Available via license: CC BY 4.0
Content may be subject to copyright.
Second language acquisition
of depicting signs
A corpus-based account
Krister Schönström and Johanna Mesch
Stockholm University
This paper concerns the acquisition of the sign lexicon in L learners of
Swedish Sign Language. Sampled data (conversation and narrative
retelling) from a longitudinal learner corpus with adult L signers was
analyzed and compared with data from nine L signers. The use of three
broad types of signs was analyzed: lexical signs, partly-lexical signs (i.e.
depicting signs) and non-lexical signs. The results revealed some dierences
between L and L signers, especially with regard to depicting signs. The
number of depicting signs used by L learners increased over time,
approaching the target language use. Qualitatively, we observed dierences
between L and L signers in their use of depicting signs, related to
handshape choice and sign constructions. We discuss these ndings in light
of previous research linked to L vocabulary as well as the role of gestural
knowledge in sign L acquisition.
Keywords: sign language, LM, lexicon, sign types
. Introduction
The vast body of research on Second Language Acquisition (SLA) is based on
spoken languages, and comparatively little is known about the acquisition of sign
languages. A learner of sign language (typically a hearing person), has to learn
not only a new language (L), but also a new modality (M), i.e. the gestural-
visual modality of sign languages in contrast to spoken languages’ oral-auditory
modality (Chen Pichler, ; Marshall et al., ).
In the spoken SLA literature, numerous studies have stressed the importance
of research on L vocabulary acquisition in order to understand which factors
aect word-learning. Researchers have therefore assessed lexical knowledge in
dierent ways, for example, by measuring vocabulary size (how many words a
https://doi.org/10.1075/lia.22005.sch
Language, Interaction and Acquisition 13:2 (2022), pp. 199–230. ISSN 1879-7865 |E‑ISSN 1879-7873
Available under the CC BY 4.0 license. © John Benjamins Publishing Company
person knows) or vocabulary depth (how well a person knows a word) (for a
comprehensive description, see Nation, ). Another way of assessing L vocab-
ulary knowledge is to look at lexical sophistication or lexical richness by exam-
ining the lexical proles of L learners (e.g. Bardel et al., ; Laufer & Nation,
).
Aside from the lack of knowledge about how sign SLA operates given sign
languages’ modality-dependent (e.g. spatial) and language-dependent (e.g.
morphological) features, the analysis of sign SLA comes with some challenges.
One critical challenge is that sign linguistics research is relatively young, and
many issues linked to the sign language lexicon and grammar are understudied.
For example, in this context, research on Swedish Sign Language, Svenskt tecken-
språk (STS), lacks scientic resources such as access to data on STS vocabulary
breadth and depth.
However, within sign SLA, one area that has been under investigation during
the last decades is the sign language lexicon, i.e. the signs themselves. In this
context, so-called classier constructions, or as we label them, depicting signs,
have received particular attention (e.g. Ferrara & Nilsson, ; Marshall &
Morgan, ). Depicting signs have been treated as a less conventionalized cate-
gory of signs by several sign linguists. Although they use conventionalized hand-
shapes, these are combined with improvised locations and movements, in
contrast with lexical signs, in which the phonological components of handshape,
location, and movement are all conventionalized (Cormier et al., ; Schembri,
) (See Section for a further explanation). Depicting signs are seen as a
prominent part of the sign lexicon in many sign languages, including STS, and
as a category that is acquired late by L learners (e.g. Schick, ; Simper-Allen,
). Furthermore, depicting signs occur in most kinds of discourse genres but
are especially apparent in the narrative genre as they contribute to sophistication
in narratives (e.g. Morford & MacFarlane, ).
Another area that has been under investigation is the role of gestural knowl-
edge in sign SLA. Here, several researchers have argued that L learners’ gestural
knowledge inuences sign production (e.g. Ortega & Morgan a, b; Taub
et al., ).
