From the past into the present:

From case frames to semantic frames

Hans C. Boas and Ryan Dux

Abstract: This paper first shows how Frame Semantics (Fillmore 1982) grew out of earlier work
by Charles Fillmore (1968) on Case Grammar. Then, it discusses some of the basic principles of
Frame Semantics and shows how these have been implemented in FrameNet, an online corpus-
based lexicographic database (http://framenet.icsi.berkeley.edu). Using semantic frames to
structure the lexicon of English, FrameNet provides a wealth of information showing how frame
elements (situation-specific semantic roles) are realized syntactically (valence patterns). Finally,
the paper provides an overview of how frame-semantic principles have been applied to cover
non-lexical phenomena using compatible annotation and data formats. This so-called
“constructicon” offers entries of grammatical constructions that are also based on corpus data
and that are parallel to lexical entries in FrameNet.

1. Introduction

A great number of publications dealing with the research of Charles Fillmore (1929-2014) focus
on his seminal paper The Case for Case (1968) and its impact on linguistic theory during the late
1960s and into the 1970s. In this paper, Fillmore proposes a set of so-called case frames, which
specify a verb’s semantic valency, and he lays out a research program proposing how such case
frames are mapped to syntax. While Fillmore’s seminal paper was groundbreaking at the time,
most of its core ideas were challenged during the 1970s and eventually abandoned. Interestingly,
in the present day (the year is 2017) Fillmore is mainly remembered by many linguists for his
1968 paper as well as a few subsequent publications, but not really for most of his later research
in Frame Semantics building on his original proposals. In other words, except for Fillmore’s
former students, colleagues, and associates or researchers closely affiliated with this branch of
Cognitive Linguistics, there is very little recognition in the mainstream linguistics literature of
Fillmore’s own eventual abandonment of his original ideas.
More importantly, Fillmore’s subsequent development of Frame Semantics during the
1980s and 1990s as a novel approach to structuring the lexicon (and exploring how lexical-
semantic information is realized syntactically) has received little to no recognition in the
mainstream linguistics literature. For example, in their book on argument structure, Levin and
Rappaport Hovav (2005) devote an entire chapter to reviewing Fillmore’s (1968) proposals and
its advantages and disadvantages, but they do not mention Fillmore’s (1982, 1985) theory of
Frame Semantics, which offers a very different view of many concepts presented in his 1968
paper while still preserving some of its key ideas. Even within the growing body of literature on
Construction Grammar, the sister theory of Frame Semantics, there is proportionally little
discussion of Frame Semantics. For example, Goldberg’s (1995) seminal book mentions

1
semantic frames a few times, but it neither defines frames nor discusses how they are structured
and organized. Similarly, the Oxford Handbook of Construction Grammar (Hoffmann and
Trousdale 2013) does not include a chapter on Frame Semantics, and the journal Constructions
and Frames published only very few papers devoted specifically to Frame Semantics (Boas
2014b).
This paper seeks to close this gap and to demonstrate how Fillmore’s (1982, 1985) Frame
Semantics grew out of his original ideas of the late 1960s and subsequently developed during the
1980s and 1990s, while at the same time interconnecting with a corresponding theory of
grammar that eventually became known as Construction Grammar. The paper is structured as
follows. Section 2 first provides a brief overview of some of Fillmore’s (1968) key proposals and
discusses why they were largely abandoned throughout the 1970s. It then introduces Fillmore’s
theory of Frame Semantics and moves on to a discussion of the FrameNet project, which, since
1997, has been applying the principles of Frame Semantics to a corpus-based lexicography
project seeking to construct an online lexical database of English. Section 3 first discusses how
insights from English FrameNet have been applied to the description and analysis of other
languages. Then, it provides a brief overview of the FrameNet Constructicon, an inventory of
entries of grammatical constructions similar to the types of lexical entries in FrameNet. Finally,
Section 4 summarizes our paper. 1

2. Frame Semantics and FrameNet

2.1 The Case for Case

Fillmore (1968) proposed a limited set of semantic roles (also known as deep cases) such as
Agentive, Instrumental, Dative, Locative, and Objective that are organized in a specific hierarchy
for realizing grammatical functions. For example, Agentive was at the top of the hierarchy,
followed by Instrumental, Objective, and others. This hierarchy was used to ensure proper
linking of a particular semantic role to syntax depending on the total number of roles present. For
example, in sentences such as Kim opened the door, the Agentive would be realized in subject
position because the Agentive role is the highest in the hierarchy. In contrast, in sentences such
as The key opened the door, the Instrumental would be realized in subject position because there
was no Agentive to link to subject position and the Instrumental was the next role down in the
hierarchy.
Fillmore’s proposals were different from previous approaches, because they explicitly
called for the identification of a restricted set of semantic roles that would be applicable to any
argument of any verb. In addition, semantic roles were defined independently of verb meaning,

1
Parts of this paper are based on Boas (2013b, 2017).

2
they were regarded as unanalyzable, and each semantic role was supposed to be realized by only
one argument. At the same time, each syntactic argument should bear only one semantic role. 2
During the 1970s, much research was devoted to applying Fillmore’s (1968) proposals to
a range of different phenomena and languages (see Busse 2012 for an overview), and it soon
became apparent that Fillmore’s original ideas were problematic. One of the key issues is that
there are no systematic tests for determining semantic roles. For example, in sentences such as
Kim ate dinner with a friend and Kim ate dinner with a fork, the objects of the with-phrases are
distinct semantic roles and as such grammatical markers do not seem to be precise when it comes
to identifying specific semantic roles. Another problem is the grain size of semantic roles, which
makes it difficult to distinguish between different types of semantic roles. For example, Nilsen
(1972) identifies four sub-classes of Instruments based on distributional data illustrating that they
exhibit different types of acceptability when realized in subject position (compare The cook
opened the jar with a new gadget / The new gadget opened the jar vs. Shelley ate the sliced
banana with a fork / *The fork ate the sliced banana) (Levin and Rappaport Hovav 2005: 39).
Another problem is a lack of one-to-one correspondence between syntactic arguments and
semantic roles. Sentences such as Pat rolled down the hill and Sascha resembles Lee illustrate
that one syntactic argument can be interpreted as two semantic roles (Pat causes the action
(Agent) and changes location (Theme)) and two syntactic arguments can bear a single role (both
Sascha and Lee are compared to each other). Issues such as these led many researchers to
abandon Fillmore’s original proposals as well as modifications of them in the early 1970s (see
Fillmore 1977 for a discussion).

