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 Historical content remains viewable.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

Phrasal Nouns

17 views
Skip to first unread message

DKleinecke

unread,
May 2, 2013, 8:03:39 PM5/2/13
to
We have been through phrasal verbs - now consider phrasal nouns.

"Phrasal" seems to be a term of art in language teaching with the
meaning of "more than one word acting like a single word". Bongartz
"Noun combinations in Interlanguage" (2002) has a discussion about
these issues couched in Chomskian terms which I think I follow but
there is always the chance I became confused.

"Phrasal Noun" is usually used to mean a noun made up of a preposition
and a a verb - like "onset" or "setup". This is not the meaning I
would give the term.

Bongartz seems to present the following view of noun compounding: If
you put two nouns side by side and think of them as a pair then you
have no a priori idea of their relationship and there are usually
multiple reading of what the pair means. However if you insert a
proposition between the nouns them, "the preposition assigns a
thematic role to the DP modifier."

I believe Bongartz is assuming the first determiner leading the first
noun is the head and the second is the modifier. I can't follow this
account any further because, so far as I know, only verbs have
thematic roles and I see no verb here. Bongartz explains the special
behavior of the preposition "of" as assigning the thematic role to the
underlying verb in the case the head noun is deverbal - like "driver
of the truck" transforms (or whatever is current jargon) into "X
drives the truck". There is more but I am now completely lost.

From where I sit, "of" is most often a version of a possessive and the
phrase is equivalent to a possessive "the truck's driver". BUT in many
of the occurrences of "of" it cannot be read comfortably as a
possessive. For example, "a majority of US citizens". In most of the
cases I have seen the first part is some kind of measure and it seems
that "50% of O" and the like are idiomatic. Hence I think there are
two kinds of "of" - a kind that behaves like other prepositions and a
kind that behave like a possessive.

My preliminary observation is that those prepositional "phrases" that
are part of noun phrases (there are other prepositional phrases) are
usually idiomatic for the noun. The most important exception seems to
be locatives.

So I would define "phrasal noun" to include, besides compounds,
templated noun phrase like "majority of O", "candidate for O",
"interview with O" and "devotion to O" as phrasal nouns. A lot of
these are already recognized as idioms.

Nathan Sanders

unread,
May 2, 2013, 9:20:52 PM5/2/13
to
In article
<4201eaaf-1fb8-44e6...@l2g2000pbn.googlegroups.com>,
DKleinecke <dklei...@gmail.com> wrote:

> Bongartz seems to present the following view of noun compounding: If
> you put two nouns side by side and think of them as a pair then you
> have no a priori idea of their relationship and there are usually
> multiple reading of what the pair means.

Of course. "Dog house" and "steak house" have very different
relationships between the component nouns. (The dog much prefers the
first relationship!)

> However if you insert a
> proposition between the nouns them, "the preposition assigns a
> thematic role to the DP modifier."
>
> I believe Bongartz is assuming the first determiner leading the first
> noun is the head and the second is the modifier.

You've jumped from noun-noun compounding to NP-P-NP (or DP-P-DP)
strings. Her examples are "a candle for the party" and "a candle from
the party". No noun-noun compounds here.

> I can't follow this
> account any further because, so far as I know, only verbs have
> thematic roles and I see no verb here.

What do you think assigns the instrument role to "a knife" in (1)?

(1) John opened the box with a knife.

Or the source role to "the bakery" in (2)?

(2) John carried the box from the bakery.

Or the beneficiary role to "the kids" in (3)?

(3) John bought a cake for the kids.

> Bongartz explains the special
> behavior of the preposition "of" as assigning the thematic role to the
> underlying verb in the case the head noun is deverbal - like "driver
> of the truck" transforms (or whatever is current jargon) into "X
> drives the truck". There is more but I am now completely lost.
>
> From where I sit, "of" is most often a version of a possessive and the
> phrase is equivalent to a possessive "the truck's driver". BUT in many
> of the occurrences of "of" it cannot be read comfortably as a
> possessive.

You missed a crucial detail: adjuncts versus complements. "Of"
assigns the underlying <patient> role to its object if the "of" phrase
is a complement of main noun ("possession of drugs", "destruction of
the city", etc.).

Adjunct "of" is usually has a possessive or possessive-ish
interpretation, while complement "of" is usually the patient if the
main noun is a deverbal noun. Contrast:

(4) I'm annoyed by the singing of small children.
(5) I'm annoyed by the singing of national anthems.

> For example, "a majority of US citizens".

You seem to have missed another crucial detail: deverbal nouns.
"Majority" isn't a deverbal noun, so it isn't relevant to what
Bongartz is saying.

> My preliminary observation is that those prepositional "phrases" that
> are part of noun phrases (there are other prepositional phrases) are
> usually idiomatic for the noun. The most important exception seems to
> be locatives.

But Bongartz's point is that "of" phrases for deverbal nouns are not
idiomatic; they are predictable from the source verb's thematic roles.
If the source verb has a patient role, then for a deverbal noun taking
an "of" phrase complement, the object of "of" inherits that patient
role.

(6) murder the politician
(6') murder of the politician

(7) draw a circle
(7') drawing of a circle

(8) educate the students
(8') education of the students

(9) discover a cure
(9') discovery of a cure

If you change any of the parameters (by not using deverbal nouns, not
using "of", and/or not using a complement "of" phrase), then sure, the
meaning might no longer be predictable.

Nathan

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

Adam Funk

unread,
May 3, 2013, 8:04:48 AM5/3/13
to
On 2013-05-03, Nathan Sanders wrote:

> In article
><4201eaaf-1fb8-44e6...@l2g2000pbn.googlegroups.com>,
> DKleinecke <dklei...@gmail.com> wrote:
>
>> Bongartz seems to present the following view of noun compounding: If
>> you put two nouns side by side and think of them as a pair then you
>> have no a priori idea of their relationship and there are usually
>> multiple reading of what the pair means.
>
> Of course. "Dog house" and "steak house" have very different
> relationships between the component nouns. (The dog much prefers the
> first relationship!)


I guess if you didn't know, you might ask whether a "tin opener" was
made of tin.


--
"Mrs CJ and I avoid clichés like the plague."

Peter T. Daniels

unread,
May 3, 2013, 10:29:01 AM5/3/13
to
On May 3, 8:04 am, Adam Funk <a24...@ducksburg.com> wrote:
> On 2013-05-03, Nathan Sanders wrote:
> > In article
> ><4201eaaf-1fb8-44e6-9b21-265cf6e23...@l2g2000pbn.googlegroups.com>,
> >  DKleinecke <dkleine...@gmail.com> wrote:
>
> >> Bongartz seems to present the following view of noun compounding: If
> >> you put two nouns side by side and think of them as a pair then you
> >> have no a priori idea of their relationship and there are usually
> >> multiple reading of what the pair means.
>
> > Of course.  "Dog house" and "steak house" have very different
> > relationships between the component nouns.  (The dog much prefers the
> > first relationship!)
>
> I guess if you didn't know, you might ask whether a "tin opener" was
> made of tin.

Not if you heard it spoken, you wouldn't.

DKleinecke

unread,
May 3, 2013, 8:52:45 PM5/3/13
to
On May 2, 6:20 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <4201eaaf-1fb8-44e6-9b21-265cf6e23...@l2g2000pbn.googlegroups.com>,
>

DKleinecke

unread,
May 3, 2013, 9:33:07 PM5/3/13
to
Whatever came through on my post a moment ago was a slipped mouse -
I'll start over

Thank you for the careful reply. I found Bongartz very hard to follow
and perhaps gave up too soon.

On May 2, 6:20 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
>  DKleinecke <dkleine...@gmail.com> wrote:

> > Bongartz seems to present the following view of noun compounding: If
> > you put two nouns side by side and think of them as a pair then you
> > have no a priori idea of their relationship and there are usually
> > multiple reading of what the pair means.
>
> Of course.  "Dog house" and "steak house" have very different
> relationships between the component nouns.  (The dog much prefers the
> first relationship!)

