
>> confirmed human fossils are
>> even older than previously
>> thought
>>
>> This is especially bad because in context with "fossils", "find" really
>> wants to have another meaning.
>
> I don't find it too bad - I got it in one.

That seems hardly possible. Is there something wrong with your internal
parser that it didn't recognize at "fossils" that you have a complete
sentence now? Or do you routinely jump around the sentence and took in
the "are" before processing the "oldest ever ..." part?

That might be the biggest issue - the "are" is so far out when you need
to know it'll come that you can't be expected to hold out for it. Seven
words before "are", that about fills the short-term memory.

And "Scientist find that ..." would've made the structure obvious from
the start. Why would you drop "that" and then not "are", too?

--
Quinn C
My pronouns are they/them
(or other gender-neutral ones)

Jerry Friedman

unread,
Jan 14, 2022, 2:20:07 PM1/14/22
to
On Friday, January 14, 2022 at 11:26:38 AM UTC-7, Quinn C wrote:
> * Anders D. Nygaard:
> > Den 14-01-2022 kl. 15:54 skrev Quinn C:
> >> If that isn't a garden path, I haven't seen one:
> >>
> >> [I space it the way it appeared on my phone screen]
> >>
> >> Scientists find oldest ever
> >> confirmed human fossils are
> >> even older than previously
> >> thought
> >>
> >> This is especially bad because in context with "fossils", "find" really
> >> wants to have another meaning.
> >
> > I don't find it too bad - I got it in one.

> That seems hardly possible. Is there something wrong with your internal
> parser that it didn't recognize at "fossils" that you have a complete
> sentence now? Or do you routinely jump around the sentence and took in
> the "are" before processing the "oldest ever ..." part?

Maybe you could consider making your point without suggesting that
there's something wrong with the person you're arguing with.

> That might be the biggest issue - the "are" is so far out when you need
> to know it'll come that you can't be expected to hold out for it. Seven
> words before "are", that about fills the short-term memory.

As I see it, you don't need "are" till after "fossils", which is where it is.

> And "Scientist find that ..." would've made the structure obvious from
> the start. Why would you drop "that" and then not "are", too?

I agree that "that" would have helped (and I'd delete "ever"). Deleting
"are", if that's what you're suggesting, wouldn't help me. Then the
sentence could be read "Scientists find oldest ever confirmed
human fossils [that are] even older than previously thought."

--
Jerry Friedman

Bebercito

unread,
Jan 14, 2022, 2:42:07 PM1/14/22
to
Couldn't it also be "Scientists find oldest ever confirmed
human fossils [to be] even older than previously thought"?

To me, the garden path was in first suggesting that "find"
could be a noun - funnily, with that interpretation, the
sentence can also make sense and results in a similar
meaning:

(The) scientists' find(, which is the) oldest ever(,)
confirmed human fossils are even older than
previously thought.

> --
> Jerry Friedman

Jerry Friedman

unread,
Jan 14, 2022, 2:55:59 PM1/14/22
to
Yes, it could also be that, but ruling out the wrong interpretation
from the start would be better writing.

> To me, the garden path was in first suggesting that "find"
> could be a noun - funnily, with that interpretation, the
> sentence can also make sense and results in a similar
> meaning:
>
> (The) scientists' find(, which is the) oldest ever(,)
> confirmed human fossils are even older than
> previously thought.

Hm. People do leave out apostrophes, but I don't think I'd have come
up with that.

--
Jerry Friedman

Peter T. Daniels

unread,
Jan 14, 2022, 3:52:37 PM1/14/22
to
On Friday, January 14, 2022 at 1:26:38 PM UTC-5, Quinn C wrote:
> * Anders D. Nygaard:
> > Den 14-01-2022 kl. 15:54 skrev Quinn C:

> >> If that isn't a garden path, I haven't seen one:
> >> [I space it the way it appeared on my phone screen]
> >>
> >> Scientists find oldest ever
> >> confirmed human fossils are
> >> even older than previously
> >> thought
> >>
> >> This is especially bad because in context with "fossils", "find" really
> >> wants to have another meaning.
> > I don't find it too bad - I got it in one.
>
> That seems hardly possible. Is there something wrong with your internal
> parser that it didn't recognize at "fossils" that you have a complete
> sentence now? Or do you routinely jump around the sentence and took in
> the "are" before processing the "oldest ever ..." part?

? If the line breaks are transcribed correctly, then there is no problem
at all, for the 'are' is right there where it belongs, precluding the potential.
garden-path.

But trying to use a few-cm square screen like a computer monitor
seems insanity. Though I suppose it will create plenty of business
for opticians or LASIK surgeons in the future.