2.2 Frame Semantics and FrameNet

The development of Case Grammar into Frame Semantics was motivated by Fillmore’s
increasing awareness of the shortcomings of “case roles” and growing interest in cognitive and
ethnographic semantics. Fillmore (1977) addresses several critiques of his initial formulations of
Case Grammar and marks a major step away from the assumption of primitive abstract case
roles. He recognizes that many scholars had interpreted his 1968 work as a full-fledged grammar,
whereas his original intention was to develop only a method for identifying case distinctions and
characterizing the case structure organization of sentences. The problems mentioned above with
the enumeration and identification of case roles led Fillmore to claim that case roles are not
primitive and universal, but that they are determined by events and more generally that “meaning
is relativized to scenes” (1977: 59). Specifically, rather than defining verb meanings (or
“situations”) by the semantic roles of their arguments, one must define situation types in their
own right and identify what participants (semantic roles) define the situations.
Concurrently, Fillmore (1975, 1982, 1985) sought to identify the nature of these
“events/scenes” that determine case roles. His 1977 article showed how recent developments in

2
For a related proposal, see Gruber (1965); for more details, see Petruck (1996), Ziem (2008/2014b), and Busse
(2012).

3
cognitive science, notably scenes (Fillmore 1987) and prototypes (Berlin and Kay 1969, Rosch
1973), could be applied to the study of word meaning. Subsequently, Fillmore (1982, 1985)
offers rich and detailed examples of how cultural and world knowledge motivates and is
embedded in linguistic expressions, emphasizing that solely truth-conditional semantic
approaches cannot account for these aspects of word meaning and demonstrating the need for a
“semantics of understanding” (see also Fillmore 1975). The core ideas underlying research in
Frame Semantics are summarized in the following quote:

A word’s meaning can be understood only with reference to a structured background of

experience, beliefs, or practices, constituting a kind of conceptual prerequisite for understanding
the meaning. Speakers can be said to know the meaning of the word only by first understanding
the background frames that motivate the concept that the word encodes. (Fillmore and Atkins
1992: 76-77)

During the 1990s, research in Frame Semantics focused on lexicographic and

grammatical questions surrounding the syntactic realization of participants. Fillmore and Atkins
(1992) provide a detailed investigation of how the concept of ‘risk’ is realized linguistically by
(a) identifying all participants in the risk scenario, (b) documenting how participants are formally
realized in concrete linguistic expressions, and finally (c) summarizing the various ways in
which the concept can be realized syntactically. They show, for instance, that ‘risk’ can be
construed in (at least) two ways and therefore evokes two different frames, with expressions such
as take a risk perspectivizing the risky activity carried out by the risk-taker and put at risk
perspectivizing the entity endangered by the risky activity.
Fillmore and Atkins’ (1992) seminal study not only demonstrated the rich results of
characterizing word meanings and grammar in terms of semantic frames, but also laid the
groundwork for the development of FrameNet (http://framenet.icsi.berkeley.edu), which started
at the International Computer Science Institute in Berkeley, California, in 1997. FrameNet is an
online lexical database documenting a wide variety of frame-semantic and syntactic information
for the English lexicon. 3 The most important notions around which FrameNet is structured are
semantic frames, Frame Elements, and lexical units. Frame Elements (FEs) are the
participants/roles by which semantic frames are defined. For instance, the FEs of the Taking
frame are AGENT, THEME, and SOURCE, because a taking event minimally requires that some
entity (AGENT) takes something (THEME) from somewhere (SOURCE). 4 A distinction is made
between these core FEs that are crucial for the understanding of the frame and non-core FEs that
do not define the frame but provide additional information such as time, place, and manner.
Lexical units (LUs) are linguistic expressions (including all parts of speech and multi-word
units) that evoke a given frame. LUs of the Taking frame, for instance, include specific senses
of the verbs take and grab and the noun seizure. FrameNet accounts for polysemy by positing
3
For details, see Fillmore and Baker (2010), Ruppenhofer et al. (2010), Ruppenhofer et al. (2013), Ruppenhofer et
al. (in press), and Boas (2017).
4
Following FrameNet practice, frame labels are in Courier New font and FE labels are in small capital font.

4
LUs in different frames for polysemous expressions. For instance, take is also listed as a LU of
the Taking_time and Ride_vehicle frames, among several others. Another major
organizing principle of FrameNet are frame-to-frame relations, which capture relations across
frames. A key frame-to-frame relation is Inheritance, whereby a daughter frame inherits and
further specifies information (including Frame Elements) of a mother frame. 5 The Taking
frame, for instance, inherits from a more general Getting frame, because taking is a more
specific instance of getting (i.e. volitionally/intentionally getting). In the other direction, the
Taking frame is an inheritance mother to the Theft frame, which further specifies the illegal
or unallowed nature of the more general taking event.
The information contained in the FrameNet database is the result of a workflow that
begins with selecting a target word (including multi-word expressions) and identifying the frame
it evokes by “characterizing schematically the kind of entity or situation represented by the
frame” (Fillmore et al. 2003b: 297). More specifically, FrameNet researchers use intuition and
corpus data to determine what features are necessary for the understanding of the word and
assign mnemonic labels to each of the Frame Elements defining the frame. Next, a thorough
corpus search is conducted for expressions deemed semantically similar to the target word in
order to determine whether they have the same frame semantics and Frame Elements, thereby
arriving at a full list of lexical units for the frame. For each of these lexical units, a number of
representative corpus sentences are extracted and annotated for both syntactic and (frame-
)semantic information. Specifically, the grammatical function and phrase type for each Frame
Element occurring in the sentence is documented, resulting in layered annotations such as that in
Table 1.
(Text) John took the bottle from the baby

FE AGENT target word THEME SOURCE

PT NP NP PP.from

GF Ext Obj Comp

Table 1. Layered frame-semantic annotation for John took the bottle from the baby

Table 1 illustrates one potential configuration of Frame Elements for the verb take, 6 in which the
AGENT is realized as a nominal subject (Ext), the THEME as a nominal object, and the SOURCE as
a complement prepositional phrase headed by from. Similar annotations are conducted to
document the full range of FE realization patterns for each lexical unit, resulting in two types of
valency tables: one summarizing the potential syntactic realization of each FE and one
summarizing the full set of FE constellations. FrameNet’s workflow is not strictly linear, as new

5
Other frame-to-frame relations include Causative_of, Perspective_on, Uses, and Subframe. See Baker et al. (2003)
and Petruck et al. (2004) for more details on frame-to-frame relations.
6
At least five other FE configurations are documented for take in FrameNet.

5
data may lead to the identification of finer-grained frames or the inclusion of more lexical units
for a given frame (see Fillmore et al. 2003a and Petruck 2004).
These frame-semantic analyses are documented on the Frame Description and Lexical
Entry pages in FrameNet. Frame Description pages provide information for semantic frames.
Figure 1 shows the Frame Definition for Taking as well as definitions and example sentences
for each of the core FEs in the frame.

Figure 1. Frame and Frame Element Definitions of Taking frame in FrameNet

Note how the frame definition includes the FE labels, explicitly describes the relations between
them, and associates each with a semantic type, specifying for instance that the AGENT of
Taking must be sentient. In addition to these definitions, the Frame Description pages also list
the non-core FEs compatible with the frame (such as PLACE and MEANS), frame-to-frame
relations (e.g. that Taking inherits from Getting), and a list of all lexical units evoking the
frame.
Lexical entry pages document frame-semantic and valency information for individual
lexical units. Each lexical unit is given a brief definition (either written by FrameNet researchers
or taken from a print dictionary). Valency information derived from corpus annotation (as in
Table 1 above) is also documented both for individual FEs (showing all possible realizations of
each FE) and for combinations of FEs (in Valence Pattern tables, see Fillmore 2007). Figure 2

6
shows a portion of the Valence Patterns table for the verb take in the Taking frame,
summarizing the results of the frame-semantic annotation of corpus sentences containing the
lexical unit.

Figure 2. Portion of Valence Patterns for take in the Taking frame in FrameNet.

Three combinations of Frame Elements are shown in the table, the first of which includes the
core FEs AGENT, SOURCE, and THEME, and the non-core PLACE FE, as in the sentence [<Agent>The
Ottomans] tooktgt [<Theme>land] [<Place>in what is now Turkey] [<Source>INI]. The grammatical
function and phrase type of each FE is listed below the FE name, e.g. the THEME is a nominal
object. The labels DNI and INI refer to FEs that are not overtly expressed and are interpreted
under definite or indefinite null instantiation, respectively (see Section 3.1). The numbers in the
left-hand column refer to the number of annotated corpus sentences bearing each FE
configuration. Users can click on the number to see the corpus sentence(s) for each
configuration, and all annotated corpus sentences can also be accessed on the annotation page of
the Lexical Entry.
As noted above, FrameNet also organizes frames in a hierarchical structure by means of
frame-to-frame relations, such as Inheritance, Uses, and Perspective_on, among others. These
relations account for frames and FEs that are similar but differ in terms of specificity or
perspective. These relations are documented on FrameNet’s FrameGrapher tool, with which
users can select a frame and see which frames are related to it in which ways (see Fillmore and
Baker [2010] for details). Table 3 shows a portion of the frame hierarchy surrounding the
Taking frame.

