Animacy Encoding in English: why and how
Annie Zaenen
PARC & Stanford University
3333 Coyote Hill Road
Palo Alto, CA 94304]
zaenen@parc.com
Jean Carletta
HCRC-University of Edinburgh
2, Buccleuch Place
Edinburgh EH8LW, UK
J.Carletta@Edinburgh.ac.uk
Gregory Garretson
Boston University
Program in Applied Linguistics
96 Cummington St.,
Boston, MA 02215
gregory@bu.edu
Joan Bresnan
CSLI-Stanford University
220, Panama Street
Stanford CA 94305
bresnan@stanford.edu
Andrew Koontz-Garboden
CSLI-Stanford University
220, Panama Street
Stanford CA 94305
andrewkg@csli.stanford.edu
Tatiana Nikitina
CSLI-Stanford University
220, Panama Street
Stanford CA 94305
tann@Stanford.edu
M. Catherine O’Connor
Boston University
Program in Applied Linguistics
96 Cummington St.,
Boston, MA 02215
mco@bu.edu
Tom Wasow
CSLI-Stanford University
220, Panama Street
Stanford CA 94305
wasow@csli.stanford.edu
Abstract
We report on two recent medium-scale initiatives
annotating present day English corpora for animacy
distinctions.  We discuss the relevance of animacy for
computational linguistics, specifically generation, the
annotation categories used in the two studies and the
interannotator reliability for one of the studies.
1 Introduction
It has long been known that animacy is an
important category in syntactic and morphological
natural language analysis.  It is less evident that
this dimension should play an important role in
practical natural language processing.  After
reviewing some linguistic facts, we argue that it
does play a role in natural language generation and
translation, describe a schema for the annotation of
animacy distinctions, evaluate the reliability of the
scheme and discuss some results obtained.  We
conclude with some remarks on the importance of
animacy and other accessibility dimensions for the
architecture of generation schemes
2 The animacy dimension in natural
language
The animacy hierarchy is one of the accessibility
scales that are hypothesized to influence the
grammatical prominence that is given to the
realization of entities within a discourse. Three
important scales (sometimes conflated into one,
also called animacy hierarchy in Silverstein, 1976),
are the definiteness, the person and the animacy
hierarchy proper.  We assume these are three
different hierarchies that refer to different aspects
of entity representation within language: the
definiteness dimension is linked to the status of the
entity at a particular point in the discourse, the
person hierarchy depends on the participants
within the discourse, and the animacy status is an
inherent characteristic of the entities referred to.
Each of these aspects, however, orders entities on a
scale that makes them more or less salient or
‘accessible’ when humans use their language.
The importance of accessibility scales is not
widely recognized in computational treatments of
natural language.  This contrasts with the situation
in linguistics where such scales have been
recognized as playing an important role in the
organization of sentence syntax and discourse.
Even in natural language studies, however, their
importance has been underestimated because the
role of these scales is not always to distinguish
between grammatical and ungrammatical
utterances but often that of distinguishing mainly
between felicitous and infelicitous ones, especially
in languages such as English.
Grammaticality and acceptability
As long as one’s attention is limited to the
distinction between grammatical and
ungrammatical sentences, the importance of the
animacy hierarchy in particular is mainly relevant
for languages with a richer morphology than
English.   In such languages animacy distinctions
can influence grammaticality of e.g. case-marking
and voice selection.  To give just one example, in
Navaho, a bi-form is used rather than an yi-form
whenever the patient is animate and the agent is
inanimate, whereas the yi-form is used when the
agent is animate and the patient is inanimate as
illustrated in (1) (from Comrie 1989 p. 193).
(1) (a) At’ééd nímasi yi-díílíd
      girl      potato burnt
      The girl burnt the potato.
(b) At’ééd nímasi bi-díílíd
      girl      potato burnt
      The potato burnt the girl.
