1.3 Transitive Sentences--Main Verbs
The last type of sentence discussed here is called the Transitive
Sentence--Main Verb. Consider the token sentences below:
(34) Table: Transitive Sentences--Main Verbs
Subject |
Main Verb |
Object - Predicate |
a) John |
hit |
Bill |
b) Mary |
Kissed |
John |
c) The students |
took |
the class |
d) The man |
bought |
a cake |
What is interesting about these sentences above is that they appear
very much like the copular/linking verb examples in the sense that
they need to have an overt complement--namely, they need to have
in its complement slot some kind of object. Moreover, such sentences
contrast with what was said regarding Intransitive Sentence Types
provided in (§1.1) above. Recall that in Intransitive types,
predicate objects weren't required (and could optionally project)--e.g.,
Fish swim ø. With this latter Transitive sentence
type, objects must project--e.g., John hit Bill (Bill=object).
There is no sense in which we could say the utterance John hit
ø without specifying what or whom John hit--in this
example, the Whom/object = Bill). The reason for the "Main
Verb" heading of such transitive types is to capture the fact
that here the transitivity feature is the inherent property
of the Main Verb (as opposed to the copular/linking verb "Be")
which requires an accompanying object-predicate. As stated above
in the example of the verb [hit], one could say that there's
an embedded feature in the semantics of this Action verb itself
which requires (i) one who hits (the Agent/Subject of action), and
(ii) one who gets hit (the Recipient/Object of action). Such semantic
features can be expressed as a Predicate Logic proposition (shown
below): (J)ohn hit (B)ill:
(35) Predicate logic: hit(J, B) {arguments: John = subject, Bill
= object}
In other words, one could roughly say that 50% of the verb's material
reflects the action of the subject, while the remaining 50% must
reflect and indicate the result of the hit. As stated above,
there is no notion in which transitive verbs reflect back onto the
subject all of its 100%'s worth of verbal meaning: viz., *John hit/kissed/bought/made/
ø. The ill-formed expression *"John hit"
( hit(J) ) is thus accounted for by the stipulation that its predicate
logic requires both arguments (Subject and Object): viz., the [+Intransitive]
feature requires both [+Subject/+Object] settings. Consider the
token Transitive sentences along with the Structures below:
Note here that the two DPs have no co-indexing as compared to the
co-indexing of the two DPs found in the structure (31 & 32)
above. The independent functioning of the two DPs could be captured
by the following individual indexing: [DPi + MVP + DPj] (where the
subscripts "i" and "j" show no referential relation.
(See (70) for co-indexing with regards to reflexive pronouns I>myself
, etc.)
Three-Place Predicates
There is one additional structure we need to note regarding Transitive
Types. There appears to be a certain class of verbs in English that
requires not only one object, but rather two DP objects in its predicate
place (= ditransitive predicates)--counting the required
subject as one argument, this then makes the structure a "three-place
predicate". The rule for such structures could be spelled-out
as follows:
(37) Rule: Three-place predicate => [DP+MVP+DP+(PP)/DP].
Consider the following token sentences below which have 'optional'
three-place predicates, as compared to the sentences with the verb
e.g., "put" that in fact 'requires' the third
object:
There are a number of verbs in this class which seem to overlap
and/or extend their semantic range of meaning--in other words, typical
transitive verbs share common semantic cores. For instance, the
verb 'roll' requires by definition of the very action (i)
an Agent argument (=actor/subject, John) and (ii)
a Theme argument (=undergoing action/object, ball).
It seems that this same characterization of semantic/argument roles
overlap in similar verbs--e.g., kick, hit, touch, punch, throw,
caught, deliver, and [three-place predicate verbs place,
put, etc.]--noting the special consideration we have placed
on the last two di-transitive verbs, place and put.
In order 'to kick', there must be an Agent 'kicker' and a theme--a
person or thing being kicked. There is no sense in the notion of
an Intransitive verb 'kick' e.g., *John kicked,
just as there would be no notion of a mono-transitive verb 'put'
e.g., *John put the book. There seems to be something in
the semantics (=meaning) of the verbs which requires one to ask
for (i) an additional DP for support--e.g., kicked + DP-what/who?
(for mono-transitive verbs) or (ii) an additional couplet of DPs---e.g.,
put + DP-what + (PP)/DP-where? (for di-transitive verbs) (respectively).
1.4 Summary
In sum, we can define a simple English sentence by putting down
the following stipulations on the Subject/Predicate--all simple
(declarative) sentences require at the very least (i) a subject
position (=DP), and (ii) a Main Verb position (=MVP). This simple
SV (Subject Verb) construct defines an Intransitive Type sentence
by the following rule:
(39)
1. Intransitive Type: [DP+MVP] a. Fish swim. b. A telephone
is ringing
This basic SV Intransitive sentence expresses the very bare minimum
of what is required to maintained a full sentence. In addition to
this bare minimum SV sequence, optional adverbial material can be
projected in the predicate--e.g., Fish swim...in the pond.
In this example, the optional material is expressed by the Preposition
Phrase (PP) (in the pond).