7
Figure 3. FrameNet’s Frame Grapher representing different relations between semantic frames.

Below, we discuss how the frame hierarchy and frame-to-frame relations can be used to arrive at
broader-scale linking rules which predict the syntactic realization of FEs, but we also point out
the difficulty of formulating such rules without adequate empirical data.
Additional FrameNet data not included in frame description or lexical entry pages include
the Frame SQL search interface, which allows detailed searches of FrameNet (e.g. for specific
FE-GF-PT combinations), and some full text annotations in which entire texts are annotated for
frame-semantic information (see Ziem et al. 2014).

2.3. Current research in and challenges for Frame Semantics and FrameNet

One of the issues raised by researchers not familiar with FrameNet is its coverage (see Hanks
2012). After twenty years of continuous work, members of the FrameNet project have created
lexical entries for 13,635 lexical units, evoking 1,223 frames (consisting of 10,542 unique frame
elements), together with 202,229 annotation sets (FrameNet accessed on June 17, 2017).
Compared to other lexical resources such as WordNet (Fellbaum 1990), the coverage of
FrameNet appears relatively limited in comparison. However, there are a few points to keep in
mind when comparing FrameNet with other lexical resources. First, FrameNet offers a much
greater wealth and depth of information, specifically when it comes to documenting how the
semantics of frames evoked by words are realized syntactically (see, e.g., the rich information
contained in the valence tables such as in Figure 2 above). Second, creating frame descriptions
and annotating corpus examples with semantic frames are both manual activities that require a
great deal of time and funding. Since its inception in 1997, the usability of FrameNet data
(despite its perceived lack of coverage) has been demonstrated by research on different types of
linguistic phenomena, including syntactic alternations (Baker and Ruppenhofer 2002, Boas
2002/2011b, Iwata 2008, Dux 2011), verb classification (Boas 2008c, Croft 2009, Bouveret
2012, Dux 2016), lexicography (Fillmore and Atkins 2000, Atkins 2003, Fillmore et al. 2003,
Schmidt 2009, Ruppenhofer et al. 2013, Boas 2013b), full text analysis (Ziem et al. 2014),

8
analysis of metaphors (Gargett et al. 2014, Gemmell 2015, Ziem 2015), and contrastive analyses
(Subirats and Petruck 2003, Boas 2010b, Hilpert 2010, Boas and Gonzalvez-Garcia 2014,
Hasegawa et al. 2016), to name just a few. In addition, FrameNet data have been applied to solve
issues in foreign language education (Atzler 2011, Friberg Heppin and Heppin 2012, Boas and
Dux 2013, Boas et al. 2016, VanNoy 2016, Benjamin et al. in progress) and natural language
processing, e.g. Automatic Semantic Role Labeling (Gildea and Jurafsky 2002, Das et al. 2010,
Ruppenhofer et al. 2013), Semantic Parsing (Baker et al. 2007, Schneider 2015), automatic
induction of frames (Hermann 2014), and Sentiment Analysis (Ruppenhofer and Rehbein 2012).
The breadth of research underway demonstrates that FrameNet data, despite its perceived
limitations of coverage, is not a serious issue. On the contrary, we would like to propose that the
different types of current research demonstrate the quality and usability of FrameNet data and
that the real issue is one of funding in order to reach a more complete coverage of the lexicon of
English more quickly. 7 An alternative option is to (semi-)automatically induce new frames and
to (semi-)automatically create frame annotations using computational techniques. Research in
these areas is ongoing, but it is not clear yet what the exact outcomes will be (Das et al. 2014).
Another issue with the current architecture of FrameNet pointed out in the literature
concerns the lack of systematicity and organization. Recall that in earlier versions of Fillmore’s
case theory, there was only a limited set of semantic roles, similar in spirit to the relatively small
inventory of semantic primitives found in other contemporary theories of lexical meaning (e.g.
Bierwisch (1970), Jackendoff (1990), Levin and Rappaport Hovav 2005, Wierzbicka (2006)).
Because of the issues with such a limited set of semantic roles (see Section 2 above), during the
1980s Fillmore purposely moved to a frame-based organization of the lexicon, which finds its
practical implementation in FrameNet. The critique leveled by researchers such as Osswald and
Van Valin (2014) focuses on the observation that there is “a certain lack of systematicity in the
definition of frames and frame relations, which may hinder the derivation of linking
generalizations” (2014: 125). Based on data from verbs of cutting and separation in FrameNet,
Osswald and Van Valin (2014) point out a number of inconsistencies in the organization of these
frames and the definitions of their Frame Elements.
While Osswald and Van Valin (2014) certainly point out some important issues in the
current organization of FrameNet, specifically with respect to its data-driven, purely bottom-up
methodology, we propose that this critique is both inadequate and too early. First, as pointed out
in the previous paragraph, the coverage of the English lexicon by FrameNet is not complete. This
means that there is yet no complete inventory of frames and frame relations available. As such, it
is difficult to arrive at broad-scale or “universal” linguistic insights about linking generalizations.
In other words, we believe that without having a more complete (if not totally complete)
coverage of the English lexicon by FrameNet, it is almost impossible to arrive at the types of
generalizations that Osswald and Van Valin are looking for. Up to this point, FrameNet research
7
Studies such as Baker and Fellbaum (2008) and Baker and Palmer (2009) have also compared FrameNet with the
WordNet project (Miller 1995, Fellbaum 1998), which organizes the English lexicon according to semantic relations
such as synonymy, hyponymy, and metonymy, showing how the two resources can be aligned and complement each
other in various tasks such as text understanding.

9
has focused on providing a rich and empirically sound characterization of the English lexicon,
which has led its researchers to appreciate the difficulty of establishing linking rules that
accurately account for all data. In fact, one of the key insights made by members of the
FrameNet project over the past two decades is that the exploration, definition, and description of
new frames during the project’s workflow regularly resulted in the necessity to reframe existing
frames and frame relations (see Petruck et al. 2004 and Ruppenhofer et al. 2010). In sum, broad-
scale linking rules can only be formulated with a much larger amount of (annotated corpus) data,
and even given this, it is yet unclear whether a word’s (frame) semantics is fully predictive of its
syntactic behavior, and vice versa (see Faulhaber 2011 and Dux 2016).
At the same time, now that FrameNet has documented a substantial (though not
complete) portion of the English lexicon, research has begun to identify certain smaller-scale
generalizations and insights, as evidenced by the plethora of studies drawing on FrameNet data
cited above. We believe that numerous empirically rich and detailed investigations of individual
frames or sets of related frames may eventually result in the identification of broader-scale
linking rules. Frame-to-frame relations and the frame hierarchy (see Figure 3 above) show
promise as theoretical tools for identifying and defining the linking rules desired by scholars
such as Osswald and Van Valin, but more data and careful small-scale investigations are
required at this point.

