10 Passive

• Characteristics of the passive
• A movement analysis of the passive
  • Object idiom chunks
  • Analysis
• The passive and nonfinite complementation
• Structural versus inherent case
• The passive of double object sentences
• Notes
• Exercises and problems
• Supplementary material
  • Thematic roles

In this chapter, we first present some general characteristics of the passive and then a movement analysis of it. Although superficially the passive does not resemble subject raising, our analysis of it treats the two constructions as analogous in important respects. Moreover, subject raising turns out to be analogous not just to the passive of simple sentences, but also to the passive of sentences with ECM verbs. All three phenomena (simple passive, subject raising, passive of ECM) turn out to obey a correlation called Burzio's generalization, according to which verbs that lack an agent argument also lack the ability to check case. We show further that there are two types of case, structural and inherent, and that Burzio's generalization holds only for structural case. Finally, we discuss the passive of double object sentences in English on the one hand and in German and Latin on the other.

Characteristics of the passive

Passivization has a number of effects. First and foremost, the agent argument, which is expressed as the subject of the active sentence, appears in the passive as an optional by phrase. Second, the now vacant subject position is taken over by the theme argument. In other words, the agent and theme arguments are linked to different grammatical relations in the passive than in the active. If subject ranks higher than object on a hierarchical scale of grammatical relations (as is often assumed), then the passive demotes the agent argument and promotes the theme argument. Third, passive past participles, unlike their homonymous active counterparts, can't check objective case.

Inchoative manner-of-motion verbs are not passive verbs. (i)     The vase dropped. It is true that sentences like (i) share with passives the property that their subject is a theme, rather than an agent. However, they differ from passives in not allowing the expression of an agent, as shown in (ii).1 (ii)   * The vase dropped by the child. (ungrammatical on agentive reading of by-phrase)

In English, the passive is expressed analytically by a combination of the past participle and an auxiliary verb (be or get). Other languages allow the passive to be expressed synthetically, as illustrated in (3) for Korean and in (4) for Latin.

Chulswu-ka  kyehoyk-ul helak-ha-yessta.
Chulswu-nom plans-acc  approve-active-past
Agent       Theme
Subject     Direct object
'Chulswu approved the plans.'

Kyehoyk-i (chulswu-eyuyhaye) helak-toy-essta.
plans-nom  Chulswu-by        approve-passive-past
Theme      Agent
Subject    Object of postposition
'The plans were approved (by Chulswu).'

Puer-i       port-am       claud-unt.
boy-nom.pl.m door-acc.sg.f close-3.pl.pres
Agent        Theme
Subject      Direct object
'The boys are closing the door.'

Port-a        (a  puer-is)     claud-it-ur.
door-nom.sg.f  by boy-abl.pl.m close-3.sg.pres-passive
Theme             Agent
Subject           Object of preposition
'The door is being closed (by the boys).'

In these languages, it is the bound morphemes -toy- and -ur that result in the effects of passivization mentioned above. As in English, the grammatical relations of the agent and theme arguments differ in the active and the passive, and passive verb forms cannot check the case that active verb forms can.

Kyehoyk-ul (chulswu-eyuyhaye) helak-toy-essta.
plans-acc   Chulswu-by        approve-passive-past
Intended meaning: 'The plans were approved (by Chulswu).'

Port-am       (a  puer-is)      claud-it-ur.
door-acc.sg.f. by boy-abl.pl.m. close-3.sg.pres-passive
Intended meaning: 'The door is being closed by the boys.'

Latin (among other languages) also uses the analytic strategy for the passive in certain linguistic contexts. In particular, the passive voice in the present perfect, past perfect, and future perfect is formed by combining the passive participle with the appropriate form of esse 'be', as illustrated in (6).

Port-a        (a  puer-is)     claus-a                 { est, erat, erit } .
door-nom.sg.f  by boy-abl.pl.m close-part.pass.nom.sg.f  is   was   will.be
Theme             Agent
Subject           Object of preposition
'The door { has, had, will have } been closed (by the boys).'

It is worth noting that the synthetic passive form in (4b) agrees with the subject in number, but that the participle in the analytic form in (6) shows additional case and gender agreement. Case and gender are ordinarily features of the nominal system (nouns and their canonical modifiers, adjectives), so participles share features of both verbs and adjectives. Indeed, the term 'participle' implies as much, as it derives from Latin particeps 'sharing'.

