Proceedings of the 8th International Workshop on Tree Adjoining Grammar and Related Formalisms, pages 65–72,
Sydney, July 2006. c©2006 Association for Computational Linguistics
Binding of Anaphors in LTAG
Neville Ryant
Department of Linguistics
619 Williams Hall
University of Pennsylvania
Philadelphia, PA, 19104-6305
nryant@ling.upenn.edu
Tatjana Scheffler
Department of Linguistics
619 Williams Hall
University of Pennsylvania
Philadelphia, PA, 19104-6305
tatjana@ling.upenn.edu
Abstract
This paper presents an LTAG account
for binding of reflexives and recipro-
cals in English. For these anaphors,
a multi-component lexical entry is pro-
posed, whose first component is a degener-
ate NP-tree that adjoins into the anaphor’s
binder. This establishes the local structural
relationship needed to ensure coreference
and agreement. The analysis also allows
a parallel treatment of reflexives and re-
ciprocals, which is desirable because their
behavior is very similar.
In order to account for non-local bind-
ing phenomena, as in raising and ECM
cases, we employ flexible composition,
constrained by a subject intervention con-
straint between the two components of the
anaphor’s lexical entry. Finally, the paper
discusses further data such as extraction
and picture-NP examples.
1 Introduction
Binding Theory (B¨uring, 2005; Reuland and Ev-
eraert, 2001) is an issue at the interface of syntax
and semantics which has previously been avoided
in the LTAG literature. While LTAGs were ini-
tially concerned only with the syntax of natural
languages, recent accounts of semantic computa-
tion in the LTAG framework (Kallmeyer and Joshi,
2003; Kallmeyer and Romero, 2004) allow us now
to tackle interface phenomena. An appropriate
formulation of Binding Theory (BT) is needed to
explain the pattern exhibited in (1–3).
(1) Johni likes himselfi.
(2) * Johni likes herselfi.
(3) * Himselfi likes himselfi / Johni.
Due to the incredible complexity of the data in
question, we will focus here on English reflex-
ives (himself, herself) and reciprocals (each other),
typically subsumed under Condition A (Chomsky,
1981).
This paper proposes a new two-component lex-
ical entry for reflexive pronouns that takes care of
the syntactic and semantic dependencies involved
in binding (agreement and coreference). In this ap-
proach, different binding options (e.g., in a ditran-
sitive sentence) follow from different derivations.
In section 3, we show how our analysis ex-
tends straightforwardly to reciprocals. Section 4
presents the extension of our account to anaphors
with nonlocal antecedents, such as the experi-
encers of raising verbs, and ECM subjects. Fur-
ther issues, including extraction, are discussed in
section 5. Section 6 concludes.
2 Basic Anaphor Binding
In traditional accounts, binding is defined rep-
resentationally: an antecedent binds an anaphor
iff they are are coindexed and in a certain struc-
tural relationship. In an LTAG, binding cannot be
viewed in this way as the notion of coindexation is
foreign to the formalism. An LTAG analysis can
therefore not be a mere translation of a previous
account.
Although the phenomenon is very complex, the
basic properties of binding are quite well under-
stood. Binding of an anaphor by an antecedent
consists of coreference and agreement between the
two items. Furthermore, it is well known that
binding of English anaphors is an asymmetrical,
local, structural relationship. The asymmetry of
binding can be easily observed in examples (1)
65
versus (3). Locality is reflected by the fact that
(1) is grammatical, but not (4).
(4) * Johni knows that Mary likes himselfi.
Finally, the binding relationship is known to be
structural because the positions of binder and
anaphor play a crucial role. This is discussed in
more detail below.
2.1 Lexical Entry
The domain of locality that LTAG provides en-
ables us to encode a local structural relationship,
such as the one between the anaphor and its an-
tecedent, very directly. We understand binding as
a lexical requirement of the anaphor: that it must
be bound. Thus, we propose the lexical entry in
Figure 1 for reflexives. It is a multicomponent set
whose second component is the anaphor. The first
component is a degenerate auxiliary tree which ad-
joins into the elementary tree of the antecedent.
In LTAG, elementary trees encode both syn-
tax and semantics. Thus, the two components of
binding, coreference and agreement, are simulta-
neously guaranteed by the coindexations between
the feature structures of binder and anaphor. Fur-
thermore, since the derivation must be tree-local,
locality is also ensured. A c-command constraint
between the two components accounts for the
asymmetry between the antecedent and anaphor as
shown in examples (1) and (3). This constraint is
checked when the two components are composed
into an elementary tree (by tree-locality).
2.2 Example Derivation
Consider (5), where himself has two possible an-
tecedents, John and Bill. Our analysis derives both
readings, given a standard tree inventory as in Fig-
ure 2.
(5) Johni showed Billj himselfi/j.
Sentence (5) is syntactically ambiguous under
this analysis, since two different derivations lead
to distinct readings. This seems to reflect our in-
tuitions about this sentence well, although it con-
trasts with the traditional vew of BT, where the
coindexation between binder and anaphor is part
of the syntactic structure for the sentence, and thus
no ambiguity arises.
2.3 Flexible Composition
Tree-locality requires the first component of him-
self to adjoin into a higher NP substitution node.
ts: Sa80
a80a80a80a16a16a16a16
NP↓ VPa97
a97a97a33a33a33
V
showed
VP
a81a81a17a17
NP↓ VP
a64a64a0a0
ε VP
NP↓
tj: NP
John
tb: NP
Bill
Figure 2: Tree inventory.
However, adjunction into substitution nodes is
generally disallowed. Adjunction of the first com-
ponent of himself into the root node of the John-
tree tj or the Bill-tree tb is, however, not tree-local.
Therefore, we employ flexible composition (Joshi
et al., 2003) to compose tj with the first compo-
nent of th (t1h), yielding a derived multicompo-
nent set. Composition of th with ts is then tree-
local. This yields the reading where John is the
antecedent of himself.
Alternatively, tb composes with th first, which
derives the other reading. The two derivation trees
representing these readings are shown in Figure 3.
ts
tb
<ts,221>
th
<<ts,1>,<ts,2222>>
tj
<t1h,0>
ts
tj
<ts,1>
th
<<ts,221>,<ts,2222>>
tb
<t1h,0>
Figure 3: Derivation trees for “John showed Bill
himself.”
66
th:











