(134) Passive Voice Rule => [Be + Verb + Past Participle + by]

The Passive voice is also carried out by Aux "Be". One of the properties of the Passive Voice is that it allows the Object of an otherwise Active sentence to become the Subject of a Passive sentence. This is particularly handy when one wants to emphasis or put focus onto the object for discourse reasons. Consider the contrast between the active vs. passive voice in the paradigm below noting that the Subject of an Active Sentence becomes the Object of a Passive (and vice versa).:

(135) Table: Active vs. Passive Voice

In all the above examples, the Passive Rule [Be+Verb+Past Part.+by] applies. Passive (OVS) constructions can then be said to originate from out of Declarative (SVO) sentences. In other words, a passive construction is formed via the following procedure: (i) a Declarative (SVO) sentence + (ii) Passive rule => passive sentence. The passive rule entails movement of a sort here in that the Subjects and Objects reverse their positions. This movement could be viewed as existing at a Sentence Level since the whole sentence has been up-rooted (as opposed to just a word or phrase). (For a summary of movement, see §4 below.)

Perhaps the most complicated part of the Passive Rule is the Past Participle (Past Part.) component plus "by". For starters, the "by" is not a prepositional "by" since it cannot be substituted by any other preposition: e.g., *Mary was kissed under John, The ball was kick between Jim and Fred, etc.. Of course, these prepositions are indeed available to be used as prepositions in derived passive sentences, the matter here is that they in no way serve as a replacement for the "by" component of the passive rule--e.g., Mary was kissed under John by Jim, The ball was kicked between Jim and Fred by John, etc. The "by" component of the passive rule should be thought of as actually that, part of the rule--it is a lexical item projecting an Adverbial Phrase (AdvP) that is invariably fixed and doesn't in any way come to represent a class of words, say 'preposition'. One other note here that needs to be mentioned is that the "by" component can get and often gets deleted: the derived passive Mary was kissed is fine enough a sentence without any mention of the culprit (= the one doing the kissing). In fact, passives quite often go without any overt object of which to speak (i.e., their intrinsic subject gets omitted). Such elliptical sentence structure places the heavy burden of understanding on pragmatics and/or context. For instance, The boy was arrested implies that The police arrested him => (by the police), President Clinton was impeached implies that Congress impeached him => (by congress), etc.

The Past Participle morphology surfaces as an Inflection onto the Main Verb of the sentence. Recall, that this Past Part. inflection in no way denotes grammatical tense since, by rule, only the first Main Verb of a sentence takes on the role of projecting Tense via Inflection: 3P, Sing/Present {s} and Regular Past {ed}. Rather, the past participle should be viewed as a Particle whose Inflection beings about some change in Aspect--a time referential of Duration with a non-grammatical [-Tense] status (much in the same manner as how the Infinitive/Particle "to" / "ing" maintains a [-Tense] status). The table below shows Past Participle Inflections:

(Note that for clarity, the Regular Past Participle morphology is labeled herein as the {en} suffix, so that there lies no confusion between (i) regular past tense {ed} and (ii) past participle {ed}--the former being regular while the latter is irregular).

Consider the tree diagramming of passive constructions below showing Passive Rule functional Features/Inflection:

(141) Have
The Auxiliary "Have" is responsible for forming the Perfect or Past Participle constructions. This type of grammatical construction often gives one the impression of having a dual tense since it is possible to denote a quasi present-past or past-past reference to grammatical time. However, what is important to realize here, as a grammatical rule, is that only the First Main Verb of a sentences gets the task of representing grammatical Tense/Time--or, as the saying goes-- "The first verb gets the Time". So, even though it may appear that a second verb (positioned in the second verb slot) has an easily identifiable tense inflection e.g., such as a past tense inflection {-ed}, this second verb's inflection doesn't represent the grammatical tense for the sentence, but rather only marks this second order grammatical function of Past Participle/Perfect Aspect. The Perfect aspect is very similar in nature to the Progressive (imperfect) aspect in the sense that the role of Aspect is not to mark tense (per se), but merely to denote the "time" and "manner" of duration of the activity described by the verb: viz., the Progressive aspect denotes continuity of action, a sort of present participle (showing that some action is not yet completed), while Perfect denotes a kind of completion of action.

Consider the Perfect rule below:

(142) Perfect Rule => [Have + Verb + Past Participle {en}]

The standard form of the Past Participle Rule is: [Verb => add {-en}] (as in has writt-en). However, the rule, just like its Past Tense rule counterpart (Verb => add {-ed}], has its irregular forms as well. Consider the regular and irregular forms below with the rule [Verb => add {-ed}] defined as the "Irregular rule" if only to separate it from the "Regular" Past Tense rule [Verb => add {ed}]. In any case, the inflection {-en} should be properly thought of as the regular form of the Perfect construction: [Verb + {en}].