I wasn't done with summarizing Bongartz yet

> > However if you insert a
> > proposition between the nouns them, "the preposition assigns a
> > thematic role to the DP modifier."
>
> > I believe Bongartz is assuming the first determiner leading the first
> > noun is the head and the second is the modifier.
>
> You've jumped from noun-noun compounding to NP-P-NP (or DP-P-DP)
> strings.  Her examples are "a candle for the party" and "a candle from
> the party".  No noun-noun compounds here.

I read that as Bongartz making exactly that movement and suggesting
that in this way we are resolving the ambiguity of a noun-noun
"compound" (I don't think she was talking compounds - just sequences).
That is, I thought I was following her jump.

>
> > I can't follow this
> > account any further because, so far as I know, only verbs have
> > thematic roles and I see no verb here.
>
> What do you think assigns the instrument role to "a knife" in (1)?
>
>      (1) John opened the box with a knife.

You mean it isn't the verb "open"? That would mean Bongartz thinks
the preposition "with" assigns instrument. How is this possible -
"with" has many meanings (in the dictionary sense - I am inclined to
say it has no meaning at all except in so far as it occurs in set
construction where it signals the construction).

> Or the source role to "the bakery" in (2)?
>
>      (2) John carried the box from the bakery.
>
> Or the beneficiary role to "the kids" in (3)?
>
>      (3) John bought a cake for the kids.
>
To all of these ditto. There are locative and purpose slots in the
sentence for discourse reasons and they filled with prepositional
phrase in these example. A locative phrase need not be prepositional -
eg "here", "there' and "everywhere". I can'r think of a non-
prepositional purpose slot filler just now.

> > Bongartz explains the special
> > behavior of the preposition "of" as assigning the thematic role to the
> > underlying verb in the case the head noun is deverbal - like "driver
> > of the truck" transforms (or whatever is current jargon) into "X
> > drives the truck".  There is more but I am now completely lost.
>
> > From where I sit, "of" is most often a version of a possessive and the
> > phrase is equivalent to a possessive "the truck's driver". BUT in many
> > of the occurrences of "of" it cannot be read comfortably as a
> > possessive.

> You missed a crucial detail: adjuncts versus complements.  "Of"
> assigns the underlying <patient> role to its object if the "of" phrase
> is a complement of main noun ("possession of drugs", "destruction of
> the city", etc.).

I reject any distinction between adjuncts and compliments (and the all
the rest of x-bar theory. With the words "From where I sit" I have
switched from following Bongartz to my own position which makes no
such distinction and does not view sentence architecture in x-bar
fashion

> Adjunct "of" is usually has a possessive or possessive-ish
> interpretation, while complement "of" is usually the patient if the
> main noun is a deverbal noun.  Contrast:
>
>      (4) I'm annoyed by the singing of small children.
>      (5) I'm annoyed by the singing of national anthems.
>

I am forced to make the same conclusion - there are two kinds of "of".
I just used a description you are not familiar with

> > For example,  "a majority of US citizens".

>
> You seem to have missed another crucial detail: deverbal nouns.
> "Majority" isn't a deverbal noun, so it isn't relevant to what
> Bongartz is saying.
>
I wasn't talking about Bongartz. She dropped out several paragraphs
above. I observed that her comment was limited to deverbal nouns but
made no comment on it.

> > My preliminary observation is that those prepositional "phrases" that
> > are part of noun phrases (there are other prepositional phrases) are
> > usually idiomatic for the noun. The most important exception seems to
> > be locatives.
>
> But Bongartz's point is that "of" phrases for deverbal nouns are not
> idiomatic; they are predictable from the source verb's thematic roles.
> If the source verb has a patient role, then for a deverbal noun taking
> an "of" phrase complement, the object of "of" inherits that patient
> role.

I wasn't discussing Bongartz. If you must - I consider the argument
that deverbal "of'" is not idiomatic because of thematic
considerations to be circular. It just leaves the thematic
relationship as what I would call an idiom. This all depends on what
one means by "idomatic". Those of us how see language as a rather
orderly collection of forms echoed by a speaker rather than rule-based
will not, of course, see introducing a rule as anything we want to do.
There is more - but this isn't what I was discussing

>      (6) murder the politician
>      (6') murder of the politician
>
>      (7) draw a circle
>      (7') drawing of a circle
>
>      (8) educate the students
>      (8') education of the students
>
>      (9) discover a cure
>      (9') discovery of a cure
>

> If you change any of the parameters (by not using deverbal nouns, not
> using "of", and/or not using a complement "of" phrase), then sure, the
> meaning might no longer be predictable.

(6) the politician's murder
(7') a circle's drawing
(8) the student's education
(9) a cure's discovery

These need to be explicated - but that is not what I was trying to do.

Nathan Sanders

unread,
May 4, 2013, 1:08:43 AM5/4/13
to
In article
<4586be0c-40b5-4e84...@ul7g2000pbc.googlegroups.com>,
DKleinecke <dklei...@gmail.com> wrote:

> On May 2, 6:20 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> >  DKleinecke <dkleine...@gmail.com> wrote:
>
> > > I can't follow this
> > > account any further because, so far as I know, only verbs have
> > > thematic roles and I see no verb here.
> >
> > What do you think assigns the instrument role to "a knife" in (1)?
> >
> >      (1) John opened the box with a knife.
>
> You mean it isn't the verb "open"?

Correct. Nearly any non-stative verb you can think of can be combined
with an instrument phrase headed by "with":

John ran with his new legs.
John smiled with his eyes.
John ate with his fingers.
John thinks with his stomach.

It's quite redundant to say that this fact must be encoded as a
property of nearly every non-stative verb, separately, rather than
saying it's encoded once as a property of "with".

> That would mean Bongartz thinks
> the preposition "with" assigns instrument.

Yes, but not just Bongartz; this is the standard analysis.

> How is this possible - "with" has many meanings

Are you suggesting that there are no such things as homophones? Why
can't there be multiple "with"s (or perhaps a single "with" that has
multiple meanings), each one assigning their own different semantic
roles?

In fact, we know this must be the case, since "with" in English is
ambiguous between the instrumental and the comitative, in pretty much
every case. Do you want to encode that ambiguity in every verb in
English? Or encode that ambiguity once, by having two different
"with"s?

> > Or the source role to "the bakery" in (2)?
> >
> >      (2) John carried the box from the bakery.
> >
> > Or the beneficiary role to "the kids" in (3)?
> >
> >      (3) John bought a cake for the kids.
> >
> To all of these ditto. There are locative and purpose slots in the
> sentence for discourse reasons and they filled with prepositional
> phrase in these example.

But how do you know whether a given PP fills a source "slot" rather
than a beneficiary "slot"? By the choice of preposition, of course!
The verb doesn't help. Notice that I can switch the two prepositional
phrases, putting them with completely different verbs, and yet their
retain their semantic contribution:

John carried the box for the kids.
John bought a cake from the bakery.

The obvious, simplest explanation is that the consistent semantic
contribution of a PP is inherent to the preposition, not to the verb,
since the semantics follow the preposition when it moves to a new
environment.

> A locative phrase need not be prepositional -
> eg "here", "there' and "everywhere".

None of those can be sources without a preposition.

> I can'r think of a non-
> prepositional purpose slot filler just now.

I don't know about "purpose", but if you mean beneficiary, that role
can seemingly be filled by a noun phrase, but only for some verbs
(however, see below). In those cases, it's obvious that the verb is
assigning the relevant role:

John bought the kids some ice cream.
John told the kids a story.

Note only a small number of verbs are able to assign this role
directly to an NP by themselves. Most verbs can't do it at all, and
require a "for" PP for any sort of beneficiary:

*John died me.
John died for me.

*John explained me the answer.
John explained the answer for me.

