Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS (linguistics.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 22 November 2017

Displacement in Syntax

Summary and Keywords

Displacement is a ubiquitous phenomenon in natural languages. Grammarians often speak of displacement in cases where the rules for the canonical word order of a language lead to the expectation of finding a word or phrase in a particular position in the sentence whereas it surfaces instead in a different position and the canonical position remains empty: ‘Which book did you buy?’ is an example of displacement because the noun phrase ‘which book’, which acts as the grammatical object in the question, does not occur in the canonical object position, which in English is after the verb. Instead, it surfaces at the beginning of the sentence and the object position remains empty. Displacement is often used as a diagnostic for constituent structure because it affects only (but not all) constituents. In the clear cases, displaced constituents show properties associated with two distinct linear and hierarchical positions. Typically, one of these two positions c-commands the other and the displaced element is pronounced in the c-commanding position. Displacement also shows strong interactions with the path between the empty canonical position and the position where the element is pronounced: one often encounters morphological changes along this path and evidence for structural placement of the displaced constituent, as well as constraints on displacement induced by the path.

The exact scope of displacement as an analytically unified phenomenon varies from theory to theory. If more then one type of syntactic displacement is recognized, the question of the interaction between movement types arises. Displacement phenomena are extensively studied by syntacticians. Their enduring interest derives from the fact that the complex interactions between displacement and other aspects of syntax offer a powerful probe into the inner workings and architecture of the human syntactic faculty.

Keywords: word order, syntax, constituency, hierarchical structure, linear order, filler–gap dependency, extraction, island effects

1. Displacement: The Phenomenon

Displacement is a ubiquitous phenomenon in natural languages. Grammarians often speak of displacement in cases where the rules for the canonical or unmarked word order of a language (see Dryer, 2007, for common ways of diagnosing unmarked word order) lead to the expectation of finding a word or phrase in a particular position in the sentence whereas it surfaces instead in a different position and the canonical position remains empty. To take a concrete example, the canonical order for English sentences is subject–auxiliary–verb–object, (1a). Under the characterization of displacement just given, all of the examples in (1b)–(1d) involve displacement of the object and (1e) involves displacement of the auxiliary. The underscore in the examples marks the canonical position of the displaced element. A theory-neutral way of referring to the displaced constituent is as the filler and the canonical position as the gap.

(1)

Displacement in Syntax

In Section 4 we will see that this rough-and-ready characterization of displacement in terms of canonical/non-canonical word order alternations only works as a first approximation, the determination of the scope of the phenomenon of displacement in syntax being ultimately made by the syntactic theory. In many theories certain word order alternations are explained without displacement, while at the same time displacement might be invoked for cases of non-alternating word order.

In much of the rest of the article, except for Section 4, we will focus on the properties of the displacement of interrogative pronouns and interrogative phrases (wh-phrases), mostly in English. One characteristic property of wh-displacement is that it can be unbounded, i.e., there is no grammatically determined upper limit to the linear and structural distance between filler and gap, (2). In (2a) the linear distance between filler and gap is seven words, in (2b) it is ten words, in (2c) — 12, and in (2d) — 15. The structural distance between filler and gap, as measured by how deeply embedded the gap is compared to the filler is one clause in (2a), two in (2b), three in (2c), and zero in (2d). In the latter, filler and gap are both in the main clause. Clausal constituents in (2) are delimited by square brackets.

(2)

Displacement in Syntax

Syntactic displacement is structure dependent in several ways. Firstly, only constituents can be displaced. To see this note that example (3) is structurally ambiguous. The two readings can be paraphrased as follows: ‘He introduced the guest and he did it with a hoarse voice.’ and ‘He introduced the guest who had a hoarse voice.’ Their structures are given in (3a) and (3b), respectively.

(3)

Displacement in Syntax

As can be seen in (4a)–(4b), displacement leads to a loss of the ambiguity. Example (4a) must be based on (3a), since the displaced string of words ‘the guest’ is a constituent only in (3a); thus, (4a) only has the reading corresponding to (3a). Likewise for (4b). (4b) must be based on (3b), since the displaced string of words ‘the guest with a hoarse voice’ is a constituent only in (3b); thus, (4b) has only the reading corresponding to (3b). Example (4c) is ungrammatical: the displaced string of words doesn’t correspond to a constituent in any of the structures.

(4)

Displacement in Syntax

A second way in which displacement is structure dependent concerns the observation that the displaced element always c-commands the gap. (This generalization is called the Proper Binding Condition, and it goes back to Fiengo, 1977.) C-command is a relation defined hierarchically on syntactic trees (for discussion of the family of syntactic command relations, of which c-command as defined here is the simplest because it makes no reference to category labels on nodes, see Barker & Pullum, 1990): A node α c-commands another node β iff β is a sister of α or is dominated by a sister of α. Consider the embedded clauses in (5). (5a) is a clause introduced by the complementizer ‘that’, followed by a very complex subject (‘the fact … cancer’), the auxiliary ‘will’, the verb ‘shock’, and the object ‘Emily’. The structure is sketched in (5b). In (5c), the object has been displaced to a c-commanding position in front of the subject with accompanying loss of the complementizer ‘that’. A rough structure for (5c) is given in (5d). Notice that the displaced element is pronounced in the c-commanding (high) position.

(5)

Displacement in Syntax

The sister of ‘who’ in (5d) is the constituent that makes up the entire rest of the clause. It contains the gap, as required.

It is clear from the examples that the verb ‘know’ can combine with declarative clauses, (5a), and with indirect questions, (5c). It might therefore be expected that example (6a) is well formed. As shown in (6b), ‘who’ (the object of ‘shock’) is displaced to the position between ‘know’ and ‘you’. The structural sister of ‘who’ is the constituent ‘you know that you have cancer’. Obviously, ‘who’ in its displaced position does not c-command the gap; the example violates the generalization that displaced elements must c-command the gap and is, therefore, complete gibberish.

(6)

Displacement in Syntax

A third demonstration of the structure dependence of displacement comes from constraints on displacement and the interaction between properties of the structural path between filler and gap and the wellformedness of displacement. Such effects will be discussed in 3.2.

The discussion above introduced a canoncial example of displacement and illustrated its structure dependence. Sections 2 and 3 discuss further properties of the displacement of wh-phrases, mostly in English. Section 2.1 shows that for certain phenomena displaced wh-phrases have the characteristic behavior of phrases occupying the gap (low behavior), and Section 2.2 shows that for other phenomena they show characteristic properties of occupying the structurally higher, c-commanding position of the filler (high behavior). Section 3 introduces the notion of a structural displacement path, Section 3.1, and discusses interactions between displacement and the path, Section 3.2. Section 4 turns to the question of how to delimit the role of displacement in syntax without, however, attempting to answer the question.

2. Low and High Behavior of Displaced Constituents

This section briefly introduces evidence to the effect that a displaced constituent can have properties associated with the apparent hierarchical position of the gap (subsection 2.1) and with the hierarchical position of the filler (subsection 2.2).

Sentence structure is characterized in terms of the linear order, hierarchical arrangement, and the category of elements making up the sentence. In the next subsection we will see evidence that a displaced constituent must be associated in some way with the site of the gap despite their different linear position. This might suggest that displacement is a purely linear phenomenon, that is, it raises the possibility that the displaced constituent occupies the position of the gap but is linearized discontinuously with its structural sister. Above, we already illustrated the important generalization that fillers always c-command gaps. This fact alone already demonstrates that displacement cannot be given a full characterization in terms of linear order of words alone but must take constituent structure into account. However, even a characterization of displacement as anomalous linearization mindful of syntactic structure is insufficient, and standard theories of displacement associate the displaced constituent with (at least) two distinct hierarchical positions: the position occupied by the filler and that of the gap, as in structure (5d) above. Mindful of how syntactic structures are usually depicted, we can then characterize behavior characteristic of the gap site as ‘low’ behavior and behavior characteristic of the filler site as ‘high’ behavior.

2.1. Low Behavior of Displaced Constituents

Consider example (7). The interpretation of ‘this song’ with respect to the event described by the embedded clause varies systematically with the verb. ‘This song’ is the object created, hated, or banned. These interpretations characterize very specifically the object of the subordinate clauses, as its subject ‘Sally’ receives a different set of interpretations.

(7)

Displacement in Syntax

In example (8), ‘this song’ has been replaced by ‘which song’ and displaced to the front of the embedded clause. ‘Which song’ receives exactly the same characteristic interpretations (of object created, hated, and banned, respectively) as the in situ objects in (7). The displaced phrase inherits the semantic (thematic) interpretation of the gap site.