Other phenomena discussed in the literature are
agreement limited to animate noun phrases
(Comrie, 1989), overt case-marking of subject
limited to inanimates or overt case-marking of
objects limited to animates (Aissen, 2003, see also
Bossong 1985 and 1991), object agreement in
Bantu languages (see e.g. Woolford, 1999), choice
between direct and inverse voice in Menominee
(Bloomfield, 1962, see also, Trommer, n.d.).
Recent linguistic studies have highlighted the
importance of the category in languages such as
English. For instance the choice between the Saxon
genitive and the of-genitive (Rosenbach, 2002,
2003, O’Connor et al. 2004, Leech et al. 1994),
between the double NP and the prepositional
dative (Cueni et al, work in progress) and between
active and passive (Rosenbach et al. 2002, Bock et
al. 1992, McDonald et al. 1993) and between
pronominal and full noun reference (Dahl and
Fraurud, 1996, based on Swedish data) have all
been shown to be sensitive to the difference
between animate and inanimate referents for the
arguments with variable realization.  In these cases
the difference between animate and inanimate does
not lead to a difference between a grammatical or
an ungrammatical sentence as in the cases cited in
the previous paragraph but to a difference in
acceptability.
Interaction between animacy and other scales
As mentioned above, the term ‘animacy hierarchy’
is used in two ways, one to refer to an ordering
imposed on definiteness, animacy and person and
the other where it refers to animacy proper.  The
reason for this lies in the interaction between the
different factors that determine accessibility.
It is conceptually desirable to distinguish between
animacy and definiteness but in practice it is
frequently the case that a linguistic phenomenon is
conditioned by more than one of these
conceptually independent factors (see e.g. Comrie,
1989 for discussion).  The projects in the context
of which the annotation tasks described here were
performed (see description below, section 5) also
encode some of these interacting factors.  The
LINK-project encodes information status (see
Nissim et al, 2004) and the Boston project encodes
definiteness and expression type (i.e., NP form) as
proxies for information status.
3 Animacy as a factor in generation and
translation
As long as animacy was discussed as a relevant
grammatical category in languages that had not
been studied from a computational point of view,
its importance for computational linguistics was
perceived as rather limited.  The fact that it
permeates the choice between constructions in
languages such as English changes this perception.
The category is of obvious importance for high
quality generation and translation.
For instance, if one is faced with the task of
generating a sentence from a three place predicate,
P (a,b,c), and one has the choice of rendering either
b or c as the direct object, knowing that c is human
and b is abstract would lead one to choose c ceteris
paribus.  However, everything else is rarely equal
and the challenge for generation will be to assign
the exact relative weights to factors such as
animacy, person distinction and recency.
Moreover, the importance of these factors needs to
be combined with that of heterogeneous
considerations such as the length of the resulting
expression.
In the context of translation we need also to keep
in mind the possibility that the details of the
animacy ranking might be different from language
to language and that the relative impact of animacy
and other accessibility scale factors might be
different from construction to construction.
4 The Animacy Hierarchy
Given the pervasive importance of animacy
information in human languages one might expect
it to be a well-understood linguistic category.
Nothing could be farther from the truth.  Linguistic
descriptions appeal to a hierarchy that in its
minimal form distinguishes between human, non-
human animate and inanimate but can contain
more distinctions, such as distinctions between
higher and lower animals (see Yamamoto, 1999 for
a particularly elaborated scheme).
What makes it difficult to develop clear categories
of animacy is that the linguistically relevant
distinctions between animate and non-animate and
between human and non-human are not the same
as the biological distinctions.  Part of this research
is devoted to discovering the principles that
underlie the distinctions; and the type of
distinctions proposed depend on the assumptions
that a researcher makes about the underlying
motivation for them, e.g. as a reflection of the
language user’s empathy with living beings (e.g.
Yamamoto, 1999).  What is of particular interest
for natural language processing is the observation
that the distinctions are most likely not the same
across languages (cf. Comrie, 1989) and can even
change over time in a given language. They are
similar to other scalar phenomena such as voicing
onset times that play a role in different languages
but where the categorization into voiced and
unvoiced does not correspond to the same physical
boundary in each language. But whereas voicing
onset times can be physically measured, we do not
have an objective measure of animacy. The
categories involved correspond to the degree to
which various entities are construed as human-like
by a given group of speakers and at this point we
have no language independent measure for this.