A movement analysis of the passive

The restriction of object idiom chunks to the complement position of the licensing verb is thrown into striking relief by the contrast between nearly synonymous expressions such as attention and heed, or progress and headway (Radford 1988:423). The variants with the ordinary expressions (attention, progress) are fine, but those with the idiom chunks (heed, headway) are not since they are not licensed by the verbs in red.

Given their licensing requirements, it isn't surprising that object idiom chunks are generally ungrammatical in subject position.

They are, however, able to occur in subject position under one condition - in passive sentences where the passive participle is that of the licensing verb. This is illustrated by the contrast between (11) and (12) (Radford 1988:423).

As we have just seen, object idiom chunks in active sentences are locally licensed as well - namely, as the complements of a licensing verb, and this local relationship is extended in passive sentences. In other words, although the licensing relationship differs in both cases (spec-predicate for subject raising, head-comp for passive), the passive is analogous to subject raising in that both constructions superficially extend a local licensing relationship. This fundamental similarity motivates our treatment of the passive as another instance of syntactic movement.

Our movement analysis of the passive is based on the premise that theme arguments originate in the same structural position in both the active and the passive. This means that the elementary trees for active and passive participles both contain a complement position. However, the relevant elementary trees differ in two important ways. First, in the active, the agent argument is obligatorily linked to (= expressed in) Spec(VP), whereas in the passive, it is linked to an optional adjunct by phrase. We will represent this by omitting Spec(VP) in the elementary tree for the passive participle. Second, passive participles in English cannot check objective case; recall the ungrammaticality of (2b). (We will generally not represent this property in the elementary tree explicitly.) The elementary trees we propose for active and passive participles are thus as shown in (13).

(13)	a.				b.
			Active verb: [+obj]				Passive verb: [-obj]

Burzio 1986 was the first to note the correlation between the two above-mentioned properties and to propose the generalization in (14).

In what follows, we illustrate the derivation of a passive sentence like (15).

First, we substitute the theme argument the proposal in the elementary tree for the passive main verb in (13b). This yields (16a). We next substitute (16a) as the complement of the passive auxiliary verb be, as in (16b). In order to distinguish the auxiliary verb (be) from the main verb (adopted), we give it the syntactic category Aux, but we could just as easily treat it as another instance of V. We will treat auxiliary verbs, like raising verbs and passive participles, as lacking specifier positions. We then substitute the resulting structure as the complement of I, as in (16c).

(16)	a.				b.				c.
			Substitute theme argument in (14b)				Substitute (16a) as complement of passive auxiliary				Substitute (16b) as complement of I

Because of the inability of the passive participle to check objective case, the theme argument's case feature cannot be checked in the complement position. Since every case feature must be checked, the theme argument must move to the closest position in which case can be checked. This position is Spec(IP), where it is possible for the theme argument to check nominative case with finite I. The resulting final structure is shown in (17). (For simplicity, we omit the verb movement of the passive auxiliary be from V to I.)

(17)

According to the analysis just presented, the passive is analogous to subject raising in the following way. In both cases, a noun phrase (the subject of the complement clause with subject raising, the object with passive) originates in a position where case can't be checked. This forces the relevant noun phrase (complement subject, object) to move to the closest position where case can be checked. Of course, subject raising and the passive aren't identical in every respect. There are two important differences between the two instances of movement. First, the path from the head of the chain to its tail contains a clause boundary (IP) in the case of subject raising, but not in the case of the passive.

Recall that the term head has two completely different meanings that shouldn't be confused. The head of an X' structure is the syntactic category that immediately dominates a word or morpheme and projects an intermediate and a maximal projection. The head of a movement chain is the highest element in the chain.

(18)	a.				b.
			Subject raising chain				Passive chain

Second, the grammatical relation of the moved noun phrase doesn't change in subject raising; it starts out as a subject and ends up as one. In the passive, the grammatical relation of the moved noun phrase does change - namely, from object to subject.

The passive and nonfinite complementation

This section focuses on the passive of ECM verbs like expect, which were introduced in connection with head-spec licensing in Chapter 8. (19a) shows an active ECM verb sentence, and (19b) shows the corresponding passive.