> That might be the biggest issue - the "are" is so far out when you need
> to know it'll come that you can't be expected to hold out for it. Seven
> words before "are", that about fills the short-term memory.
>
> And "Scientist find that ..." would've made the structure obvious from
> the start. Why would you drop "that" and then not "are", too?

It's a headline, isn't it?

Ken Blake

unread,
Jan 14, 2022, 3:58:24 PM1/14/22
to
To me that's a poor choice. It looks to me like "find" means
"discover": Scientists find oldest ever confirmed human fossils ...."

To me, much better would be "Scientists find that the oldest ever
confirmed human fossils are even older than previously thought."

"Find" by itself could mean "discover" or "establish." "Find that"
takes away the "discover" meaning, and clarifies the sentence.

Quinn C

unread,
Jan 14, 2022, 5:29:01 PM1/14/22
to
* Jerry Friedman:

> On Friday, January 14, 2022 at 11:26:38 AM UTC-7, Quinn C wrote:
>> * Anders D. Nygaard:
>>> Den 14-01-2022 kl. 15:54 skrev Quinn C:
>>>> If that isn't a garden path, I haven't seen one:
>>>>
>>>> [I space it the way it appeared on my phone screen]
>>>>
>>>> Scientists find oldest ever
>>>> confirmed human fossils are
>>>> even older than previously
>>>> thought
>>>>
>>>> This is especially bad because in context with "fossils", "find" really
>>>> wants to have another meaning.
>>>
>>> I don't find it too bad - I got it in one.
>
>> That seems hardly possible. Is there something wrong with your internal
>> parser that it didn't recognize at "fossils" that you have a complete
>> sentence now? Or do you routinely jump around the sentence and took in
>> the "are" before processing the "oldest ever ..." part?
>
> Maybe you could consider making your point without suggesting that
> there's something wrong with the person you're arguing with.

Would you allow this: You can't run? Is there something wrong with your
heart (or legs)?

I didn't conceptualize the above as a personal failing, as it's not
something I think we have conscious control over. At best we may be able
to improve its working by training, but that's not something most people
do intentionally, either, certainly not in a native language.

>> That might be the biggest issue - the "are" is so far out when you need
>> to know it'll come that you can't be expected to hold out for it. Seven
>> words before "are", that about fills the short-term memory.
>
> As I see it, you don't need "are" till after "fossils", which is where it is.

You need to know it's coming in order to realize that "fossils" is not
the object of "find" and to get to the right meaning of "find": "find
out", not "unearth", the latter being primed by "scientists find
fossils".

So when I arrived at "fossils", my parser jumped on the obvious
interpretation and then at "are", I needed to a) reinterpret the
structure of the sentence and b) reinterpret "find".

The challenge is to keep multiple decisions of that kind open. So
really, what is "wrong" with Anders' parser as opposed to mine seems to
be that his is able to keep more ambiguities undecided. A bit like
there's something wrong with my mental mathematics module in that I
often don't see the continuation of a sequence that is obvious to most
people, because I still see other good options.

"What's wrong with you?" in casual conversation can mean "you're not
normal", and that can mean you're extraordinary.

>> And "Scientist find that ..." would've made the structure obvious from
>> the start.

And narrowed down the specific meaning of "find".

>> Why would you drop "that" and then not "are", too?
>
> I agree that "that" would have helped (and I'd delete "ever"). Deleting
> "are", if that's what you're suggesting, wouldn't help me.

I didn't intend to say it'd help, I just thought delete both if the goal
is to be as short as possible, or leave both in if you want it to be
easier to understand.

> Then the
> sentence could be read "Scientists find oldest ever confirmed
> human fossils [that are] even older than previously thought."

True. I didn't see that. Semantically a bit convoluted, but of course
grammatically fine.

Quinn C

unread,
Jan 14, 2022, 6:04:12 PM1/14/22
to
* Peter T. Daniels:

> On Friday, January 14, 2022 at 1:26:38 PM UTC-5, Quinn C wrote:
>> * Anders D. Nygaard:
>>> Den 14-01-2022 kl. 15:54 skrev Quinn C:
>
>>>> If that isn't a garden path, I haven't seen one:
>>>> [I space it the way it appeared on my phone screen]
>>>>
>>>> Scientists find oldest ever
>>>> confirmed human fossils are
>>>> even older than previously
>>>> thought
>>>>
>>>> This is especially bad because in context with "fossils", "find" really
>>>> wants to have another meaning.
>>> I don't find it too bad - I got it in one.
>>
>> That seems hardly possible. Is there something wrong with your internal
>> parser that it didn't recognize at "fossils" that you have a complete
>> sentence now? Or do you routinely jump around the sentence and took in
>> the "are" before processing the "oldest ever ..." part?
>
> ? If the line breaks are transcribed correctly, then there is no problem
> at all, for the 'are' is right there where it belongs,

Of course it's where it belongs; that's not what being a garden-path
sentence is about. A garden-path sentence is one where up to a certain
point, some way into the sentence, a different structure than the actual
one is suggested, but then breaks down.