Moreover, languages make ample use of metaphor
and metonomy.  The intent of an animacy coding is
to encode the animacy status of the referent of the
linguistic expression. But sometimes in figurative
language it is not clear what the referent it.
Especially prevalent cases of metonomy are the
use of names to refer both to organizations (e.g.
IBM) and to characteristic members of them, and
the use of place names (e.g. Russia) to refer both to
organizational entities and geographical places or
inhabitants of them. Terms belonging to these
semantic classes are systematically ambiguous.
Whereas it is true that animacy can be determined
by looking at the entity an expression refers to, in
practice it is not always clear what the referent of
an expression is.
The notions that the animacy hierarchy appeals to,
then, are not a priori well defined. And work is
necessary on two levels: to better define which
distinctions play a role in English and to determine
where they play a role.  Conceptually, it might be
desirable to replace the idea of a hierarchy with
discrete classes by a partial ordering of entities.
This is, however, not the place to pursue this idea.
Fortunately, one doesn’t need to wait until the first
problem is solved completely to tackle the second.
The results obtained in certain linguistic contexts
are robust for the top and the bottom of the
hierarchy.  Uncertainty about the middle does not
prevent us from establishing the importance of the
dimension as such.  Refining the definition of
animacy will, however, be important for more
detailed studies of the interaction between the
various accessibility hierarchies. This more precise
notion will be needed for cross-linguistic studies,
and, in the context of natural language processing,
for high quality generation and translation.
5 Animacy Annotation
As we have discussed above, the animacy scale is
an important factor in the choice of certain
construction in English. But it is only a soft
constraint and as such outside of the realm of
things that native speakers have clear judgments
about.  The best ways to study such phenomena are
psychological experiments and corpus analysis.
The annotation exercise we engaged in is meant to
facilitate the latter.
Given the situation described with respect to
animacy categories, a natural way to proceed is to
start with a commonsensical approach and see
where it leads.  In 2000-2002, two rather similar
initiatives led to the need for animacy annotations:
one, the paraphrase-project, a collaboration
between Stanford and Edinburgh, concentrating on
the factors that influence the choice between
different sentence level syntactic paraphrases
(Bresnan et al. 2002) and another concentrating on
the possessive alternation (O’Connor, 2000).  The
two projects used a very similar animacy
annotation scheme, developed in the context of the
O’Connor project.
The scheme was used in two different ways. The
Boston team coded 20,000 noun phrases in
‘possessive’ constructions from the Brown Corpus.
The first round of coding was automated, with the
animacy annotation based primarily on word lists
and morphological information. The second round
was performed manually by pairs of coders using a
decision tree. The two coders were required to
agree on each code; every case in which there was
not complete agreement was discussed by the rest
of the team, until a choice of code was made. This
way of annotating does not lend itself to a study of
reliability, except between the automated coder
and the human coders as a group. For more
information on this use of the coding system, see
Garretson & O’Connor (2004).
In what follows we concentrate on the use of the
coding scheme in the Stanford-Edinburgh
paraphrase project.
The overall aim of the paraphrase project is to
provide the community of linguists and
computational linguists with a corpus that can be
used to calculate the impact of the various factors
on different constructions.  The annotation scheme
assumes that the main distinction is three-way:
human, other animates and inanimates, but the two
latter categories are subdivided further as follows:
- Other animates: organizations, animals,
intelligent machines and vehicles.
- Inanimates: concrete inanimate, non-concrete
inanimate, place and time.
The category ‘organization’ is important because
organizations are often presented as groups of
humans engaging in actions that are typically
associated with humans (they make
pronouncements, decisions, etc.).  The categories
place and time are especially important for the
possessive encoding as it has often been observed
that some spatial and temporal expressions are
realized as Saxon genitives (see e.g. Rosenbach
(2002)).
For the cases in which no clear decision could be
made, a category ‘variable animacy’ was invented,
and the coders were also given the option to defer
the decision by marking an item with ‘oanim’.