The elementary tree for expect in (19a) is given in (20a). In accordance with the previous discussion, the elementary tree for the passive participle expected is as in (20b). The difference between the two trees is analogous to that between the trees in (13); the only difference is the syntactic category of the complement (DP in the case of ordinary verbs, IP in the case of ECM verbs). Like (14b), the elementary tree (20b) is missing a specifier position, and in accordance with Burzio's generalization, the verb lacks the ability to check objective case.

(20)	a.				b.
			Active verb: [+obj]				Passive verb: [-obj]

In what follows, we illustrate the derivation of (19b). The derivation of the complement clause is shown in (21); we assume that the complement subject moves from Spec(VP) to Spec(IP) to provide the complement clause with a subject (recall the subject requirement discussed in Chapter 3).

(21)	a.				b.				c.
			Structure of complement VP				Substitute (21a) as complement of nonfinite I				Move subject of complement clause

The subsequent steps of the derivation involving the matrix clause are as shown in (22).

(22)	a.				b.				c.
			Substitute (21c) as complement of passive participle of ECM verb				Substitute (22a) as complement of passive auxiliary				Substitute (22b) as complement of matrix I

In (22c), the subject of the complement clause cannot check case with the participle of the ECM verb in the head-spec configuration because the participle, being passive, lacks the ability to check case. Therefore, the complement subject must move to the nearest position where case can be checked. This is the matrix Spec(IP), where nominative case is checked. The resulting structure is shown in (23). (Once again, we omit V to I raising of the passive auxiliary for simplicity.)

(23)

Consider now the chain headed by you in (23). The chain consists of three links, which occupy the matrix Spec(IP), the complement Spec(IP), and the complement Spec(VP). Chains with links in these same three positions are also the result of subject raising, as we saw most recently in (18a). In this respect, then, the passive of ECM verbs is analogous to subject raising, as is evident from comparing the schematic structures in (24). The only difference is that the structure of ECM passives is slightly more complex because the matrix clause contains a passive auxiliary in addition to the main verb.

(24)	a.				b.
			ECM passive chain				Subject raising chain

The common source for the analogous chains in (18) and (24) is that ordinary passive verbs, subject raising verbs, and passive ECM verbs all lack a specifier position, as shown in (25), and therefore (by Burzio's Generalization) the ability to check case.

(25)

a.

b.

c.

Although Burzio's generalization itself remains to be explained, it does allow us to understand the ungrammaticality of all three sentences in (26) as stemming from a single source. In all three sentences, the head (highlighted in italics), fails to be associated with an agent and hence fails to project Spec(VP),³ yet case would have to be checked on the underlined noun phrases contrary to Burzio's generalization, whether in the head-comp configuration, as in (26a), or in the head-spec configuration, as in (26b,c).

Structural versus inherent case

This section will not be part of the exam. Resume reading at The passive of double object sentences.

In languages with morphologically richer case systems than English, there is evidence for a distinction between two types of case: structural and inherent. For instance, the following German examples show that the active participle unterstützt 'supported' checks accusative case, but that the homonymous passive participle cannot. Instead, in the passive, nominative case on the theme argument is checked by finite I, as in English.

The German examples in this section are given in the form of subordinate clauses in order to abstract away from a syntactic constraint on main clauses in the Germanic languages other than English that is irrelevant for present purposes; this so-called V2 constraint is discussed in Chapter 14.

dass wir dies-en  Kandidat-en   unterstützt haben
that we  this-acc candidate-acc supported   have
'that we have supported this candidate'

dass dies-er  Kandidat      unterstützt wurde
that this-nom candidate.nom supported   became
'that this candidate was supported'

dass dies-en  Kandidat-en   unterstützt wurde
that this-acc candidate-acc supported   became
Intended meaning: 'that this candidate was supported'

German also allows verbs to check the dative. Such verbs - for instance, helfen 'help' - continue to check the dative even in passive sentences, as shown in (29) and (30).⁴

dass wir dies-em  Kandidat-en   geholfen haben
that we  this-dat candidate-dat helped   have
'that we helped this candidate'

dass dies-er  Kandidat      geholfen wurde
that this-nom candidate.nom helped   became
Intended meaning: 'that this candidate was helped'

dass dies-em  Kandidat-en   geholfen wurde
that this-dat candidate-dat helped   became
'that this candidate was helped'

The accusative and the dative are both checked in the head-comp configuration by the verbs unterstützen 'support' and helfen 'help', respectively, but they differ in that the accusative alternates with the nominative, whereas the dative doesn't. What we mean by 'alternate' is simply that an accusative object in the active corresponds to a nominative subject in the passive, whereas a dative object in the active remains dative in the passive. The alternating cases (nominative, accusative) are known as structural cases (because which one occurs depends on the structural position of the argument bearing the case), whereas the nonalternating case (dative) is known as an inherent case.