Are you saying that for you - and possibly for most native speakers -
it's easy to take in "Scientists find oldest ever confirmed human
fossils are" without deciding on a structure yet? Then it may be a
non-native issue. I find that a lot of words to deal with
un-tree-structured.

> precluding the potential.
> garden-path.
>
> But trying to use a few-cm square screen like a computer monitor
> seems insanity. Though I suppose it will create plenty of business
> for opticians or LASIK surgeons in the future.

The headline filled almost my whole screen (or at least a square about
its width), so it wasn't really small.

Besides, my phone has a higher resolution screen than my laptop.

>> That might be the biggest issue - the "are" is so far out when you need
>> to know it'll come that you can't be expected to hold out for it. Seven
>> words before "are", that about fills the short-term memory.
>>
>> And "Scientist find that ..." would've made the structure obvious from
>> the start. Why would you drop "that" and then not "are", too?
>
> It's a headline, isn't it?

That's why you could also drop "are", if dropping stuff is your thing.

Peter Moylan

unread,
Jan 14, 2022, 8:16:56 PM1/14/22
to
On 15/01/22 07:52, Peter T. Daniels wrote:

> But trying to use a few-cm square screen like a computer monitor
> seems insanity. Though I suppose it will create plenty of business
> for opticians or LASIK surgeons in the future.

Over the last week I've had trouble booking my covid booster shot. The
medical authorities want me to do it on my phone, which is crazy giving
the amount of data entry that is needed to create an account. To make
matters worse, at some stage I had to activate a link sent to me by
e-mail. The e-mail comes to my computer, not to my phone, so the
activation was a pain.

The younger generation obviously has sharper eyes than mine. On top of
that they must have narrower fingers. My index finger is about three
characters wide on the phone "keyboard". And you can't touch-type on
such a keyboard, you have to go back to one-finger typing.

Some time last year, Google decided that my computer was a phone, and
since then Google search results have been presented in a narrow format
that is a pain to read.

--
Peter Moylan Newcastle, NSW http://www.pmoylan.org

Quinn C

unread,
Jan 14, 2022, 8:32:18 PM1/14/22
to
* Peter Moylan:

> On 15/01/22 07:52, Peter T. Daniels wrote:
>
>> But trying to use a few-cm square screen like a computer monitor
>> seems insanity. Though I suppose it will create plenty of business
>> for opticians or LASIK surgeons in the future.
>
> Over the last week I've had trouble booking my covid booster shot. The
> medical authorities want me to do it on my phone, which is crazy giving
> the amount of data entry that is needed to create an account.

Do you have to use an app?

It seems impossible to enforce on a Web page.

--
- It's the title search for the Rachel property.
Guess who owns it?
- Tell me it's not that bastard Donald Trump.
-- Gilmore Girls, S02E08 (2001)

Tony Cooper

unread,
Jan 14, 2022, 8:46:20 PM1/14/22
to
I have the same problem. My fat fingers hit multiple letters and the
wrong one appears on the screen.

I always carry a ballpoint pen with me, so I purchase six pens with
hollow rubber tips at the non-writing end. They are used like a
stylus to tap out the letters on my phone when sending text.

The pens look like this:

https://i5.walmartimages.com/asr/13d62e15-e3ba-4a7b-b933-5b942e191688.8032f4c967b7f42b40c914300b0bbb68.jpeg
--

Tony Cooper Orlando Florida

Sam Plusnet

unread,
Jan 14, 2022, 9:17:42 PM1/14/22
to
On 14-Jan-22 14:54, Quinn C wrote:
> If that isn't a garden path, I haven't seen one:
>
> [I space it the way it appeared on my phone screen]
>
> Scientists find oldest ever
> confirmed human fossils are
> even older than previously
> thought
>
> This is especially bad because in context with "fossils", "find" really
> wants to have another meaning.
>
> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>

Men in white coats claim: Major setback for oldest human fossils.

--
Sam Plusnet

Dingbat

unread,
Jan 14, 2022, 10:17:39 PM1/14/22
to
On Friday, January 14, 2022 at 6:54:15 AM UTC-8, Quinn C wrote:
> If that isn't a garden path, I haven't seen one:
>
> [I space it the way it appeared on my phone screen]
>
> Scientists find oldest ever
> confirmed human fossils are
> even older than previously
> thought
>
> This is especially bad because in context with "fossils", "find" really
> wants to have another meaning.
>
> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
>
Even in the world of fossils, it's a fossil!

A spectacularly old fossil seems a rare find.
What is the other meaning you think "find" should have?