(8)

Displacement in Syntax

We will now see that the displaced phrase also has important syntactic properties of the gap site. First, consider selection. Example (9a)–(9c) show that ‘applaud’, ‘admire’, and ‘sleep’ respectively select an object optionally, obliagatorily, and not at all. None of them select two objects, (9d).

(9)

Displacement in Syntax

Example (10) shows that the displaced wh-phrase plays the role of the optional, obligatory, or prohibited object. In other words, the displaced constituent fulfills the selectional properties associated with the gap.

(10)

Displacement in Syntax

In terms of agreement, displaced wh-phrases also behave as though they were in situ. In English, the verb shows (some) morphological agreement with the local subject. Thus, (11a) requires ‘is’ and (11b) requires ‘are’ in the embedded clause. Notice that the presence of a singular subject ‘Fred’ in the main clause does not influence agreement in the embedded clause.

(11)

Displacement in Syntax

Displaced wh-phrases trigger agreement just as if they occupied the position of the gap. This is illustrated in (12), where the displaced wh-phrase ‘which player(s)’ relates to the gap in the subject position of the embedded clause.

(12)

Displacement in Syntax

We see the characteristic agreement effect as though the wh-phrase were in situ.

A further illustration of the low behavior of a displaced wh-phrase comes from languages with case morphology on fillers. In such languages the form of the filler varies in the same way with the verb as an element filling the gap would; (13) shows that the German verb ‘unterstützen’ (‘support’) takes an accusative object while ‘helfen’ (‘help’) takes a dative object. Questions about those being supported or helped must be introduced by interrogative pronouns in the appropriate form: accusative for ‘unterstützen’ (‘support’) and dative for ‘helfen’ (‘help’), (14).

(13)

Displacement in Syntax

(14)

Displacement in Syntax

A final illustration that displaced constituents may behave as though they occupied the gap position comes from their interpretation, specifically, the interpretation of pronouns within the displaced phrase. Pronouns have various uses. In the use relevant to the present discussion, a pronoun does not pick out a particular, stable referent, but instead depends for its reference on another expression and the referent varies with that other expression.

Example (15) illustrates this. The possessive pronoun ‘his’ has three easily accessible interpretations: First, ‘his’ can refer back to Joe (… no boy has found Joe’s mother). Second, ‘his’ can pick out a referent not mentioned in the sentence at all; for example, if we have been talking about a boy named Fred, then ‘his’ can refer back to Fred (… no boy has found Fred’s mother). Finally and most importantly, ‘his’ can depend on ‘no boy’ (… no boy has found his own mother).

(15)

Displacement in Syntax

(16)

Displacement in Syntax

This example allows the first two interpretations but the interpretation where ‘his’ depends for its reference on ‘no boy’ is not available (or only at a stretch).

Examples (15)–(16) illustrated the behavior of a pronoun within a phrase that has not been displaced. Now consider examples (17) and (18). The examples are constructed on the basis of (15)–(16) by replacing ‘his mother’ by ‘which of his relatives’ and displacing the wh-phrase to the beginning of the clause. The wh-phrase appears in exactly the same sentence-initial position in both of the examples. We are interested in the question of whether the possessive pronoun within the displaced phrase in examples (17) and (18) has the referentially dependent reading.

(17)

Displacement in Syntax

(18)

Displacement in Syntax

As can be seen, the reading where `his’ depends referentially on `no boy’ is available in example (17) but not (or only at a stretch) in example (18). Thus, the interpretation of the possessive pronoun is not determined by the position where the phrase is pronounced, which is the same position in both examples. Instead, the interpretation is determined by the position of the gap: The possessive pronoun in (17) has the same interpretive range as that in (15) and the possessive pronoun in (18) has the same interpretive range as the one in (16). This is what we have been calling the low behavior of the displaced phrase.

In this subsection we have seen five ways in which a displaced constituent behaves as though it were in situ, that is, as though it occupied the position of the gap. This kind of evidence is strong motivation for structurally associating the displaced constituent with the position of the gap. It should be noted that not all types of displacement behave identically. While all kinds of displacement show low behavior for selection, some displacement operations show low and others high behavior when it comes to case and pronominal interpretation. See Section 4 for further discussion.

2.2. High Behavior of Displaced Constituents

The previous subsection looked at the low behavior of constituents displaced by wh-movement. We now turn to their high behavior.

As mentioned in the introduction, the Proper Binding Condition demands that dislocation target a c-commanding position. This leads to the straightforward prediction that we should be able to find processes that apply to constituents including the gap but excluding the filler. Such a prediction cannot be derived from a hypothetical theory (though see Blevins, 1990; Kathol, 2000) that explained the low behavior of displacement in terms of a low hierarchical position of the filler and its linear position in terms of discontinuous constituency: such a theory crucially would not make available any constituents including the gap but excluding the filler. Prima facie evidence that the theory associating displaced constituents with two hierarchical position is correct comes from sluicing and other types of ellipsis. In sluicing, the meaning of a wh-question is expressed by a truncated expression consisting of just a wh-phrase. Example (19) is possible both with the material in angled brackets pronounced and with this material left out. When the material is left out, we are apparently dealing with ellipsis of a clausal constituent containing the gap but excluding the filler.1

(19)

Displacement in Syntax

The example shows that the expectations of a theory associating displaced constituents with a high hierarchical positions are borne out.2

Another piece of evidence for the assumption that displaced constituents occupy a position hierarchically higher than the gap comes from anti-reconstruction effects. The previous subsection introduced a number of reconstruction effects, that is, effects where where a displaced constituent behaves as though it occupied the position of the gap. There are also instances where this is not the case: anti-reconstruction effects.

Consider the examples in (20a) and (20b). In (20a) ‘him’ can refer to the same person as ‘Fred’, but in (20b) ‘he’ cannot refer to the same person; ‘he’ and ‘Fred’ must be different people.

(20)

Displacement in Syntax

The principle behind this observation is, as a first approximation, that an expression which can directly refer to an individual (like ‘Fred’) must not be c-commanded by a pronoun referring to the same individual. As structure (21a) shows, ‘Fred’ in (20a) is not c-commanded by the pronoun and the two can refer to the same person. However, as structure (21b) shows, ‘Fred’ is c-commanded by ‘he’ in (20b) and co-reference is impossible. (Note that an alternative, purely linear formulation of the principle, whereby a referential expression cannot come after a pronoun picking out the same referent, would work for examples (20a)–(20b) but would fail for (22a), immediately below.)

(21)

Displacement in Syntax

The crucial observation is that when the object is displaced, the difference between the examples disappears. ‘Fred’ and ‘him’/‘he’ can refer to the same person both in (22a) and in (22b). Example (22a) is based on (20a) with ‘the’ replaced by ‘which’ and the entire object displaced to the beginning. Examples (22b) and (20b) are related in just the same way.

(22)

Displacement in Syntax

We can understand why the examples allow the interpretations they do, by inspecting their structures, (23). In (23a) ‘him’ does not c-command ‘Fred’, and therefore the two can refer to the same person. In (23b), ‘he’ does not c-command ‘Fred’, and therefore the two can refer to the same person.

(23)

Displacement in Syntax

To achieve this result, we have to apply the principle that blocks coreferential interpretation in (20b) (with the structure in (21b)) to the wh-phrase in the displaced position in (22b) (with the structure in (23b)). In other words, the displaced constituent shows high behavior.3

A final illustration that displaced constituents behave unlike their non-displaced counterparts comes from the phenomenon of parasitic gaps (see Culicover & Postal, 2001; Engdahl, 1983). A basic example (Engdahl, 1983, ex. 1) is given in (24).

(24)

Displacement in Syntax

There are two gaps in this sentence: one after ‘file’ and another after ‘reading’. The content of both gaps depends on the displaced constituent. The gap after ‘reading’ is parasitic on the regular gap after ‘file’ in the sense that the parasitic gap cannot exist without the regular gap, as (25b) shows. The regular gap on the other hand can exist without the parasitic gap, (25a).

(25)

Displacement in Syntax

A second important property of parasitic gaps is that they are possible only when there is displacement. Example (26) has the same wh-phrase (‘which articles’) functioning as the object of the verb as in example (24). But while it is displaced in (24) and does license the parasitic gap, it is not displaced in (26) and does not license the parasitic gap.

(26)

Displacement in Syntax

This shows again that in-situ and displaced phrases have different syntactic properties. Crucially, the high position of the wh-phrase determines the possible structural positions for the parasitic gap, since the wh-phrase in the high position must c-command the parasitic gap. To see this, consider the following sentences.