3. Extending FrameNet 8

3.1 Multilingual FrameNets

One of the ideas found in Fillmore’s earlier work during the 1960s was that his early case frames
could be used to investigate universal aspects of language such as the subject selection rule
(Fillmore 1968). Later research in Frame Semantics also explored the idea of applying frame-
semantic insights to study aspects of language that could possibly reveal interesting universal
aspects. Early research during the 1990s by Heid (1996) and Fontenelle (1997) investigated how
frame-semantic insights derived on the basis of English could be applied to the systematic
analysis of the lexicons of languages other than English. Subsequent research demonstrated in
greater detail how English-based semantic frames could be employed for frame-semantic
analysis of other languages (Fillmore and Atkins 2000, Boas 2001/2002/2005b, Petruck and
Boas 2003). The results of this research inspired the creation of FrameNets for other languages,
most notably Spanish (Subirats and Petruck 2003; Subirats 2009), Japanese (Ohara et al. 2003,
Ohara 2009), German (Burchardt et al. 2009), Swedish (Borin et al. 2010), and Brazilian
Portuguese (Torrent et al. 2014). These projects provide vital information about how the
semantics of a given frame are realized syntactically by different LUs evoking that frame in
different languages, allowing researchers to systematically conduct contrastive and comparative

8
Parts of this section are based on Boas (2017).

10
research on both lexical-semantic and argument realization topics such as polysemy, profiling
properties, and the interface between lexicon and syntax (see Petruck 2009, Hasegawa et al.
2016, Leino 2010, Timyam and Bergen 2010, Bouveret 2012, and Willems 2012). 9

3.2 The Constructicon

Fillmore’s research in Frame Semantics repeatedly pointed to the interconnection between

semantic frames and grammatical construction, and Fillmore et al.’s (1988) account of the ‘let
alone’ construction and Fillmore’s (1988) concurrent formulation of Construction Grammar
established the close relation between these theories. However, it was not until 2008, when a
pilot project housed at FrameNet sought to explore systematically the possibility of identifying
grammatical constructions and describing them using a FrameNet-type format. 10 The goal of this
effort was to identify, analyze, and annotate constructions in a very similar way as LUs in
FrameNet.
Consider an example such as the construct Kim doesn’t like citrus fruit, let alone
grapefruit. A construct is a linguistic form that instantiates one or more constructions. In this
example, the construct instantiates the Let-alone construction (Fillmore et al. 1988), 11 in
which the phrase let alone functions as a conjunction with very specific semantic-pragmatic
constraints on the pieces that it joins. Other constructions contributing to the construct include
the non-lexical Subject-predicate and Negation constructions and the individual words
(except let alone), which are lexical constructions (i.e. LUs evoking a particular semantic frame).
Finding and annotating constructions follows roughly the same steps as the lexicographic
workflow in FrameNet, i.e. just like lexicographers first choose a frame to analyze, construction
grammarians first choose a construction and then search for corpus data that enables them to
annotate the data and arrive at an adequate description (see Fillmore et al. 2012 for details).
Table 2 compares the similarities between lexical and constructional description (and
analysis). Descriptions of constructions are formulated in prose (similar to lexical frame
descriptions), together with a definition of construct elements (CEs). In construction annotation,
annotators often look for a so-called construction-evoking element (CEE) (parallel to a frame-
evoking LU), which is specific lexical material central for evoking the construction, such as the
phrase let alone. Then, annotators identify and use the FrameNet desktop software to mark CEs
such as, in the case of the Let-alone construction, First_conjunct (citrus fruit) and
Second_conjunct (grapefruit), which are constituent parts of a construction, similar to FEs in
lexical annotation. In some cases, however, there may not be any CEE, as in abstract

9
In fall of 2016, the Berkeley FrameNet project received a grant for a three-year long project to systematically set
up a multilingual FrameNet infrastructure. The first goal of this project is to have FrameNet teams working on at
least eight different languages annotate the same texts in different languages in order to take a first step towards
systematically describing and linking semantic frames across different languages.
10
Constructional entries developed in this pilot project can be found at: http://sato.fm.senshu-
u.ac.jp/frameSQL/cxn/CxNeng/cxn00/21colorTag/index.html.
11
Following Fillmore et al. (2012), names of constructions are represented in italicized Courier New font.

11
constructions such as Subject_Predicate, Gapping, and Right_Node_Raising,
which have no overt lexical material signaling the presence of a construction. In such cases,
annotators only employ the CE labels to identify the different parts of the construction. Besides
the identification of CEs, annotations on different layers may also include information about
grammatical functions and phrase types, parallel to FrameNet’s lexical annotation. These added
annotation layers are intended to capture possible variations in the realization of a construction.

Table 2. Lexical and constructional description and annotation compared (Fillmore 2008: 9).

After the annotation process is complete, the construction descriptions, together with their
annotated example sentences, are stored in the Constructicon, an extension of the original FN
database. It currently consists of roughly 75 construction entries documenting different types of
constructions according to the kinds of constructs they create. These include frame-bearing
constructions, valence-augmenting constructions, constructions without meanings, contextually
bound constructs, pumping constructions, exocentric and headless constructions, and clause-
defining constructions (see Fillmore et al. 2012 for details). To illustrate, consider Figure 4
below, which is the first part (the description) of the entry of the Way_manner construction in

12
the Constructicon. Above the construction description we see that this construction evokes the
Motion frame, and it inherits from the Way_neutral construction. This information is
followed by a general prose description, including the semantics of the construction. Beneath the
description we find references to publications on the Way_manner construction.

Figure 4. First part of Way_manner construction: Construction description.

The second part of a construction entry, seen in Figure 5, contains the definitions of CEE(s) and
CEs. In the case of the Way_manner construction, this is the noun phrase one’s way, where
one’s is co-indexed to the Theme. One special feature of the Way_manner construction is the
fact that its CEs are directly linked to the FEs of the Motion frame.

13
Figure 5. Second part of Way_manner construction entry (partial)

Finally, the summary of how the construction’s CEs are realized syntactically can be found in the
last part of a construction entry. In the case of the Way_manner construction, Figure 6 shows
that the Theme is always realized as an external NP, and that the Intransitive_manner_verb
appears in different forms such as finite (VPfin) and the progressive form (VPing). The CEE is
always a NP, while the Direction is realized as a dependent ADVP or PP. 12

12
Note that while the types and granularity of information displayed differs from construction to construction, they
are still parallel to the valence tables found in the FN lexical entries.

14
Figure 6. Third part of Way_manner construction entry: partial summary

While the Berkeley Constructicon for English currently consists of about 75 construction entries,
more work is currently under way at FrameNet to enlarge the inventory of constructions. In the
meantime, various multilingual FrameNet projects have begun to also assemble construction
entries for Swedish, Japanese, Brazilian Portuguese, and German (Lyngfelt et al. 2012, Ohara
2013, Boas 2014, Ziem 2014a, Ziem and Boas 2017). The goal is to determine to what degree it
is possible to arrive at corresponding grammatical constructions across multiple languages and to
see whether such construction entries can be linked to each other (Boas 2010b). While it is too
early to arrive at any definite conclusions about the correspondence of constructions across
languages, first results seem to suggest that it is possible to arrive at a fair amount of
corresponding constructions for English constructions, although typically with some minor
differences. The closest equivalents are typically relative general grammatical constructions,
whereas those containing specific lexical elements tend to differ more (Bäckstrom et al.,
submitted). As the inventories of corpus-based construction entries grow for each language, we
expect these to be of more than descriptive value and thereby also to become more relevant for a
more general theory of language.