*John finished me the assignment.
John finished the assignment for me.

But pretty much every verb you can think of allows a beneficiary "for"
PP, suggesting again that it is the "for", not the verb, that assigns
the beneficiary role in those cases (again, to avoid redundantly
giving every single verb a beneficiary role marked by "for"):

John ran for me.
John smiled for me.
John ate for me.
John thinks for me.

Note that the beneficiary role assigned by "for" is subtly different
from the similar role assigned by those verbs that can assign a role
directly, because we can use both, and the relevant noun phrases play
different roles in the event:

John bought the kids some ice cream for me.
John told the kids a story for me.

In these cases, "me" is the true beneficiary, while "the kids" are
assigned similar but different roles (recipient, goal). Again, the
consistent role (beneficiary) corresponds to a consistent word
("for"), regardless of the rest of the words of the words, so the
simplest explanation is that "for" assigns the beneficiary role, and
that the verb-dependent role (recipient, goal) is assigned by the verb.

> > > Bongartz explains the special
> > > behavior of the preposition "of" as assigning the thematic role to the
> > > underlying verb in the case the head noun is deverbal - like "driver
> > > of the truck" transforms (or whatever is current jargon) into "X
> > > drives the truck".  There is more but I am now completely lost.
> >
> > > From where I sit, "of" is most often a version of a possessive and the
> > > phrase is equivalent to a possessive "the truck's driver". BUT in many
> > > of the occurrences of "of" it cannot be read comfortably as a
> > > possessive.
>
> > You missed a crucial detail: adjuncts versus complements.  "Of"
> > assigns the underlying <patient> role to its object if the "of" phrase
> > is a complement of main noun ("possession of drugs", "destruction of
> > the city", etc.).
>
> I reject any distinction between adjuncts and compliments

Then you reject mounds of cross-linguistic data showing that there is
a difference.

> With the words "From where I sit" I have
> switched from following Bongartz to my own position which makes no
> such distinction and does not view sentence architecture in x-bar
> fashion

X-bar theory isn't relevant. The different behavior of adjuncts and
complements is a fact about language; X-bar theory happens to provide
one way to formalize that distinction, but it's not required.
Dependency grammar , LFG, role and reference grammar, etc., all encode
the distinction, too.

> > Adjunct "of" is usually has a possessive or possessive-ish
> > interpretation, while complement "of" is usually the patient if the
> > main noun is a deverbal noun.  Contrast:
> >
> >      (4) I'm annoyed by the singing of small children.
> >      (5) I'm annoyed by the singing of national anthems.
>
> I am forced to make the same conclusion - there are two kinds of "of".

You seemed to want to avoid having two "with"s, but now you're okay
with two "of"s?

DKleinecke

unread,
May 4, 2013, 7:14:42 PM5/4/13
to
On May 3, 10:08 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <4586be0c-40b5-4e84-83ac-8272c7966...@ul7g2000pbc.googlegroups.com>,
On the whole I think it useless to go any further. You have stated
frequently that, as a phonologist, you have no special interest or
expertise in grammar. So I assume you are stating what you imagine is
"standard" mainstream linguistics. But your arguments are to me
"standard" Chomskian thought - not linguistic thought. Practically
every sentence involves preconceptions I do not share. You seem to be
bemused that you are considered a Chomskian. Look at this exchange for
why that might be so.

Nathan Sanders

unread,
May 4, 2013, 9:03:11 PM5/4/13
to
In article
<67195cc2-b7b0-4fd2...@oj8g2000pbb.googlegroups.com>,
> On the whole I think it useless to go any further. You have stated
> frequently that, as a phonologist, you have no special interest or
> expertise in grammar. So I assume you are stating what you imagine is
> "standard" mainstream linguistics. But your arguments are to me
> "standard" Chomskian thought - not linguistic thought. Practically
> every sentence involves preconceptions I do not share. You seem to be
> bemused that you are considered a Chomskian. Look at this exchange for
> why that might be so.

If you think dependency grammar, LFG, or role and reference grammar
are Chomskyan, then you know even less about syntax than I thought.

Nathan Sanders

unread,
May 5, 2013, 12:52:47 AM5/5/13
to
In article
<sanders-3DBE77...@news.eternal-september.org>,
Nathan Sanders <san...@alum.mit.edu> wrote:

> In article
> <67195cc2-b7b0-4fd2...@oj8g2000pbb.googlegroups.com>,
> DKleinecke <dklei...@gmail.com> wrote:
>
> > On May 3, 10:08�pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article
> > > <4586be0c-40b5-4e84-83ac-8272c7966...@ul7g2000pbc.googlegroups.com>,
> > >
> > > �DKleinecke <dkleine...@gmail.com> wrote:
> > > > On May 2, 6:20 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > > �DKleinecke <dkleine...@gmail.com> wrote:
> > >
> > > > I reject any distinction between adjuncts and compliments
> > >
> > > Then you reject mounds of cross-linguistic data showing that there is
> > > a difference.
> > >
> > > > With the words "From where I sit" I have
> > > > switched from following Bongartz to my own position which makes no
> > > > such distinction and does not view sentence architecture in x-bar
> > > > fashion
> > >
> > > X-bar theory isn't relevant. �The different behavior of adjuncts and
> > > complements is a fact about language; X-bar theory happens to provide
> > > one way to formalize that distinction, but it's not required.
> > > Dependency grammar , LFG, role and reference grammar, etc., all encode
> > > the distinction, too.

It's probably also worth pointing out that X-bar theory is about 20
years obsolete; Chomskyan syntax has been using bare phrase structure
instead since the early 90s.

DKleinecke

unread,
May 5, 2013, 9:25:23 PM5/5/13
to
On May 4, 9:52 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <sanders-3DBE77.21031104052...@news.eternal-september.org>,
>  Nathan Sanders <sand...@alum.mit.edu> wrote:
>
>
>
>
>
>
>
>
>
> > In article
> > <67195cc2-b7b0-4fd2-8721-eb5d24082...@oj8g2000pbb.googlegroups.com>,
By Chomskian I do not mean Chomsky's current theories - I mean
theories that arose from Chomskian preoccupations - responding to them
either positively or negatively. Dependency grammar indeed started
outside the Chomskian camp but has long since been absorbed by the
Chomskian theories about thematic relations. LFG is a fork from the
Chomskian mainstream. I don't know much about role and reference
grammar but it seems to be a development from the same thematic
relationships thinking that I called x-bar theory. As Peter has
pointed out the history of linguistics has not been well served by
linguists.

Nathan Sanders

unread,
May 5, 2013, 10:18:08 PM5/5/13
to
In article
<95588cbe-832d-4437...@hc4g2000pbb.googlegroups.com>,
> By Chomskian I do not mean Chomsky's current theories - I mean
> theories that arose from Chomskian preoccupations - responding to them
> either positively or negatively.

So even if the theory in question was explicitly a negative response
to Chomskyan theories, you're calling it Chomskyan? Even if Chomsky
himself argues against the theory?

That seems like an utterly useless definition of "Chomskyan", and not
one I had ever encountered before. I doubt you'll find many other
people who use that word that way either.

Fine, I guess by that overly broad definition, I'm probably a
Chomskyan, since I advocate plenty of ideas that I know Chomsky
disagrees with.

Brian M. Scott

unread,
May 5, 2013, 10:55:51 PM5/5/13
to
On Sun, 5 May 2013 18:25:23 -0700 (PDT), DKleinecke
<dklei...@gmail.com> wrote in
<news:95588cbe-832d-4437...@hc4g2000pbb.googlegroups.com>
in sci.lang:

[...]

> By Chomskian I do not mean Chomsky's current theories - I
> mean theories that arose from Chomskian preoccupations -
> responding to them either positively or negatively.

Using 'Chomskian' to mean 'either Chomskian or
anti-Chomskian' seems perverse.