(27)

Displacement in Syntax

The difference between the two examples is that while example (27a) contains a pronoun as the object of ‘opening’, example (27b) contains a parasitic gap. Example (27a) is ambiguous. One reading can be paraphrased as follows: I can tell which books were shelved by you without me having to open them to make my determination. The other reading can be paraphrased thus: I can tell which books you shelved without you having read them. Example (27b) on the other hand is unambiguous; it only has the latter type of reading. The ambiguity of example (27a) is a structural attachment ambiguity. The first reading corresponds to the structure in (28a), while the second reading corresponds to that in (28b).

(28)

Displacement in Syntax

Just like the real gap, the parasitic gap must be c-commanded by the displaced phrase in its high position. This explains why the parasitic gap is impossible in the structure in (28a) and why, therefore, (27b) is unambiguous. As we have seen, the theory of parasitic gaps makes crucial reference both to the fact of displacement and to the high structural position of the displaced phrase and therefore furnishes another example of the high behavior of the displaced phrase.

In this section we have briefly surveyed some evidence showing that in certain respects displaced constituents behave as though they occupied the position of the gap (their low behavior) and that in other respects they behave as though they occupied a different hierarchical position (their high behavior). In the simple cases discussed so far, displacement was always characterized by the presence of a pronounced filler and an unpronounced gap. Things are not necessarily this simple. It is often necessary to postulate abstract, unpronounced fillers (syntactic operators, null pronouns, etc.) and sometimes even pronounced gaps (Alexandre, 2012; Muysken, 1980). (The question of whether resumptive pronouns are the residue of displacement is a vexing one, and I shall not discuss it here.) In such cases, other properties of displacement, especially those relating to reconstruction and path effects, to be discussed in the next section, are used to diagnose displacement.

3. Displacement Paths

This section discusses the notion of displacement path and illustrates three types of interaction between a displaced element and the displacement path: morphological reflexes of displacement along the path, binding and scope effects, and constraints on displacement.

3.1. Modeling Paths

The discovery of unbounded filler–gap dependencies gives rise to the question of how unbounded dependencies are implemented in the grammar. Their unbounded nature is somewhat unusual within the grammar, and syntacticians have sought to understand the unboundedness in terms of the iterated application of local operations. We saw in the previous section that the displaced constituent is located in a structurally high position, c-commanding the gap, while at the same time retaining morphological, syntactic, and semantic properties of the gap site. How do the grammatical properties associated with the gap end up being expressed on the filler?

Two main metaphors have guided theorizing about the question of how filler–gap dependencies are mediated: The first metaphor is movement, the second metaphor is percolation.

The movement metaphor says that, as part of the process that derives sentences with filler–gap dependencies, the filler moves from the position of the gap to the position of the filler, occupying each at distinct points of a syntactic derivation. In such models unbounded displacement is implemented as an arbitrarily long sequence of short, bounded steps. With the ancillary assumption that movement targets the edge of VP and the edge of CP (Chomsky, 1986, 2000, 2001), example (29) would roughly have the representation in (30). It is important to note how the filler leapfrogs through the structure, entering very local relations with some nodes along the path, those boxed in the diagram, but not with others, particularly the TP nodes in the structure in (30). This metaphor underlies the majority of studies of displacement (see Chomsky, 1973, 1977, 1986, 2001; Lasnik & Saito, 1992; McCloskey, 2002; Ross, 1967).

(29)

Displacement in Syntax

(30)

Displacement in Syntax

Under the percolation metaphor, the filler never occupies the site of the gap. Instead, the information that there is a gap, what its category and interpretation are, and what restrictions are placed on it, percolates through the syntactic structure locally, from node to immediately dominating node (see Bach, 1984; Cinque, 1990; Kayne, 1983; Neeleman & van de Koot, 2002; Pesetsky, 1982; Pollard, 1984; Pollard & Sag, 1994; Steedman, 1996; among others). This percolation of information is depicted by circling the nodes through which the information percolates in (31).

(31)

Displacement in Syntax

The distinctions between the two ways of thinking about syntactic displacement are somewhat subtle and need not concern us here.

3.2. Path Effects

We find that movement along a path can exert an influence on the material crossed, as both types of model straightforwardly predict. The most famous illustration of this type of effect comes from the complementizer alternation in Irish along the path of movement (see McCloskey, 1979, 1990, 2002; Noonan, 1997). The basic fact is illustrated in (32). While in example (32a) no relevant displacement has taken place and we find the complementizer ‘go’, in (32b) the most deeply embedded object, a wh-phrase, has been displaced across three clauses forcing the complementizers along the path to show up in the alternative form ‘a’ instead. The subscripts N and L on the complementizers indicate that the form has either a nasalizing or leniting effect on the following word, which accounts for the observed additional alternations.

(32)

Displacement in Syntax

The alternating complementizers here lie along the path of movement. Complementizers that do not lie along the movement path in the sense of the previous subsection do not participate in the alternation.

We also find that the displaced element can behave for purposes of binding and scope as though it occupied a position along the movement path which is neither the site at which it surfaces nor the gap site. This is illustrated by the examples in (33) (from Barss, 2001, p. 676). Example (33a) features an object with a referentially dependent item in the embedded clause: ‘these portraits of {themselves | each other}’. The referentially dependent items are ‘themselves’ and ‘each other’. The example illustrates the requirement that these items have a local (c-commanding) antecedent (though see Pollard & Sag, 1992). The matrix subject ‘the men’ is too far away to serve as the antecedent for ‘themselves’ or ‘each other’.

(33)

Displacement in Syntax

When the object is displaced to the beginning of the sentence, the referential dependency of ‘themselves’ and ‘each other’ on ‘the men’ becomes possible, (33b). Curiously, neither the high nor the low position for the displaced element will serve to establish the dependency: the low position is the position of the object in (33a), which, as we saw, is too far away from ‘the men’ to establish the dependency; in its high position, the wh-phrase is not c-commanded by ‘the men’, making it again impossible to establish a referential dependency. The correct prediction can be made only if we assume that the displaced constituent can behave as though it is located along the displacement path.

The third and most famous type of interaction between displacement and displacement path are constraints on displacement, known in the trade as island effects. Recall that the distance between filler and gap can be unbounded. This does not by any means entail that displacement is entirely unconstrained. Ross (1967) was the first to systematically catalogue constraints on displacement. The discovery of the existence of systematic structural constraints on filler–gap dependencies is among the most important empirical discoveries of modern syntactic theory. Ross (1967) called a particular structural configuration an island if the gap cannot grammatically be situated inside of the configuration while the displaced element is outside. (For excellent introductions to the topic, see Richards, 2014; Szabolcsi, 2001.)

Among the island constraints operative in English, I illustrate three that have played a particularly important role in the history of the theory of syntax. The Subject Condition bans extraction from subjects (see Ross, 1967, p. 243, for clausal and Chomsky, 1973, pp. 249–250, for DP subjects). The Subject Condition is illustrated in (34). In each case, the grammatical subject is enclosed in square brackets. The grammatical and ungrammatical examples differ in what constituent acts as the subject. The grammatical controls show that extraction from those constituents is possible in principle, while the ungrammatical versions show that the sentence becomes degraded when the constituent in question acts as the subject. There are two examples of DP subjects, (34b) and (34d), and one of a clausal subject, (34f).

(34)

Displacement in Syntax

The Complex NP Constraint (going back to Ross, 1967, p. 127) bans movement of elements contained in a clause dominated by a noun phrase (DP) out of that noun phrase. The Complex NP Constraint is illustrated in (35a), the remaining examples in (35) serve as controls.

(35)

Displacement in Syntax

Finally, the wh-island condition, which varies substantially in the severity of the degradation along several dimensions, bans displacement from an indirect question. It is illustrated in (36) with the only mildly deviant displacement of an object from a non-finite indirect question, (36a), and the very severely degraded displacement of a manner adverb from a finite indirect question, (36b).

(36)

Displacement in Syntax

We have seen that displacement of wh-phrases in English is characterized simultaneously by its unbounded nature and its sensitivity to island configurations, i.e., to specific properties of the displacement path. This is the third illustration of the structure dependence of displacement mentioned in Section 1. The same simultaneous characterization of unboundedness with island sensitivity also describes a substantial number of other constructions in English involving gaps but not always overt fillers. Their unbounded nature and sensitivity to islands has led to their classification not only as displacement phenomena but as displacement phenomena of the same type as wh-movement (see Chomsky, 1977). Significantly, this characterization also describes long-distance dependencies in many other languages (see already Ross, 1967).