Peter T. Daniels

unread,
Jan 15, 2022, 12:24:16 AM1/15/22
to
That seems like it might be it -- you're reading the sentence word
by word instead of taking in whole phrases at a time.

But this particular sentence ought to be easier for a German-
speaker, because you're used to having the verb at the end,

> un-tree-structured.
> > precluding the potential.
> > garden-path.
> >
> > But trying to use a few-cm square screen like a computer monitor
> > seems insanity. Though I suppose it will create plenty of business
> > for opticians or LASIK surgeons in the future.
>
> The headline filled almost my whole screen (or at least a square about
> its width), so it wasn't really small.
>
> Besides, my phone has a higher resolution screen than my laptop.
>
> >> That might be the biggest issue - the "are" is so far out when you need
> >> to know it'll come that you can't be expected to hold out for it. Seven
> >> words before "are", that about fills the short-term memory.
> >> And "Scientist find that ..." would've made the structure obvious from
> >> the start. Why would you drop "that" and then not "are", too?

Maybe it was written by a Brit, for whom such a "that" is far more omissible
than for an American.

Kerr-Mudd, John

unread,
Jan 15, 2022, 5:58:20 AM1/15/22
to
On Fri, 14 Jan 2022 20:46:14 -0500
Tony Cooper <tonyco...@gmail.com> wrote:

[small virtual keyboards]
>
> I always carry a ballpoint pen with me, so I purchase six pens with
> hollow rubber tips at the non-writing end. They are used like a
> stylus to tap out the letters on my phone when sending text.
>
> The pens look like this:
>
> https://i5.walmartimages.com/asr/13d62e15-e3ba-4a7b-b933-5b942e191688.8032f4c967b7f42b40c914300b0bbb68.jpeg

So what colour was the 6th one? :-)

--
Bah, and indeed Humbug.

Tak To

unread,
Jan 15, 2022, 8:34:40 AM1/15/22
to
Both "to find <obj>" and "find that ..." have the "discover"
meaning, only that in the latter case what is discovered is an
abstract notion.

Sometimes "find out" is used to emphasize the "hitherto unknown"
aspect.

--
Tak
----------------------------------------------------------------+-----
Tak To ta...@alum.mit.eduxx
--------------------------------------------------------------------^^
[taode takto ~{LU5B~}] NB: trim the xx to get my real email addr

Quinn C

unread,
Jan 15, 2022, 10:51:47 AM1/15/22
to
* Dingbat:
Those fossils were *dug up* 50 years ago. They were now *determined* to
be not less than 200,000, but over 230,000 years old.

Hint: "human", not "humanoid".
--
For since no male
Has ruled me or has fed,
I think my own thoughts
In my woman's head. -- Lesbia Harford, Fatherless

Quinn C

unread,
Jan 15, 2022, 11:43:30 AM1/15/22
to
* Peter T. Daniels:

>>>>> Den 14-01-2022 kl. 15:54 skrev Quinn C:
>
>>>>>> If that isn't a garden path, I haven't seen one:
>>>>>> [I space it the way it appeared on my phone screen]
>>>>>>
>>>>>> Scientists find oldest ever
>>>>>> confirmed human fossils are
>>>>>> even older than previously
>>>>>> thought
>>>>>>
>>>> That might be the biggest issue - the "are" is so far out when you need
>>>> to know it'll come that you can't be expected to hold out for it. Seven
>>>> words before "are", that about fills the short-term memory.
>>>> And "Scientist find that ..." would've made the structure obvious from
>>>> the start. Why would you drop "that" and then not "are", too?
>
> Maybe it was written by a Brit, for whom such a "that" is far more omissible
> than for an American.

It's from a British outlet.

bruce bowser

unread,
Jan 15, 2022, 12:24:21 PM1/15/22
to
On Saturday, January 15, 2022 at 10:51:47 AM UTC-5, Quinn C wrote:
> * Dingbat:
> > On Friday, January 14, 2022 at 6:54:15 AM UTC-8, Quinn C wrote:
> >> If that isn't a garden path, I haven't seen one:
> >>
> >> [I space it the way it appeared on my phone screen]
> >>
> >> Scientists find oldest ever
> >> confirmed human fossils are
> >> even older than previously
> >> thought
> >>
> >> This is especially bad because in context with "fossils", "find" really
> >> wants to have another meaning.
> >>
> >> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
> >>
> > Even in the world of fossils, it's a fossil!
> >
> > A spectacularly old fossil seems a rare find.
> > What is the other meaning you think "find" should have?
> Those fossils were *dug up* 50 years ago. They were now *determined* to
> be not less than 200,000, but over 230,000 years old.
>
> Hint: "human", not "humanoid".