> Dependency grammar indeed started outside the Chomskian
> camp but has long since been absorbed by the Chomskian
> theories about thematic relations. LFG is a fork from the
> Chomskian mainstream. I don't know much about role and
> reference grammar but it seems to be a development from
> the same thematic relationships thinking that I called
> x-bar theory. [...]

Which suggests to me that you have a private meaning for
'x-bar theory' as well.

Peter T. Daniels

unread,
May 6, 2013, 7:19:26 AM5/6/13
to
On May 5, 10:18 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <95588cbe-832d-4437-9f27-2d0858191...@hc4g2000pbb.googlegroups.com>,
The very fact that you are unaware that Chomsky co-opts and absorbs
"theories" originally presented in opposition to his own proposals
proves David's point about my point.

Has Chomsky ever explicitly said he was wrong about anything and was
changing his position? No. He always claims that each new attempt to
plug the holes in the "theory" is merely a "development" of it. The
term "Revised Extended Standard Theory" is worthy of any Comintern
apparatchik.

Nathan Sanders

unread,
May 6, 2013, 10:34:21 AM5/6/13
to
In article
<ff8eb83d-f4a8-44f3...@h13g2000yqe.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> Has Chomsky ever explicitly said he was wrong about anything

*splork*

Peter T. Daniels

unread,
May 6, 2013, 1:06:46 PM5/6/13
to
On May 6, 10:34 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <ff8eb83d-f4a8-44f3-9a4d-c65b0cd7d...@h13g2000yqe.googlegroups.com>,
>  "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> > Has Chomsky ever explicitly said he was wrong about anything
>
> *splork*

That doesn't answer the question.

Though it does seem to reveal the utter childishness of your approach
to the problem.

Nathan Sanders

unread,
May 6, 2013, 1:21:17 PM5/6/13
to
In article
<b40ee283-5b4e-400d...@n11g2000yqe.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> On May 6, 10:34�am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <ff8eb83d-f4a8-44f3-9a4d-c65b0cd7d...@h13g2000yqe.googlegroups.com>,
> > �"Peter T. Daniels" <gramma...@verizon.net> wrote:
> >
> > > Has Chomsky ever explicitly said he was wrong about anything
> >
> > *splork*
>
> That doesn't answer the question.

Because I was so shocked that you thought it was a question worth
asking. It never occurs to you to say you were wrong, so why would
you care whether other people do?

Peter T. Daniels

unread,
May 6, 2013, 4:40:54 PM5/6/13
to
On May 6, 1:21 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <b40ee283-5b4e-400d-964d-56474663d...@n11g2000yqe.googlegroups.com>,
>  "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> > On May 6, 10:34 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article
> > > <ff8eb83d-f4a8-44f3-9a4d-c65b0cd7d...@h13g2000yqe.googlegroups.com>,
> > > "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> > > > Has Chomsky ever explicitly said he was wrong about anything
>
> > > *splork*
>
> > That doesn't answer the question.
>
> Because I was so shocked that you thought it was a question worth
> asking.   It never occurs to you to say you were wrong, so why would
> you care whether other people do?

Do you simply not read what David writes to you, but just go on
autopilot?

Give an example where Chomsky explicitly disavowed something he
previously embraced, as opposed to claiming that B is a
"refinement" (or revision or extension) of A.

Nathan Sanders

unread,
May 6, 2013, 5:46:19 PM5/6/13
to
In article
<69049fb0-68a5-4b70...@i3g2000yqf.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> On May 6, 1:21�pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <b40ee283-5b4e-400d-964d-56474663d...@n11g2000yqe.googlegroups.com>,
> > �"Peter T. Daniels" <gramma...@verizon.net> wrote:
> >
> > > On May 6, 10:34 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > In article
> > > > <ff8eb83d-f4a8-44f3-9a4d-c65b0cd7d...@h13g2000yqe.googlegroups.com>,
> > > > "Peter T. Daniels" <gramma...@verizon.net> wrote:
> >
> > > > > Has Chomsky ever explicitly said he was wrong about anything
> >
> > > > *splork*
> >
> > > That doesn't answer the question.
> >
> > Because I was so shocked that you thought it was a question worth
> > asking. � It never occurs to you to say you were wrong, so why would
> > you care whether other people do?
>
> Do you simply not read what David writes to you, but just go on
> autopilot?

Do you still not see the irony oozing from your first question, but
just go on autopilot?

> Give an example where Chomsky explicitly disavowed something he
> previously embraced, as opposed to claiming that B is a
> "refinement" (or revision or extension) of A.

Bare phrase structure replacing X-bar theory is probably the most
obvious recent example:

"In the papers on economy and minimalism cited earlier, I took X'
theory to be given, with specific stipulated properties. Let's now
subject these assumptions to critical analysis, asking what the theory
of phrase structure should look like on minimalist assumptions and
what the consequences are for the theory of movement. [many pages of
critical analysis later] To summarize, it seems that we can hope to
eliminate the theory of phrase structure almost entirely, deriving its
properties on highly principled grounds."
<http://biolinguistica.files.wordpress.com/2010/03/chomsky-bare-phrase-
structure.pdf>

DKleinecke

unread,
May 6, 2013, 8:00:36 PM5/6/13
to
On May 5, 7:55 pm, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
> On Sun, 5 May 2013 18:25:23 -0700 (PDT), DKleinecke
> <dkleine...@gmail.com> wrote in
I think a better presentation of my point of view goes like this:
Many people have claimed that there was a Kuhnian Paradigm Shift in
linguistics during the decade centered on 1960 and that serious
linguistics began with Syntactic Structures, I think that historians
of language have proven that this idea is both wrong and misleading -
but the historians are largely ignored. I call anyone who explicitly
or implicitly demonstrates a belief in the alleged Paradigm Shift a
Chomskian. Is there a better name (that is neutral)? But I must
insist that I am talking about grammar - not about phonology or
semantics.

There was brief generative phonology but I think it is now long dead.
The best exposition I know of it is Eugene Loos' "The Phonology of
Capanahua" (SIL #20) which was published before SPE. I shouldn't
comment on semantics in a Chomskian World without re-reading "The
Linguistic Wars" but my reaction is that the generative semantics
crowd were onto something but rather than make a fork from Chomsky's
own thoughts they tried to take over the mainstream. They failed and
are now more-or-less forgotten. Thirty years ago I reviewed Horn's
"Lexical-Functional Grammar" for some journal (I have forgotten which)
but I still have the review copy. Glancing through it I see no reason
to change my old opinion - LFG is based on one particular position of
Chomsky's and a conscious decision to differ with the evolution that
Chomsky choose to follow.

"X-bar Theory" can be used in either narrow sense or a broad sense - I
intended a very broad sense.

Nathan Sanders

unread,
May 6, 2013, 9:50:33 PM5/6/13
to
In article
<97e1719a-e4fb-4f00...@k8g2000pbf.googlegroups.com>,
Well, since I don't hold the belief that "serious linguistics began
with Syntactic Structures", I am not a Chomskyan by your definition.

> "X-bar Theory" can be used in either narrow sense or a broad sense - I
> intended a very broad sense.

Broad or narrow (whatever that's supposed to mean; you keep redefining
well-established jargon in unorthodox ways, to suit the moment, so
it's hard to tell what you really mean), X-bar theory is 20 years out
of date with current Chomskyan phrase structure. Chomsky himself
explicitly argues against X-bar theory (see the quote I gave PTD).

Peter T. Daniels

unread,
May 6, 2013, 11:15:45 PM5/6/13
to
On May 6, 5:46 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <69049fb0-68a5-4b70-92bc-b1222eec3...@i3g2000yqf.googlegroups.com>,
He can simultaneously "eliminate" and "derive the properties of"
something?

And you can take that seriously?