While there can be no doubt about the importance of island effects to linguistic theory, the question of the ultimate cause of island effects remains disputed (see Boeckx, 2008a, 2012). Theoreticians hold a broad range of views on how to characterize islands. There are processing accounts (Hofmeister & Sag, 2010; Hofmeister, Staum Casasanto, & Sag, 2013; Kluender, 1998; Kluender & Gieselman, 2013; Pritchett, 1991; Sag, Hofmeister, & Snider, 2007; among others), semantic accounts (e.g., Abrusán, 2014; Szabolcsi & Zwarts, 1992; among others), accounts that locate island effects at the interface of syntax with semantics (Chomsky & Lasnik, 1993; Lasnik & Saito, 1992; among others), the interface of syntax with information structure (e.g., Erteschik-Shir, 1998, 2007; Erteschik-Shir & Lappin, 1979; Goldberg, 2013; Van Valin & LaPolla, 1997), or the interface of syntax with phonology (e.g., Abe & Hornstein, 2012; Chomsky, 1973; Hornstein, Lasnik, & Uriagereka, 2007[2003]; Uriagereka, 1999). Finally, there are many accounts that locate island effects in the syntax proper (Chomsky, 1977, 1986; Cinque, 1990; Huang, 1982; Manzini, 1992; Müller, 2011; Pesetsky, 1987; Rizzi, 1990; Starke, 2001; Stepanov, 2007; among others).

4. The Scope of Displacement

The preceding sections briefly illustrated properties of displaced wh-phrases, mostly in English. Wh-displacement is a standard example of syntactic displacement: There is clear evidence for two distinct positions (filler and gap); they are distinct both linearly and hierarchically; the filler strongly exhibits properties associated with both positions; there are non-trivial path effects. Wh-displacement belongs to a larger family of structures that have very similar properties. The constructions that are part of this empirical cluster include topicalization, (37a), relative clause formation, (37b)–(37c), and comparative clauses, (37d).

(37)

Displacement in Syntax

In each case above, the gap is the object of ‘read’. In (37a)–(37b) there is a clear filler: ‘this book’ and ‘which’, respectively.4 In (37c) and (37d), there is no obvious filler. Nevertheless, the clustering of properties mentioned in the next paragraph strongly suggests a unified treatment of all four types of structures. A unified analysis is often implemented by assuming that, despite superficial appearances, there is a filler present in (37c) and (37d) after all: the inaudible element referred to as Op (for operator).

The family of structures that behave essentially like wh-displacement belongs to the class of displacement phenomena dubbed non-A movement in Postal (1971). Postal’s work classified operations on the basis of how they interact with pronominal reference. Section 2 illustrated that wh-displacement shows low behavior when it comes to pronominal reference but high behavior when it comes to disjoint reference effects. Postal’s classification is based on similar facts and, for English, classifies displacement operations into two non-overlapping classes, which Postal called, for lack of a better term, A-operations and non-A-operations.5 In his seminal paper “On wh-movement,” Chomsky (1977) demonstrated that there is a subset of Postal’s non-A-operations including wh-displacement that behave essentially identically when it comes to the position of the filler in the clause and to path effects: the filler is always at the edge of the clause; the path is unbounded; and it can span multiple finite clauses but is sensitive to island effects. McCloskey (1979) showed that all and only the operations studied by Chomsky give rise to identical morphological path effects in Irish. The operations include wh-displacement, relative clause formation, and comparative clause formation (as well as a range of further structures that cannot be discussed here for reasons of space). These form in many ways the core case of displacement in syntax.

The reminder of this section will mention some other constructions that are more or less similar to these canonical cases of displacement, as well as some of their core properties. There is no agreement on the extent to which the analyses of all of these cases should be assimilated to each other. We discuss what is at stake briefly in the concluding section. We will content ourselves with simply illustrating the range of potential further cases of syntactic displacement here without, however, any attempt at comprehensive coverage.

As mentioned above, Postal, 1971 identified two classes of movement operations. We have looked at some examples of non-A-movement above and now turn to A-movement. Examples of A-movement include the passive and raising-to-subject. In the passive, the thematic object appears preverbally in the canonical subject position. In raising-to-subject, the thematic subject of one verb precedes a different verb and agrees with it.

(38)

Displacement in Syntax

(39)

Displacement in Syntax

The above examples show an active-passive alternation and the alternation between a canonical structure and a subject-to-subject raising structure. Underscores indicate the position of the gap. A-operations have some important properties of displacement: There is a filler and a gap; just like in wh-displacement, the filler c-commands the gap; the operation is sensitive to constituent structure. These properties are familiar from Section 1. Furthermore, there are path effects: The filler can be an unbounded distance away from the gap but may not be separated from it by an island or a finite clause boundary; we often find agreement (for example in gender and number) along the path. These are variations on the properties we illustrated in 3. Clearly low behavior is seen in the fact that the filler must obey selectional requirements imposed on the gap. In stark contrast to wh-displacement, other low behavior is difficult to induce (for discussion see Boeckx, 2001; Chomsky, 1993, 1995; Fox, 1999; Lasnik, 1999; Postal, 1971). By contrast, high behavior of A-movement is easy to demonstrate using constituency diagnostics, semantic properties like scope, variable binding, and disjoint reference effects. A-movement also displays high behavior when it comes to case properties of the filler and person–number agreement: The filler bears the case associated with the high position and enters agreement relations there. This is probably connected with the nature of the high positions targeted by A-movement. The relative paucity of evidence for low behavior of A-movement has led some researchers to propose fundamentally different mechanisms for the analysis of A-movement on the one hand and wh-displacement on the other. Other researchers consider the existing similarities as compelling evidence for a unifying treatment.

A third class of displacement phenomena is head movement. The difference in word order between (40a) and (40b) illustrates head movement. The position of the gap in (40b) corresponds to the position of the verb in (40a). What is being displaced here is a syntactic head. In this example, the finite verb of the clause is displaced from the final position to the initial position.

(40)

Displacement in Syntax

Head movement again is sensitive to syntactic structure and can cross linearly arbitrarily long distances. The deliberate complexity of the example above with a complex subject hints at this. We find strong path effects with head movement, which not only obeys island constraints, but also obeys the very stringent condition that no intervening heads can be crossed (see Travis, 1984; Koopman, 1984; Rivero, 1991 for potential counterexamples). For head movement, it is easy to demonstrate low behavior since heads always select, assign case, and so forth in the low position and, arguably (see Hall, 2015; Lechner, 2005, and references cited there for discussion), also takes scope in the low position. Evidence for high behavior comes mainly from word order, though Baker’s (1988) Government Transparency Corollary and certain analysis of verb-stranding VP ellipsis (Goldberg, 2005) provide additional arguments for a high syntactic position of displaced heads. A further property that has attracted much attention (see Matushansky, 2006, and references cited there) is that, at least under the standard analysis, the filler does not, strictly speaking, c-command the gap, being instead adjoined to an element that does c-command the gap. This last property, together with the notable lack of semantic evidence for the high syntactic position of displaced heads, has given rise to a strand of analyses that sharply divide head movement from syntactic displacement phenomena in terms of analytical approach and, indeed, deny that there is a high position involved in head movement at all (Adger, 2013; Boeckx & Stjepanović, 2001; Brody, 2000; Bye & Svenonius, 2012; Hall, 2015; among others).

The three types of displacement discussed so far (non-A-movement, A-movement, and head movement) differ vastly in their properties. They differ so much that some theories deny either A-movement or head movement or both the status of displacement in syntax. Nevertheless, all three are considered syntactic displacement by many theoreticians working within the Principles & Parameters and Minimalist frameworks. As discovered by Rizzi (1990, see also Chomsky, 1995; Cinque, 1990; Rizzi, 2001; Starke, 2001), all three types obey the same shape conservation constraint known as Relativized Minimality. Relativized Minimality says that a filler in a structural position of type α cannot be separated6 from its gap by another element β, which is of the same structural type as α. Rizzi recognized three structural types: A-positions, non-A-positions, and head positions. Relativized Minimality can be expressed as the following slogan: Likes do not cross likes. The fact that this shape conservation principle governs all three types of structures suggests that they are all instances of the same general kind: syntactic displacement.

In the displacement phenomena mentioned so far, the filler always precedes the gap. This is not necessarily the case, though. The standard analysis of example (41), where the object, uncharacteristically for English, is separated from the verb by a prepositional phrase involves the gap preceding the filler.7

(41)

Displacement in Syntax

Ross first noticed that, at least in English, gap and filler are never separated by a finite clause boundary when the gap precedes the filler. This makes (41), which exemplifies a structure called heavy NP shift, similar to A-movement. However, in most other respects, heavy NP shift patterns with wh-dislocation, a case of non-A-movement. Again, the similarities to syntactic displacement have inspired unifying analysis and the differences have been used to motivate substantially different approaches.