We therefore explore how L signers use and develop the sign lexicon, with
a particular focus on their use of depicting signs, through a longitudinal corpus-
based account of the use of the sign lexicon in their STS production. To the best
of our knowledge, this developmental perspective on sign learning is unique, as
most of the results from earlier studies on L signers have been based on novice or
non-signers (cf. Schönström, ; see also Marshall & Morgan, ). With this
aim, even though our study is data-driven and merely exploratory and restricted
to the use of lexicon in STS, we hope to contribute to a better understanding of
Krister Schönström and Johanna Mesch
the status of signs, and in particular depicting signs, in relation to the entire sign
language lexicon of L learners. Furthermore, we hope to contribute to the area of
SLA in general.
. The sign language lexicon and depicting signs
As in most sign languages, the lexicon of STS consists of dierent subcategories of
signs. As well as the basic classication into content (open) and function (closed)
signs (equivalent to the wide and traditional concept and denition of words)
many sign languages share similar subcategories.1In the sign linguistics literature,
there is a set of dierent labels for dierent types of signs based on the physical
form of a sign and its degree of lexicalization; such types include lexical signs,
ngerspelling, pointing, and depicting signs. Signs can also be classied based
on their origins, i.e. so-called native vs. non-native signs (Brentari & Padden,
). In this paper, we borrow the terminology used for the Auslan Corpus by
Johnston and Schembri () and Johnston and Ferrara () (see also Hodge
& Johnston, ): fully lexical, partly-lexical and non-lexical signs. Taking as our
starting point a more functional division of the sign types found in our corpus
data, we are therefore able to adopt an exploratory and usage-based perspective
on the signs used by the L signers in our data. Lexical signs are manual fully
lexical signs – so-called “frozen” signs – which are listed in the STS dictionary
(Öqvist et al., ; Riemer Kankkonen et al., ). This category is most aligned
with the notion of “word” in spoken languages and is thus the most convention-
alized (Hodge & Johnston, ). Partly-lexical signs include signs that have some
degree of conventionalization in form and meaning, and may require a degree
of contextualization to be understood. Here, pointing signs and depicting signs
are counted as partly-lexical signs. The nal category, non-lexical signs, constitutes
gestures and other manual or non-manual acts that are the least conventionalized
in form and meaning.
The descriptions and labels of depicting signs have varied over time and
between researchers and dierent theoretical frameworks. In our article, we do
not attempt to summarize this research but instead refer the reader to earlier
extensive works (Cormier et al., ; Emmorey, ) which provide an
overview of research on depicting signs and their counterpart labels (such as
polycomponential signs, polysynthetic signs, and classier constructions), as well
as the dierent theoretical accounts for these signs. Within sign language studies
. It should, however, be noted that there are alternative views on this kind of categorization of
signs and lexicalization in sign languages in general (see e.g. Lepic, ).
Second language acquisition of depicting signs
there has been a great deal of research on depicting signs or so-called classier
constructions, with researchers describing them as complex morphosyntactic
constructions and similar to the classiers of spoken languages. This assumption
has been challenged by some researchers taking a more cognitive/functional view
(e.g. Liddell, ). In STS linguistics research, depicting signs (under the label
of polysynthetic signs) have been described as multi-morphemic constructions
(Wallin, ), using Talmy’s semantic model for motion verbs expressing loca-
tion and movement (Talmy, ).
In earlier descriptions of depicting signs, handshape types were a key compo-
nent in the depicting sign structure, and the number of handshape categories or
types has varied in the literature (Schembri, ; Simper-Allen, ). In our
study, we adopt Schembri’s () three main handshape types of depicting signs,
which have been consistently applied in the literature (although under dierent
labels): entity, handle, and descriptor handshapes, as outlined below.