Second order types then stipulate that Object(s) (or second, third
arguments) are required. This simple SVO construct defines a Transitive
Type sentence by the following rule:
2. Transitive Type-linking [DP+MVP+AdvP]: a. Mary is outside
b. John is tired
3. Transitive Type-linking [DP+MVP+AdjP]: a. The music is loud
b. I feel bad
4. Transitive Type-link [DP+MVP+DP]: a. She is a teacher b.
Pat is a postman
5. Transitive Type-non-link [DP+MVP+DP]: a. John hit the ball
b. I kissed Mary
6. Di-Transitive Type [DP+MVP+DP+(PP)/DP]: a. They put the book
on the table
2. The Phrase
In this section, we turn our attention to the Phrase level of language.
The 'Elephant Sentence'
One of the reasons we are able to recognize (=parse) the novel 'elephant
sentence' in (21a)--and conversely, why we can't for (23)--is that
certain words need to 'link-up' together to form larger strings
called phrases. For example, as soon as one says the (indefinite)
article 'A', our mind is set-up to perceive the following
word as a Noun (or Adjective+Noun frame). In fact, using here as
an example the (definite) article 'The', any word that
follows 'The' will instantly be construed as a Noun--even,
as we saw with a made-up novel sentence, made-up novel words can
become highly productive nouns: e.g., The 'pringle' was put
on the table (pringle => some kind of noun/object that could
be handled, say like a book), etc. Now, 'pringle' here is a novel
made-up word, yet in the configuration of the string, it projects
a word-class distinction (Noun). In fact, verbs are often turned
into nouns by this very process. Consider the Verb => Noun formulation
as found with Gerunds:
(40)
Verb |
=> |
Noun |
a. To walk |
The walking |
b. To cook |
The cooking |
This same novel word could likewise be construed as a Verb--e.g.,
John was pringling all day. Mary can pringle in the morning.
etc. One thing that can be said here, and will be further developed
in latter sections, is that (i) the role of the Determiner/article
(A/The) is to introduce a Noun, while (ii) the role of an Auxiliary/modal
is to introduce a Verb. This productive and highly creative aspect
of language stems from our innate necessity to think of language
not in terms of individually memorized and isolated words, but rather
to think of words as strings, chunks or phrases which build-up specific
constituencies.
There are a couple of main points we can now address regarding
phrases here--they are outlined below.
(41) A Seven-Step Guide to Phrases
1. Determiners (DP) (Articles, a/the;
Demonstratives, this/that/these/those; Genitives my/our/your/their,
etc.) precede Nouns: e.g., The book, A car, This pen, etc. This
type of phrase is referred to as a Determiner Phrase since the determiner
(the first word of the phrase) heads and projects the phrase. (NB.
We use DP throughout in the place of the otherwise prosaic Noun
Phrase (NP) since theory internal considerations demand that all
Nouns must have at least an abstract functional Determiner).
2. Adjectives (AdjP) (modifiers of Nouns e.g.,
red, good, fast, etc.) precede and generally describe nouns [(Det)+Adj+N]--(e.g.,
(The) read shoes, (A) good boy, (My) fast car, etc.). Attributive
Adjectives which modify following Nouns are best understood as being
embedded within larger DPs since such adjectives must accompany
Nouns. On the other hand, there are Predicative Adjectives which
need not accompany a noun (e.g., The man is tall/big/fat).
These Adjectives are not embedded in larger DPs and are thus diagrammed
as AdjPs. Their modifications tend to be antecedent to the subject
of the linking verb.
3. Main Verbs (MVP) (Tensed Verbs such as goes/went,
walks/walked, keeps/kept, etc.) typically follow the subject of
declarative sentences (adhering to the English SVO--Subject-Verb-Object
word order). Tensed (=Finite) verbs serve to introduce the overall
predicate of a sentence. In fact, part of the requirements for a
complete sentence is to have a Tensed Main Verb housed in the predicate
as they generate the predicate information of the subject.
4. Auxiliary/Modals (AuxP) serve to introduce
Main Verbs (MVPs). All functional features associated with Verbs
{Tense, and Agreement features of Person/Number} are borne out of
the Aux. The Aux also houses any inflection that might be derived
onto a verb stem such as past tense inflections {-ed}, participles
{-en}, {-ing} as well as housing the usual class of auxiliary elements
themselves (e.g., Do/Be/Have and Modals can/will/shall, etc.).
5. Verb Phrase (VP) (Infinitive Phrase) unlike
the MVP is a Non-Tensed Verb Phrase. Such VPs tend to project after
an already positioned MVP. These phrases include all three Infinitive
types/forms--e.g., I like--to cook (=Infinitive
'to'), I like-cooking (Infinitive 'ing'), I can cook
(Infinitive 'bare verb stem' ).
6. Adverb Phrase (AdvP), like adjectives for nouns,
modifies verbs--e.g, softly touched, quickly ran, etc., (Adv+V).
7. SVO/Head Initial Phrase: In addition to English
being an SVO word order, English stipulates that the Head of a Phrase
must be in the first initial position within the phrase (i.e., that
word which labels the phrase--such as Determiner, Adjective, Main
Verb, etc.--must come first in forming the phrase, and not last
(moving left to right). The word being introduced serves as the
Complement of the Head (or Comp).