Another word order phenomenon raising intricate classificatory and analytical problems is A-scrambling, a process characteristic of OV languages. It is exemplified here by the alternation in (42). The order in (42a) arguably represents the canonical order (Frey, 1993, 2015; Frey & Pittner, 1998; Lenerz, 1977), while in (42b) the indirect object has scrambled across the locative prepositional phrase and in (42c) the direct object has scrambled across the indirect object.8

(42)

Displacement in Syntax

A-scrambling, like the other displacement phenomena we have seen, is sensitive to structure, the filler c-commands the gap, and so forth. (for an overview see Abels, 2015; Frey, 2015). Like in wh-displacement, the fillers in (42b) and (42c) bear the case associated with the gap. Like A-movement and heavy NP shift, A-scrambling cannot cross finite clause boundaries. When we look at the interaction of A-scrambling with referential properties, a mixed A/non-A pattern emerges. Indeed, any attempt to classify scrambling as either A- or non-A-displacement is doomed to fail. This is known as Webelhuth’s paradox (Webelhuth, 1992), which was widely discussed in the literature starting in the 1990s (Corver & Riemsdijk, 1994; Karimi, 2003; Mahajan, 1990, and references cited there). Again, while some researchers have found the similarities between scrambling and core syntactic displacement compelling and have brought the same tools to bear on the analysis of both, others have emphasized the differences and used substantially different theoretical tools in the analysis of both (Haider, 2006; Haider & Rosengren, 1998; Kuthy & Meurers, 2001; Neeleman, 1994; Reape, 1994; among others).

While the phenomena mentioned so far all share the property of having a silent gap, not all analyses agree that this is a necessary property of syntactic displacement. Thus, Alexandre (2012) and Muysken (1982) suggest that in some languages gaps may receive an invariant but audible pronunciation. More vexing problems are raised by resumptive pronouns. We mentioned above that relative clauses in English are a typical example of syntactic displacement. However, in many languages the position of the gap in relative clauses is occupied by a pronoun, a so-called resumptive pronoun. While such structures are still sensitive to structure and the filler c-commands the resumptive pronoun, they can behave in a way substantially different from canonical syntactic displacement structures. For example, the case of the filler and the resumptive pronoun may mismatch (Boeckx, 2003), the filler can relate to a resumptive pronoun inside a syntactic island (Ross, 1967; see also Boeckx, 2003 and references cited there), and referential and semantic effects can be substantially different from those found with wh-displacement (see Aoun, Choueiri, & Hornstein, 2001; Doron, 1982; Sells, 1984; Sharvit, 1999; among many others). Such observational differences are often, though by no means always, taken to speak against a unified account of syntactic displacement and pronominal resumption.

A case that may look even more exotic at first blush is provided by cases like (43) (and other so-called ‘covert movement’ constructions, Huang, 1982; May, 1977, 1985), where there is no gap and no visible displacement at all.

(43)

Displacement in Syntax

Note that the scope of the quantifier (‘every building’) includes the subject. It has been observed that by default, one element takes another in its scope only if the first c-commands the second. In order to achieve this here, we need to invoke displacement of ‘every building’ to a position c-commanding the subject (‘a guard’). Support for an analysis assimilating the grammar of scope taking to the grammar of displacement comes from the observation that covert scope shifting is sensitive to structure, targets a c-commanding position, is subject to syntactic islands and, according to recent experimental work (Syrett & Lidz, 2009, 2011; Tanaka, 2015), is unbounded. Again, analyses differ.

This section has provided a brief and superficial sampling of the different types of constructions that have been analyzed by one theory or another as syntactic displacement, along with wh-displacement discussed in the first three sections. The list is by no means exhaustive. It is intended merely to give a sense of the range of properties that these constructions have and to convey a sense of the theoretical challenges posed by these structures: How to explain the commonalities? How to explain the differences?

5. Theoretical Outlook

The first three sections of the present article introduce the phenomenon of displacement in syntax on the basis of wh-displacement in English. Displacement is a diagnostic for constituent structure in the sense that it affects only (but not all) constituents. In the simplest and clearest cases, displacement phenomena involve a single constituent that shows properties associated with two distinct linear and hierarchical positions. Typically, one of these two positions c-commands the other and the displaced phrase is pronounced in the (high) c-commanding position. Displacement also shows strong accompanying path effects: morphological changes along the path and evidence for structural placement of the displaced constituent along the path, as well as island effects and other constraints on displacement.

Section 4 surveyed a number of constructions and construction types with properties that can be vastly different from those of wh-displacement but which, nevertheless, have been analyzed as displacement, that is, the analysis uses the same theoretical tools as the analysis of wh-displacement. The question whether such unifying analyses are truly explanatory remains open at present. An analysis counts as explanatory to the extent that properties of the particular structure follow from simple, general principles. For example, all structures we have surveyed (except for head movement) relate two positions, one of which strictly c-commands the other. Minimalist analyses claim that the c-command property can be explained by the assumption that we are dealing with displacement implemented uniformly as ‘internal merge’. This is so because internal merge obeys the c-command property for principled reasons (Chomsky, 1995, 2008). However, according to Koster (1986) and Neeleman and van de Koot (2002), the c-command property characterizes not only syntactic displacement but all syntactic dependencies. This undermines the argument for a unification in terms of displacement substantially. As another example of an explanation made available by a unifying account (of at least some of the processes studied here), we can mention Relativized Minimality (Rizzi, 1990). Recall that Relativized Minimality is the principle according to which likes do not cross likes. The principle explains a large number of facts on the assumption that A-movement, non-A movement, and head movement are all modeled as syntactic displacement.

Another important issue that a theory of syntactic displacement will have to deal with is the way various operations interact. For example, can one and the same element undergo two separate displacement operations? And if so, in what order? In other words, can the expected filler site of one dependency act as the gap for another? Similarly, can a displaced element host a gap and, if so, under what circumstances? Syntactic theory is only beginning to address these questions. Some facts have been uncovered, but at the time of this writing there is no consensus on what the main generalizations are, let alone how to explain them. For some conflicting views see Abels (2007), Grewendorf (2003), May (1979), Müller (1998, 2010), Rizzi (2006), Takano (2000), Wexler and Culicover (1980), Williams (2002), and references cited there. The touchstone of an explanatory theory will be the extent to which the varied properties of the constructions surveyed in Section 4 and the interactions between them can be predicted. In this sense, Williams (2002) (despite the empirical shortcomings noted in Abels, 2007) stands out as the most explanatory of the existing approaches: Under Williams’ theory all properties of a displacement operation can be deduced on the basis of the relative structural height of the high position alone.

Displacement phenomena have been and will continue to be extensively studied by syntacticians. Their enduring interest derives from the fact that the complex interactions between displacement and other aspects of syntax offer a powerful probe into the inner workings and architecture of the human syntactic faculty.

Further reading

An understanding of the more technical literature cited throughout this paper will require previous knowledge of the relevant syntactic framework at a level equivalent to at least a graduate level textbook.

Richards, (2014) provides an overview of non-A movement, that is, the kind of displacement structures that the first three sections are based on.

Background to the various diagnostics for high and low behavior of displaced wh-phrases can be found in the following sources. Truswell (2014) is an introduction to the generalizations about variable binding and disjoint reference and theories thereof. Barss (2001) and Sportiche (2006) discuss the interaction displacement with variable binding and reference. Polinsky and Preminger (2014) provide background for the understanding of the case effect discussed in Section 2.1; Merchant (2006) discusses the elliptical construction used to illustrate high behavior in Section 2.2.

Background on the various theories mentioned in Section 3 can be found in Alexiadou and Kiss, (2015, Volume 2, Section IV) or Carnie, Sato, and Siddiqi (2014, Part V). A good introduction to island effects can be found in Roberts (1997, chapter 4) and Boeckx (2008b, chapter 2) provides a useful introduction to morphological path effects.

Handbook level introductions to some of the main phenomena mentioned in Section 4 are provided in Baltin and Collins (2001) for A-movement, Roberts (2001) for head movement, Szabolcsi (2001) for quantifier raising, and Abels (2015) for scrambling.

Rizzi (2001) and Corver (2006) discuss two issues mentioned in Section 5: the issue of Relativized Minimality and one type of interaction between different displacement phenomena (‘freezing effects’), respectively.