Peter T. Daniels

unread,
May 6, 2013, 11:18:26 PM5/6/13
to
You might want to look at *Chomskyan (R)evolutions*, the Festschrift
for Konrad Koerner, edited by Douglas Kibbee (2010), which has my
chapter on the revision of the MA thesis, for an on the whole
Koernerian approach to Chomsky.

If you can get at JSTOR, you can at least consult a number of
substantive review articles.

Nathan Sanders

unread,
May 6, 2013, 11:21:27 PM5/6/13
to
In article
<896d3e31-b20e-4c25...@o2g2000yqb.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

Eliminate the theory, derive the properties of phrase structure.

> And you can take that seriously?

I can read it.

Nathan Sanders

unread,
May 6, 2013, 11:40:45 PM5/6/13
to
In article
<sanders-81A688...@news.eternal-september.org>,
Here's more of Chomsky arguing against his own earlier theories, from
_The Minimalist Program_ (1995):

"Concepts and principles regarded as fundamental in one chapter are
challenged and eliminated in those that follow. These include the
basic ideas of the Extended Standard Theory that were adopted in the
P&P approaches: D-Structure; S-Structure; government; the Projection
Principle and the theta-Criterion; other conditions held to apply at
D- and S-Structure; the Empty Category Principle; X-bar theory
generally; the operation Move alpha; the split-I hypothesis; and
others." (p.10)

He spends an entire book explicitly challenging his own previous
theories, and then eliminating them. That counts as saying he was
wrong.

Peter T. Daniels

unread,
May 7, 2013, 7:30:55 AM5/7/13
to
On May 6, 11:40 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <sanders-81A688.23212706052...@news.eternal-september.org>,
>  Nathan Sanders <sand...@alum.mit.edu> wrote:
>
>
>
>
>
>
>
>
>
> > In article
> > <896d3e31-b20e-4c25-aa68-43d178472...@o2g2000yqb.googlegroups.com>,
True, no one has ever suggested that he writes well.

If that's how you want to assign the antecedent of "its," fine. But
it's not the first reading that comes to mind.

> > > And you can take that seriously?
>
> > I can read it.
>
> Here's more of Chomsky arguing against his own earlier theories, from
> _The Minimalist Program_ (1995):
>
> "Concepts and principles regarded as fundamental in one chapter are
> challenged and eliminated in those that follow. These include the
> basic ideas of the Extended Standard Theory that were adopted in the
> P&P approaches: D-Structure; S-Structure; government; the Projection
> Principle and the theta-Criterion; other conditions held to apply at
> D- and S-Structure; the Empty Category Principle; X-bar theory
> generally; the operation Move alpha; the split-I hypothesis; and
> others." (p.10)
>
> He spends an entire book explicitly challenging his own previous
> theories, and then eliminating them.  That counts as saying he was
> wrong.

Good! You found an example! (Almost.)

Except, of course, that this could simply be dialectic: the concepts
are "regarded as fundamental in one chapter" but "challenged and
eliminated" in others.

What is the ultimate _synthesis_? Are all those "basic ideas"
rejected, or are they simply _recast_ in a different notation?

(The notion of "notational variant" was explicated -- and ridiculed --
at NAAHoLS in Boston by Fritz Newmeyer: he claimed that when Jim
McCawley demonstrated that Interpretive Semantics was a "notational
variant" of Generative Semantics, he was merely throwing Chomsky's
favored term of abuse back at him.)

Nathan Sanders

unread,
May 7, 2013, 8:24:25 AM5/7/13
to
In article
<63b19897-8803-4b51...@i3g2000yqf.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

It's the only reading that makes given the difference between X-bar
theory and bare phrase structure, and given the fundamental concept
behind the minimalist program.

> > > > And you can take that seriously?
> >
> > > I can read it.
> >
> > Here's more of Chomsky arguing against his own earlier theories, from
> > _The Minimalist Program_ (1995):
> >
> > "Concepts and principles regarded as fundamental in one chapter are
> > challenged and eliminated in those that follow. These include the
> > basic ideas of the Extended Standard Theory that were adopted in the
> > P&P approaches: D-Structure; S-Structure; government; the Projection
> > Principle and the theta-Criterion; other conditions held to apply at
> > D- and S-Structure; the Empty Category Principle; X-bar theory
> > generally; the operation Move alpha; the split-I hypothesis; and
> > others." (p.10)
> >
> > He spends an entire book explicitly challenging his own previous
> > theories, and then eliminating them. �That counts as saying he was
> > wrong.
>
> Good! You found an example! (Almost.)
>
> Except, of course, that this could simply be dialectic: the concepts
> are "regarded as fundamental in one chapter" but "challenged and
> eliminated" in others.
>
> What is the ultimate _synthesis_? Are all those "basic ideas"
> rejected, or are they simply _recast_ in a different notation?

They are rejected. Read the book.

(What do you think "eliminated" means?)

Peter T. Daniels

unread,
May 7, 2013, 11:08:12 AM5/7/13
to
On May 7, 8:24 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <63b19897-8803-4b51-9bee-d7b2c864c...@i3g2000yqf.googlegroups.com>,
Unfortunately, it's not what it says. Clearly he's one of those ueber-
prolific writers who never look back over what they have produced
before sending it off to be printed.

A superb example is found in a volume by Jacob Neusner and Andrew
Greeley (The Bible and Us: A Priest and A Rabbi Read Scripture
Together [New York 1990]), ostensibly a Jewish-Catholic "dialogue," by
two scholars of that ilk, who simply sent chapters to each other where
each man's chapters do not maintain a coherent position on issues that
are discussed. Each one simply read what the other had most recently
sent and never looked back at what he himself had previously written.
> > > > > And you can take that seriously?
>
> > > > I can read it.
>
> > > Here's more of Chomsky arguing against his own earlier theories, from
> > > _The Minimalist Program_ (1995):
>
> > > "Concepts and principles regarded as fundamental in one chapter are
> > > challenged and eliminated in those that follow. These include the
> > > basic ideas of the Extended Standard Theory that were adopted in the
> > > P&P approaches: D-Structure; S-Structure; government; the Projection
> > > Principle and the theta-Criterion; other conditions held to apply at
> > > D- and S-Structure; the Empty Category Principle; X-bar theory
> > > generally; the operation Move alpha; the split-I hypothesis; and
> > > others." (p.10)
>
> > > He spends an entire book explicitly challenging his own previous
> > > theories, and then eliminating them. That counts as saying he was
> > > wrong.
>
> > Good! You found an example! (Almost.)
>
> > Except, of course, that this could simply be dialectic: the concepts
> > are "regarded as fundamental in one chapter" but "challenged and
> > eliminated" in others.
>
> > What is the ultimate _synthesis_? Are all those "basic ideas"
> > rejected, or are they simply _recast_ in a different notation?
>
> They are rejected.  Read the book.
>
> (What do you think "eliminated" means?)

What do you think "fundamental in one chapter" and "challenged and
eliminated" mean?

Maybe you're not familiar with the concept of (Marxist) dialectic so
you didn't understand what I was saying. In dialectic, one presents a
thesis and an antithesis and arrives at a synthesis. Chomsky is
familiar with Marxist rhetoric.

Nathan Sanders

unread,
May 7, 2013, 12:29:51 PM5/7/13
to
In article
<01e8eb80-ed43-4cf2...@g9g2000vbl.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

Read the book: it is arranged roughly historically.

Bare phrase structure is not a "synthesis" of X-bar theory. It is an
outright replacement.

Adam Funk

unread,
May 7, 2013, 3:40:17 PM5/7/13
to
On 2013-05-06, Peter T. Daniels wrote:

> Has Chomsky ever explicitly said he was wrong about anything and was
> changing his position?


HA!!!


--
But the government always tries to coax well-known writers into the
Establishment; it makes them feel educated. [Robert Graves]

Peter T. Daniels

unread,
May 7, 2013, 3:46:26 PM5/7/13
to
On May 7, 12:29 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <01e8eb80-ed43-4cf2-a26f-c4819ef77...@g9g2000vbl.googlegroups.com>,
Yes, you have not understood the concept of dialectic. A synthesis is
the outcome of tension between thesis and antithesis. It is not "of"
anything.