Having laid out some of these general principles, let's consider
what the template of an English Phrase looks like, and then turn
our attention to the real matter at hand--to see just how we were
able to process (parse) our now infamous "Elephant Sentence":
(42)
Consider the series of phrases which build-up the structure of the
"Elephant Sentence" (restated below):
(43) "Yesterday, I saw a pink and yellow elephant roller-skating
down flower lane"
(43.a) < Yesterday>
Although yesterday is the first word of the sentence, and
since we have said that English is an SVO word order language, at
first blush our intuitions might have us say that 'Yesterday'
is therefore the subject. But as it turns out, it indeed is not.
Two things must be pointed out here. Firstly, proto-type subjects
must be Nouns--of the traditional 'Person-Place-Thing' sort. Yesterday
simply functions here as a time reference (adverbial element) in
giving additional (Time) information regarding the predicate. Secondly,
Yesterday doesn't originate in the first position in any
event--it has, in fact, moved from out of the very last position
of the sentence and has positioned itself sentence-initially (sometimes
referred to as fronting--see Movement
in §4 below). If you read the sentence above, yesterday
naturally fits in the finally position (the comma here shows the
actual movement).
(43.b) < I >
I is in fact the subject of the sentence (the person)
and takes the form of a Pronoun. So, having returned the adverb
(yesterday) to its original place (final position), the
Subject or Topic naturally surfaces as the first word of the sentence--preserving
our SVO order. The phrase "I" would be spelled out as
a DP since I is a ProNoun and all nouns must be introduced by a
structure-class functional determiner--yielding a DP:
DP |
(the zero allomorph ø is used here) |
/
\ |
(see DP below for full DP features) |
D |
N |
|
|
ø |
I |
|
|
(43.c) < saw>
Saw functions as the Main Verb (MVP) due to the fact that it
wields [+ Tense]: the two morphological inflection markers of English
Tense being (i) the present 3Person Singular {-s}, and (ii) the
regular past tense marker {-ed}. English has only two grammatical
Tenses: Present and Past. An optional future mode is not available
as a grammatical tense marker and so doesn't form from out of a
main verb--being taken up instead by the modals will/may/could
etc: e.g., John starts next Monday, The President speaks tomorrow
tonight, (= Present tense marker {s} is opted though with future
interpretation) vs. John will start next Monday, The President
may speak tomorrow night (= Modal is opted). (See (158) regarding
Tense & Modals). Given that saw is the main verb, all other
subsequent material is said to form the predicate (verb included)--devising
the subject + predicate template. The Main Verb Phrase is created
below:
VP |
( => [+lexical/-functional] verb phrase) |
| |
|
MVP |
(=>[-lexical/+functional] main verb phrase |
/ |
\ |
|
|
Aux |
MV |
|
[+Tense] |
saw |
(see MVP below for full Aux features) |
(43.d) <a pink and (a) yellow elephant>
are twin DP expressions each with an embedded Adjective Phrase (for
color). The two DPs are then joined by a simple conjunctive element
<and> as diagrammed in isolation below:
(43.e) <roller-skating>
We have now come up against a second-order verb type in this sentence.
Roller-skating is certainly a verb here taking on the role
of some action; however, there seems to be no clear Tense indicated
on the verb. Well, rightly so--for there should be at all times
only one grammatical Tensed verb per clause/simple-sentence and
the first-order [+Tense] main verb <see> has already
been marked for Tense. For now, this second-order verb type will
be referred to here as an "ing- infinitive" (present participle)
verb (whereas participles don't maintain tense). It should not be
understood as a gerund since gerunds are typically referred to as
(i) Verbs with an "ing" inflection that (ii) occupy a
noun slot (e.g., [DP The shopping] was fun, etc.). Here,
the "ing" verb remains a verb albeit without tense--hence,
the label Infinitive Verb. (See 3-Types of Infinitive Verbs in (86)
below). This verb exemplifies the distinction spelled out above
(in 43c) between [+Lexical/-Functional] VPs--where there are no
functional features of which to speak--and [-Lexical/+Functional]
MVPs--where there are functional features as carried out by the
Aux. Clearly, second-order Infinitive verbs wield none of the typical
functional features associated with a Main Verb [Person/Number/Tense]
and could be simply diagrammed as a prosaic VP:
VP (+Lexical/-Functional) |
| |
V |
roller-skating |
One further implication here is that the VP (roller-skating) could
actually be substituted--again, calling on our favored linguistic
substitution test--as a Progressive aspect of an
Elliptical Clause:
(a) I saw the elephant roller-skating...
=> (b) I saw the elephant: <it <was roller-skating>>...
(43.f). <down> Heads a straightforward Prepositional
Phrase:
One very important aspect of the Preposition is that it must always
introduce a DP (in prescribed grammar). This notion is what is behind
the often 'prescribed rule' of "preposition stranding"--mandating
never to leave a preposition standing at the end of a sentence (see
§2.4 for Prep). Though, I must say that there are plenty of
occasions when one can only save a sentence by otherwise standing
a preposition: e.g.,
(44)
(a) This is the man for whom my son
works.