Depicting signs with entity handshapes
Here the hands represent entities. The handshapes are determined by the proto-
typical shapes of the (animate and inanimate) entities being referred to. They can
be “whole entity” or “part entity” as well as objects or metaphorical representa-
tions of entities. In the STS data from the Swedish Sign Language Corpus (Mesch
& Wallin, ), the glossing of depicting signs including entity handshapes is
divided into two types: (Figure ) and (Figure ), diering
in terms of animacy. These signs can share the same handshape, e.g. an index
nger can be used to refer to ‘tree’ or ‘person’. This division was made in order to
easily nd signs representing animate/inanimate entities in the corpus.
Figure . ( ) ‘two persons are ying’ (SSLC_, ::.)
Krister Schönström and Johanna Mesch
Figure . ‘a car is coming’ (SSLC_, ::.)
Depicting signs with handle handshapes
Here the hands represent functions that are instrumental, manipulative or
handling dierent objects. The shape of the objects being handled can contribute
to the form of the handshapes. Also, the movement can depend on the kind of
handling expressed. An L signer describes how a man is rolling a newspaper,
Figure .
Figure . ‘a man is rolling a newspaper’ (SSLC_, ::.)
Depicting signs with descriptor (size- and shape-specifying) handshapes
The third category includes handshapes whose function is to describe the size and
shape of objects. Handshapes and movements represent the physical dimensions
of any object. In Figure , an L signer describes the shape of a mobile phone.
Figure . ‘the shape of a mobile phone’ (SSLC_, ::.)
Second language acquisition of depicting signs
As mentioned above, depicting signs are productive, partly-lexical signs,
as opposed to more conventionalized lexical signs. Another category which
has been investigated in terms of linguistic conventionalization is the manual
gestures used in both spoken and sign languages. A relevant question is whether
there are similarities and/or dierences between depicting signs (which are
less conventional than lexical signs) and speakers’ gestures. Several researchers
have investigated whether signers using depicting signs and non-signers using
gestures dier in formational aspects (e.g. Schembri et al., ). Schembri et al.
() concluded that handshapes in depicting signs were conventionalized but
not location and movement. Furthermore, Brentari et al. () compared the
handshapes of signers with gesturers and found, perhaps not surprisingly, that
although signs and gestures have similarities in formation, the handshapes used
by the signers were more complex phonologically. Another study compared the
symmetric use of handshapes and movements in two-handed signs, i.e. the so-
called Symmetry Condition, in signers and gesturers (Kita et al., ). The
symmetry condition has been suggested to be language-specic and part of
sign language phonology, but Kita et al. found no dierence in the use of this
condition in signers and gesturers suggesting that this is a matter of cognitive
constraint rather than being language-specic. Clearly, there are many questions
regarding the formational aspects of signs and how they are acquired. With this
as a backdrop, it is interesting to study from a developmental perspective how L
learners acquire and use the dierent types of signs, in particular those partly-
lexical signs such as depicting signs, in a relatively spontaneous and naturalistic
conversational situation.
. Sign second language acquisition
The discipline of sign SLA is young. While L signers were included in some indi-
vidual studies during the th century, it was not until the s that the eld
really began to contribute to the theory and practice of SLA. However, despite
the growing interest, research focusing on sign SLA is still very sparse (Boers-
Visker, ; Chen Pichler & Koulidobrova, ; Schönström, ). In general,
most studies involving L signers, beginner signers, and non-signers, have not had
these as key participants but as a control cohort. Oen, they have only been used
as comparison data in (mostly experimental) studies focusing on deaf L signers
for dierent purposes. Furthermore, the few such studies have mostly looked at
beginners or non-signers, with less focus on intermediate or advanced signers.
Here we will describe the studies most relevant to our study, as to our knowledge
no study has yet looked at lexical use of intermediate L signers in a broader sense.
Krister Schönström and Johanna Mesch
One experimental study looked at hearing L British Sign Language (BSL)
signers’ use of depicting signs (called in that paper ‘classier constructions’) and
compared their use with that of non-signers (Marshall & Morgan, ). The
authors found that the L signers showed a high rate of depicting signs (with a
focus on entity classiers) but that the signs oen included formational errors.