The overall coding scheme, with a summary of the
instructions given to the coders, looks as follows
1
HUMAN
Refers to one or more humans; this includes
imaginary entities that are presented as human,
gods, elves, ghosts, etc.: things that look human
and act like humans.
ORG
This tag was proposed for collectivities of humans
when displaying some degree of group identity.
The properties that are deemed relevant can be
represented by the following implicational
hierarchy:
+/- chartered/official
+/- temporally stable
+/- collective voice/purpose
+/- collective action
+/- collective
The cut-off point between HUMAN and ORG was
put at ‘having a collective voice/purpose’: so a
group with collective voice and purpose is deemed
to be an ORG, a group with collective action, such
as a mob, is not an ORG.
                                                       
1
 For a more extensive description of the annotation
scheme see Garretson et al. 2004.
ANIMAL
Non-human animates, including viruses and
bacteria.
PLACE
The tag is used for nominals that ‘refer to a place
as a place’. There are two different problems with
the delimitation of place. On the one hand, any
location can be a place, e.g. a table, a drawer, a
pinhead, … The coding scheme takes the view that
only potential locations for humans are thought of
as ‘places’.  On the other hand some places can be
thought of as ORGs.  The tag was applied in a
rather restricted way, for instance in a sentence
such as ‘my house was built in 1960’, ‘my house’
is coded as CONC (see below), whereas in ‘I was
at my house’, it would be a PLACE.
TIME
This tag is meant to be applied to expressions
referring to periods of time.  It was applied rather
liberally.
CONCRETE
This tag is restricted to ‘prototypical’ concrete
objects or substances.  Excluded are things like air,
voice, wind and other intangibles. Body parts are
concrete.
NONCONC
This is the default category.  It is used for events,
and anything else that is not prototypically
concrete but clearly inanimate.
MAC
A minor tag used for intelligent machines, such as
computers or robots.
VEH
Another minor category used for vehicles as it
has been observed that these are treated as living
beings in some linguistic contexts (e.g. pronoun
selection in languages such as English where
normal gender distinctions only apply to animates).
OANIM
This tag is used when the coder is completely
unsure and wants to come back to the example
later.
VANIM
This tag can be used in conjunction with another
one to indicate that the coder is not entirely sure of
the code and thinks there are reasons to give
another code too.
Finally, NOT-UNDERSTOOD was supposed to
be used when the text as a whole was not clear.
Three coders coded the parsed part of the
Switchboard corpus (Godfrey et al. 1992) over the
summer of 2003.  The corpus consists of around
600 transcribed dialogues on various
predetermined topics among speakers of American
English.  Before the annotation exercise began, the
dialogues were converted into XML (Carletta et al.
2004).  The entities that needed to be annotated
(the NPs and possessives determiners) were
automatically selected and filtered for the coders.
The three coders were undergraduate students at
Stanford University who were paid for the work.
The schema presented above was discussed with
them and presented in the form of a decision tree.
Difficult cases were discussed but eventually each
coder worked independently.  599 dialogues were
annotated.
6 Coding reliability
The reliability of the annotation was evaluated
using the kappa statistic (Carletta, 1996).
Although there are no hard and fast rules about
what makes an acceptable kappa coefficient—it
depends on the use to which the data will be
put—many researchers in the computational
linguistics community have adopted the rule of
thumb that discourse annotation should have a
kappa of at least .8.
For the reliability study, we had three
individuals work separately to code the same four
dialogues with the animacy scheme.  Markables (in
this case NPs and possessives) had been extracted
automatically from the data, leading the coders to
mark around 10% of the overall set with a category
that indicated that they were not proper markables
and therefore not to be coded.  Omitting these
(non-) markables, for the data set overall, K=.92
(k=3, N=1081).
In general, coders did not agree about which cases
were problematic enough to mark as VANIM, and
omitting the markables that any coder marked as
problematic using the VANIM code leads to a
slight improvement (K=.96, k=3,N=1135).