On Earth, the oldest humanoid might be 16.6 million year old Ramapithecus "fossil primate dating from the Middle and Late Miocene epochs (about 16.6 million to 5.3 million years ago)".
-- https://www.britannica.com/topic/Ramapithecus

Yet, Xenon 129 (129Xe) atoms (only produced by air burst thermo nuclear weapon discharge?) are found on Mars all dating back 300 million years.

-- https://arc.aiaa.org/doi/abs/10.2514/6.2016-5529

Dingbat

unread,
Jan 15, 2022, 12:35:18 PM1/15/22
to
On Saturday, January 15, 2022 at 7:51:47 AM UTC-8, Quinn C wrote:
> * Dingbat:
> > On Friday, January 14, 2022 at 6:54:15 AM UTC-8, Quinn C wrote:
> >> If that isn't a garden path, I haven't seen one:
> >>
> >> [I space it the way it appeared on my phone screen]
> >>
> >> Scientists find oldest ever
> >> confirmed human fossils are
> >> even older than previously
> >> thought
> >>
> >> This is especially bad because in context with "fossils", "find" really
> >> wants to have another meaning.
> >>
> >> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
> >>
> > Even in the world of fossils, it's a fossil!
> >
> > A spectacularly old fossil seems a rare find.
> > What is the other meaning you think "find" should have?
> Those fossils were *dug up* 50 years ago. They were now *determined* to
> be not less than 200,000, but over 230,000 years old.
>
> Hint: "human", not "humanoid".
> --
Yes, that age would make them Homo. Were they Sapiens too?

Jerry Friedman

unread,
Jan 15, 2022, 12:56:45 PM1/15/22
to
On Friday, January 14, 2022 at 3:29:01 PM UTC-7, Quinn C wrote:
> * Jerry Friedman:
> > On Friday, January 14, 2022 at 11:26:38 AM UTC-7, Quinn C wrote:
> >> * Anders D. Nygaard:
> >>> Den 14-01-2022 kl. 15:54 skrev Quinn C:
> >>>> If that isn't a garden path, I haven't seen one:
> >>>>
> >>>> [I space it the way it appeared on my phone screen]
> >>>>
> >>>> Scientists find oldest ever
> >>>> confirmed human fossils are
> >>>> even older than previously
> >>>> thought
> >>>>
> >>>> This is especially bad because in context with "fossils", "find" really
> >>>> wants to have another meaning.
> >>>
> >>> I don't find it too bad - I got it in one.
> >
> >> That seems hardly possible. Is there something wrong with your internal
> >> parser that it didn't recognize at "fossils" that you have a complete
> >> sentence now? Or do you routinely jump around the sentence and took in
> >> the "are" before processing the "oldest ever ..." part?
> >
> > Maybe you could consider making your point without suggesting that
> > there's something wrong with the person you're arguing with.

> Would you allow this: You can't run? Is there something wrong with your
> heart (or legs)?

Maybe in certain settings (though I can't allow or disallow it).

When Ross Howard, a respected regular, mentioned here that he was in a
wheelchair, no one asked him what the cause was, as I recall.

But the main thing I want to say is that I see a big difference between
suggesting a physical disability and suggesting even a minor and localized
mental one. In my opinion the latter is far more offensive to most people.
(I'm not speaking for Anders, but offering you advice in general.)

And see below.

> I didn't conceptualize the above as a personal failing, as it's not
> something I think we have conscious control over. At best we may be able
> to improve its working by training, but that's not something most people
> do intentionally, either, certainly not in a native language.

Calling someone a moron, a retard, etc., is offensive though we don't have
conscious control over it.

> >> That might be the biggest issue - the "are" is so far out when you need
> >> to know it'll come that you can't be expected to hold out for it. Seven
> >> words before "are", that about fills the short-term memory.
> >
> > As I see it, you don't need "are" till after "fossils", which is where it is.

> You need to know it's coming in order to realize that "fossils" is not
> the object of "find" and to get to the right meaning of "find": "find
> out", not "unearth", the latter being primed by "scientists find
> fossils".
>
> So when I arrived at "fossils", my parser jumped on the obvious
> interpretation and then at "are", I needed to a) reinterpret the
> structure of the sentence and b) reinterpret "find".

I agree with all that.

> The challenge is to keep multiple decisions of that kind open. So
> really, what is "wrong" with Anders' parser as opposed to mine seems to
> be that his is able to keep more ambiguities undecided. A bit like
> there's something wrong with my mental mathematics module in that I
> often don't see the continuation of a sequence that is obvious to most
> people, because I still see other good options.

Yes, exactly, so there wasn't even any basis for you to think something
was wrong (without scare quotes) with his English parser.

> "What's wrong with you?" in casual conversation can mean "you're not
> normal", and that can mean you're extraordinary.