NP* syn:hAGR 1i
sem:2
4NP
»
T
h
I 2
i–3
5
,
NP
himself
syn:2
66
4AGR 1
2
64PERS 3NUM sg
GEN masc
3
75
3
77
5
sem:2
4NP
»
T
h
I 2
i–3
5











c-command
Figure 1: Lexical entry for himself.
2.4 Advantages
The different binding options (e.g., in double-
object sentences) follow directly from the deriva-
tion and do not have to be hardcoded. Further-
more, the reflexive itself is responsible for agree-
ment and coreference with its antecedent.
2.5 Alternative Analysis
There is at least one obvious alternative analysis
for BT in LTAG. In this case, features are em-
ployed instead of a multicomponent set to derive
the binding relationship. Features on each verbal
elementary tree would encode whether an argu-
ment is an anaphor, and if so, what it is bound
to. Just like in our analysis introduced above, a
certain locality necessary for binding can be en-
sured under this approach. However, this approach
is very stipulative. It is merely an accident that
agreement and coreference go hand in hand: Two
separate feature equations have to ensure agree-
ment between the binder and anaphor, and coref-
erence between them. Furthermore, a number of
verbal trees is added; and the reflexive itself be-
comes syntactically and semanticially vacuous.
3 Reciprocals
Another advantage of the proposed account is that
it allows an analogous treatment of reciprocals like
each other in (6).
(6) [John and Mary]i like each otheri.
This is desirable given that the syntactic behavior
of reciprocals resembles reflexives. Semantically,
though, reciprocals are very complex (Dimitriadis,
2000). The meaning of “each other” roughly cor-
responds to its parts, “each”, and “other”. That
is, “John and Mary love each other” means some-
thing like “Of John and Mary, each loves the
other”.1
These properties are neatly accounted for with
our analysis of each other that is syntactically
analogous to himself, but contributes additional
operators in the semantics2. The proposed lexical
entry is spelled out in Figure 4.
The fact that each other contributes two dis-
tinct quantifiers corresponds directly to its syntac-
tic analysis as a two-part multicomponent set.
4 Nonlocal Antecedents
The discussion of anaphoric pronoun binding dis-
cussed in the previous section demonstrated how
certain locality (7) and configurational restrictions
(8) on anaphoric pronouns follow from TAG’s
constrained mechanisms of structural composition
coupled with a multicomponent analysis of reflex-
1It is sometimes claimed that “long-distance” reciprocals
require non-local adjunction of “each”:
(i) The boxers thought they would defeat each other.
check each # each
The LTAG analysis proposed here does not allow this. This
may constitute independent evidence for Dimitriadis’ (2000)
analysis of reciprocals in which “each” is not as high as it
seems in these kinds of examples.
2The exact semantics of each other is a matter of ongoing
discussion. We assume for simplicity that each other corre-
sponds to each+the other, as reflected in the lexical entry.
3subsetsqequalA= “is an atomic part of”.
In the absence of a complete analysis of plural semantics
in LTAG, we assume here that plural noun phrases like “John
and Mary” or “the boys” contribute at least a group (G) vari-
able. This variable is used by certain collective predicates, for
example in “The boys surrounded the castle.” It corresponds
to the plural individual contributed by the NP.
The semantics given here predicts strongly distributive
“each other”. Some adjustment is needed to account for lesser
forms of distributivity.
67
tea:











NP* syn:hAGR 1i
sem:2
66
66
64NP
2
66
66
4
T
"
I x
G 2
#
B
h
G 2
i
3
77
77
5
3
77
77
75
,
NP
each other
syn:h
AGR 1
i
sem:2
64NP
2
4T
"
I y
P 3
#3
5
3
75
l1 : ∀(x,x subsetsqequalA 2,l2) l2 : ∀(y,y subsetsqequalA 2 ∧ y negationslash= x,h2),h
2 ≥ 3











c-command
Figure 4: Lexical entry for each other.3
ives and reciprocals.
(7) a. Johni likes himselfi.
b. *Johni thinks that Mary believes that
Kate likes himselfi.
(8) a. John believes Maryi to like herselfi.
b. *John believes herselfi to like Maryi.
A significant problem with this analysis as
stands, however, is that it works too well, denying
the grammaticality of certain raising (9) and ECM
constructions (10) and constructions in which the
anaphor is embedded within a subject (11). Un-
der current assumptions, the antecedent-anaphor
dependency must be established within an ele-
mentary tree (by adjunction of a single multi-
component set). However, for both of these con-
structions the anaphor and its antecedent lie in dif-
ferent elementary trees. In (9) the auxiliary tree
of the raising verb seems contains no local argu-
ment for the degenerate NP* component to com-
bine with. In (10) himself occurs as an argument
of like while its antecedent occurs in another ele-
mentary tree, believe. In each case, generating the
examples requires that we relax some of our cur-
rent assumptions.
(9) Johni seems to himselfi to be a decent guy.
(10) Johni believes himselfi to be a decent guy.
(11) Johni thought that the pictures of himselfi
were wonderful.
4.1 Raising
We see from (9) that anaphors can occur as ex-
periencer arguments of raising verbs providing
they are c-commanded by a possible antecedent.
Though predicted to be ungrammatical under the
current proposal, (9) can be generated if we relax
the requirement that the two parts of the multicom-
ponent set of the anaphor attach to the same ele-
mentary tree. This relaxation could take the form
of simply allowing non-local adjunction for spe-
cific classes of multicomponent sets, those with
a degenerate components. Alternately, we retain
the restriction to tree-local MC-TAG but achieve
nonlocality through more extensive use of flexible
composition, already adopted for independent rea-
sons.
Under a flexible composition analysis (Figure
6), the John-tree composes with the degenerate
NP* member of the reflexive set as before. This
yields a derived multicomponent set consisting of
one derived part, John, and one underived part,
himself. The seems-tree then composes with the
himself component of the reflexive set, yielding a
derived set (Figure 5) containing the components
John and seems to himself. Finally, this derived
multicomponent set combines with the like-tree,
the John component substituting into the open NP
slot and the seems to himself component adjoining
at VP.
4.2 ECM
In ECM constructions such as (10) the anaphor ap-
pears as the subject of the embedded to be a decent
guy-tree while its antecedent appears as subject of
the matrix believes-tree. A derivation for this sen-
tence under our account is shown in Figure 7. As
before, the John-tree first composes with the de-
generate NP* component of the reflexive tree, fol-
lowed by the the substitution of the himself-tree
68
Sa97a97
a97a33a33a33
NP↓ VPa81
a81a17a17
believes S*
Sa97
a97a97a33a33a33
NP↓ VPa80a80
a80a80a16a16a16a16
to be a decent guy



NP* NP
himself



NP
John
Derivation tree:
tbelieves
th
tj
<t1h,0>
tdg
<t2h,0>
Figure 7: Derivation of “John believes himself to be a decent guy.”