It is important to note that these kappa coefficients
are so high primarily because two categories which
are easy to differentiate from each other, HUMAN
and NONCONC, swamp the rest of the categories.
The cross-coder reliability for them is satisfactory
but the intermediate categories were not defined
well enough to allow reliable coding.
Figure 1 shows the confusion matrix for the data
including markables that any coder marker
additionally as problematic using the VANIM
notation.  Considering the coders one pair at a
time, the matrix records the frequency with which
one coder of the pair chose the category named in
the row header whilst the other chose the category
named in the column header for the same
markable.
Although we were aware of the less than formal
definitions given for the categories, we had hoped
that the coders would share the intuitive
understanding of the developers of the categories.
This is obviously not the case for all categories.
What was also surprising was that allowing coders
to mark cases as problematic using the VANIM
code was not worthwhile, since the coders did not
often take advantage of this option and taking the
VANIM codes into account during analysis has
little effect.
Analysis of the four annotated dialogues points to
several sources for the intercoder disagreement.
• The categories TIME and PLACE were
defined in a way that did not coincide with the
coders’ intuitive understanding of them.  The
tag TIME was supposed to refer to ‘periods of
time’. This led to some wavering
interpretations for temporal expressions that do
not designate a once-occurring period of time.
For instance ‘this time’ and ‘next time’ were
coded as TIME by two coders but as
‘NONCONC” by the third one.  Clearer
training on what was meant could have helped
here.
• As mentioned above, the choice
between HUMAN, ORG and NONCONC
depended on how the coders interpreted the
referent of the expression.  Although
guidelines were given about the difference
between HUMAN and ORG (see above), the
cut-off point wasn’t always clear
2
. The
distinction between ORGs as proposed in our
schema and less organized human groups
seems too fluctuating to be useful.
                                                       
2 All coders agreed that Vulcans are HUMAN.
Figure 1
• The vagueness of pronominal
reference: for instance a school as an
organization can be marked as ORG by the
coders but later in the dialogue there is
discussion about the what is done with napping
children in the school and one speaker says ‘if
they (the children) fall asleep they kind of let
them sleep’, one coder interpreted that the
second ‘they’ as simply referring to the school
organization and marked it as ORG, whereas
another interpreted it as referring to a rather
vague group of humans, presumably some
teachers, and marked it as HUMAN.  This
vagueness of reference is quite prevalent in
spoken language, especially with the pronoun
‘they’.
• Attention errors, e.g. vehicles were
supposed to get a special code but, presumably
because there were so few, this was sometimes
forgotten.  One coder coded ‘a couple of
weeks’ as HUMAN.  These kinds of mistakes
are unavoidable and the very tools that make
the encoding easier (e.g. the automatic
advancing from one markable to another)
might make them more frequent.
While the problems with ORG and HUMAN don’t
come as a surprise, the difficulties with PLACE,
TIME and CONCRETE are more surprising.  The
two minor classes, MAC and VEH and the
ANIMAL class occurred so seldom that no
significant results were obtained in this sample.
They were equally rare in the corpus as a whole.
7 Conclusion
We are not aware of any other medium-scale
attempts to annotate corpora of contemporary
English for animacy information apart from the
two mentioned here.  There are smaller efforts
concentrating on the genitive alternation (e.g.
Leech et al., 1994, Anschutz, 1997, Stefanowitsch,
2000)
3
.  The resources that have been created give
robust results for the opposition ‘human’ versus
‘nonconcrete’ entities in the large sense (as the
category was used as a catch-all).  This should
suffice for further inquiries about linguistic
processes that are sensitive to a binary opposition
in this dimension.  Moreover the Stanford-
Edinburgh effort is integrated in a corpus that has
already been marked up for syntactic information,
so correlations between syntactic constructions and
animacy (and information status) should be easy to
calculate.  It is also the first effort that studies
inter-annotator reliability.