> It is an
> outright replacement.

It does not cease being the essence of the same theory presented by
Chomsky in 1951.

(Which, of course, is what David was referring to when you denied
being a Chomskyan.)

Has Chomsky recanted his position on the "magic mutation" that
suddenly yielded the language faculty?

Is it worth noting that Marc Hauser, as in Hauser, Chomsky, and Fitch
2002, which attempts to justify that position, was exposed as a data-
faker and is no longer employed by Harvard University?

Neurolinguistic investigation is making that position less and less
tenable, just as the development of chemistry in the late 18th century
made the notion of phlogisthon, and the development of physics in the
late 19th century made the notion of ether, eventually untenable.

Nathan Sanders

unread,
May 7, 2013, 3:59:15 PM5/7/13
to
In article
<e345c4a7-e2d3-4e1c...@e13g2000yqp.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

I understand X-bar theory and bare phrase structure.

> >�It is an
> > outright replacement.
>
> It does not cease being the essence of the same theory presented by
> Chomsky in 1951.

Since you don't even know what bare phrase structure is, let alone
understand it (and I doubt you really understand X-bar theory either),
you're in no position to determine that.

[snip irrelevancies]

Chomsky, unlike you, has said he was wrong.

DKleinecke

unread,
May 7, 2013, 10:20:17 PM5/7/13
to
On May 7, 12:59 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <e345c4a7-e2d3-4e1c-b2e1-ca527438b...@e13g2000yqp.googlegroups.com>,
I view bare phrase structure as a notational variant of x-bar theory.
About all that it adds is that every merge is binary. I also view the
notions of adjunct and complement as another notational variant of
that theory. I call the theory x-bar theory because I know of no
better name. The fact that Chomsky has changed the name should in no
way obscure the identity of all these things.

Nathan Sanders

unread,
May 7, 2013, 11:13:36 PM5/7/13
to
In article
<6f49fec2-dcc5-4445...@k8g2000pbf.googlegroups.com>,
DKleinecke <dklei...@gmail.com> wrote:

> On May 7, 12:59�pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <e345c4a7-e2d3-4e1c-b2e1-ca527438b...@e13g2000yqp.googlegroups.com>,
> > �"Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > It does not cease being the essence of the same theory presented by
> > > Chomsky in 1951.
> >
> > Since you don't even know what bare phrase structure is, let alone
> > understand it (and I doubt you really understand X-bar theory either),
> > you're in no position to determine that.
>
> I view bare phrase structure as a notational variant of x-bar theory.

Then you clearly don't understand them.

> About all that it adds is that every merge is binary. I also view the
> notions of adjunct and complement as another notational variant of
> that theory. I call the theory x-bar theory because I know of no
> better name.

"Bare phrase structure" is a better name for BPS.

> The fact that Chomsky has changed the name should in no
> way obscure the identity of all these things.

The fact that you don't understand the fundamental differences between
them doesn't mean those differences don't exist.

(Of course, I don't believe in either X-bar theory or BPS...)

Peter T. Daniels

unread,
May 7, 2013, 11:16:48 PM5/7/13
to
On May 7, 11:13 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <6f49fec2-dcc5-4445-8ee4-cbceff380...@k8g2000pbf.googlegroups.com>,
>  DKleinecke <dkleine...@gmail.com> wrote:
> > On May 7, 12:59 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article
> > > <e345c4a7-e2d3-4e1c-b2e1-ca527438b...@e13g2000yqp.googlegroups.com>,
> > > "Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > > It does not cease being the essence of the same theory presented by
> > > > Chomsky in 1951.
>
> > > Since you don't even know what bare phrase structure is, let alone
> > > understand it (and I doubt you really understand X-bar theory either),
> > > you're in no position to determine that.
>
> > I view bare phrase structure as a notational variant of x-bar theory.
>
> Then you clearly don't understand them.
>
> > About all that it adds is that every merge is binary. I also view the
> > notions of adjunct and complement as another notational variant of
> > that theory. I call the theory x-bar theory because I know of no
> > better name.
>
> "Bare phrase structure" is a better name for BPS.
>
> > The fact that Chomsky has changed the name should in no
> > way obscure the identity of all these things.
>
> The fact that you don't understand the fundamental differences between
> them doesn't mean those differences don't exist.
>
> (Of course, I don't believe in either X-bar theory or BPS...)

So they are equivalent.

(That's Aristotelian logic, which IIRC you have previously
demonstrated less than a perfect grasp of.)

So why are you denying it, other than being a devoted Chomskyan?

Nathan Sanders

unread,
May 7, 2013, 11:24:05 PM5/7/13
to
In article
<927b5b9b-5eac-42e1...@w13g2000vbn.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

I don't believe in glottochronology or Greenbergian mass comparison,
but they aren't "equivalent", because they don't make the same
predictions.

> So why are you denying it,

Because I've read and understood the relevant work.

DKleinecke

unread,
May 8, 2013, 8:44:39 PM5/8/13
to
On May 7, 8:24 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <927b5b9b-5eac-42e1-89e4-3061d1fd7...@w13g2000vbn.googlegroups.com>,
Now that I think about I believe I could take the position that
Chomsky has gone full circle and the minimalist program is nothing but
a notational variant of Syntactic Structures. There's phrase
structure (merge) and transformations (move). I might even have tried
to demonstrate that if the minimalist program weren't so incoherent.

What would you require me to demonstrate to allow me to call two
theories notational variants?

Nathan Sanders

unread,
May 9, 2013, 12:05:34 AM5/9/13
to
In article
<2c1472c8-0a5e-42af...@h9g2000pbr.googlegroups.com>,
> Now that I think about I believe I could take the position that
> Chomsky has gone full circle and the minimalist program is nothing but
> a notational variant of Syntactic Structures. There's phrase
> structure (merge) and transformations (move).

You have a deeply flawed conception of the minimalist program.

There are no transformations in minimalism: Move is just a particular
instantiation of Merge (see Chomsky 2001).

> I might even have tried
> to demonstrate that if the minimalist program weren't so incoherent.
>
> What would you require me to demonstrate to allow me to call two
> theories notational variants?

Demonstrate how a unary branching node (as allowed in X-bar theory)
can be represented in bare phrase structure, which happens to only
allow binary branching.

(The inability of either system to allow ternary branching is a
primary reason I dislike both.)

Peter T. Daniels

unread,
May 9, 2013, 7:16:32 AM5/9/13
to
On May 9, 12:05 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <2c1472c8-0a5e-42af-93b3-9b86a91b1...@h9g2000pbr.googlegroups.com>,
What a bizarre thing to complain about. A "unary branching node" is
neither a branch nor a node, it's just a change of label.

> can be represented in bare phrase structure, which happens to only
> allow binary branching.
>
> (The inability of either system to allow ternary branching is a
> primary reason I dislike both.)

Yes, you've already said bizarre things about "In Defense of the
Number Two."

Nathan Sanders

unread,
May 9, 2013, 12:36:19 PM5/9/13
to
In article
<0ab4e6e6-404a-4e54...@gb2g2000vbb.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> On May 9, 12:05�am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <2c1472c8-0a5e-42af-93b3-9b86a91b1...@h9g2000pbr.googlegroups.com>,
> > �DKleinecke <dkleine...@gmail.com> wrote:
> > > What would you require me to demonstrate to allow me to call two
> > > theories notational variants?
> >
> > Demonstrate how a unary branching node (as allowed in X-bar theory)
>
> What a bizarre thing to complain about.

It's one of many key distinctions between X-bar theory and bare phrase
structure.

> A "unary branching node" is
> neither a branch nor a node, it's just a change of label.

Huh? A unary branching node is a node that immediately dominates
exactly one node.