NP
John
,
VPa80
a80a80a16a16a16
seems VPa97
a97a33a33
PP
a81a81a17a17
to NP
himself
VP*









Figure 5: Derived multicomponent set for (9).
into the to be a decent guy-tree, yielding the de-
rived multicomponent set containing John and be-
lieves himself, which locally composes with the to
be a decent guy-tree.
4.3 Subject Embedding
Anaphors contained within embedded subjects4
(12) cause the binding domain to be minimally ex-
panded. Again, it is transparent that these cases
can be derived successfully from the lexical entry
in Figure 1 and repeated use of flexible composi-
tion.
(12) a. The meni knew that pictures of each
otheri were on sale.
b. The meni felt that the pictures of
themselvesi were horrid.
c. The meni knew that each otheri’s
pictures were on sale.
4The absence of nonlocal binding of reflexive subjects
(e.g. John thinks that himself is grand.) is assumed to de-
rive from an inability of reflexives to take nominative case.
tdg: Sa80
a80a80a16a16a16
NP↓ VPa96a96
a96a96a32a32a32a32
to be a decent guy
tseems: VPa80
a80a80a16a16a16
seems VPa72
a72a8a8
PP
to NP↓
VP*
Derivation tree: tdg
th
<<tdg,1>,<tdg,2>>
tj
<t1h,0>
tseems
<t2h,0>
Figure 6: Derivation of “John seems to himself to
be a decent guy.”
69
4.4 Constraints on Flexible Composition
The use of flexible composition with tree-local
MC-TAG is very powerful, thus able to account
for the non-local binding in (9), (10), and (12).
However, it is too powerful if unconstrained as it
will also generate (13). It is necessary to constrain
the derivation such that in the derived tree no sub-
ject node intervenes between the antecedent and
anaphor (Chomsky’s Subject Intervention Con-
straint). This is obtained by strengthening the
link between NP and himself in the lexical en-
try s.t. when the two trees connected by the link
are adjoined, a requirement that NP* c-command
himself and no subject NP intervenes between the
two (c-commanding himself and c-commanded by
NP* ) is checked. This constraint formalizes the
descriptive account given in the linguistic litera-
ture. Note that a similar account may be active
in other places in the grammar as well, due to the
pervasiveness of left-edge phenomena (see section
5.4).
Computationally, this constraint can be checked
as soon as the multicomponent set which con-
tains it attaches into another elementary tree. C-
command as well as subject intervention cannot
be disturbed by later operations on the outer tree,
if they are valid at the time of composition.
(13) * Johni believes me to like himselfi.
5 Further Issues
5.1 Exempt Pronouns
As it currently stands, the proposal follows heav-
ily in the footsteps of traditional configurational
approaches to BT. As such, it mirrors the more tra-
ditional BT of Chomsky in it’s inability to license
such examples as (17b), where the antecedent does
not c-command the anaphor and (14) and (15),
where binding is possible despite presence of an
intervening subject along the c-command path.
(14) a. I spoke to [John and Bill]i about each
otheri.
b. Pictures of myselfi frighten mei.
c. Johni’s greatest problem is a fear of
himselfi.
(15) [John and Mary]i are letting the honey drip
on each otheri’s feet.
(16) Clones of each other annoy the children.
The examples in (14) can be accommodated by
having the prepositions appearing before the ar-
guments be surface case markers rather than real
prepositions (as suggested in (Jackendoff, 1972)).
Even so, (15) and (16) remain and seem to present
an intractable problem for an LTAG account, as
well as traditional accounts of English binding
phenomena. This may in fact be the case and
prove firm support for claims by numerous authors
(Pollard and Sag, 1994; Postal, 1971; Kuroda,
1965) that at least part of the data subsumed un-
der BT (the “exempt pronouns”) is governed by
pragmatic constraints such as point-of-view rather
than purely structural constraints. In fact, the
LTAG analysis proposed here is a relatively clean
structural account of English binding data. The
(un)availability of a derivation for certain exam-
ples may thus point to their classification into “ex-
empt” and regular anaphora. These considerations
are left for further work.
5.2 Extraction
A potential problem for the proposed analysis is
presented by extraction phenomena, as in wh-
movement or topicalization. Extraction of a
phrase containing an anaphor, whether topicaliza-
tion or (17) or wh-movement (18), does not induce
a Condition A violation. The current proposal
predicts the grammaticality of (17a) and (18a)
given that in each case the reflexive is locally c-
commanded by its antecedent. However, in (17b)
and (18b) the reflexive fails to be c-commanded by
its antecedent, hence these examples are predicted
to be ungrammatical although they are clearly ac-
ceptable.
(17) a. Johni saw himselfi.
b. Himselfi John saw ti.
(18) a. Johni liked the color pictures of
himselfi.
b. [Which pictures of himselfi] did Johni
like ε?
A classical solution to these facts involves re-
construction of the Aprime-moved element to its origi-
nal site for the purposes of binding. Clearly, syn-
tactic reconstruction is untenable in LTAG. How-
ever, it is possible to emulate it through an en-
try akin to that in Figure 8, which is capable
of deriving the topicalization examples in (17).
The first component is the extracted reflexive
70
th2:







NP
himself
syn:"
AGR 1
EXTRACTED +
#
sem:2
4NP
»
T
h
I 2
i–3
5
,
NP* syn:hAGR 1i
sem:2
4NP
»
T
h
I 2
i–3
5
,
NP*







c-command
Figure 8: Lexical entry for extracted reflexive himself.
(Aprime-moved constituents are marked by extraction-
features (XTAG Group, 2001)), the second com-
ponent is the binder, and the third component is
the position that the reflexive has been extracted
from. The requirement that the antecedent locally
c-command the trace of movement has the effect
of emulating reconstruction.
Note, furthermore, that even if some manner of
reconstruction operation were to be implemented
in LTAG, we are faced with the reality of cases
such as (19), which demonstrate that extraction
of an element can alter its set of possible binders.
GB accounts (van Riemsdijk and Williams, 1986;
Clark, 1982) have explained the opposition in (19)
by allowing partial reconstruction to an interme-
diate trace from which the matrix subject is an ac-
cessible binder of the anaphor. The LTAG analysis
of wh-movement, though, neither exploits inter-
mediate traces nor allows transformational move-
ment over domains larger than an elementary tree,
meaning that such intermediate traces are simply
unavailable to us.
(19) a. *Marshai thought that I painted a
picture of herselfi.
b. [Which pictures of herselfi] did
Marshai think that I painted ε?
Instead, we suggest that Spec,IP subjects of
clauses are able to bind into Spec,CP of the same
clause as proposed by Reinhart (1991) and Frank
and Kroch (1995). Rather than being a disadvan-
tage, though, this seems to be a strength, predict-
ing as it does that (20) is bad where reconstruction
to a posited intermediate trace would predict ac-
ceptability.
(20) *[Which pictures of himselfi] did Mary
think that Johni believed that Sally wanted?
Future work should attempt to determine the
correct form of this lexical entry as well as
whether or not it is possible to collapse it with the
previously proposed Figure 8.
5.3 Conditions B,C
It is often assumed that the analyses for anaphors
and regular pronouns should be related, because of
a certain complementarity in distribution: While
anaphors must be locally bound, pronouns must be
locally free. In English, however, this complemen-
tarity is not absolute (cf. 21–22). Furthermore, a
negative locality constraint seems to be discour-
aged by the LTAG framework. This suggests that
the analysis of pronouns is independent of our ac-
count of anaphors. We leave pronouns, as well as
r-expressions (Mary, the man) for further work.
(21) Johni pulled the blanket over himi /
himselfi.
(22) a. Theyi saw each otheri’s friends.
b. Theyi saw theiri friends.
5.4 Importance of the Left Edge
Examination of language exposes the left edge to
be special with regards to certain phenomena. In
Binding Theory, this is revealed in the guise of a
Subject Intervention Constraint. Case assignment
represents a similar case. We see that verbs can as-
sign accusative case to objects, and subjects of the
next lowest clause (ECM), but no further. Ideally,
a new analysis of left-edge effects would clarify
the relationship between the two components of
the lexical entry proposed above.
5.5 Inherent Reflexives
English has a small number of inherently reflexive
verbs, such as behave:
(23) John behaves himself.5
Note that this verb requires its object to be a
reflexive pronoun which is coreferent with its sub-
ject:
5We would like to thank one anonymous reviewer for
bringing this example to our attention.
71
(24) * John behaves her.
We conclude from this that behave has a specific
lexical constraint, namely that its object should be
[+ reflexive]. Since there can be no other binder
for this reflexive pronoun, it must be bound by the
subject of the sentence.
6 Conclusion
In conclusion, we have presented an account of
the syntax and semantics of anaphoric expres-
sions that covers basic binding as well as raising,
ECM, and extraction data. Our analysis employs a
multicomponent lexical entry whose first compo-
nent corresponds to the anaphor’s binder, thus es-
tablishing a local relationship between antecedent
and anaphor. A structural constraint that links the
two components accounts for the basic asymmetry
that is observed in the binding of reflexives and re-
ciprocals in English.
7 Acknowledgements
We would like to thank the members of the XTAG
group, as well as the CIS 630 class of Spring 2006
at Penn for comments and discussion.