15
4. Conclusions

FrameNet was originally designed as a corpus-based project that sought to apply the principles of
Fillmore’s Frame Semantics towards constructing a lexicographic database for English. Over the
past twenty years, FrameNet has grown significantly in coverage, but much work remains to be
done. This paper provided an overview of the layout and methodology underlying FrameNet, and
how its principles have been applied to the construction of FrameNets for other languages. In
addition, we have shown how FrameNet has been expanded to cover grammatical constructions
of English, and multilingual FrameNets around the world are now also working on assembling
constructicons for other languages. We hope that the interested reader will spend some time
familiarizing himself with the rich content that FrameNet has to offer in order to see how it may
inform the analysis of specific linguistics problems.

References
Atkins, Sue, Michael Rundell, and Hiroaki Sato. 2003. The contribution of FrameNet to practical
lexicography. International Journal of Lexicography. 16(3). 333–357.
Atzler, Judith K. 2011. Twist in the list: Frame Semantics as vocabulary teaching and learning
tool: University of Texas at Austin dissertation.
Bäckstrom, Linnea, Lyngfelt, Benjamin, and Emma Sköldberg. Submitted. Towards interlingual
constructicography. On correspondence between construction resources for English and
Swedish. In B. Lyngfelt and T. Torrent (eds.), Interlingual Constructicography.
Amsterdam/Philadelphia: John Benjamins.
Baker, Collin and Josef Ruppenhofer. 2002. FrameNet’s Frames vs. Levin’s Verb Classes. In J.
Larson and M. Paster (eds.), Proceedings of the 28th Annual Meeting of the Berkeley Linguistics
Society, 27–38. UC Berkeley: Berkeley Linguistics Department.
Baker, Collin, Charles J. Fillmore, and Beau Cronin. 2003. "The structure of the FrameNet
database." International Journal of Lexicography 16(3). 281-296.
Baker, Collin, Michael Ellsworth, and Katrin Erk. 2007. SemEval’07 task 19: frame semantic
structure extraction. In Proceedings of the 4th International Workshop on Semantic Evaluations
(SemEval’07, ACL), 9–104, Stroudsburg, PA, USA.
Baker, Collin F. and Christiane Fellbaum. 2009. WordNet and FrameNet as complementary
resources for annotation. In Proceedings of the third linguistic annotation workshop, 125–129.
Association for Computational Linguistics.
Barðdal, Jóhanna. 2008. Productivity. Evidence from Case and Argument Structure in Icelandic.
Amsterdam/Philadelphia: John Benjamins.

Berlin, Brent and Paul Kay. 1969. Basic Color Terms: Their Universality and Evolution.
Berkeley: University of California Press.

16
Bierwisch, Manfred. 1970. Semantics. In J. Lyons (ed.), New horizons in linguistics, 166–184.
Harmondsworth: Penguin.
Boas, Hans C. 2001. Frame Semantics as a framework for describing polysemy and syntactic
structures of English and German motion verbs in contrastive computational lexicography. In P.
Rayson, A. Wilson, T. McEnery, A. Hardie & S. Khoja (eds.), Proceedings of Corpus
Linguistics 2001, 64–73.
Boas, Hans C. 2002. Bilingual FrameNet dictionaries for machine translation. In M. González
Rodríguez & C. Paz Suárez Araujo (eds.), Proceedings of the Third International Conference on
Language Resources and Evaluation, Vol. IV, 1364–1371. Las Palmas, Spain.
Boas, Hans C. 2003a. A constructional approach to resultatives. Stanford: CSLI Publications.
Boas, Hans C. 2003b. A lexical-constructional account of the locative alternation. In L.
Carmichael, C.-H. Huang, and V. Samiian (eds.), Proceedings of the 2001 Western Conference
in Linguistics. Vol. 13, 27–42
Boas, Hans C. 2005a. From theory to practice: Frame Semantics and the design of FrameNet. In
S. Langer and D. Schnorbusch (eds.), Semantik im Lexikon, 129–160. Tübingen: Narr.
Boas, Hans C. 2005b. Semantic frames as interlingual representations for multilingual lexical
databases. International Journal of Lexicography 18(4). 445–478.
Boas, Hans C. 2008a. Resolving Form-meaning Discrepancies in Construction Grammar. In J.
Leino (ed.), Constructional Reorganization, 11–36. Amsterdam/Philadelphia: Benjamins.
Boas, Hans C. 2008b. Determining the structure of lexical entries and grammatical constructions
in Construction Grammar. Annual Review of Cognitive Linguistics 6. 113–144.
Boas, Hans C. 2008c. Towards a frame-constructional approach to verb classification. In E.
Sosa Acevedo and F. J. Cortés Rodríguez (eds.), Grammar, Constructions, and Interfaces.
Special Issue of Revista Canaria de Estudios Ingleses 57. 17–48.
Boas, Hans C. (ed.) 2009. Multilingual FrameNets in Computational Lexicography. Berlin/New
York: Mouton de Gruyter.
Boas, Hans C. 2010a. Linguistically relevant meaning elements of English communication verbs.
Belgian Journal of Linguistics 2. 54–82.
Boas, Hans C. 2010b. Comparing constructions across languages. In H.C. Boas (ed.),
Contrastive Studies in Construction Grammar, 1–20. Amsterdam/Philadelphia: Benjamins.
Boas, Hans C. 2011a. Zum Abstraktionsgrad von Resultativkonstruktionen. In S. Engelberg, K.
Proost, and A. Holler (eds.), Sprachliches Wissen zwischen Lexikon und Grammatik, 37–69.
Berlin/New York: Mouton de Gruyter.
Boas, Hans C. 2011b. A frame-semantic approach to syntactic alternations with build-verbs. In
P. Guerrero Medina (ed.), Morphosyntactic alternations in English, 207–234. London: Equinox.
Boas, Hans C. 2013a. Cognitive Construction Grammar. In T. Hoffmann and G. Trousdale
(eds.), The Oxford Handbook of Construction Grammar, 233–254. Oxford: Oxford University
Press.
Boas, Hans C. 2013b. Wie viel Wissen steckt in Wörterbüchern? Eine frame-semantische
Perspektive. Zeitschrift für Angewandte Linguistik 57. 75-97.