> > can be represented in bare phrase structure, which happens to only
> > allow binary branching.
> >
> > (The inability of either system to allow ternary branching is a
> > primary reason I dislike both.)
>
> Yes, you've already said bizarre things about "In Defense of the
> Number Two."

Which things? That I disagreed with it? I don't believe that
phonological features need to be binary any more than I believe that
syntactic structure needs to be binary.

Peter T. Daniels

unread,
May 9, 2013, 4:26:49 PM5/9/13
to
On May 9, 12:36 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <0ab4e6e6-404a-4e54-be00-2d5e97a26...@gb2g2000vbb.googlegroups.com>,
>  "Peter T. Daniels" <gramma...@verizon.net> wrote:
> > On May 9, 12:05 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > In article
> > > <2c1472c8-0a5e-42af-93b3-9b86a91b1...@h9g2000pbr.googlegroups.com>,
> > >  DKleinecke <dkleine...@gmail.com> wrote:
> > > > What would you require me to demonstrate to allow me to call two
> > > > theories notational variants?
>
> > > Demonstrate how a unary branching node (as allowed in X-bar theory)
>
> > What a bizarre thing to complain about.
>
> It's one of many key distinctions between X-bar theory and bare phrase
> structure.
>
> > A "unary branching node" is
> > neither a branch nor a node, it's just a change of label.
>
> Huh?  A unary branching node is a node that immediately dominates
> exactly one node.

What does "node" or "branch" mean? And you have the gall to complain
when _David_ reuses terminology??

> > > can be represented in bare phrase structure, which happens to only
> > > allow binary branching.
>
> > > (The inability of either system to allow ternary branching is a
> > > primary reason I dislike both.)
>
> > Yes, you've already said bizarre things about "In Defense of the
> > Number Two."
>
> Which things?  That I disagreed with it?  I don't believe that
> phonological features need to be binary any more than I believe that
> syntactic structure needs to be binary.

Any structure (other than plain coordination, of course) can be
binarily analyzed. (I don't know where "needs to" comes into it.)

The Hebrew Masoretes knew this more than 1000 years ago -- see
Aronoff's 1985 *Language* article. The entire text of the Hebrew Bible
incorporates an unexceptioned binary syntactic analysis.

Nathan Sanders

unread,
May 9, 2013, 5:13:34 PM5/9/13
to
In article
<ca370850-6ea0-4649...@v3g2000vbr.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> On May 9, 12:36�pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <0ab4e6e6-404a-4e54-be00-2d5e97a26...@gb2g2000vbb.googlegroups.com>,
> > �"Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > On May 9, 12:05�am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > In article
> > > > <2c1472c8-0a5e-42af-93b3-9b86a91b1...@h9g2000pbr.googlegroups.com>,
> > > > �DKleinecke <dkleine...@gmail.com> wrote:
> > > > > What would you require me to demonstrate to allow me to call two
> > > > > theories notational variants?
> >
> > > > Demonstrate how a unary branching node (as allowed in X-bar theory)
> >
> > > What a bizarre thing to complain about.
> >
> > It's one of many key distinctions between X-bar theory and bare phrase
> > structure.
> >
> > > A "unary branching node" is
> > > neither a branch nor a node, it's just a change of label.
> >
> > Huh? �A unary branching node is a node that immediately dominates
> > exactly one node.
>
> What does "node" or "branch" mean?

I'm using the terms with their standard meaning in linguistics, which
is usually the same as their meaning in graph theory.

A node is a point in a graph.

A branch (or edge) is a connection between two nodes in a graph.

A directed graph (or digraph) is a graph in which the edges have
direction (i.e., directed edges are ordered pairs). For a given
directed graph <a,b>, a is the mother node and b is the daughter node.

A linguistic tree is usually taken to be a directed acyclic graph,
with various other restrictions to get the necessary properties for
the specific kind of tree in question (no line crossing, the single
mother condition, root nodes, etc.).

The outdegree (or branching factor) of a node in a directed graph is
the number directed edges in which it is the first member (i.e., the
number of daughter nodes it has).

In X-bar theory, nodes may have outdegree of 0, 1, or 2, because X-bar
theory demands that every XP also have separate X' and X nodes, even
if the full phrase only contains one word.

In bare phrase structure, nodes may have outdegree of either 0 or 2,
but crucially, never 1. There is no longer a requirement on having
extra abstract nodes for one-word phrases.

So, a simple mass noun like "water" would have at least three nodes in
X-bar theory, two of which have unary branching, which cannot exist in
bare phrase structure, where "water" would be just a single node.

> > > > can be represented in bare phrase structure, which happens to only
> > > > allow binary branching.
> >
> > > > (The inability of either system to allow ternary branching is a
> > > > primary reason I dislike both.)
> >
> > > Yes, you've already said bizarre things about "In Defense of the
> > > Number Two."
> >
> > Which things? �That I disagreed with it? �I don't believe that
> > phonological features need to be binary any more than I believe that
> > syntactic structure needs to be binary.
>
> Any structure (other than plain coordination, of course)

That's a pretty big exception, and it's one of the primary reasons I
disagree with any theory which rejects ternary branching (i.e.,
rejects nodes with outdegree of 3), as both X-bar theory and bare
phrase structure do.

> can be binarily analyzed.

Which yields a different internal structure, and thus, has different
implications for various syntactic properties and makes different
predictions about learnability, parsing, etc. A ternary analysis and
an embedded binary analysis are thus not identical.

DKleinecke

unread,
May 9, 2013, 8:03:01 PM5/9/13
to
On May 9, 2:13 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <ca370850-6ea0-4649-aa0e-1752be630...@v3g2000vbr.googlegroups.com>,
There is no difficulty with the non-branching nodes of x-bar theory in
bare phrase structure - just merge x with itself and label the result
x. Sure, you can edict this out of the theory - but that wouldn't be
very minimalist. The minimalist program is not so well defined that
it requires such an edict.

Move is defined by different writers in different ways but the general
statement seems to be that move is less economical than merge because
it involves two steps - copy and merge. It is, of course, the copy
part of move that corresponds to a transformation. The machinery
transformation provides the merge automatically.

Chomsky seems to be saying that the minimalist program is needed
because in order for universal grammar to be innate in humans it must
be based on simple mental mechanisms. But, it seems to me, he has
taken simplest to mean simplest among a set of grammars bounded by a
number of preconceived and mostly implicit constraints. But I don't
think I need to argue that point because almost everybody except
Chomsky thinks we don't know enough about how the brain works to know
what is "simple" for it to do.

Nathan Sanders

unread,
May 9, 2013, 8:21:14 PM5/9/13
to
In article
<0620c395-f8ab-476d...@wg15g2000pbb.googlegroups.com>,
> There is no difficulty with the non-branching nodes of x-bar theory in
> bare phrase structure - just merge x with itself and label the result
> x.

It's certainly been attempted (though not with labels: bare phrase
structure doesn't have labels; that's what "bare" means). See
Guimaraes 2000 for example.

> Move is defined by different writers in different ways but the general
> statement seems to be that move is less economical than merge because
> it involves two steps - copy and merge. It is, of course, the copy
> part of move that corresponds to a transformation. The machinery
> transformation provides the merge automatically.
>
> Chomsky seems to be saying that the minimalist program is needed
> because in order for universal grammar to be innate in humans it must
> be based on simple mental mechanisms. But, it seems to me, he has
> taken simplest to mean simplest among a set of grammars bounded by a
> number of preconceived and mostly implicit constraints. But I don't
> think I need to argue that point because almost everybody except
> Chomsky thinks we don't know enough about how the brain works to know
> what is "simple" for it to do.

On this point, we agree. There is very little about language (if any
of it) that I think needs to be innately specific to language. I
think it's quite possible that most, if not everything, about language
can be derived from other cognitive processes, but I hold a functional
view of language that runs in direct opposition to Chomsky.