Included among the pattern of steady rules, there is some overlap of irregular forms: e.g., got/gotten, dived/dove, knelt/kneeled, etc. and in particular, whenever an irregular verb is used in a novel setting, the regular rule process kicks in and acts as the default: e.g., He has drived (vs. driven) the ball over the fence (U.S. baseball), The man was hanged (vs. hung) etc. The {-ed} form is overextended on such verbs. (NB. This same process is observed with the N+{s} default rule as well--e.g., I need two computer mouses / *mice). In this sense, the {-ed} forms for both regular (past) and irregular (perfect) seem to be the default inflection. In addition to the {-en} inflection of regular, the Perfect also entertains a host of Irregular inflections including (i) no change (or zero allomorph) put>put>put, and (ii) vowel change sing<sang>sung.

Tense
Of course, by simply changing a Present (=Pres(ent) 3 (Person) Sing(ular)) inflection to a Past Inflection [+Past] (a feature selection as controlled by the Tense node under Aux) we would get had run/used/written (respectively). Note that the Participle inflections don't change with the changing of tense from present to past--"only the first verb gets the time" so only the first Auxiliary/verb changes tense. The examples below show how a Main Verb (in second position-- V2) and an Auxiliary Verb (in first position--V1) may show up as non-contrasting homophones (cf., 146, 148) though with distinct grammatical roles.

Recall that these examples typify the dual status of "Do-Be-Have" as belonging both to a Main Verb class and an Auxiliary class--depending on its role and structure within the given sentence.

Using a reduced tree notation, you may think of the hierarchical syntactic structure for the three Aux verbs as follows (Reduced Tree Representation for Declaratives):

Note that for all configurations, the Auxiliary Inversion rule (as shown in (a, ii) as well as the Negation rule applies.

(151) Combinations of Auxiliary Constructions "Be" & "Have"

Keeping to our now reduced syntactic trees (as drawn above), we see how the latter two Aux Verbs (Be & Have), which form complex grammars (as opposed to the simple grammar of Aux "Do"), can merge to form Auxiliary Combinations:

A Rule of thumb on Combination orders:
(i) Perfect always before Progressive/Passive
(ii) Progressive always before Passive

(a) Perfect Aux "Have" with Progressive Aux "Be"

i. [Have + V + Past Part]   => [has + be + {en}]
ii. [Be + V + ing]               => [be + study +{ing} ]

Diagram:	She has be -en study -ing

(NB. Note that while the verb "Be" serves as a Lexical Main Verb for the first diagram (a, i)--since it occupies the verb-slot of the structure--it, at the same time, also serves as true functional Aux. Verb for the second diagram (a, ii)--since it occupies the Auxiliary slot. Recall, that since Verbs "Do-Be-Have" hold a dual status as functioning either in the capacity of a Main Verb or Aux Verb (depending on the structure or slot the verb occupies), "recycling" might be a nice way to think about their 'switching' of roles here. Also see (153) below for differences between Aux and Main verbs--recalling that "be"/"have" can function as a linking/main verb (respectively) in the sentences (She is a teacher vs. she has been very good & She has my book vs. she has had a bad day, etc.). This same style of recycling applies across the board for all merged combinations of Auxiliary construction. Also note that the rules of Aux. Inversion and negation continue to apply.)

Diagram:	She [ was    be  -ing    kiss -ed   by ] John

(c) Perfect Aux "Have" with Passive Aux "Be"

Diagram:	She [ has     be  -en   take -en   by ] John

As an exercise, see if you can diagram the following three tier Aux construction (you should notice that the structure involves the recycling of two Aux/Main Verbs "be" ):

e.g., !? She [ has been being beaten by] her husband for several years.

Why not toss in a modal for good measure?:

e.g., !? She [could have been being beaten by] her husband for several years

We must address one last note before we leave the Auxiliary constructions behind. One interesting way to show that the Aux class reflects a functional highly abstract class par excellence is to see what happens to it in colloquial English. The abstract nature of Aux shows up in colloquial English in ways that suggest it may form a general proto-class of its own. For instance, consider how the Negative form "ain't" can overlap as a general abstract verb to cover both "Be" and "Have" counterparts. I believe this demonstrates more than anything the non-concrete (non-substantive) nature of the Aux. verb.

(152) Usage if "Ain't"

In other words, "Ain't" in the above cases seems to serve as a sort of prototype formal/functional category that makes use of an overlapping category--say, [+Aux] with two spell-out forms: Be and Have. The rule might look something like the following: "ain't" = [Be or Have + (n't)].