The distinction between structural and inherent case has consequences for the formulation of Burzio's generalization. In view of the facts just presented, it holds only for verbs that check structural case.

An important question that arises in connection with (30b) is what element checks the nominative case feature of the finite I. It is generally assumed that German has a silent expletive element, corresponding to English expletive it or there, that can check nominative case in Spec(IP). The structure of (30b) is then as in (32). (As usual, we omit the V to I movement of the passive auxiliary werden 'become'.)

(32)

The silent expletive bears a number feature with the default value 'singular'. Evidence for this comes from the number agreement pattern in (33): the overt plural subject in (33a) agrees with a plural verb form, whereas the silent expletive in (33b) agrees with the corresponding singular verb form.

dass dies-e    Kandidat-en    unterstützt { wurd-en,  *wurd-e }
that these-nom candidates-nom supported     became-pl became-sg
'that these candidates were supported'

dass dies-en   Kandidat-en    geholfen { wurd-e,   *wurd-en }
that these-dat candidates-dat helped     became-sg  became-pl
'that these candidates were helped'

Like German, Latin is a verb-final language with rich morphological case, and it patterns like German with regard to structural versus inherent case. As in German, the structural cases are nominative and accusative. The inherent cases include the remaining cases (apart from the vocative) - that is, the dative, genitive, and ablative. (34)-(37) illustrate the parallel with German, using the dative as a representative for the inherent cases.

Avuncul-us   puer-os    adiuva-t.
uncle-nom.sg boy-acc.pl help-3.sg.pres
'The uncle is helping the boys.'

Puer-i     adiuva-nt-ur.
boy-nom.pl help-3.pl.pres-passive
'The boys are being helped.'

Puer-os    adiuva-t-ur.
boy-acc.pl help-3.sg.pres-passive
Intended meaning: 'The boys are being helped.'

Avuncul-us    puer-is    subveni-t.
uncle-nom.sg  boy-dat.pl help-3.sg.pres
'The uncle is helping the boys.

Puer-i      subveni-unt-ur.
boy-nom.pl  help-3.pl.pres passive
Intended meaning: 'The boys are being helped.'

Puer-is    subveni-t-ur.
boy-dat.pl help-3.sg.pres-passive
'The boys are being helped.'

Resume reading here.

The grammaticality contrast for English in (39) can be derived straightforwardly as follows. The active and passive sentences have the schematic structures in (40) and (41), respectively. For clarity, (41) shows the structure before and after verb movement.

(40)
(41)	a.				b.

In (41a), the higher VP shell lacks a specifier position for the agent, and therefore, by Burzio's generalization, the head is unable to check structural case. This property is preserved in the complex head that results from adjoining the lower verb to the higher one, as highlighted by the boxes. The recipient DP, which in the corresponding active sentence would check case with the complex verb in the head-spec configuration, must therefore move out of the VP shell structure. This yields (39a). In order to derive the ungrammatical (39b), the theme would have to move, rather than the recipient. But then the resulting chain would check case twice: the foot of the chain with the objective case feature on GET and the head of the chain with the nominative case feature on finite Infl. Meanwhile, the recipient DP would remain in its original location, unable to check case, exactly as before. As a result, the passive in (39b) is ungrammatical.

The rest of this section will not be part of the exam. Resume reading at Notes.

Not all languages show the pattern just discussed. German and Latin, for instance, have exactly the reverse pattern of that in (39); they allow the theme passive, but not the recipient passive. After illustrating the pattern for both languages, we propose an explanation that relies on the distinction just introduced between structural and inherent case.