Peter T. Daniels

unread,
May 9, 2013, 11:11:41 PM5/9/13
to
On May 9, 5:13 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <ca370850-6ea0-4649-aa0e-1752be630...@v3g2000vbr.googlegroups.com>,
Would those all be the definitions Chomsky uses?

> In X-bar theory, nodes may have outdegree of 0, 1, or 2, because X-bar
> theory demands that every XP also have separate X' and X nodes, even
> if the full phrase only contains one word.
>
> In bare phrase structure, nodes may have outdegree of either 0 or 2,
> but crucially, never 1.  There is no longer a requirement on having
> extra abstract nodes for one-word phrases.
>
> So, a simple mass noun like "water" would have at least three nodes in
> X-bar theory, two of which have unary branching, which cannot exist in
> bare phrase structure, where "water" would be just a single node.
>
> > > > > can be represented in bare phrase structure, which happens to only
> > > > > allow binary branching.
>
> > > > > (The inability of either system to allow ternary branching is a
> > > > > primary reason I dislike both.)
>
> > > > Yes, you've already said bizarre things about "In Defense of the
> > > > Number Two."
>
> > > Which things? That I disagreed with it? I don't believe that
> > > phonological features need to be binary any more than I believe that
> > > syntactic structure needs to be binary.
>
> > Any structure (other than plain coordination, of course)
>
> That's a pretty big exception, and it's one of the primary reasons I
> disagree with any theory which rejects ternary branching (i.e.,
> rejects nodes with outdegree of 3), as both X-bar theory and bare
> phrase structure do.

Wasn't it convenient of you to pretend that the Hebrew Bible contains
no lists of more than two items.

Nathan Sanders

unread,
May 9, 2013, 11:15:37 PM5/9/13
to
In article
<8da50506-a280-4e7e...@h1g2000vbx.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

They are definitions that graph theorists use.

Peter T. Daniels

unread,
May 9, 2013, 11:18:44 PM5/9/13
to
On May 9, 11:15 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <8da50506-a280-4e7e-8b76-e5686379d...@h1g2000vbx.googlegroups.com>,
That's a different claim from "I'm using the terms with their standard
meaning in linguistics."

Nathan Sanders

unread,
May 9, 2013, 11:38:34 PM5/9/13
to
In article
<45f89f0b-e71c-4359...@a6g2000vbf.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

Wasn't it convenient of you to skip the clarification "which is
usually the same as their meaning in graph theory"?

I think the only significant point of divergence between the two is in
what a "tree" is (it's directed in linguistics, but not in graph
theory).

Peter T. Daniels

unread,
May 10, 2013, 7:28:51 AM5/10/13
to
On May 9, 11:38 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <45f89f0b-e71c-4359-a612-adf772911...@a6g2000vbf.googlegroups.com>,
In the English language, a node has branches, and branches have each
other.

Brian M. Scott

unread,
May 10, 2013, 10:55:09 AM5/10/13
to
On Thu, 09 May 2013 23:38:34 -0400, Nathan Sanders
<san...@alum.mit.edu> wrote in
<news:sanders-CEAE44...@news.eternal-september.org>
in sci.lang:

[...]

> Wasn't it convenient of you to skip the clarification
> "which is usually the same as their meaning in graph
> theory"?

> I think the only significant point of divergence between
> the two is in what a "tree" is (it's directed in
> linguistics, but not in graph theory).

And this is merely because linguistics does not in general
deal with anything but directed, rooted trees, generally
planar as well, so it might as well save a few words. There
are also branches of mathematics in which trees are by
default rooted and directed.

Brian

Nathan Sanders

unread,
May 10, 2013, 11:22:19 AM5/10/13
to
In article
<db6f0d3d-a9f9-4e3a...@q8g2000vbl.googlegroups.com>,
Well then, I'm happy to have taught you some new technical jargon from
linguistics.

You're welcome!

Nathan Sanders

unread,
May 10, 2013, 11:29:31 AM5/10/13
to
In article <bo1mhibmf68i$.zymce39zem0m$.d...@40tude.net>,
"Brian M. Scott" <b.s...@csuohio.edu> wrote:

> On Thu, 09 May 2013 23:38:34 -0400, Nathan Sanders
> <san...@alum.mit.edu> wrote in
> <news:sanders-CEAE44...@news.eternal-september.org>
> in sci.lang:
>
> [...]
>
> > Wasn't it convenient of you to skip the clarification
> > "which is usually the same as their meaning in graph
> > theory"?
>
> > I think the only significant point of divergence between
> > the two is in what a "tree" is (it's directed in
> > linguistics, but not in graph theory).
>
> And this is merely because linguistics does not in general
> deal with anything but directed, rooted trees, generally
> planar as well, so it might as well save a few words.

Right, good point.

> There
> are also branches of mathematics in which trees are by
> default rooted and directed.

Ah, good to know!

Having transitioned from math to linguistics, I was initially bothered
by the linguistics use of "tree", but now I've been using it for so
long now that the mathematics use befuddles me!

Peter T. Daniels

unread,
May 10, 2013, 2:13:37 PM5/10/13
to
On May 10, 11:22 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <db6f0d3d-a9f9-4e3a-93d4-3e450abc7...@q8g2000vbl.googlegroups.com>,
I'm not thanking you for useless misuse of ordinary language for
"concepts" I have no use for.

Show me that there are "trees" with "nodes" and "branches" in human
brains, among the neurons and synapses (as Chomsky would like there to
be), and I might find the terms useful.

Nathan Sanders

unread,
May 10, 2013, 2:19:52 PM5/10/13
to
In article
<6ecde477-2beb-4f95...@a3g2000vbr.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

[snip]

I said you're welcome.

Peter T. Daniels

unread,
May 10, 2013, 2:24:02 PM5/10/13
to
On May 10, 2:19 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> In article
> <6ecde477-2beb-4f95-8657-42bef2771...@a3g2000vbr.googlegroups.com>,
The only thing worse than an asshole is a snarky asshole.

(Especially one who denies the foundations of his own "belief" (which
is how he characterized his theoretical orientation).)

You still have failed to acknowledge that there are lists of more than
two items in the Hebrew Bible.

Nathan Sanders

unread,
May 10, 2013, 2:37:22 PM5/10/13
to
In article
<6beab1a1-ba38-45cd...@b2g2000yqe.googlegroups.com>,
"Peter T. Daniels" <gram...@verizon.net> wrote:

> On May 10, 2:19�pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > In article
> > <6ecde477-2beb-4f95-8657-42bef2771...@a3g2000vbr.googlegroups.com>,
> > �"Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > On May 10, 11:22 am, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > In article
> > > > <db6f0d3d-a9f9-4e3a-93d4-3e450abc7...@q8g2000vbl.googlegroups.com>,
> > > > "Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > > > On May 9, 11:38 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > > > In article
> > > > > > <45f89f0b-e71c-4359-a612-adf772911...@a6g2000vbf.googlegroups.com>,
> > > > > > "Peter T. Daniels" <gramma...@verizon.net> wrote:
> > > > > > > On May 9, 11:15 pm, Nathan Sanders <sand...@alum.mit.edu> wrote:
> > > > > > > > In article
> > > > > > > > <8da50506-a280-4e7e-8b76-e5686379d...@h1g2000vbx.googlegroups.co
Then stop being one.

Marvin J. Mooney

unread,
May 29, 2013, 10:00:00 PM5/29/13
to
Peter T. Daniels wrote:

> What a bizarre thing to complain about. A "urinary branching node" is
> neither a branch nor a node, it's just a change of label.
>
>> can be represented in bare phrase structure, which happens to only
>> allow binary branching.
>>
>> (The inability of either system to allow ternary branching is a
>> primary reason I dislike both.)
>
> Yes, you've already said bizarre things about "In Defense of the
> Number Two."

You can't even make up your mind whether you're talking about urinary
functions or #2s! I don't even want to know how you got a seabird
involved.
0 new messages