(b) This is the man (who) my
son works for
(c) This is the man (-) my
son works for
(d) *This is the man for who my
son works
(e) *This is the man for (-) my
son works
(43.g) <flower lane>
brings us to the close of our beloved "elephant sentence".
The Phrase can be drawn as a Proper Noun DP with all the typical
trappings of a DP:
DP |
/ |
\ |
D |
N |
| |
/
\ |
ø |
flower lane |
As an exercise, you should now be able to link all the isolated
phrases above into one completely diagrammed sentence--starting
with the DP <I> and ending with the DP <flower
lane> ( recall that <yesterday> serves as
an adverbial adjunct which can either be joined at the beginning
or at the end of the sentence.
In the ensuing sections, we will be turning our attention to more
detailed analyses of specific functional features having to do with
DPs (§2.1), MVPs (§2.3) and to some extent PPs (§2.4).
2.1 Determiner Phrase (DP)
(45) Table: Determiners
Summary of Determiners |
Articles:
Demonstratives:
Possessives:
Indefinites:
Cardinal Numbers:
Ordinal Numbers: |
a/an, the
this, that, these, those
my, your, his, her, its, our, their
some, any, no, every, other, another, many, more, most, enough,
few, less, much, either, neither, several, all, both, each,
one, two, three, four,...
first, second, third,...last |
Definition :
Features:
Phrase Structure: |
A Determiner is a functional structure-class word that
precedes and modifies a Noun.
Definiteness, Case, Person, Number, (Gender)
D + N => DP |
There are a number of facts we need to spell out about the DP.
Firstly, it must be said that a lively debate still persists among
linguists regarding whether or not a DP even exists. For a number
of purely theoretical reasons, mostly being driven by theory internal
factors, we take it that DPs do exist: the alternative view--and
a view mostly in line with traditional grammar pedagogy--is that
they are simply Noun Phrases (NPs) with a determiner modifier at
the Head of the Phrase. However, as has been unfolding throughout
this text, our story is crafted around the idea of Lexical vs. Functional
Grammar and how such Lexical items (Nouns, Verbs) might merge with
Functional formal features. In order to capture this story to its
fullest extent possible, we must classify (all) NPs as DPs. There
are a number of reasons for doing this. Let's flesh them out below.
First, there is a natural, and I think quite elegant notion behind
lexical vs. functional grammar. However, in order to conceptualize
the clear distinctions, they must manifest at the Phrase-level.
It's not enough to simply say the some words are born with more
formal meanings than others. The question must be how does such
formal grammar get onto the lexical items themselves. Such an oversimplified
lexicon analysis would not go far enough in accounting for the syntactic
phenomena. In other words, although lexical item distinctions are
part of what is behind the labeling of lexical vs. functional grammar,
more is needed in order to capture where and how such distinctions
take place. Our first line of reasoning is that for every Lexical
item, there is some Functional counterpart. The relationships are
by no means perfect and there are interceding gaps, but for all
intents and purposes, a partnership of "Functional-to-Lexical"
does emerge (in that order). The most basic partnerships--as presented
in the first sections--are between (i) (D)eterminer and (N)oun and
between (ii) (Aux)iliary and (V)erb. Whenever there is a Verb, there
must be an Aux in order to deliver the appropriate functional/formal
grammar onto that Verb. Likewise, for the Noun: whenever there is
a Noun, there must be a Determiner to deliver the appropriate functional/formal
features onto that Noun. Hence the most basic four square partnerships
are: D => N (D introduces N) and Aux => V (Aux introduces
V). Prepositions too have some sort of partnership. In this section,
let's spell out in more detail the exact functional features for
D => N (DP).
(46)
Functional Features: |
The standard class of Functional Features having
to do with D => N are the following: |
(a) Definiteness [+/-Def] |
( [+Def] The book vs. [-Def] A book) |
(b) Case (i) [+/-Nom] |
(Nominative I/she/they/ vs. Accusative me/her/them) |
(ii) [+Gen] |
(Genitive my/her/their) |
(c) Person [1,2,3P] |
(I =1P, you=2P, she/he/it=3P) |
(d) Number [+/-Plural] |
(I/she [-Pl], we/they [+Pl], book [-Pl], books [+Pl] ) |
(e) Gender (+/-Fem) |
(He vs. She) |
(NB. Interestingly, the Gender feature for DPs in English only
seems to reflect 3P Pronouns and a handful of lexical nouns--e.g.,
actor vs. actress, etc. In Romance languages (e.g., Spanish), this
Gender feature is more prevalent where agreement between the D and
N must take place).
What is important to realize here is that every Noun must select
and host all relevant functional features in order to ensure a proper
projection of the phrase. (When a phrase doesn't properly project,
it is said that the features Crash--but we'll return to that a little
later on). To make matters more concrete, let's draw-up some differing
examples of DPs along the lines of the features spelled out above
and see just where and how the relevant features project.
[Def]-Feature
Clearly, the two phrases below hold an important grammatical difference
that must be stated both in the semantics (meaning) as well as in
the syntax:
(47)
(a) John has finally written [DP [D the] [N book] ]. =>[+Def]
(b) John has finally written [DP [D a ] [N book ] ]. => [-Def]
The above examples pin down two very different meanings of the
"book": the first sentence reveals that John has finally
written "the" book that we all expected him to write.