In Chapter 8, we saw that German double object sentences surface with a dative-marked recipient and an accusative-marked theme. Given the distinction between structural and inherent case, the question arises of what happens in the passive of double object sentences in German. As shown in (42) and (43), the argument with structural case (the theme) can be promoted to subject, but the argument with inherent case (the recipient) cannot.

dass ich   dem     Jungen  den     Roman gebe
that I.nom the.dat boy.dat the.acc novel give
'that I am giving the boy the novel'

dass der     Junge den     Roman gegeben wird
that the.nom boy   the.acc novel given   becomes
Intended meaning: 'that the boy is being given the novel'

dass der     Roman  dem     Jungen gegeben wird
that the.nom novel  the.dat boy    given   becomes
'that the novel is being given to the boy'

Latin exhibits the same pattern, as shown in (44) and 45).

Avuncul-us    puer-is    libell-um          mitt-it.
uncle-nom.sg  boy-dat.pl little.book-acc.sg send-3.sg.pres
'The uncle is sending the boys a little book.'

Puer-i     libell-um   mitt-unt-ur.
boy-nom.pl book-acc.sg send-3.pl.pres-passive
Intended meaning: 'The boys are being sent a little book.'

Libell-us          puer-is    mitt-it-ur.
little.book-nom.sg boy-dat.pl send-3.sg.pres-passive
'The boys are being sent a little book.'

As already noted, the German and Latin pattern is exactly the reverse of that in English: English allows the recipient passive, but not the theme passive, whereas German and Latin allow the theme passive, but not the recipient passive.

The German and Latin pattern can be derived as follows. The schematic structures for the active and the passive are shown in (46) and (47), respectively. As in (41), (47) shows the structure for the passive before and after verb movement.

(46)	a.
(47)	a.				b.

The passive morpheme on the higher V head in (47a) removes the verb's ability to check structural case (though not its ability to check inherent case). As before, the properties of the higher head are preserved after verb movement. What the grammaticality of the theme passive suggests is that in German and Latin, the complex head's inability to check structural case is passed down to the adjoined lower verb. In other words, the passive morpheme affects the ability of the lower verb (via its trace) to check structural case in the head-comp configuration, as highlighted by the boxes in (47b). Since the theme argument cannot check structural case with its sister, it must move. Unlike in English, the recipient argument can check inherent case with the complex head, so the resulting (45b) is grammatical.

Resume reading here.

Notes

1. By the criterion just mentioned, participial constructions as in (i) are true passive constructions, despite the absence of a passive auxiliary.

Note the close parallel with the constructions in (ii), which contain a passive auxiliary.

2. Strictly speaking, the verb-object combinations under discussion might better be referred to as fixed expressions than as idioms. The so-called idiom chunk has a very restricted distribution, but the interpretation of the entire expressions is compositional, unlike in a true idiom.

3. We assume that expletive it and there in (26) would substitute directly into Spec(IP).

4. Certain archaic German verbs govern the genitive, which patterns as an inherent case, as shown in (i) and (ii), which are parallel to (29) and (30).

dass er der     Ahnen     gedachte
that he the.gen ancestors commemorated
'that he commemorated the ancestors'

dass die     Ahnen     gedacht      wurden
that the.nom ancestors commemorated became.pl
Intended meaning: 'that the ancestors were commemorated'

dass der     Ahnen     gedacht      wurde
that the.gen ancestors commemorated became.sg
'that the ancestors were commemorated'

5. At least some speakers of British English allow the theme passive in (39b), which was possible in earlier stages of English. In American English, the theme passive has been entirely replaced by the recipient passive, whereas British English is not quite as far along in the replacement process.

B. In addition to be, English uses get as a passive auxiliary, as shown in (2).

There is a purely syntactic difference betweeen the two passive auxiliaries. What is it?

Exercise 10.2

B. Based on the evidence in (1), compare the syntactic properties of Standard English versus Pittsburgh English need. Be explicit but brief.

Exercise 10.3

B. One of the sentences in (2) is structurally ambiguous. Which one is it, and why?

Exercise 10.4

B. Which case is checked on the underlined noun phrases? How do you know?

C. In which licensing configuration(s) is case checked on the underlined noun phrases?

D. The syntactic properties of prove differ in (1a) and (1b). Briefly explain how.

Exercise 10.5

Hint: See Exercise 9.2 C, D.