In addition, they showed that the learners had diculties with the handshape
parameter. In contrast, the location parameter was more accurately used, and
suggested that dierent error rates in the parameters can be linked to iconic
mapping and conventionalization, i.e. locations are more transparent iconically
than handshapes, which are less transparent and more conventionalized in form.
In a study on L acquisition of Norwegian Sign Language (NTS), Ferrara &
Nilsson () explored the production and use of depicting signs in L NTS
students through elicited tasks, and found that learners preferred lexical signs
over depicting signs. The authors showed that learners’ use of depicting signs was
lower, at , in comparison to the L group at . In addition, L signers most
oen used signs depicting location entities, while L signers most oen used signs
depicting movement of entities, as well as size and shape of the entities. In addi-
tion, the authors observed that the L signers tended to rely on lexical signs rather
than depicting signs in the signing space. The task was however more focused on
describing spatial layouts, and the authors found that L signers struggled speci-
cally with the placement and coordination of signs in the signing space (Ferrara &
Nilsson, ). In one recent study on Sign Language of the Netherlands (NGT),
Boers-Visker and colleagues studied L students’ production of modality-specic
devices, i.e. the use of space for linguistic information (Boers-Visker, ; Boers-
Visker & van den Bogaerde, ). They noticed that depicting signs (labeled as
classier predicates) appeared relatively late in a free production study of two
NGT L learners (Boers-Visker & van den Bogaerde, ). In a follow-up study,
Boers-Visker () administered a production task to elicit depicting signs from
a group of NGT L learners, and concluded that depicting signs appeared at
early stages in the acquisition, although the formational characteristics of these
signs were erroneous and inconsistent. However, the learners showed a signicant
development at the end of the rst year of their sign language education.
The L acquisition and use of a related linguistic structure, constructed action
(or enactment), has been studied for British Sign Language (BSL), American Sign
Language (ASL) and Quebec Sign Language (LSQ). Gulamani et al. (), for
BSL, found that there are dierences regarding the use of constructed action, and
more specically viewpoints, in L learners compared to deaf L signers. BSL
learners use character viewpoint, which requires use of constructed action, less
oen than L signers, and they spend more time in observer viewpoint compared
to the L signers. Observer viewpoint was therefore suggested to be easier to
Second language acquisition of depicting signs
acquire for the learners (Gulamani et al., ). In this context, their results are
interesting and relevant to this study, as the use of observer viewpoint is more
reliant on the use of depicting signs. Furthermore, Gulamani et al. observed that
the learners did not use the same conventional handshapes in the depicting signs
as the L signers. For ASL, Kurz et al. (), in a small qualitative study of L
learners, found that the use of observer viewpoint seems to increase with time.
The authors suggest that this may be a result of greater sensitivity to depicting
signs (as in observer viewpoints) in L ASL signers compared to co-gesture-like
constructions as constructed action (as in character viewpoints). Sanders and
Parisot () found that L signers of LSQ use constructed action less frequently
compared to L signers.
Transfer or crosslinguistic inuence as a concept is widely discussed in SLA
(see Jarvis for an overview). Not surprisingly, sign SLA provides no excep-
tion; there has been much discussion of the role of iconicity and learners’ gestural
knowledge in sign L acquisition with respect to the issue of transfer (Chen
Pichler & Koulidobrova, ), specically considering what structures are
“transferable” between spoken languages and sign languages and to what degree.
On the one hand, given the dierent physical forms of the phonology in spoken
languages (sound-based) and sign languages (visual-based), some researchers
have concluded that the transfer of phonology between L spoken language and
L sign language is not possible (Bochner et al., ; Rosen, ). On the other
hand, researchers have discussed the possibility of a transfer of gestural knowl-
edge to the signed modality. Several studies have pointed out gestural knowledge
as an important source for sign L acquisition (Chen Pichler, ; Ortega &
Morgan, b). According to Ortega and colleagues, learners’ gestural knowledge
may facilitate their sign language production, but also contributes to formational
errors in the phonology of signs (Ortega, ; Ortega & Morgan, a).