Some studies based on the annotations are
currently being conducted.  The study by Cueni,
Bresnan, Nikitina and Baayen (2004) supplements
partial data from the work described here with
further annotations.  The work reported by
O’Connor et al. (2004) derives from the Boston
use of the encoding described here. Within the
paraphrase project we are currently investigation
                                                       
3 Some related work is done in the context of entity
tracking sponsored by various US government programs
(ACE, TIDES, etc.).  The proposed annotation schemes
have problems in distinguishing between persons and
organizations or geo-political entities that are similar to
ours, but the basic categories and the aims of these
enterprises are different.  We have not reviewed them
here.
concr
ete
hum
an
non-
conc
not-
understood
oan
im
o
rg
pl
ace
ti
me
v
eh
concrete31 9 19 100 0 5 0 1
human148
9
27 110 3
3
0 4 0
nonconc 1256 3 3 3
3
235
3
1
notundersto
od
0 0 0 0 0 0
oanim 2 2 0 1 0
org 9
1
0 0 0
place 660 2
time 7
0
0
veh 3
the possible effect of animacy on constructions
such as Left-Dislocation and Topicalization.
Further work remains to be done, however, to
determine the exact nature of the distinctions in the
animacy dimension that are important for English
and for other languages. The annotations we
provide do not settle this issue. In that sense they
are insufficient to guide generation and translation
precisely. To investigate this further we will need
to devise more careful annotation schemes and
approach the problem via experiments where the
hypothesized relative animacy of various entities
can be carefully controlled. As mentioned above, it
might be better not to think in terms of robust large
categories but rather try to rank specific entities or
small categories relative to each other and to
gradually build up a more precise picture.  This is
most likely better done through controlled
experiments than through corpus annotation.
The annotated corpus, however, will be helpful to
determine where animacy plays a role and which
other factors it interacts with. This knowledge will
help devise more adequate generation and
translation architectures.
The Boston University noun Phrase Corpus is
publicly available at http://np.corpus.bu.edu.  The
paraphrase corpus will be made available to
subscribers to the Switchboard Corpus.
8 Acknowledgements
This work was in part supported by a Scottish
Enterprise Edinburgh-Stanford Link Grant
(265000-3102-R36766) and by NSF grant BCS-
0080377.  Thanks to Toni Jeanine Harris, Rebecca
Regos and Anna Cueni for the encoding work and
to Richard Crouch and Neal Snider for comments
and help. The usual disclaimers obtain.

References
Aissen, Judith, 2002. Differential object marking:
Iconicity vs. economy. NLLT, 21 435-483.
Anschutz, A. 1997. "How to Choose a Possessive
Noun Phrase Construction in Four Easy
Steps." Studies in Language 21, 1, 1-35.
Bock, J. K., Loebell, H. and Morey, R. (1992)
From conceptual roles to structural relations:
Bridging the syntactic cleft.  Psychological
Review 99: 150--171.
Bossong, Georg. 1985. Differentielle
Objektmarkierung in den Neuiranischen
Sprachen,Gunter Narr Verlag, Tübingen.
Bossong, Georg. 1991. 'Differential Object
Marking in Romance and Beyond', in D.
Wanner and D.Kibbee (eds.), New Analyses in
Romance Linguistics: Selected Papers from
theXVIII Linguistic Symposium on Romance
Languages, John Benjamins, Amsterdam, pp.
143-170.
Bresnan, Joan, Dingare, Shipra, and Manning,
Christopher D. 2001.   Soft Constraints Mirror
Hard Constraints: Voice and Person in English
and Lummi.  Proceedings of the LFG '01
Conference. CSLI Online. 20 pages.
Bresnan, Joan, Carletta, Jean, Crouch, Richard,
Nissim, Malvina, Steedman, Mark, Wasow,
Tom and  Zaenen, Annie. 2002. Paraphrase
analysis for improved generation, LINK
project, HCRC Edinburgh-CLSI Stanford.
Carletta, Jean. 1996. Assessing agreement on
classification task: the kappa statstic.
Computational linguistics, 22 (2): 249-254
Carletta, Jean, Shipra Dingare, Malvina Nissim,
and Tatiana Nikitina. 2004. Using the NITE
XML Toolkit on the Switchboard Corpus to
study syntactic choice: a case study. In
Proceedings of the 4th International
Conference on Language Resources and
Evaluation (LREC2004), Lisbon, May 2004.