It is a very specific book--hence, one of its features must mark
for [+Def]. The nature of this "book" is a very different
concept than the generic book in (b)--e.g., we all know John could
write a book (any book) and so he has finally done it--he has succeeded
in writing a book. The general nature of "a book" highlights
the task of writing the book only, it doesn't specify the actual
book itself. This conceptual difference is captured in functional
grammar--viz., the D in sentence (a) is marked for [+Def] while
the D in sentence (b) marks for [-Def].
Another interesting Definiteness phenomenon found an the syntactic
level is the following pair of sentences:
(48)
Def affecting number on
Verb |
D-feature |
Aux-feature |
(a) A number of students *is/are... |
(D: [-Def] ) => |
(Aux: [+Pl] ) |
(b) The number of students is/*are... |
(D: [+Def] ) => |
(Aux: [-Pl] ) |
(NB. Adverbials can have a [+Def] pragmatic effect. Consider the
following distributions marking [+/-Def] on the DP-object--e.g.,
John has never read a book ('a book' = [-Def]) (*and
he never will read it) (no expletive "it" insert)
vs. John read a book yesterday (and he enjoyed "it")
('a book' = [+Def] and can be replaced by a [-Pl] expletive "it").
What we find here is that the definiteness feature on the D (marking
for specificity) holds an agreement relationship with its Aux (MVP)
so that when a D is [+Def], the verb must be singular (or minus
plural [-Pl] ) and vice versa. One number, "The number",
say 2 or 200 is specific and thus constitutes a singular matching
verb, whereas "A number" more or less constitutes a non-specific
group--hence, plural verb: (e.g., The number of students enrolling
in Grammar is dropping/has dropped from 200 to 180.) Also note that
students, (the complement of the genitive particle of)
is plural. Many people are too quick to assume that this plural
noun acts as a defunct subject, and so plural agreement of the verb
are must ensue--clearly, this is wrong. In this case, it
is the Determiner (not even the Noun) that determiners whether or
not the agreeing verb is singular or plural. In any event, the nouns
students keeps its plural marker {s} in both examples,
and still the distinctions on number hold. There would be no way
to capture this interesting correlation without somehow addressing
the notion that a particular feature embedded in the D has something
to do with the number agreement on the verb. By saying that both
sentences in (48) above are instantiated by NPs and not DPs, we
forgo any meaningful discussion on the nature of functional features.
(See (56) below). Mainly speaking, what Feature Theory allows us
to do is break down the components of "parts-of-speech"
words to a finer grained analyses--this lets us tinker with certain
sub-particles of the word in order to see how one isolated feature
might project and contribute to a phrase over another. Consider
here how the [Def] feature is incorporated into the two DPs below:
(49)
DP |
/ \ |
D |
N |
Feature: [-Def] |
| |
| |
| |
A |
number...are |
DP |
/ \ |
D |
N |
[-Def] |
| |
| |
| |
The |
number...is |
Case-Feature
Let's turn now to the next D-feature--Case. Along with Person and
Number features (see sections below), Case builds up a very intricate
paradigm:
(50) Table: Case--Personal Pronoun
Nominative (Subject)
Case |
Accusative (Object)
Case |
First Person:
Second Person:
Third Person: |
Singular:
I
you
he/she/it |
Plural:
we
you
they |
First Person:
Second Person:
Third Person: |
Singular:
me
you
him/her/it |
Plural:
us
you
them |
Perhaps the most important aspect of Case to understand is that
it is Structural--meaning, that Case is recognized
dependent upon where the pronoun sits in the sentence: viz., if
subject, then Nominative [Nom], if object then Accusative [Acc].
In English, Case is confined to the Personal Pronoun. In Latin,
for instance, Case was crucial in determiner whether or not a Pronoun
was a subject or an object--this was owing to the fact that Latin
was somewhat of a free word order language where words could have
a relatively mixed arrangement. In order to distinguish if a Noun
(Pronoun) was a subject or not, one had to look to the Case of the
word (marked as an Inflection). English too has remnants of this
type of Case Inflection: (e.g., the inflection {-m} has a similar
Latin-based history in that it marked Accusative (Object) case as
in e.g., he vs. hi-m, they vs. the-m). (Note in Latin the noun "love"
"amor"--Nominative case amor-ø,
Accusative case amor-em, Genitive case amor-is). Case no
longer indicates word order for English--English has secured for
some time now an SVO (Subject-Verb-Object) order (parting from an
earlier Germanic mix of SOV & SVO), so that functional case
is no longer a crucial grammatical marker of word order. (NB. We
have seen in our own lifetime the approaching abandonment in standard
English of the use of who vs. whom--where the former marks
for Subject Case and the latter for Object Case: noting the {-m}
inflection once again). In addition, part of this overzealous Latin
bias for Case is still with us today. For instance, consider Pronoun
case confusion below:
(51)
Formal: |
Informal: |
(a) It is I |
(c) It's me |
(b) This is she |
(d) This is her |
The pressure from Latin-based schooling has succeeded in making
us at least more self-conscious toward the Latin style of Case;
however, given that this pressure must be externally reinforced
(by prescribed grammarians having their roots in the 19th century),
from time to time our own English intuitions win out. Note the following
problems with (51) above:
Latin is a Pro-drop (pronoun drop) language (meaning that the subject,
like in Spanish/Italian) can go missing. English is not a Pro-drop
language. However French, which too is a Latin based language, is
also not a Pro-drop language and must resort to the English equivalent
"It is me" (C'est moi). In addition, the Latin
based paradigm seems to falter in the following examples: There
is you,...There is I,... but one never seems to say *?There
is we...