If there are clues to the irony.

> >> And "Scientist find that ..." would've made the structure obvious from
> >> the start.

> And narrowed down the specific meaning of "find".

I agree.

> >> Why would you drop "that" and then not "are", too?
> >
> > I agree that "that" would have helped (and I'd delete "ever"). Deleting
> > "are", if that's what you're suggesting, wouldn't help me.

> I didn't intend to say it'd help, I just thought delete both if the goal
> is to be as short as possible, or leave both in if you want it to be
> easier to understand.

OK, I wasn't sure.

> > Then the
> > sentence could be read "Scientists find oldest ever confirmed
> > human fossils [that are] even older than previously thought."

> True. I didn't see that. Semantically a bit convoluted, but of course
> grammatically fine.

--
Jerry Friedman

Tony Cooper

unread,
Jan 15, 2022, 1:02:41 PM1/15/22
to
On Sat, 15 Jan 2022 09:56:42 -0800 (PST), Jerry Friedman
<jerry_f...@yahoo.com> wrote:



>Calling someone a moron, a retard, etc., is offensive though we don't have
>conscious control over it.

Or "coward", but no one with the mental maturity to be interested in
a.u.e. would use such terms.

Quinn C

unread,
Jan 15, 2022, 5:32:57 PM1/15/22
to
* Dingbat:
Yes, that's what it's about. Oldest homo sapiens fossils.

--
Please stop treating gender as though it were a set menu.
Gender is an a la carte arrangement.
-- S. Bear Bergman, The Field Guide to Transmasculine Creatures

Kerr-Mudd, John

unread,
Jan 16, 2022, 5:23:10 AM1/16/22
to

Dingbat

unread,
Jan 16, 2022, 7:45:46 AM1/16/22
to
On Friday, January 14, 2022 at 6:54:15 AM UTC-8, Quinn C wrote:
> If that isn't a garden path, I haven't seen one:
>
> [I space it the way it appeared on my phone screen]
>
> Scientists find oldest ever
> confirmed human fossils are
> even older than previously
> thought
>
> This is especially bad because in context with "fossils", "find" really
> wants to have another meaning.
>
> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
> --
> ... while there are people who are consecrated, chronic
> assholes--like Donald Trump for example, or General Patton--
> it's a condition that all of us are liable to.
> -- Geoffrey Nunberg, 2012 interview

I just got another link:
<https://www.inverse.com/science/inverse-daily-011422>

CDB

unread,
Jan 16, 2022, 8:29:13 AM1/16/22
to
On 1/15/2022 12:35 PM, Dingbat wrote:
> Quinn C wrote:
>> * Dingbat:
>>> Quinn C wrote:

>>>> If that isn't a garden path, I haven't seen one:

>>>> [I space it the way it appeared on my phone screen]

>>>> Scientists find oldest ever confirmed human fossils are even
>>>> older than previously thought

>>>> This is especially bad because in context with "fossils",
>>>> "find" really wants to have another meaning.

>>>> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>

>>>
>>>>
Even in the world of fossils, it's a fossil!

>>> A spectacularly old fossil seems a rare find. What is the other
>>> meaning you think "find" should have?
>> Those fossils were *dug up* 50 years ago. They were now
>> *determined* to be not less than 200,000, but over 230,000 years
>> old.

>> Hint: "human", not "humanoid".

> Yes, that age would make them Homo. Were they Sapiens too?

Are we there yet, Homie?

I'm leaning towards "Homo fabulans" as being more accurate.

Dingbat

unread,
Jan 17, 2022, 3:49:57 AM1/17/22
to
On Saturday, January 15, 2022 at 2:32:57 PM UTC-8, Quinn C wrote:
> * Dingbat:
> > On Saturday, January 15, 2022 at 7:51:47 AM UTC-8, Quinn C wrote:
> >> * Dingbat:
> >>> On Friday, January 14, 2022 at 6:54:15 AM UTC-8, Quinn C wrote:
> >>>> If that isn't a garden path, I haven't seen one:
> >>>>
> >>>> [I space it the way it appeared on my phone screen]
> >>>>
> >>>> Scientists find oldest ever
> >>>> confirmed human fossils are
> >>>> even older than previously
> >>>> thought
> >>>>
> >>>> This is especially bad because in context with "fossils", "find" really
> >>>> wants to have another meaning.
> >>>>
> >>>> <https://news.sky.com/story/scientists-find-oldest-ever-confirmed-human-fossils-are-even-older-than-previously-thought-12514360>
> >>>>
> >>> Even in the world of fossils, it's a fossil!
> >>>
> >>> A spectacularly old fossil seems a rare find.
> >>> What is the other meaning you think "find" should have?
> >> Those fossils were *dug up* 50 years ago. They were now *determined* to
> >> be not less than 200,000, but over 230,000 years old.
> >>
> >> Hint: "human", not "humanoid".
> >> --
> > Yes, that age would make them Homo. Were they Sapiens too?
> Yes, that's what it's about. Oldest homo sapiens fossils.
>
> --
Thanks. Re. Homo but not Sapiens, here's an older find in Morocco:
https://www.sciencealert.com/the-earliest-unequivocal-evidence-of-our-species-might-be-even-older-than-we-realized