(153) A Recap: Differences between Auxiliary and Main Verbs "DO-BE-HAVE"

As a nice recap, perhaps the easiest way to understand the inherent differences between the homophones (same-sounding, but grammatically different) Auxiliary/Main Verbs is to consider the logic behind the three forms of propositions below:

(157) Modals

On the heels of the Auxiliary Verb, we have a class of verb-like items called modals (or modality verbs). While these verb are also functional (and hence somewhat abstract) like their Auxilairy verb counterparts, they however cannot take on Verbal INFLection such as Tense and/or Agreement. For example, consider the ungrammaticality of the sentences below:

The class of modals tends to denote abstract states such as--e.g., the giving of advice (should), possibility (might/may), potential (can), non-grammatical future time reference (will) etc. We shouldn't think of modal "will" being our Future tense in English since, as a grammatical rule, only the verbs (not Modals) take on Tense and Inflection (see (102) above). Besides, "will" seems to be used for a number of possible modalities aside from our commonly conceived future reference--e.g. (cited from Palmer 1984: p. 198):

Recollecting our badly conceived notion that modal "will" provided English with a future grammatical Tense, consider the counter examples below which similarly provide future reference with or without the modal "will".

The overall syntactic functions of the Aux/Modal is that they introduce Verbs. Recall, in our earlier discussions, that Auxiliaries are viewed as playing a functional/grammatical role in that they introduce Lexical Verbs, [Aux =>V] and that Determiners are said to function in a similar way in that they introduce Lexical Nouns [Det => N]. So, here we have gone full circle in expressing the roles of the two functional items. One side note is in order here. Since Modals seem to have their own word classification, they are entitled to link-up with their Auxiliary counterparts to form two types of modality expressions:

In example (i), the modal might expresses possibility within a progressive grammar. In (ii), should expresses advice within a (Negative) Perfect Passive structure. And (iii) expresses a future reference within a Interrogative (Question) Perfect structure--denoting that idea that the action of the verb "studying" will be completed (marking a ten year span, perfect grammar) at some future date. Also note that the same rules apply to Modals as they do to Aux Verbs regarding Inversion (for Questions) and Negative "not" (for Negation).

2.4 Prepositional Phrase (PP)

(161) Table: Prepositions

Earlier on in this text, we had tentatively established the word-class of Preposition as having a Lexical Categorical status. One argument in favor of a lexical status could be based on the fact that since (at times) Prep(ositions) formulate opposites, there must be a certain amount of inherent meaning involved--since only proper meaningful properties could ever hope to derive opposites--e.g., on-top vs. under, on vs. off, for vs. from, etc. However, as it turns out, this keen and somewhat intuitive observation regarding semantics is often what is at the heart of our misguided analyses of Prep as a lexical category. In fact, there are a number of good reasons for considering Prepositions as having a functional status (and not a lexical status). One reason has to do with this quasi inherent meaning. It is indeed true that Prepositions do communicate a certain amount of meaning, but at a closer look, one discovers that all derived 'meaning' is rather dependent on structure (an element pointing more to a functional status): clearly, there is no sense of meaning in the words with/in/on/between/etc. except that they establish a structural relationship with the preceding nominal (DP-subject) and the following Determiner Phrase (DP-object) as in J(ohn) walks with M(ary) [ walk(J & M) ] showing that the DP-subject John and the DP-object Mary are conjoined in the act of walking. As was presented earlier in our discussion of lexical/substantive words (lexical categories: Nouns, Verbs, and Adjectives), Prepositions carry very little in the way of substantive/conceptual meaning. Ask yourself the following question: what does--(i) Car [+N], or (ii) Red car [+Adj, +N], or (iii) Red cars go fast look like in your mental eye? Certainly, you can formulate some type of mental substantive description of the word categories that make up the sentence above. Now, ask yourself what a preposition would look like in your mind's eye--say, "with" in the sentence below: Girls "with" red cars drive fast. While all other words bear and contribute a fair amount of substantive meaning, the preposition "with" lacks any sort of meaning and is inserted into the structure in order to maintain an abstract structural/grammatical relation (expressing location or manner) to the substantive words. In a more finer-grained analysis of Feature Theory however, the level of functional abstraction becomes obvious. Recall that our earlier discussion of Functional categories brought us to the topic of lexical counterparts--recapping, we stated that Functional Determiners work alongside their Lexical Noun counterparts (merging into a DP) in providing formal abstract functional material for purposes of feature checking, and that the Functional Auxiliaries work alongside their Lexical Verb counterparts apparently for the same reasons. Well, it seems that Prepositions likewise enter into a functional partnership--viz., Preps introduce DPs [Prep => DP-object]. That is, whenever you find a Preposition, a following DP-Object shouldn't be far behind. Again, as stated earlier, this is what's behind the notion of preposition standing--you can never leave a Prep standing alone without it properly introducing a DP. Feature Theory nicely captures this PP to DP relation by stipulating that Prepositions are indeed functional in that they hold at least one functional feature that must be checked by its adjacent DP. The feature to be considered here involves Case. Recall, Case is a grammatical realization of a given Pronoun and/or Prenominal forming the following paradigm: Nominative/Subject "I" vs. Accusative/Object "me" vs. Genitive "my"(+ N) etc. (See (50) above for a recap of the full Case paradigm). What we are on about here is that Preps hold at least one functional feature specific to Case, and that this one feature forces the two Phrases (PP & DP) to merge together (for reasons having to do with formal feature checking). Since Preps hold this functional feature, we must now reanalyze the whole class of Prepositions as a Functional Category. Considered the PP projections below showing both proper and erroneous feature spell-outs:

As we see in (162c)-prime above, a feature crash occurs because the Prep's obliged [-Nom] feature crashes with an improper projected [+Nom] DP (a DP which instead should have a [-Nom] Accusative status). This one feature projection originating in the Preposition of a Prepositional Phrase accounts for the ungrammaticality of the sentences below:

(163) PP Feature Mismatch

(a) [ pp because] [ DP you and *I/ (= me)]     (d) [ pp with] [ DP you and *she/ (=her)]
(b) [ pp like] [ DP you and *I/ (= me)]            (e) [ pp alongside] [ DP *we (=us)]
(c) [ pp before] [ DP Mary and *he (= him)]   (f) [ pp between] [ DP you and *I/ (= me)

Preposition Phrases
Prepositional Phrases can function in one of two ways--(i) Adverbial, or (ii) Adjectivally. When Preps function Adverbially, they are called adverbial modifiers, and when they function Adjectivally, they are called adjectival modifiers. One must keep in mind, however, that the word class "Preposition" doesn't change under these circumstances: viz., there are no hybrid word classes such as Adj(ective)-Prep or Adv(erb)-Prep (respectively). Perhaps a better way to think about it is to say that the Prep changes are not systematic, but rather have to do with bringing about a certain flavor of modification--something Preps are not typically associated with. Recall that the classic functional definition of Prep more or less involves the expression of manner or location between the two associated DPs. The notion behind any form of modification is not an inherent feature of Preposition; nonetheless, Preps do seem to enter into such modification. Consider the examples of Adj/Adv-Prep modification below:

(164) Adjectival Prepositional Phrases: The modification of a DP/Noun

(a) The Professor often teaches classes full of Freshmen.		(Adjectival)
	=> Freshmen classes
	[DP [D ø] [Adj Freshmen] [N classes]]
		Adj-modification

(b) The President owns a boat with a red, white, and blue sail.		(Adjectival)
	=> A red/white and blue sailed boat
	[DP [D A] [Adj red-white-blue sailed] [N boat]]
		Adj-modification

(d) The car with an electric motor was too expensive.		(Adjectival)
	=> The electric motor car
	[DP [D The] [Adj electric motor] [N car]]
		Adj-modification

(165) Adverbial Prepositional Phrases: The modification of a VP/Verb

(a) John and Mary walked along the beach		(Adverbial: Place)
	=> walked along
	[VP [V walked] [Adv along]]
		Adv-modification

(b) The lecture began after lunch (and) without interruption
	=> began after (&) => began without	(Adv Time & Manner)
	[VP [V began] [Adv after] [Adv without]]
		Adv-modification

2.5 Summary of Features

In summary, having now Spelled-Out a certain amount of Functional Features in the pages above, what we can conclude are the following points:

(0) The most general observation that can be drawn here is that there exist two types of words:

(i) Lexical Words (=Noun, Verbs, Adjective, and Adverb)-these words are substantive in nature and thus contribute to a full range of meaning; and

(ii) Functional Words (=Determiners, Auxiliary/Modal, Preposition)--these words are not substantive, but rather abstract in meaning and thus contribute only abstract grammatical relations.

(1) There seems to arise a general grammatical framework that stipulates what kinds of words can sit amongst other words. So, for all intense and purposes, what has been presented in the pages above is more or less a theory which stipulates a general adjacency condition. This condition is said to apply to Functional words as they work alongside their Lexical word counterparts--a kind of Structure Class-to-Form Class syntactic co-operation. In fact, one of the ways we were able to determine functional vs. lexical word class distinctions was to see if an adjacency condition applied. This test later allowed us to reconsider the status of the word class "Preposition" and to adjust its status from lexical to functional--since its own adjacency condition called for it to precede a DP-Object.

(2) D=>N
One of the first class of Functional features looked at was the class of DP-features. This class included the following specific DP-features: Definiteness, Person, Number, and Case. These four main DP-features contributed to an abstract (formal) grammatical relation reflecting how substantive Nouns get interpreted in the grammar.

(3) Aux=>V
Another class of functional features we looked at involved how an Auxiliary verb worked alongside a Main Verb. The class included the following specific AUX-features: Tense, and Person & Number (=Agreement).