The problems of choosing the correct Case can at times be affected
by something as small as a conjunctive and. For instance, the sentence
My dog and I went for a walk may have alternative case
values for some of us whereas the Nom Case I becomes Acc Case me--e.g.,
*My dog and me... Of course, while the latter is 'grammatically
incorrect', the instability, nevertheless, demonstrates just how
abstract Case can be. Recall, the substitution test can always help
you find your way in these matters: by substituting 'My dog
and I' we get the Pronoun We (=Nom Case): We went for a
walk (my dog and I). Another important syntactic fact regarding
case has to do with Preposition (see PP §2.4). It seems that
Prepositions specify for Acc case for its complement. In other words,
when a Pronoun-DP follows a Prep, it must always be marked for Acc
(object) Case. This also may give us some trouble. Take the Prep
between, if a pronoun follows, it must have Acc case: e.g.,
Between you and me,... vs. *Between you and I,...
The latter version is a very popular error of Case marking. Take
other Prepositions for example, one would never say e.g., I
want to sit near/with/next-to *she/he/they... Clearly, these
pronouns occupy the object position of the sentence and so must
reflect Acc case. A very simplistic picture now emerges with the
Case paradigm captured in the following exchange:
(52)
S |
V |
O |
(word order) |
a) |
John |
kissed |
Mary |
(proper names not case marked) |
b) |
He |
kissed |
her |
(Nom + V + Acc) |
c) |
She |
was kissed |
by him |
(passive voice) |
Let's now turn to see how we can incorporate Case as a Feature
of D within a DP. By starting with the simple Pronoun exchange I
& me in a DP, we can begin to employ the two D-features
examined above (Def and Case).
(53)
|
Phrase: |
|
|
|
DP |
(= DP-subject) |
|
/ |
\ |
|
|
D |
N |
|
ø |
| |
features: |
[+Def] |
| |
|
[+Nom] |
| |
|
|
|
I.... |
....kissed Mary |
Token Sentence: |
I kissed Mary |
(54)
|
Phrase: |
|
|
|
DP |
(= DP-subject) |
|
/ |
\ |
|
|
D |
N |
|
ø |
| |
features: |
[+Def] |
| |
|
[-Nom] |
| |
|
|
Mary kissed |
...me |
|
Token Sentence: |
Mary kissed me |
Binary notation:
Note that we are marking Accusative (object) Case in a binary manner
as [-Nom]. As stated earlier, Feature Theory makes good use of Binary
Notation via +/- (plus or minus value settings). E.g., [-Nom] is
the same as stating [+Acc]. It is common practice in binary code,
whenever possible, to share one common denominator and simply place
a [+/-] setting to its value. This is done throughout the theory--hence,
[-Pl] marks for singular, [-Nom] marks for Accusative, [-Def] marks
for Indefiniteness, etc.
Zero allomoph
This is as good a place as any to digress and examine the role of
the zero allomorph {ø} found in our DPs above. First of all,
the question should come up--why do we need it if it is simply zero
or has a zero value? Good point! We could just as well return to
our more prosaic version of an NP as eluded to above and omit all
this superfluous abstract material. Right you say, let's keep it
as simple as possible. Well, while your heart may be in the right
place, the theory mandates that we have Functional projections alongside
lexical ones--remember? If it weren't for our little D, there would
be no place to house all that abstract functional grammar in the
first place. It's crucial to remember that lexical items (Nouns,
Verbs, etc.) can't house functional features--while lexical items
may take them on as inflections to their stems, these inflections
are born from out of the (functional) D/Aux and are then
delivered onto the (lexical) N/V (respectively). I think
our 'born & deliver service' makes for a nice movement analogy
here: formal features (usually in the guise of inflection) are born
from the functional projection <x> (of a proto-type functional
XP) and are then delivered onto their lexical counterpart <y>
(using variables <x> to mark the Head and <y> to mark
its Complement). (See §2.3 for a similar discussion of Inflectional
Movement regarding verb features such as Tense and Participle Inflection).
For example, a)-prime below shows the inflectional process of plural
number [+Pl] onto the Noun stem [N+{s}].
The zero allomorph more than anything else serves as a kind of
(theory internal) place holder in keeping the D projection active--much
like zero place holders serve us in mathematics e.g., 1, 10, 100...
Although there is no phonological reality to it, the zero place
holder allows us to maintain our D-projection. This idea of a grammatical
placeholder for otherwise non-existent phonological material is
similar to what we find in the Pro-drop languages cited above--Spanish
being one of them. Consider the Pro-drop effect of the following
Spanish sentence ø habl-o ingles ( = (I) speak English).