Quinn C

unread,
Jan 17, 2022, 7:07:52 PM1/17/22
to
* Jerry Friedman:

> On Friday, January 14, 2022 at 3:29:01 PM UTC-7, Quinn C wrote:
>> * Jerry Friedman:
>>> On Friday, January 14, 2022 at 11:26:38 AM UTC-7, Quinn C wrote:
>>>> * Anders D. Nygaard:
>>>>> Den 14-01-2022 kl. 15:54 skrev Quinn C:
>>>>>> If that isn't a garden path, I haven't seen one:
>>>>>>
>>>>>> [I space it the way it appeared on my phone screen]
>>>>>>
>>>>>> Scientists find oldest ever
>>>>>> confirmed human fossils are
>>>>>> even older than previously
>>>>>> thought
>>>>>>
>>>>>> This is especially bad because in context with "fossils", "find" really
>>>>>> wants to have another meaning.
>>>>>
>>>>> I don't find it too bad - I got it in one.
>>>
>>>> That seems hardly possible. Is there something wrong with your internal
>>>> parser that it didn't recognize at "fossils" that you have a complete
>>>> sentence now? Or do you routinely jump around the sentence and took in
>>>> the "are" before processing the "oldest ever ..." part?
>>>
>>> Maybe you could consider making your point without suggesting that
>>> there's something wrong with the person you're arguing with.
>
>> Would you allow this: You can't run? Is there something wrong with your
>> heart (or legs)?
>
> Maybe in certain settings (though I can't allow or disallow it).

I think you can, once you land on the right meaning of "allow".

> When Ross Howard, a respected regular, mentioned here that he was in a
> wheelchair, no one asked him what the cause was, as I recall.

It might be inappropriate to ask for causes this way, because it's quite
a personal matter. But that's a different point from the one I was
making, whether a peculiarity is a personal failing or not.

> But the main thing I want to say is that I see a big difference between
> suggesting a physical disability and suggesting even a minor and localized
> mental one. In my opinion the latter is far more offensive to most people.
> (I'm not speaking for Anders, but offering you advice in general.)

I truly appreciate the advice. Even if it's just to know that my words
are received differently from how I intended them, or how I believe I'd
receive them (insofar as I can tell), that's valuable information, and I
I'm not offended by the way you present it.

The rest of my comments is also not meant as "but I'm right". I'm not
saying there's something wrong with you (or others) because you receive
the words differently than intended. I'm working through how I can apply
the lessons I take from your advice.

----

I'd not classify an unusual working of someone's language comprehension
as a mental disability, unless it's a repeated impediment to
communication.

The crux here may be that I have training in linguistics, and from that,
a certain idea of how language processing works in the mind/brain that
may differ from the general public. The result may be a kind of
misunderstanding that is vaguely similar to that which often happens
with accusations of acting in racist ways. Both are variations on the
theme that the majority of your mental processes aren't accessible to
your consciousness (not even to introspection, even less to conscious
control), an idea that many aren't comfortable accepting.

> And see below.
>
>> I didn't conceptualize the above as a personal failing, as it's not
>> something I think we have conscious control over. At best we may be able
>> to improve its working by training, but that's not something most people
>> do intentionally, either, certainly not in a native language.
>
> Calling someone a moron, a retard, etc., is offensive though we don't have
> conscious control over it.

Because those words have a history of being used to disparage. And given
existing prejudices, it's hard to protect any word from that fate.

And then there's the privacy issue, which makes it difficult to ask
about disabilities even when the motive is genuine empathy and a
willingness to make accommodations.

>>>> That might be the biggest issue - the "are" is so far out when you need
>>>> to know it'll come that you can't be expected to hold out for it. Seven
>>>> words before "are", that about fills the short-term memory.
>>>
>>> As I see it, you don't need "are" till after "fossils", which is where it is.
>
>> You need to know it's coming in order to realize that "fossils" is not
>> the object of "find" and to get to the right meaning of "find": "find
>> out", not "unearth", the latter being primed by "scientists find
>> fossils".
>>
>> So when I arrived at "fossils", my parser jumped on the obvious
>> interpretation and then at "are", I needed to a) reinterpret the
>> structure of the sentence and b) reinterpret "find".
>
> I agree with all that.
>
>> The challenge is to keep multiple decisions of that kind open. So
>> really, what is "wrong" with Anders' parser as opposed to mine seems to
>> be that his is able to keep more ambiguities undecided. A bit like
>> there's something wrong with my mental mathematics module in that I
>> often don't see the continuation of a sequence that is obvious to most
>> people, because I still see other good options.
>
> Yes, exactly, so there wasn't even any basis for you to think something
> was wrong (without scare quotes) with his English parser.
>
>> "What's wrong with you?" in casual conversation can mean "you're not
>> normal", and that can mean you're extraordinary.
>
> If there are clues to the irony.