Comrie, Bernard, 1989, Language Universals and
Linguistic Typology, The University of Chicago
Press
Cueni, Anna, Joan Bresnan, Tatiana Nikitina and Harald
Baayen. 2004. Predicting the Dative Alteration,
Stanford University, ms. in preparation.
Dahl, Östen and  Fraurud, Kari. 1996 Animacy in
Grammar and Discourse. In Thorstein
Fretheim & Jeanette K. Gundel (eds.),
Reference and Referent Accessibility, 47-64.
Amsterdam/Philadelphia: John Benjamins.
Garretson, Gregory, O'Connor, M. Catherine;
Skarabela, Barbora; & Hogan, Marjorie.
March 2004. Coding practices used in the
project Optimal Typology of Determiner
Phrases.
http://npcorpus.bu.edu/documentation/index.ht
ml
Garretson, Gregory, and O’Connor, M. Catherine.
2004. A combined automatic-and-manual
method for studying discourse features in
corpora. Paper to be presented at the Fifth
North American Symposium on Corpus
Linguistics, May 21-23, Montclair State
University, NJ.
Godfrey, J. Holliman, E. & McDaniel J. 1992
SWITCHBOARD: Telephone speech corpus
for research and development. Proceedings of
ICASSP-92, 517-520.
Leech, G., B. Francis, et al. (1994). The Use of
Computer Corpora in the Textual
Demonstrability of Gradience in Linguistic
Categories. Continuity In Linguistic Semantics.
C. Fuchs and B. Victorri. Amsterdam, John
Benjamins Publishing Company: 57-76.
McDonald, J.L., Bock, K. and M. Kelly. 1993.
Word and world order: semantic, phonological,
and metrical determinants of serial position.
Cognitive Psychology 25: 188-230.
Nissum, Malvina, Shipra Dingare, Jean Carletta
and Mark Steedman. 2004. An annotation
scheme for information status in dialogue.
Submitted to LREC 2004
O’Connor, M. Catherine. 2000. Optimality
typology of the DP: Markedness within the
nominal. NSP-grant. BCS-0080377.
O’Connor, M. Catherine., Anttila, Arto, Fong,
Vivienne and   Maling, Joan (2004).
Differential possessor expression in English:
Re-evaluating animacy and topicality effects.
Paper presented at the Annual Meeting of the
Linguistic Society of America, January 9-11,
Boston, MA.
Rosenbach, Anette (2002) Genitive Variation in
English. Conceptual Factors in Synchronic
and Diachronic Studies. Berlin/New York:
Walter de Gruyter.)
Rosenbach, Anette. 2003. Aspects of iconicity and
economy in the choice between the s-genitive
and the of-genitive in English. In: Günter
Rohdenburg & Britta Mondorf (eds.).
Determinants of Grammatical Variation in
English Berlin/New York: de Gruyter.
Rosenbach, A., J. Aissen, and J. Bresnan. 2002.
Pilot study of the influence of animacy on
subject choice in a reading task.  (Heinrich
Heine University, UCSC, and Stanford)
Silverstein, Michael. 1976. Hierarchy of features
and ergativity. In Richard Dixon, editor,
Grammatical Categories in Australian
Languages. Australian Institute of Aboriginal
Studies.
Stefanowitsch, Anatol. 2000. Constructional
semantics as a limit to grammatical alternation:
The two genitives of English. in CLEAR
(Cognitive Linguistics: Explorations,
Applications, Research), 3.
Trommer, Jochen. s.d. Direction Marking and Case
in Menominee, http://www.ling.uni-
osnabrueck.de/trommer/nim.pdf
Woolford, Ellen. 1999. Animacy hierarchy effects
on object agreement, in Paul Kotey ed. New
Dimensions in African Linguistics and
Languages (Trends in African Linguistics 3),
203-216.
Yamamoto, Mutsumi, 1999, Animacy and
Reference: a cognitive approach to corpus
linguistics , John Benjamins.