References

Abe, J. (2015). The in-situ approach to sluicing. Linguistik Aktuell 222. John Benjamins.Find this resource:

Abe, J., & Hornstein, N. (2012). “Lasnik-effects” and string vacuous ATB movement. In M. Uribe-Extebarria & V. Valmala (Eds.), Ways of structure building (pp. 169–205). Oxford University Press.Find this resource:

Abels, K. (2007). Towards a restrictive theory of (remnant) movement. In Linguistic Variation Yearbook 7, pp. 53–120.Find this resource:

Abels, K. (2015). Word order. In A. Alexiadou & T. Kiss (Eds.), Syntax—theory and analysis: An international handbook (Vol. 2, chapter 40, pp. 1400–1445). Handbooks of Linguistics and Communication Science 42. Berlin: Mouton de Gruyter.Find this resource:

Abels, K. (to appear). Movement and islands. In J. van Craenenbroeck & T. Temmerman (Eds.), The Oxford handbook of ellipsis (chapter 15). Oxford: Oxford University Press.Find this resource:

Abrusán, M. (2014). Weak island semantics. Vol. 3. Oxford Studies in Semantics and Pragmatics. Oxford: Oxford University Press.Find this resource:

Adger, D. (2013). A syntax of substance. Vol. 64. Linguistic Inquiry Monograph. Cambridge, MA: MIT Press.Find this resource:

Alexandre, N. (2012). The defective copy theory of movement: Evidence from wh-constructions in Cape Verdean Creole. Vol. 41. Creole Language Library. Amsterdam: John Benjamins.Find this resource:

Alexiadou, A., & Kiss, T. (Eds.). (2015). Syntax: theory and analysis. An international handbook. Vol. 1–3. Handbooks of linguistics and communication science, 42. Berlin: Mouton de Gruyter.Find this resource:

Aoun, J., Choueiri, L., & Hornstein, N. (2001). Resumption, movement, and derivational economy. Linguistic Inquiry, 32(3), 371–403.Find this resource:

Bach, E. (1984). Some generalizations of categorial grammars. In F. Landman & F. Velman (Eds.), Varieties of formal semantics (pp. 1–23). Dordrecht, The Netherlands: Foris Publications.Find this resource:

Baker, M. (1988). Incorporation: A theory of grammatical function. Chicago: University of Chicago Press.Find this resource:

Baltin, M., & Collins, C. (Eds.). (2001). The handbook of contemporary syntactic theory. Oxford: Blackwell.Find this resource:

Barker, C., & Pullum, G. K. (1990). A theory of command relations. Linguistics and Philosophy, 15, 1–34.Find this resource:

Barss, A. (1986). Chains and anaphoric dependence: On reconstruction and its implications. PhD diss., MIT.Find this resource:

Barss, A. (2001). Syntactic reconstruction effects. In M. Baltin & C. Collins (Eds.), The handbook of syntactic theory (chapter 21, pp. 670–696). Oxford: Blackwell.Find this resource:

Beerman, D., LeBlanc, D., & van Riemsdijk, H. (Eds.). (1997). Rightward movement. Vol. 17. Linguistik Aktuell. Amsterdam: John Benjamins.Find this resource:

Blevins, J. P. (1990). Syntactic complexity: Evidence for discontinuity and multidomination. PhD diss., University of Massachusetts.Find this resource:

Boeckx, C. (2001). Scope reconstruction and A-movement. Natural Language and Linguistic Theory, 19(3), 503–548.Find this resource:

Boeckx, C. (2003). Islands and chains: resumption as stranding. John Benjamins.Find this resource:

Boeckx, C. (2008a). Islands. Language and Linguistics Compass, 2(1), 151–167.Find this resource:

Boeckx, C. (2008b). Understanding minimalist syntax: Lessons from locality in long-distance dependencies. Vol. 9. Generative Syntax. Malden, MA: Blackwell Publishing.Find this resource:

Boeckx, C. (2012). Syntactic islands. Cambridge, U.K.: Cambridge University Press.Find this resource:

Boeckx, C., & Stjepanović, S. (2001). Heading toward PF. Linguistic Inquiry, 31(2), 345–355.Find this resource:

Bresnan, J. W. (2001). Lexical-functional syntax. Oxford: Blackwell.Find this resource:

Brody, M. (2000). Mirror theory: Syntactic representation in perfect syntax. Linguistic Inquiry, 31(1), 29–57.Find this resource:

Büring, D. (2005). Binding theory. Cambridge textbooks in linguistics. Cambridge, U.K.: Cambridge University Press.Find this resource:

Bye, P., & Svenonius, P. (2012). Non-concatenative phonology as epiphenomenon. In J. Trommer (Ed.), The morphology and phonology of exponence (pp. 427–495). Oxford: Oxford University Press.Find this resource:

Carnie, A., Sato, Y., & Siddiqi, D. (Eds.). (2014). The Routledge handbook of syntax. Routledge Handbooks. Abingdon, U.K.: Routledge.Find this resource:

Chomsky, N. (1973). Conditions on transformations. In S. R. Anderson & P. Kiparsky (Eds.), A festschrift for Morris Halle (pp. 232–286). New York: Holt, Rinehart and Winston.Find this resource:

Chomsky, N. (1977). On wh-movement. In P. W. Culicover, T. Wasow, & A. Akmajian (Eds.), Formal syntax (pp. 71–132). New York: Academic Press.Find this resource:

Chomsky, N. (1986). Barriers. Linguistic Inquiry Monograph 13. Cambridge, MA: MIT Press.Find this resource:

Chomsky, N. (1993). A minimalist program for linguistic theory. In K. Hale & S. J. Keyser: The view from Building 20 (pp. 1–52). Cambridge, MA: MIT Press.Find this resource:

Chomsky, N. (1995). Categories and transformations. In The minimalist program (chapter 4, pp. 219–394). Cambridge, MA: MIT Press.Find this resource:

Chomsky, N. (2000). Minimalist inquiries: The framework. In R. Martin, D. Michaels, & J. Uriagereka (Eds.), Step by step: Essays on minimalism in honor of Howard Lasnik (pp. 89–155). Cambridge, MA: MIT Press.Find this resource:

Chomsky, N. (2001). Derivation by phase. In M. Kenstowicz (Ed.), Ken Hale: A life in language (pp. 1–52). Cambridge, MA.: MIT Press.Find this resource:

Chomsky, N. (2008). On phases. In R. Freidin, C. Otero, & M.-L. Zubizarreta (Ed.), Foundational issues in linguistic theory: Essays in honor of Jean-Roger Vergnaud (pp. 133–166). Cambridge, MA: MIT Press.Find this resource:

Chomsky, N., & Lasnik, H. (1993). The theory of principles and parameters. In J. Jacobs et al. (Eds.), Syntax: An international handbook of contemporary research (Vol. 1, pp. 506–569). Berlin: Walter de Gruyter.Find this resource:

Cinque, G. (1990). Types of A′-dependencies. Cambridge, MA: MIT Press.Find this resource:

Corver, N. (2006). The Blackwell companion to syntax. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax (Vol. II, chapter 28, pp. 383–406). Malden, MA: Blackwell Publishers.Find this resource:

Corver, N., & van Riemsdijk, H. (Eds.). (1994). Studies on scrambling. Studies in Generative Grammar 41. Berlin: De Gruyter.Find this resource:

Culicover, P. W., & Jackendoff, R. S. (2005). Simpler syntax. Oxford University Press.Find this resource:

Culicover, P. W., & Postal, P. (Eds.). (2001). Parasitic gaps. Cambridge, MA: MIT Press.Find this resource:

Doron, E. (1982). On the syntax and semantics of resumptive pronouns. In R. Bley-Vroman (Ed.), Texas Linguistic Forum, 19, 1–48.Find this resource:

Dryer, M. S. (2007). Word order. In T. Shopen (Ed.), Language typology and syntactic description, Vol. 1, Clause Structure (2d ed., chapter 2, pp. 61–131). Cambridge, U.K.: Cambridge University Press.Find this resource:

Engdahl, E. (1983). Parasitic gaps. Linguistics and Philosophy, 6, 5–34.Find this resource:

Erteschik-Shir, N. (1998). The syntax–focus structure interface. In P. W. Culicover & L. McNally (Eds.), Syntax and semantics 29: The limits of syntax (pp. 211–240). Academic Press.Find this resource:

Erteschik-Shir, N. (2007). The syntax–discourse interface. Oxford Surveys in Syntax and Morphology. Oxford: Oxford University Press.Find this resource:

Erteschik-Shir, N., & Lappin, S. (1979). Dominance and the functional explanation of island phenomena. Theoretical Linguistics, 6, 41–86.Find this resource:

Everaert, M., & van Riemsdijk, H. (Eds.). (2006). The Blackwell companion to syntax. Malden, MA: Blackwell Publishers.Find this resource:

Frey, W. (2015). Word order. In A. Alexiadou & T. Kiss (Eds.), Syntax—theory and analysis: An international handbook(Vol. 1–3, chapter 18). Handbooks of Linguistics and Communication Science 42. Berlin: Mouton de Gruyter.Find this resource:

Fiengo, R. (1977). On trace theory. Linguistic Inquiry, 8, 35–62.Find this resource:

Fox, D. (1999). Reconstruction, binding theory, and the interpretation of chains. Linguistic Inquiry, 30(2), 157–196.Find this resource:

Fox, D. (2000). Economy and semantic representation. Cambridge, MA: MIT Press and MITWPL.Find this resource:

Frey, W. (1993). Syntaktische Bedingungen für die semantische Interpretation. Studia Grammatica 35. Berlin: Akademie Verlag.Find this resource:

Fox, W. (2015). Word order. In A. Alexiadou & T. Kiss (Eds.), Syntax—theory and analysis: An international handbook (Vol. 1–3, chapter 18). Handbooks of Linguistics and Communication Science 42. Berlin: Mouton de Gruyter.Find this resource:

Frey, W., & Pittner, K. (1998). Zur Positionierung der Adverbiale im deutschen Mittelfeld. Linguistische Berichte, 176, 489–534.Find this resource:

Goldberg, A. E. (2013). Backgrounded constituents cannot be “extracted”. In J. Sprouse & N. Hornstein (Eds.), Experimental syntax and island effects (pp. 221–238). Cambridge University Press.Find this resource:

Goldberg, L. M. (2005). Verb-stranding VP ellipsis: A cross-linguistic study. PhD thesis, McGill University, Montreal.Find this resource:

Grewendorf, G. (2003). Improper remnant movement. Gengo Kenkyu, 123, 47–94.Find this resource:

Haegeman, L. (1994). Introduction to government and binding theory (2d ed.). Blackwell Textbooks in Linguistics. Oxford: Blackwell.Find this resource:

Haider, H. (2005). How to turn German into Icelandic—and derive the OV-VO contrasts. The Journal of Comparative Germanic Linguistics, 8(1–2), 1–56.Find this resource:

Haider, H. (2006). Mittelfeld phenomena (Scrambling in Germanic). In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax (Vol. 3, chapter 43, pp. 204–274). Malden, MA: Blackwell Publishers.Find this resource:

Haider, H., & Rosengren, I. (1998). Scrambling. Sprache und Pragmatik, 49, 1–104.Find this resource:

Hall, D. (2015). Spelling out the noun phrase: Interpretation, word order, and the problems of “meaningless movement.” PhD diss., Queen Mary, University of London.Find this resource:

Hofmeister, P., & Sag, I. A. (2010). Cognitive constraints and island effects. Language, 86(2), 366–415.Find this resource:

Hofmeister, P., Staum Casasanto, L., & Sag, I. A. (2013). Islands in the grammar? Standards of evidence. In J. Sprouse & N. Hornstein (Eds.), Experimental syntax and island effects (pp. 42–63). Cambridge University Press.Find this resource:

Hornstein, N., Lasnik, H., & Uriagereka, J. (2007[2003]). The dynamics of islands: Speculations on the locality of movement. Linguistic Analysis, 33, 149–175.Find this resource:

Huang, C.-T. J. (1982). Logical relations in Chinese and the theory of grammar. PhD diss., MIT.Find this resource:

Karimi, S. (Ed.). (2003). Word order and scrambling. Malden, MA: Blackwell Publishing.Find this resource:

Kathol, A. (2000). Linear syntax. Oxford University Press, USA.Find this resource:

Kayne, R. (1983). Connectedness. Linguistic Inquiry, 14, 223–249. Reprinted in Kayne, 1984.Find this resource:

Kluender, R. (1998). On the distinction between strong and weak islands: A processing perspective. In P. W. Culicover & L. McNally (Eds.), Syntax and semantics 29: The limits of syntax (pp. 241–279). Academic Press.Find this resource:

Kluender, R., & Gieselman, S. (2013). What’s negative about negative islands? A re-evulation of extraction from weak island contexts. In J. Sprouse & N. Hornstein (Eds.), Experimental syntax and island effects (pp. 186–207). Cambridge University Press.Find this resource:

Koopman, H. (1984). The syntax of verbs: From verb movement rules in the Kru languages to Universal Grammar. Studies in Generative Grammar; 15. Dordrecht: Foris.Find this resource:

Koster, J. (1986). Domains and dynasties: The radical autonomy of syntax. Studies in Generative Grammar 30. Dordrecht: Foris.Find this resource:

Kuthy, K. de, & Meurers, W. D. (2001). On partial constituent fronting in German. The Journal of Comparative Germanic Linguistics, 3(3), 143–205.Find this resource:

Lasnik, H. (1999). Chains of arguments. In S. D. Epstein & N. Hornstein (Eds.), Working minimalism (pp. 189–215). Cambridge, MA: MIT Press.Find this resource:

Lasnik, H., & Saito, M. (1992). Move α: Conditions on its application and output. Cambridge, MA: MIT Press.Find this resource:

Lebeaux, D. (2009). Where does the Binding Theory apply? Linguistic Inquiry Monograph 50. Cambridge, MA: MIT Press.Find this resource:

Lechner, W. (2005). Interpretive effects of head movement. Available at http://ling.auf.net/lingbuzz/000178.

Lenerz, J. (1977). Zur Abfolge nominaler Satzglieder im Deutschen. Tübingen: Narr.Find this resource:

Mahajan, A. (1990). The A/A-bar distinction and movement theory. PhD diss. MIT.Find this resource:

Manzini, M. R. (1992). Locality: A theory and some of its empirical consequences. Linguistic Inquiry Monograph 19. Cambridge, MA: MIT Press.Find this resource:

Matushansky, O. (2006). Head-movement in Linguistic Theory. Linguistic Inquiry, 37(1), 69–109.Find this resource:

May, R. (1977). Logical Form and conditions on rules. In J. Kegl, D. Nash, & A. Zaanen (Eds.), Papers from the regional meeting of the North Eastern Linguistic Society (Vol. 8, pp. 189–207).Find this resource:

May, R. (1979). Must COMP-to-COMP movement be stipulated? Linguistic Inquiry, 10(4), 719–725.Find this resource:

May, R. (1985). Logical Form: Its structure and derivation. Cambridge, MA: MIT Press.Find this resource:

McCloskey, J. (1979). Transformational syntax and model theoretic semantics: A case study in Modern Irish. Synthese Language Library. Dordrecht, The Netherlands: D. Reidel Publishing Company.Find this resource:

McCloskey, J. (1990). Resumptive pronouns, A′-binding, and levels of representation in Irish. In R. Hendrick (Ed.), The syntax of the modern Celtic languages (pp. 199–248). Syntax and Semantics, 23. San Diego, CA: Academic Press.Find this resource:

McCloskey, J. (2002). Resumption, successive cyclicity, and the locality of operations. In S. D. Epstein & T. D. Seely (Eds.), Derivation and explanation in the Minimalist Program (pp. 184–236). Malden, MA: Blackwell Publishing.Find this resource:

Merchant, J. (2006). Sluicing. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax (Vol. IV, chapter 60, pp. 271–291). Malden, MA: Blackwell Publishers.Find this resource:

Müller, G. (1998). Incomplete category fronting: A derivational approach to remnant movement in German. Dordrecht, The Netherlands: Kluwer Academic Publishers.Find this resource:

Müller, G. (2010). On deriving CED effects from the PIC. Linguistic Inquiry, 41(1), 35–82.Find this resource:

Müller, G. (2011). Constraints on displacement: A phase-based approach. Vol. 7. Language Faculty and Beyond. John Benjamins.Find this resource:

Muysken, P. (1980). Papiamentu prepositions stranding and the typology of empty categories. Talk presented at the LSA Summer Meeting. Alberquerque.Find this resource:

Muysken, P. (1982). Parametrizing the notion Head. Journal of Linguistic Research, 2(3), 57–76.Find this resource:

Neeleman, A. (1994). Scrambling as a D-structure phenomenon. In N. Corver & Henk van Riemsdijk (Eds.), Studies on scrambling (pp. 387–429). Studies in Generative Grammar 41. Berlin: De Gruyter.Find this resource:

Neeleman, A., & van de Koot, H. (2002). The configurational matrix. Linguistic Inquiry, 33(4), 529–574.Find this resource:

Noonan, M. (1997). Functional architecture and wh-movement: Irish as a case in point. Canadian Journal of Linguistics, 42, 111–139.Find this resource:

Pesetsky, D. (1982). Paths and categories. PhD diss. MIT.Find this resource:

Pesetsky, D. (1987). Wh-in situ: Movement and unselective binding. In E. J. Reuland & A. G. B. ter Meulen (Eds.), The representation of (in)definiteness (pp. 98–129). Cambridge, MA: MIT Press.Find this resource:

Polinsky, M., & Preminger, O. (2014). Case and grammatical relations. In A. Carnie, Y. Sato, & D. Siddiqi (Eds.), The Routledge handbook of syntax (pp. 150–166). Routledge Handbooks. Abingdon, U.K.: Routledge.Find this resource:

Pollard, C. (1984). Generalized Phrase Structure Grammar. PhD thesis. Stanford University.Find this resource:

Pollard, C., & Sag, I. A. (1992). Anaphors in English and the scope of Binding Theory. Linguistic Inquiry, 23, 261–303.Find this resource:

Pollard, C., & Sag, I. A. (1994). Head-Driven Phrase Structure Grammar. Volume 1: Fundamentals. CSLI Lecture Notes Series No. 13. Stanford, CA: Center for the Study of Language and Information.Find this resource:

Postal, P. (1971). Cross-over phenomena. New York: Holt, Rinehart, and Winston.Find this resource:

Pritchett, B. (1991). Subjacency in a principle-based parser. In Principle-based parsing: Computation and psycholinguistics (pp. 301–345). Dordrecht, The Netherlands: Kluwer.Find this resource:

Reape, M. (1994). Domain union and word order variation in German. In R. C. Berwick, S. P. Abney, & C. Tenny (Eds.), German in head-driven phrase structure grammar (pp. 151–197). Stanford, CA: CSLI.Find this resource:

Richards, N. W. III (2014). A-bar movement. In A. Carnie, Y. Sato, & D. Siddiqi (Eds.), The Routledge handbook of syntax (pp. 167–191). Routledge Handbooks. Abingdon, U.K.: Routledge.Find this resource:

Rightward movement in a comparative perspective (2013). Vol. 200. Linguistik Aktuell. Amsterdam: John Benjamins.Find this resource:

Rivero, M.-L. (1991). Long head movement and negation: Serbo-Croatian vs. Slovak and Czech. The Linguistic Review, 8, 319–351.Find this resource:

Rizzi, L. (1990). Relativized minimality. Linguistic Inquiry Monograph 16. Cambridge, MA: MIT Press.Find this resource:

Rizzi, L. (2001). Relativized minimality effects. In M. Baltin & C. Collins (Eds.), The handbook of syntactic theory (pp. 89–110). Oxford: Blackwell.Find this resource:

Rizzi, L. (2006). On the form of chains: Criterial positions and ECP Effects. In L. L. S. Cheng Eds.), Wh-movement: Moving on (pp. 97–134). Cambridge, MA: MIT Press.Find this resource:

Roberts, I. (1997). Comparative syntax. London: Arnold.Find this resource:

Roberts, I. (2001). Head movement. In M. Baltin & C. Collins (Eds.), The handbook of syntactic theory (chapter 5, pp. 113–147). Oxford: Blackwell.Find this resource:

Ross, J. R. (1967). Constraints on variables in syntax. PhD diss. MIT.Find this resource:

Sag, I. A., Hofmeister, P., & Snider, N. (2007). Processing complexity in subjacency violations: The complex noun phrase constraint. Manuscript, Stanford University.Find this resource:

Sells, P. (1984). Syntax and semantics of resumptive pronouns. PhD diss. University of Massachusetts.Find this resource:

Sharvit, Y. (1999). Resumptive pronouns in relative clauses. Natural Language and Linguistic Theory, 17(3), 587–612.Find this resource:

Sportiche, D. (2006). Reconstruction, binding, and scope. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax (Vol. 4, chapter 53, pp. 35–93). Malden, MA: Blackwell Publishers.Find this resource:

Sprouse, J., & Hornstein, N. (Eds.), (2013). Experimental syntax and island effects. Cambridge University Press.Find this resource:

Starke, M. (2001). Move dissolves into Merge: A theory of locality. PhD diss. University of Geneva.Find this resource:

Steedman, M. (1996). Surface structure and interpretation. Cambridge, MA: MIT Press.Find this resource:

Stepanov, A. (2007). The end of CED? Minimalism and extraction domains. Syntax, 10(1), 80–126.Find this resource:

Syrett, K., & Lidz, J. (2009). QR in child grammar: Evidence from antecedent-contained deletion. Language Acquisition, 16(2), 67–81.Find this resource:

Syrett, K., & Lidz, J. (2011). Competence, performance, and the locality of quantifier raising: Evidence from 4-year-old children. Linguistic Inquiry, 42(2), 305–337.Find this resource:

Szabolcsi, A. (2001). The syntax of scope. In M. Baltin & C. Collins (Eds.), The handbook of syntactic theory (chapter 19, pp. 607–633). Oxford: Blackwell.Find this resource:

Syrett, K., & Lidz, J. (2006). Strong vs. weak islands. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax (Vol. 4, chapter 64, pp. 479–531). Malden, MA: Blackwell Publishers.Find this resource:

Szabolcsi, A., & Zwarts, F. (1992). Weak islands and an algebraic semantics for scope taking. Natural Language Semantics, 1(3), 235–284.Find this resource:

Takano, Y. (2000). Illicit remnant movement: An argument for feature-driven movement. Linguistic Inquiry, 31(1), 141–156.Find this resource:

Tanaka, M. (2015). Scoping out of adjuncts: Evidence for the parallelism between QR and wh-movement. PhD thesis. UCL.Find this resource:

Travis, L. (1984). Parameters and effects of word order variation. PhD diss. MIT.Find this resource:

Truswell, R. (2014). Binding theory. In A. Carnie, Y. Sato, & D. Siddiqi (Eds.), The Routledge handbook of syntax (pp. 214–238). Routledge Handbooks. Abingdon, U.K.: Routledge.Find this resource:

Uriagereka, J. (1999). “Multiple Spell-Out”. In N. Hornstein & S. D. Epstein (Eds.), Working minimalism (pp. 251–282). Cambridge, MA: MIT Press.Find this resource:

Van Valin, R. D., & LaPolla, R. J. (1997). Syntax: Structure, meaning, and function. Cambridge, U.K.: Cambridge University Press.Find this resource:

Webelhuth, G. (1992). Principles and parameters of syntactic saturation. New York: Oxford University Press.Find this resource:

Wexler, K., & Culicover, P. W. (1980). Formal principles of language acquisition. Cambridge, MA: MIT Press.Find this resource:

Williams, E. S. (2002). Representation Theory. Cambridge, MA: MIT Press.Find this resource:

Notes:

(1.) Ellipsis is but one constituency diagnostic that can be brought to bear on this question. Displacement of constituents containing a gap but not the corresponding filler, so-called remnant movement, provides converging evidence. The interested reader is referred to Müller (1998).

(2.) It should be noted that the argument is not entirely uncontroversial, since the involvement of unpronounced syntactic structure in ellipsis has sometimes been disputed, as has the claim that ellipsis targets complete constituents. See Abe (2015); Abels to appear; Culicover and Jackendoff (2005) for relevant discussion.

(3.) There is an interesting theoretical tension created by the observation that the interpretive properties of the possessive pronoun in (17) are computed in terms of the low position of the displaced phrase while the interpretive properties of the referential expression in (22b) are computed in terms of the high position of the displaced phrase. The literature on reconstruction effects contains a lively debate about a deeper theoretical understanding of the relevant phenomena (see Barss, 1986; Chomsky, 1993; Fox, 1999, 2000; Lebeaux, 2009; Postal, 1971; and for treatment of the issues in handbooks and textbooks see Barss, 2001; Bresnan, 2001, part IV; Büring, 2005, chapter 12; Haegeman, 1994, chapter 9; Pollard & Sag, 1994, chapter 6). This literature also shows that the conclusion drawn in the text is overly simplistic: A more careful formulation would have said that the displaced constituent behaves as though it were higher in the structure than the gap while not necessarily as high as the filler.

(4.) Note that ‘the book’ in (37b) should not be treated as the filler, since, in languages where there is visible case morphology, the relative pronoun bears the case associated with the gap while the head of the relative clause bears the case associated with it in the superordinate clause. This is illustrated with the German example (i), where ‘den’ (‘whichACC’) is the relative pronoun showing the accusative case associated with the gap.

((i))

Displacement in Syntax

(5.) For notation, Postal wrote non-A as A¯, borrowing set theoretic complement notation. Because of the bar indicating the set theoretic complement, this class of operations is also called A-bar operations. It is sometimes written as A′.

(6.) β separates α from α’s gap iff α c-commands β and β c-commands α’s gap.

(7.) See Beerman, LeBlanc, and Riemsdijk (1997), and Rightward movement in a comparative perspective (2013) for further examples of gap–filler structures and a sampling of syntactic analyses.

(8.) All OV languages appear to have scrambling across adverbials (Haider, 2005) while inversion of this type of objects and subjects with respect to each other is more restricted and is virtually absent in Dutch, for example.