Certainly, notwithstanding the fact that there is no audible subject
(at the phonological level), we would still want to have a DP-subject
here in order to trigger the {-o} verb inflection for the features
[1Person, -Pl, +Def, +Nom] of the pronoun I. The zero allomorph
is a sort of safety valve: its role is to secure this grammatical
level of representation in the event of a lack of any overt phonological
material. Returning to the NP analogy, one would be hard pressed
to account for the whereabouts of functional features given that
an NP can't manifest or house such formal material. Consider what
goes wrong with an NP-analysis below:
(56)
NP -analysis |
NP |
| |
N |
| |
I......kissed Mary |
DP-analysis |
DP |
/
\ |
D |
N |
ø |
| |
{features} |
I... |
Clearly, in theory, there would be no place to house the functional
features in the NP-analysis. There is a bit more to it than I am
making out here, but for our purposes, and in keeping with the spirit
of our theory, the problem is indeed real for the NP-analysis.
Genitive Case
(57) Table: Genitive Case (DP)
'Singular' |
Person:
1. my
2. your
3. his, her, its |
'Plural' |
Person:
1. our
2. your
3. their |
whose, John's,
(see §2.2.2 for DP structure) |
The Genitive (or Possessive Case) is another Case marker that functions
via a DP. Consider the tree diagram and sentences below incorporating
the features presented thus far:
(58) (a) My/the book is on the desk. =>
My/the book.....= subject [+Nom]
(b) Did your read my/the
book? =>....my/the book. =
object [-Nom]
(59)
Notice that the Subject/Object [+/-Nom] features are preserved
alongside the additional feature of possessiveness [+Gen]. Also,
note that in both sentences, the [+Def] features is spelled out
since all Genitive Determiners are specific and mark for definiteness.
One notational confusion however does emerge when we try to consider
the Person Feature (see below). How should we notate the Genitive
feature alongside the Person feature--for instance, should the [+Gen]
Determiner My in "My book" be marked as First
Person [1P]? One might say "My" certainly reflects first
person (me). In a manner of speaking, you are right. However, again
if we utilize our beloved substitution test and throw in a pronoun
(for good measure), one would see right away that "My book"
gets reduced to a third person pronoun "it" e.g., (Did
you read it? where "it" refers to "my book"
). So, how should we mark Person on Genitive DPs? In this sense,
one could claim the [Gen] feature serves as a special overlap unlike
any other Case marker. Namely, whereas it is theoretically assumed
that no pronoun can carry two positive case features (viz., pronouns
in their formal sense cannot be both [+Gen] and [+Nom] ), once a
Gen Determiner is reduced to a pronoun (via our substitution), its
[+/-Nom] feature reappears. Hence, some dual marking seems to be
warranted here and I propose that Gen DPs mark both for [+Gen] and
[+/-Nom] dependent upon where the Gen DP sits in the sentence. So,
after all is said and done, perhaps the best way to settle the Person/Genitive
issue is by compromise: (i) if only the Determiner (and not the
DP) is marked for person, then [1Per] could be maintained since
"My" refers to "I" (First Person). Noting that
a sole D can't be reduced to a pronoun in a substitution test: ("my"
doesn't reduce to "it" ). However, we are not considering
the D in isolation here, and once we analyze the DP as a complete
phrase, the pronoun substitution mandates that it be Third Person
[3P] (reducible to "it" ). And since we are analyzing
the complete DP and not just the D, there seems to be some support
in marking it [3P], but clearly, as you can see for yourself, much
of argument here is largely centered on a notation quibbles (trivial
sticking points that more often than not succeed in forming a wedge
between some of our finest linguists).
Since the Person Feature has caused such a commotion here, let's
address it next.
Person
(61) Table: Person
Number:
Case: |
Singular:
Nom-Acc : |
Plural:
Nom-Acc : |
1st [1P]
2nd [2P]
3rd [3p] |
I-me
you-you
he/she/it-him/her/it |
we-us
you-you
they-them |
In traditional grammar, English is said to have three grammatical
persons: first, second and third. The first person
[1P] expression (I-we) denotes the person(s) speaking.
The second person [2P] (you) denotes the person(s) being
spoken to (addressed) by the first person. The third person [3P]
(he/she/it/-they) denotes the person(s) outside the immediate
circle of speech activity and who becomes a referent excluded from
the first and second person--the [3P] refers to someone other than
the speaker(s) or addressee(s) We take it that this defining property
between speaker and addressee is (i) real in the pragmatics of language
discourse, and (ii) constitutes a formal abstraction in the syntax
worthy of having the status of a formal functional feature.
Number
Number [+/-Pl] (plus or minus Plural) is an additional functional
feature which denotes the contrast between "grammatical"
singular and plural forms. Note that we use "grammatical"
here as a way of showing that such seemingly inherently real notions
as say number may not maintain true values as say, in real numbers
of math--for instance, the noun "family" denotes an inherently
plural notion (in the sense that "family" means more than
one person making up a nuclear social unit), however, it is grammatically
marked as being singular in number (cf., My family is/*are).