17
Boas, Hans C. 2013c. Frame Semantics and translation. In A. Rojo and I. Ibarretxte-Antuñano
(eds.), Cognitive Linguistics and Translation, 125–158. Berlin/New York: Mouton de Gruyter.
Boas, Hans C. 2014a. Zur Architektur einer konstruktionsbasierten Grammatik des Deutschen. In
A. Ziem and A. Lasch (eds.), Grammatik als Inventar von Konstruktionen? Sprachliches Wissen
im Fokus der Konstruktionsgrammatik, 37-63. Berlin/New York: de Gruyter.
Boas, Hans C. 2014b. What happened to Frame Semantics? Invited plenary lecture at the 8th
International Conference on Construction Grammar, Osnabrück, Sept. 3-6, 2014.
Boas, Hans C. 2017. Computational Resources: FrameNet and Constructicon. In B. Dancygier
(ed.), The Cambridge Handbook of Cognitive Linguistics, 549-573. Cambridge: Cambridge
University Press.
Boas, Hans C. and Ryan Dux. 2013. Semantic frames for foreign language education: Towards a
German frame-based dictionary. Veridas On-line. Special Issue on Frame Semantics and its
Technological Applications. 82–100. http://www.ufjf.br/revistaveredas/edicoes-2013
Boas, Hans C. and Francisco Gonzalvez-Garcia. 2014. Romance perspectives on Construction
Grammar. Amsterdam/Philadelphia: John Benjamins.
Boas, Hans C., Ryan Dux, and Alexander Ziem. 2016. Frames and constructions in an online
learner’s dictionary of German. In S. De Knop and G. Gilquin (eds.), Applied Construction
Grammar, 303-326. Berlin: de Gruyter.
Boas, Hans C. and Ivan A. Sag (eds.). 2012. Sign-based Construction Grammar. Stanford: CSLI
Publications.
Borin, Lars., Dannells, D., Forsberg, M., Toporowska Gronostaj, M., & Kokkinakis, D. 2010.
The past meets the present in the Swedish FrameNet++. 14th EURALEX International Congress,
269-281.
Bouveret, Myriam. 2012. GIVE frames and constructions in French. In: M. Bouveret and D.
Legallois (eds.), Constructions in French, 99–126. Amsterdam/Philadelphia: John Benjamins.
Burchardt, Aljoscha, Katrin Erk, Anette Frank, Andrea Kowalski, Sebastian Padó, and Manfred
Pinkal. 2009. Using FrameNet for the semantic analysis of German: annotation, representation,
and automation. In Hans C. Boas (ed.), Multilingual FrameNets: Methods and Applications,
209–244. Berlin/New York: Mouton de Gruyter.
Busse, Dietrich. 2012. Frame-Semantik. Ein Kompendium. Berlin/New York: Mouton de
Gruyter.
Croft, William. 2003. Lexical rules vs. constructions: A false dichotomy. In H. Cuyckens, T.
Berg, R. Dirven and K.-U. Panther (eds.), Motivation in Language: Studies in Honor of Günther
Radden, 49–68. Amsterdam/Philadelphia: John Benjamins.
Croft, William A. 2009. Connecting frames and constructions: A case study of eat and feed.
Constructions and Frames 1(1). 7–28.
Croft, William. 2012. Verbs. Oxford: Oxford University Press.
Croft, William. 2013. Radical Construction Grammar. In T. Hoffmann and G. Trousdale (eds.),
The Oxford Handbook of Construction Grammar, 211–232. Oxford: Oxford University Press.

18
Croft, William and Alan Cruse. 2004. Cognitive Linguistics. Cambridge: Cambridge University
Press.
Das, Dipenjan, Schneider, Nathan, Chen, Desai, and Noah A. Smith. 2010. Probabalistic frame-
semantic parsing. In Proceedings of HLT ‘10 Human Language Technologies: The 2010 Annual
Conference of the North American Chapter of the Association of Computational Linguistics,
948-956.
Das, Dipanjan. Desai Chen, André FT Martins, Nathan Schneider, and Noah A Smith. 2014.
Frame-semantic Parsing. Computational Linguistics 40(1). 9-56.
Dolbey, Andrew. 2009. BioFrameNet: A FrameNet extension to the domain of molecular
biology. Unpublished Ph.D. dissertation, University of California, Berkeley.
Dux, Ryan. 2011. A frame-semantic analysis of five English verbs evoking the Theft frame. M.A.
Thesis, The University of Texas at Austin.
Dux, Ryan. 2016. A usage-based approach to verb classes in English and German. Ph.D.
dissertation, The University of Texas at Austin.
Dux, Ryan. In press. Frames, verbs, and constructions: German constructions with verbs of
stealing. In: Alexander Ziem and Hans C. Boas (eds.): Constructional Approaches to Argument
Structure in German. Berlin/New York: Mouton de Gruyter.
Engelberg, Stefan, König, Svenja, Proost, Kristel, and Edeltraud Winkler. 2011.
Argumentstrukturmuster als Konstruktionen? In S. Engelberg, A. Holler, and K. Proost (eds.),
Sprachliches Wissen zwischen Lexikon und Grammatik, 71–112. Berlin/New York: De Gruyter.
Faulhaber, Susen. 2011. Verb valency patterns. Berlin/New York: Mouton de Gruyter.
Fellbaum, Christiane. 1990. English verbs as a semantic net. International Journal of
Lexicography 3(4). 278-301.
Fellbaum, Christiane (ed.). 1998. WordNet: An electronic lexical database. Cambridge, MA: The
MIT Press.
Fillmore, Charles J. 1968. The case for case. In E. Bach & R. Harms (Eds.), Universals in
Linguistic Theory, 1–90. New York: Rinehart & Winston.
Fillmore, Charles J. 1975. An alternative to checklist theories of meaning. Proceedings of the
First Annual Meeting of the Berkeley Linguistic Society (BLS), 123–131.
Fillmore, Charles J. 1976. Frame semantics and the nature of language. Annals of the New York
Academy of Sciences: Conference on the origin and development of language and speech, 280,
20–32.
Fillmore, Charles J. 1977a. Topics in Lexical Semantics. In P. Cole (ed.), Current Issues in
Linguistic Theory, 76–136. Bloomington: Indiana University Press.
Fillmore, Charles J. 1977b. Scenes-and-frames semantics. In A. Zampolli (ed.), Fundamental
Studies in Computer Science, 55–88. Dordrecht: North Holland Publishing.
Fillmore, Charles J. 1978. On the organization of semantic information in the lexicon. Papers
from the Parasession on the Lexicon, Chicago Linguistic Society, 148–173.
Fillmore, Charles J. 1982. Frame Semantics. In: Linguistics in the Morning Calm, ed. Linguistic
Society of Korea, 111–38. Seoul: Hanshin.

19
Fillmore, Charles J. 1985. Frames and the Semantics of Understanding. Quaderni di Semantica
6. 222–254.
Fillmore, Charles J. 1986. Pragmatically controlled zero anaphora. Proceedings of the Berkeley
Linguistics Society, 95–107.
Fillmore, Charles, J. 1987. A private history of the concept ‘frame.’ In R. Dirven and G. Radden
(eds.), Concepts of Case, 28-36. Tübingen: Gunter Narr Verlag.
Fillmore, Charles J. 1988. The mechanisms of “Construction Grammar”. Proceedings of the
Fourteenth Annual Meeting of the Berkeley Linguistic Society (BLS), 35–55.
Fillmore, Charles J. 2007. Valency issues in FrameNet. In T. Herbst and K. Götz-Vetteler (eds.),
Valency: theoretical, descriptive, and cognitive issues, 129–160. Berlin/New York: Mouton de
Gruyter.
Fillmore, Charles J. 2008. Border Conflicts: FrameNet meets Construction Grammar.
Proceedings of the XIII EURALEX International Congress (Barcelona, 15–19 July 2008), 49–68.
Fillmore, Charles J. 2013. Berkeley Construction Grammar. In T. Hoffmann and G. Trousdale
(eds.), The Oxford Handbook of Construction Grammar, 111–132. Oxford: Oxford University
Press.
Fillmore, Charles J. & Beryl T.S. Atkins. 1992. Toward a Frame-based Lexicon: The Semantics
of RISK and its Neighbors. In: A. Lehrer and E. Kittay (eds.), Frames, Fields and Contrasts:
New Essays in Semantic and Lexical Organization, 75–102. Hillsdale: Erlbaum.
Fillmore, Charles J. and B.T.S. Atkins. 2000. Describing polysemy: The case of crawl. In Y.
Ravin and C. Leacock (eds.), Polysemy: Linguistic and computational approaches, 91–110.
Oxford: Oxford University Press.
Fillmore, Charles J. and Colin Baker. 2010. A frames approach to semantic analysis. In B. Heine
and H. Narrog (Eds.), The Oxford Handbook of Linguistic Analysis, 313–340. Oxford: Oxford
University Press.
Fillmore, Charles J. and Paul Kay. 1993. Construction Grammar Course Book. UC Berkeley:
Department of Linguistics.
Fillmore, Charles J., Paul Kay, and Mary O’Connor. 1988. Regularity and idiomaticity in
grammatical constructions: The case of ‘let alone.’ Language 64, 501–538.
Fillmore, Charles J., Johnson, Chris, and Miriam Petruck. 2003. Background to FrameNet.
International Journal of Lexicography 16, 235–251.
Fillmore, Charles, Lee-Goldman, Russell, and Russell Rhomieux. 2012. The FrameNet
Constructicon. In: H.C. Boas and I. Sag (eds.), Sign-based Construction Grammar, 309–372.
Stanford: CSLI Publications.
Fontenelle, Thierry. 1997. Using a bilingual dictionary to create semantic networks.
International Journal of Lexicography 10(4). 275–303.
Forsbert, Markus, Johansson,Richard, Bäckström, Linnea, Borin, Lars, Lyngfelt, Benjamin,
Olofsson, Joel, and Julia Prentice. From construction candidates to constructicon entries.
Constructions and Frames 6(1). 114–135.