In comparison to spoken languages, sign languages are argued to have a
higher degree of iconicity: that is, a large proportion of signs and linguistic
structures of sign languages are more or less transparent between the linguistic
form and its referent, which has consequences for sign language acquisition (see
Ortega, for an overview). Ortega and Morgan (a) studied hearing novice
signers performing a sign repetition task, and found a higher accuracy in arbitrary
signs compared to iconic signs in BSL L signers. This was explained by the
fact that the learners, despite their “gestural advantage”, transferred their gestural
knowledge in the production of iconic signs, leading to less attention to the forma-
tional parts of the target signs, while with arbitrary signs, they have to pay greater
attention to their formational parts in order to be able to sign them, leading to
better phonological accuracy. Studies on brain activation through Event Related
Potentials (ERP) in sign language comprehension tasks conrm that hearing
Krister Schönström and Johanna Mesch
learners at the onset of learning transfer their gestural knowledge as “semiotic
resources” to L sign language (Ortega et al., ).
In the early days of sign language linguistics, depicting signs were described
as multimorphemic or polymorphemic by some researchers. That is, they are
complex signs containing several morphemes, the use of which characterizes a
higher degree of linguistic competence. With this in mind, we consider acqui-
sition and use of signs in L signers to be of particular interest, given on the
one hand the role of vocabulary in sign language lexicon/morphology, and on
the other hand the role of morphology in SLA research. For STS there are now
several ongoing studies on sign SLA using corpus-based approaches (Mesch &
Schönström, , ; Schönström & Mesch, ). The present study is among
the rst of this kind.
. Purpose of the study
In this study, we aim to investigate the sign acquisition of adult L signers of
STS, according to the sign types that constitute the STS lexicon. We are interested
in particular in types of signs, such as depicting signs, that are not traditionally
dened as fully lexical signs, but which are described as partly-lexical, i.e. a
productive category of signs that need contextualization to be understood fully.
In addition, we provide a comparison to a control group of L signers. As well as
presenting results in terms of frequencies and distributions, we provide insights
into the use and development of L signers’ sign production.
We aim to answer the following questions:
– What are the patterns of longitudinal change in the use of signs in the L
group?
– How are partly-lexical signs such as depicting signs used and developed in
relation to the other sign types in the STS lexicon?
– What similarities and dierences do the L group exhibit in the sign produc-
tion compared to the L signers?
. Data and methodology
Our starting point is a functionalist perspective. We perform a data-driven,
descriptive and functional account exploring and analyzing sign production and
interlanguage, namely lexical production, of L learners, using a corpus-based
approach, i.e. Contrastive Interlanguage Analysis (CIA; Granger, ). Within
Second language acquisition of depicting signs
the CIA framework not only are the L learners compared with L signers, but
also L learners are compared at dierent stages, or between learners of dierent
prociency levels, providing a broader longitudinal picture of interlanguage vari-
eties over time.
The results presented in this article are based on a sample of video data from
the Swedish Sign Language as an L Corpus (STS L Corpus), which is a learner
corpus, and to date, unique in L sign language research (Mesch & Schönström,
, ; Schönström & Mesch, ). The whole corpus consists of a set of
longitudinal data of adult L-learners’ signed production. In total, it contains
: hours of data from learners, along with a control cohort consisting of
L signers (: hours). The learners are hearing students in a three-year sign
language interpreting program at a Swedish university.
Data were collected on four occasions over a period of a year and a half. In the
recordings, the learners are asked to participate in a short conversation with an L
user of STS, to retell stories based on a variety of dierent pictures, and lastly to
retell a short narrative from a one-minute clip from the British short movie “The
Plank”.2In this study, two kinds of data extracted from the corpus were included:
conversation data from interviews with the participants, and retelling data (“The
Plank”).