References
B¨uring, Daniel. 2005. Binding Theory. Cambridge:
Cambridge University Press.
Chomsky, Noam. 1981. Lectures on Government and
Binding. Dordrecht: Foris.
Clark, Robin. 1982. Scope assignment and modifica-
tion. Linguistic Inquiry 23: 1-28.
Copestake, Ann, Dan Flickinger, Ivan A. Sag, and Carl
Pollard. 1999. Minimal Recursion Semantics: An
introduction. Manuscript, Stanford University
Dalrymple, M., M. Kanazawa, Y. Kim, S.A. Mehombo,
and S. Peters. 1998. Reciprocal expressions and the
concept of reciprocity. Linguistics and Philosophy
21:159–210
Dimitriadis, Alexis. 2000. Syntactic locality and
tree adjoining grammar: Grammatical, acquisition,
and processing perspectives. Doctoral Dissertation,
University of Pennsylvania.
Frank, Robert. 1992. Beyond Identity: Topics in
Pronominal and Reciprocal Anaphora. Doctoral
Dissertation, University of Pennsylvania.
Frank, Robert and Anthony Kroch. 1995. Generalized
transformations and the theory of grammar. Studia
Linguistica 49(2): 103-151.
Jackendoff, Ray. 1972. Semantic Interpretation in
Generative Grammar. Cambridge, MA: MIT Press
Joshi, Aravind K., Laura Kallmeyer, and Maribel
Romero. 2003. Flexible Composition in LTAG:
Quantifier Scope and Inverse Linking. In Proceed-
ings of the International Workshop on Composi-
tional Semantics. Tilburg, The Netherlands
Joshi, Aravind K. and K. Vijay-Shanker. 1999.
Compositional Semantics with Lexicalized Tree-
Adjoining Grammar (LTAG): How Much Under-
specification is Necessary? In H.C. Blunt and
E.G.C. Thijsse, editors. Proceedings of the Third In-
ternational Workshop on Computational Semantics
(IWCS-3), pp. 131-145. Tilburg, The Netherlands
Kallmeyer, Laura and Aravind K. Joshi. 2003. Factor-
ing predicate argument and scope semantics: Under-
specified semantics with LTAG. Research on Lan-
guage and Computation 1:3–58
Kallmeyer, Laura and Maribel Romero. 2004. LTAG
semantics with semantic unification. In Proceedings
of TAG+7. Vancouver, Canada
Kuroda, S.Y.. 1965. Generative Gramamtical Studies
in the Japanese Language. MIT: PhD. dissertation.
Pollard, Carl and Ivan Sag. 1994. Head-driven Phrase
Structure Grammar. Chicago, Il: University of
Chicago Press.
Postal, Paul. 1971. Crossover Phenomena. New York:
Holt.
Reinhart, Tanya. 1991. Definite NP anaphora and c-
command domains. Linguistic Inquiry 12(4): 605-
635.
Reuland, Eric and Martin Everaert. 2001. Decon-
structing Binding. In M. Baltin and C. Collins, edi-
tors. The Handbook of Contemporary Syntactic The-
ory. Oxford
van Riemsdijk, Henk and Edwin Williams. 1986. In-
troduction to the Theory of Grammar. Cambridge,
MA: MIT Press
XTAG Group. 2001. A Lexicalized Tree Adjoining
Grammar for English. IRCS Technical Report, Uni-
versity of Pennsylvania