20
Heppin, Karin Friberg, and Håkan Friberg. (2012). Using FrameNet in communicative language
teaching. In Proceedings of the XV EURALEX International Congress. 640-647.
Gargett, Andrew, Josef Ruppenhofer, and John Barnden. 2014. Dimensions of metaphorical
meaning. In Proceedings of the 4th Workshop on Cognitive Aspects of the Lexicon
(COGALEX).166-173.
Gemmell, Maggie. 2015. Semantic role alignment in metaphor: A frame semantic approach to
metaphoric meaning. Unpublished Ph.D. dissertation, The University of Texas at Austin.
Gildea, Daniel, and Daniel Jurafsky. 2002. Automatic labeling of semantic roles. Computational
Linguistics 28(3). 245-288.
Goldberg, Adele. 1995. Constructions: A Construction Grammar Approach to Argument
Structure. Chicago: University of Chicago Press.
Goldberg, Adele. 2006. Constructions at work. Oxford: Oxford University Press.
Hanks, Patrick. 2012. Corpus Evidence and Electronic Lexicography. In: S. Granger and M.
Paquot (eds.), Electronic Lexicography. Oxford University Press.
Hasegawa, Yoko, Lee-Goldman, Russell, and Charles J. Fillmore. 2016. On the universality of
frames. In M. Hilpert and J.-O. Őstman (eds.), Constructions across Grammars, 34-66.
Amsterdam/Philadelphia: John Benjamins.
Hasegawa, Yoko, Lee-Goldman, Russell, Ohara, Kyoko Hirose, Fuji, Seiko, and Charles J.
Fillmore. 2010. On expressing measurement and comparison in English and Japanese. In H.C.
Boas (ed.), Contrastive Studies in Construction Grammar, 169–200. Amsterdam/ Philadelphia:
John Benjamins.
Heid, Ulrich. 1996. Creating a multilingual data collection for bilingual lexicography from
parallel monolingual lexicons. In Proceedings of the VIIth EURALEX International Congress,
Gothenburg 1996, 573–559.
Herbst, Thomas. 2011. The status of generalizations: Valency and argument structure
constructions. In T. Herbst and A. Stefanowitsch (eds.), Argument structure: valency and/or
construction? Special issue of Zeitschrift für Anglistik und Amerikanistik, 59(4). 347–367.
Herbst, Thomas. 2014. The valency approach to argument structure constructions. In Thomas
Herbst, Hans-Jörg Schmid, Susen Faulhaber (eds.), Constructions – Collocations – Patterns.
Berlin: de Gruyter Mouton.
Hermann, Karl Moritz. 2014. Distributed Representations for Compositional Semantics. Ph.D.
thesis, University of Oxford.
Hilpert, Martin. 2010. What can synchronic gradience tell us about reanalysis? Verb-first
conditionals in written German and Swedish. In: E.C. Traugott and G. Trousdale (eds.),
Gradience, gradualness and grammaticalization, 181-201. Amsterdam/ Philadelphia: John
Benjamins.
Hoffmann, Thomas and Graeme Trousdale (eds.). 2013. The Oxford Handbook of Construction
Grammar. Oxford: Oxford University Press.
Iwata, Seizi. 2008. Locative alternation: a lexical-constructional approach. Amsterdam/
Philadelphia: John Benjamins.

21
Jackendoff, Ray. 1990. Semantic structures. Cambridge, MA: The MIT Press.
Kay, Paul and Charles J. Fillmore. 1999. 1999. Grammatical constructions and linguistic
generalizations: The ‘What’s X doing Y?‘ Construction. Language 75. 1–33.
Lambrecht, Knud and Kevin Lemoine. 2005. Definite null objects in (spoken) French. In M.
Fried and H.C. Boas (eds.), Grammatical Constructions – Back to the Roots, 13–56.
Amsterdam/Philadelphia: John Benjamins.
Lasch, Alexander. 2016. Nonagentive Konstruktionen des Deutschen. Berlin/Boston: De Gruyter.
Lee-Goldman, Russell and Miriam Petruck. Submitted. The FrameNet Constructicon in action.
In: B. Lyngfelt, L. Borin, K. Ohara and T. Torrent (eds.), Constructicography: Construction
development across languages.
Leino, Jaakko. 2010. Results, cases, and constructions: Argument structure constructions in
English and Finnish, In: H.C. Boas (ed.), Contrastive Studies in Construction Grammar, 103–
136. Amsterdam/Philadelphia: Benjamins.
Levin, Beth. 1993. English verb class and alternations. Chicago: Chicago University Press.
Levin, Beth and Malka Rappaport Hovav. 2005. Argument Realization. Cambridge: Cambridge
University Press.
Lyngfelt, Benjamin. 2012. Re-thinking FNI. On null instantiation and control in Construction
Grammar. Constructions and Frames 4(1). 1–23.
Lyngfelt, Benjamin, Lars Borin, Markus Forsberg, Julia Prentice, Rudolf Rydstedt, Emma
Sköldberg and Sofia Tingsell. 2012. Adding a constructicon to the Swedish resource network of
Språkbanken. – In: Proceedings of KONVENS 2012, Vienna, 452-461.
Michaelis, Laura. 2013. Sign-based Construction Grammar. In T. Hoffmann and G. Trousdale
(eds.), The Oxford Handbook of Construction Grammar, 133–152. Oxford: Oxford University
Press.
Michaelis, Laura, and Josef Ruppenhofer. 2001. Beyond Alternations. Stanford: CSLI
Publications.
Miller, George A. 1995. WordNet: A lexical database for English. Communications of the ACM
38(11). 39–41.
Nemoto, Noriko. 1998. On the polysemy of ditransitive SAVE: The role of frame semantics in
Construction Grammar. English Linguistics 15, 219-242.
Nemoto, Noriko. 2005. Verbal polysemy and Frame Semantics in Construction Grammar: some
observations about the locative alternation. In M. Fried and H.C. Boas (eds.), Grammatical
Constructions. Back to the Roots, 119–138. Amsterdam/Philadelphia: John Benjamins.
Nilsen, Don L. F. 1973. The instrumental case in English: Syntactic and semantic
considerations. The Hague: Mouton.
Ohara, Kyoko. 2009. Frame-based contrastive lexical semantics in Japanese FrameNet: The case
of risk and kakeru. In: H.C. Boas (Ed.), Multilingual FrameNets: Methods and Applications,
163–182.
Ohara, Kyoko, Fujii, Seiko, Ohori, Toshio, Suzuki, Ryoko, Saito, Hiroaki, and Shun Ishizaki.
2004. The Japanese FrameNet Project: An introduction. In: Fourth International Conference on