The sample comes from a total of learners, on four occasions: one month
aer course onset T (N=) (not including the retelling data because of the
learners’ limited sign competence), four-ve months aer onset T (N=), nine
months aer onset T (N=) and months aer onset T (N=). All
learners had no prior STS knowledge when they enrolled at university; see
Table . Average hours of sign language instruction were approximately , ,
and hours for T, T, T and T, respectively. Comparisons to an L
cohort (nine uent signers) were made; see Table . All the L signers acquired
STS from early childhood, had attended Swedish deaf schools, and are members
of the Swedish deaf community.
In total, the sample includes , sign tokens in the conversation data, and
, sign tokens in the retelling data.
. The Plank (Sykes & Penington, ) is a cult classic short and near silent lm with very
little dialogue. The one-minute clip we are using shows two men carrying, with diculty, a long
wooden plank through streets and alleys. At one point they manage to get one end of the plank
through a bar window where a man, unnoticed, manages to put his glass of beer on the plank
that is sticking into the window, whereupon the glass disappears o the window, falling outside
into a bowl with water intended for window cleaning. Shortly aer, the window cleaner arrives,
picking up the glass with confusion, and the beer man notices him holding the beer glass and
gets angry, as he thinks the window cleaner has drunk the beer. He then pulls him into the bar
through the window.
Krister Schönström and Johanna Mesch
Table . Participants in the L cohort
L group Age Gender Session Total length (mm:ss)
S F T, T :
S F T, T, T :
S F T :
S F T, T, T, T :
S F T, T, T, T :
S F T, T, T :
S F T, T, T, T :
S F T, T, T :
S F T, T, T, T :
S F T, T, T, T :
S F T, T, T, T :
S M T, T, T, T :
S M T, T, T :
S F T, T :
S F T, T, T, T :
S F T, T, T, T :
::
Table . Participants in the L cohort
L group Age Gender Total length (mm:ss)
S F :
S M :
S F :
S F :
S F :
S F :
S F :
S F :
S F :
::
Coding was done with the multimodal annotation soware ELAN . ().
Lexical signs and other sign types such as pointing, ngerspelling, depicting signs,
noun classiers, buoys, palm-up gestures, and gesture-like signs are annotated
using the annotation conventions for STS (Mesch & Wallin, , ). Besides
these glossing categories, there are also tags for various pragmatic segments such
as incomplete sign articulations, hesitation pauses and sign holds. However, target
sign glosses were selected regardless of the produced form, that is, even if they
Second language acquisition of depicting signs
come with learners’ varieties (i.e. phonological, lexical varieties or other devia-
tions) (cf. Mesch & Schönström, ). The STS L corpus does not report on
the annotator reliability of the coding. However, the research assistants followed
the annotation conventions, and the senior researcher, who was responsible for
the coding work, checked the annotation work regularly. Table shows the coding
scheme of the sign categories.
Table . Coding scheme
Category Sign type Tag example Description
Fully lexical signs Lexical signs Conventionalized lexical signs
Fingerspelling @fs Signs based on manual alphabets
of spoken words
PRO PRO Self-pointing towards the signer’s
chest, ‘rst person pronoun’
Partly lexical
signs
Depicting signs @p
@p or
@p
Context-bounded productive signs
Pointing Pointing in the signing space for
referential information
Non-lexical signs Gestures @g Gestures, e.g. palm-up
Furthermore, three broad handshape types of depicting signs, as described
above, are entity handshapes, handle handshapes, and descriptor handshapes. In
the STS corpus, annotation of entity handshapes is divided into two types for
practical purposes, and , based on their animacy, i.e. human
vs. non-human (see Figure and Figure above).