I think there's a spectrum of possible attitudes, not all of which I'd
classify as irony.

If I'd say "You enjoy classical music? What's wrong with you?", then I
might really think that's weird, in the sense that I can't imagine it or
that I don't know anyone else who enjoys it. What I'd hope in order for
the other person to receive it the right way is that they know I
wouldn't actually judge a person by their taste in music, because I
don't think it's a measure of their character or moral standing, but
something there's no accounting for. I.e., wrong in the sense of "not
working within normal parameters", but not wrong in any ethically
relevant sense.

----

On some level, I may also have internalized "what's wrong with you?"
(and variations) as a traditional way of talking on Usenet, which isn't
to be taken literally.

--
Genocide is not like gifts - the thought doesn't count.
-- 22 minutes,
on Erin O'Toole's statement that residential
schools were "meant to try and provide education."

Jerry Friedman

unread,
Jan 18, 2022, 11:31:05 PM1/18/22
to
I guess I tend to be a little literal with "allow", "accept", and some other words.

> > When Ross Howard, a respected regular, mentioned here that he was in a
> > wheelchair, no one asked him what the cause was, as I recall.

> It might be inappropriate to ask for causes this way, because it's quite
> a personal matter. But that's a different point from the one I was
> making, whether a peculiarity is a personal failing or not.

Yes, but the suggestion that it might be a personal failing is the problem I
see.

> > But the main thing I want to say is that I see a big difference between
> > suggesting a physical disability and suggesting even a minor and localized
> > mental one. In my opinion the latter is far more offensive to most people.
> > (I'm not speaking for Anders, but offering you advice in general.)

> I truly appreciate the advice. Even if it's just to know that my words
> are received differently from how I intended them, or how I believe I'd
> receive them (insofar as I can tell), that's valuable information, and I
> I'm not offended by the way you present it.
>
> The rest of my comments is also not meant as "but I'm right". I'm not
> saying there's something wrong with you (or others) because you receive
> the words differently than intended. I'm working through how I can apply
> the lessons I take from your advice.

I appreciate that, and I'm sorry to continue to disagree with you, but I'm still
hoping to say things that are helpful.

> ----
>
> I'd not classify an unusual working of someone's language comprehension
> as a mental disability, unless it's a repeated impediment to
> communication.
>
> The crux here may be that I have training in linguistics, and from that,
> a certain idea of how language processing works in the mind/brain that
> may differ from the general public. The result may be a kind of
> misunderstanding that is vaguely similar to that which often happens
> with accusations of acting in racist ways. Both are variations on the
> theme that the majority of your mental processes aren't accessible to
> your consciousness (not even to introspection, even less to conscious
> control), an idea that many aren't comfortable accepting.

Without any linguistic training, I agree with that theme. However, it
makes answering questions about one's internal parser difficult, and
the possibility of something wrong with it no less a personal failing.

> > And see below.
> >
> >> I didn't conceptualize the above as a personal failing, as it's not
> >> something I think we have conscious control over. At best we may be able
> >> to improve its working by training, but that's not something most people
> >> do intentionally, either, certainly not in a native language.
> >
> > Calling someone a moron, a retard, etc., is offensive though we don't have
> > conscious control over it.

> Because those words have a history of being used to disparage. And given
> existing prejudices, it's hard to protect any word from that fate.

I'm suggesting the reason those words and many similar ones are used to
disparage is in the concept, not the choice of words--which is the reason for
the "euphemism treadmill".

> And then there's the privacy issue, which makes it difficult to ask
> about disabilities even when the motive is genuine empathy and a
> willingness to make accommodations.

I agree.
I'm not familiar with that trope. If I say or hear "What's wrong with you?"
or anything similar, I think it means something really wrong, unless it
obviously means the opposite. "You're a sixteen-year-old boy with a
new license, and you're not taking any foolish risks? What's wrong with
you?"

(If one is still allowed to say "boy" in this context.)

> ----
>
> On some level, I may also have internalized "what's wrong with you?"
> (and variations) as a traditional way of talking on Usenet, which isn't
> to be taken literally.

OK, I'm not familiar with that.

--
Jerry Friedman
0 new messages