(Likewise, for instance, "Hair" in French ("Cheveux")
is grammatically marked as plural whereas in English it is grammatically
singular.) In Adjectival constructions utilizing Number, often Plural
inflection gets omitted--e.g., He is six-foot-five (*six-feet-five),
two-car garage (*two-cars-garage), a three-storey house (* a three-storeys
house), etc. Surely, in the above examples, our notion of 'plural
number' holds notwithstanding the fact that no grammatical plural
inflection surfaces on the noun. In real terms then, there is no
real logical notion of number other than some abstract grammatical
property that maintains itself as a formal functional feature.
(62) The Grammatical Rule for Number
(a) N+ {ø} => singular e.g., The car, The book, An
exam...
(b) N + {s} => plural e.g., The cars, The books, These exams...
Note that we are demonstrating the regular rule here for number.
Irregular grammatical number manifests in a variety of ways.
(63) Some Irregular Number inflections:
(a) vowel change (tooth>teeth, goose>geese)
(b){en}-suffix (child>children, ox>oxen),
(c) No change zero allomorph {ø} (fish>fish-ø,
sheep>sheep-ø)
Also, some nouns have an inherent singular feature and thus can't
be marked for number. These are called Mass Nouns--e.g., salt,
milk, butter, sand, furniture: (*The furniture-s).
We have now exhausted the main class of formal features associated
with the DP--let's now list them all together below as well as spell
them out in real DP circumstances:
Summary of DP features and their projections
(64) Table: Main DP-features
1. Definiteness [+/-Def]
2. Case [+/-Nom] [+Gen]
3. Person [1/2/3P]
4. Number [+/-Pl] |
Definiteness & Case contrast
Above, there are two important feature distinctions at play: (i)
the determiners the vs. any bear contrast between
definiteness and Indefiniteness, and (ii) the [3P,+Pl] Noun books
bears either subject nominative case [+Nom] or object accusative
case [-Nom].
Let's exam another pair of DPs wielding contrasts between person
and number features:
Above, the feature contrasts are between Person and Number (as
well as Case). The two Pronoun DPs (We and You) are both
manifestations of these features in that "we" marks first
person [1P] and Plural (number) [+Pl] while "you" marks
second person [2P] and singular [-Pl] (though, in the 2nd person
paradigm it is impossible to tell whether or not "you"
is singular or plural since both lexical items are homomorphous
e.g., I told you [-Pl] vs. I told you (guys) [+Pl]
). Examples (b) and (b- prime) show Third Person, Masculine Gender
[3P, -Fem] "him".
Other DP types
As an exercise, consider some other types of DPs:
(69)
(d) Recursive DPs
(See (97b) for Recursive VPs)
DP |
|
/
|
\ |
|
D |
DP |
|
| |
/
\ |
| |
D
DP |
| |
| |
/ |
\ |
| |
| |
D |
N |
(i) My |
first |
two |
years were hard....but, |
(ii) the |
last |
three |
years were wonderful |
(70) Reflective Pronouns
Lastly, in rounding off our DP study, let's see how Reflexive Pronouns
get incorporated into a DP-analysis. Perhaps the most crucial aspect
of Reflexives to understand is that they involve an anaphoric
structure--that is, there is a structural relationship between the
subject and the object of a reflexive DP. This relationship can
be notated via a co-indexing which binds
the reflexive Object-DP back to its antecedent
Subject-DP. Consider the sentences below:
(71)
(a) Shei hurt herselfi.
(b) The childreni did it for themselvesi
(c) Our Presidenti should feel proud
of himselfi
(d) Ii mailed the letter to myselfi
In all four examples, the reflexive object cannot be used to refer
directly to an entity in the outside world, but rather must be bound
by an antecedent subject within the same phrase or sentence. In
other words, "herself" is bound to "She" in
ex. (a), "themselves to "The children" in ex. (b),
"himself" to "Our President" in ex.(c), and
"myself" to "I" in ex. (d). This very close
structural and grammatical relationship is denoted by the co-indexing
{i} subscript found below both the Subject and Object Pronouns
(and is notated in tree diagrams just below the DP).
Consider the anaphoric DP diagrams below (showing only the relevant
isolated DP binding and co-indexing with no other feature spell-outs):
(72) Reflexive- DP co-indexing and binding
DPi |
|
DPi |
/
\ |
|
/
\ |
D |
N |
D |
N |
| |
| |
MVP |
PP |
| |
| |
(a) ø |
She |
hurt |
|
ø |
herself |
(b) The |
children |
did it |
for |
ø |
themselves |
(c) Our |
president |
should feel proud
|
of |
ø |
himself |
(d) ø |
I |
mailed the letter |
to |
ø |
myself |
Closing this section on Lexical Nouns and Functional DPs, let's
now turn our attention to the next main lexical category of the
sentence--the Verb Phrase. But before we tackle its lexical-functional
relation via the Main Verb Phrase (MVP), let's simply exam in more
detail the inner working of the Lexical Verb Phrase in isolation
saving its more abstract functional MVP counterpart for the subsequent
section (§2.3).
|