22
Language Resources and Evaluation (LREC 2004). Proceedings of the Satellite Workshop
“Building Lexical Resources from Semantically Annotated Corpora”, 9–11.
Ohara, Kyoko. 2013. Toward Constructicon Building for Japanese in Japanese FrameNet.
Veredas 17(1). 11-27.
Osswald, Rainer and Robert D. Van Valin Jr. 2014. Framenet, frame structure and the syntax-
semantics interface. In R. Osswald T. Gamerschlag, D. Gerland and W. Petersen (eds.), Frames
and concept types: Applications in language and philosophy, 125–156. Heidelberg: Springer.
Petruck, Miriam R.L. 1996. Frame Semantics. In: Verschueren, J., J-O Östman, J. Blommaert
and C. Bulcaen (eds.), Handbook of Pragmatics, 1–13. Amsterdam/Philadelphia: Benjamins.
Petruck, Miriam R.L. 2009. Typological considerations in constructing a Hebrew FrameNet. In
H.C. Boas (Ed.), Multilingual FrameNets: Methods and Applications, 183–208. Berlin/New
York: Mouton de Gruyter.
Petruck, Miriam R.L. and Hans C. Boas. 2003. All in a Day's Week. In E. Hajicova, A.
Kotesovcova & Jiri Mirovsky (eds.), Proceedings of CIL 17. CD-ROM. Prague: Matfyzpress.
Petruck, Miriam R. L., Charles J. Fillmore, Collin Baker, Michael Ellsworth, and Josef
Ruppenhofer. 2004. Reframing FrameNet data. In Proceedings of the 11th EURALEX
International Congress, Lorient, France, 405–416.
Pinker, Steven. 1989. Learnability and Cognition. Cambridge, Mass.: MIT Press.
Rappaport Hovav, Malka and Beth Levin. 1998. 1998. Building Verb Meaning. In M. Butt & W.
Geuder (eds.), The Projection of Arguments, 97–134. Stanford: CSLI Publications.
Rosch, Eleanor. 1973. Natural categories. Cognitive Psychology 4. 328–350.
Ruppenhofer, Josef. 2004. The interaction of valence and information structure. Unpublished
Ph.D. dissertation, University of California, Berkeley.
Ruppenhofer, Josef & Laura A. Michaelis. 2010. A Constructional Account of Genre-Based
Argument Omissions. Constructions and Frames, 2. 158–184.
Ruppenhofer, Josef, Michael Ellsworth, Miriam R. L. Petruck, Chris Johnson, and Jan
Scheffczyk. 2010. FrameNet II: Extended theory and practice. Available at
[http://framenet.icsi.berkeley.edu].
Ruppenhofer, Josef, Boas, Hans C., and Collin Baker. 2013. The FrameNet approach to relating
syntax and semantics. In R.H. Gouws, U. Heid, W. Schweickhard, and H.E. Wiegand (eds.),
Dictionaries. An International Encyclopedia of Lexicography, 1320–1329. Berlin/New York:
Mouton.
Ruppenhofer, Josef, and Ines Rehbein. 2012. Yes we can!? Annotating the senses of English
modal verbs. Proceedings of the 8th International Conference on Language Resources and
Evaluation (LREC). 1538–1545.
Ruppenhofer, Josef, Boas, Hans C., and Collin Baker. In press. FrameNet. In P. A. Fuertes-
Olivera (ed.), The Routledge Handbook of Lexicography. London: Routledge.
Sag, Ivan A. 2010. English Filler-Gap Constructions. Language 86. 486-545.
Sag, Ivan A. 2012. Sign-based Construction Grammar: An informal synopsis. In: H.C. Boas and
I.A. Sag (eds.), Sign-based Construction Grammar, 69–202. Stanford: CSLI Publications.

23
Schmidt, Thomas. 2009. The Kicktionary – A multilingual lexical resource of football language.
In H.C. Boas (ed.), Multilingual FrameNets: Methods and Applications, 101–134. Berlin/New
York: Mouton de Gruyter.
Subirats, Carlos. 2009. Spanish FrameNet: A frame-semantic analysis of the Spanish lexicon. In:
H.C. Boas (ed.), Multilingual FrameNets: Methods and Applications, 135–162. Berlin/New
York: Mouton de Gruyter.
Subirats, Carlos and Miriam Petruck. 2003. Surprise: Spanish FrameNet! Proceedings of CIL 17.
CD-ROM. Prague: Matfyzpress.
Sullivan, Karen. 2013. Frames and constructions in metaphoric language.
Amsterdam/Philadelphia: John Benjamins.
Timyan, Napasri and Benjamin Bergen. 2010. A contrastive study of the caused-motion and
ditransitive construction in English and Thai: Semantic and pragmatic constraints. In H.C. Boas
(eds.), Contrastive Studies in Construction Grammar, 137–168. Amsterdam/Philadelphia: John
Benjamins.
Torrent, Tiago, Maria Margarida Salomão, Fernanda Campos, Regina M.M. Braga, Ely Edison
Matos, Maucha Gamonal, Julia Gonçalves, Bruno Souza, Daniela Gomes, and Simone Peron.
2014. "Copa 2014 FrameNet Brasil: a frame-based trilingual electronic dictionary for the
Football World Cup." In COLING (Demos), 10-14.
VanNoy, Annika. 2017. Culture specific aspects of semantic frames in multilingual frame
descriptions. Unpublished Ph.D. dissertation, The University of Texas at Austin.
Wierzbicka, Anna. 2006. English: Meaning and culture. Oxford: Oxford University Press.
Willems, Dominique. 2012. Verb typology: Between construction and lexicon. In: M. Bouveret
and D. Legallois (eds.), Constructions in French, 23–48. Amsterdam/Philadelphia: John
Benjamins.
Ziem, Alexander. 2008. Frames und sprachliches Wissen: kognitive Aspekte der semantischen
Kompetenz. Berlin/New York: Mouton de Gruyter.
Ziem, Alexander. 2014a. Von der Konstruktionsgrammatik zum FrameNet: Frams,
Konstruktionen und die Idee eines Konstruktikons. In: A. Lasch and A. Ziem (eds.), Grammatik
als Inventar von Konstruktionen? Sprachwissen im Fokus der Konstruktionsgrammatik, 263-
290.cBerlin: De Guyter.
Ziem, Alexander. 2014b. Frames of understanding in text and discourse: Theoretical
foundations and practical applications. Amsterdam/Philadelphia: John Benjamins.
Ziem, Alexander. 2015. Metaphors meet G-FOL. Zur Integration von Metaphern in eine Lehr-
und Lernplattform für DaF. In: M. Dalmas and E. Piirainen (eds.), Figurative Sprache, 201-220.
Tübingen: Stauffenburg.
Ziem, Alexander, Boas, Hans C., and Josef Ruppenhofer. 2014. Grammatische Konstruktionen
und semantische Frames für die Textanalyse. In: J. Hagemann and S. Staffeldt (eds.),
Syntaxtheorien. Analysen im Vergleich, 297-333. Tübingen: Stauffenburg.

24
Ziem, Alexander and Hans C. Boas. 2017. Towards a Constructicon for German. In Proceedings
of The AAAI 2017 Spring Symposium on Computational Construction Grammar and Natural
Language Understanding, Technical Report SS-17-02, 274-277.

Hans C. Boas
Department of Germanic Studies
Burdine Hall 336
2505 University Drive, C3300
The University of Texas at Austin
Austin, TX 78712
U.S.A.
hcb@mail.utexas.edu

Ryan Dux
Department of Languages, Cultures, and Linguistics
Bucknell University
1 Dent Drive
Lewisburg, PA 17837
U.S.A.
ryjodux@gmail.com

25