. Results
. Analysis of the sign types
Tables and show the raw and normalized frequencies of sign types for lexical
signs, ngerspelling, pointing, where pointing representing rst-person pronoun
is counted separately, and depicting signs for conversation and retelling data,
respectively. As the number of gestures in our data was very low (a total mean of
. and . for conversation and retelling data, respectively), we will not account
for these. The rst main column accounts for the mean based on raw frequen-
cies, and the second main column accounts for the normalized frequencies based
on total signs (per signs). As can be seen in Tables and , the number of
depicting signs is much smaller in the conversation data than in the retelling data
Krister Schönström and Johanna Mesch
for the L groups, where the participants are describing events from “The Plank”
with a higher frequency of co-occurrence of body movement and non-manual
expression. We consider this to be typical of the two dierent genres.
Table shows that lexical signs are a dominant sign type in the conversation
data with a total mean of . for T–T and L. Next to lexical signs are
(total M=.), ngerspelling (.) and pointing (.). Depicting signs
are completely or almost completely absent from the conversation data, which is
expected of this discourse type. A Welch’s was performed to compare the
eects of groups on normalized frequencies of sign categories, as well as on raw
frequencies of total signs, using SPSS . The analysis revealed that there was a
statistically signicant dierence between at least two groups only for Total signs
(F(, .) = [.], p<.), but not for the sign types. A post-hoc analysis
using a Games-Howell test shows that means for total signs was signicantly
dierent between T and T (p<., CI =[−., −.]), T and T
(p<., CI = [−., −.]), T and T (p= ., CI =[−.,
−.]), and T and L (p= ., CI=[−., −.]). For pointing,
there was a marginally signicant dierence (F(,.)=[.], p=.). To
sum up the conversation data, we could conrm an increase in the number of
signs in learners’ production, but not dierences between the groups in using
dierent sign types; at least, those dierences were not statistically signicant.
However, we can note a tendency to an increase in the use of pointing in the
learners with time. In addition, the sign category ngerspelling deserves a sepa-
rate mention. Although statistical evidence is absent, ngerspelling is used mostly
in T but decreases with time, starting as soon as T. Thus, ngerspelling seems to
be a strategy for novice L learners to supplement their limited sign vocabulary in
the beginning. Fingerspelling as an initial lexical strategy has also been conrmed
for L signers of Irish Sign Language (Leeson et al., ).
Table . Raw and normalized frequencies of sign types by group in the conversation data
Raw frequencies
Normalized frequencies
(per signs)
CI
N M SD Min Max M SD S.E. Lower Upper
Pointing L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Second language acquisition of depicting signs
Table . (continued)
Raw frequencies
Normalized frequencies
(per signs)
CI
N M SD Min Max M SD S.E. Lower Upper
PRO L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Fingerspelling L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Depicting signs L-T . . . . . . .
L-T . . . . . −. .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . −. .
Total . . . . . . .
Lexical signs L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Total signs L-T . .
L-T . .
L-T . .
L-T . .
L . .
Total . .
Krister Schönström and Johanna Mesch
Table . Raw and normalized frequencies of sign types by group in the retelling data
Raw frequencies
Normalized frequencies
(per signs)
CI
N M SD Min Max M SD S.E. Lower Upper
Pointing L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
PRO L-T . . . . . −. .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . −. .
Total . . . . . . .
Fingerspelling L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Depicting signs L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Lexical signs L-T . . . . . . .
L-T . . . . . . .
L-T . . . . . . .
L . . . . . . .
Total . . . . . . .
Total signs L-T . .
L-T . .
L-T . .
L . .
Total . .
Table shows that lexical signs are used in more than half of the sign types in
the retelling data. In comparison with the conversation data, other sign types
are more apparent here. Regarding the normalized frequencies of the signs, there
are some notable dierences between the L group and the L group, especially
between learners at T and L signers. Overall, signers seem to be consistent in
their lexical use pattern, especially in the beginning. However, the use of lexical
Second language acquisition of depicting signs
signs falls from T to T, approaching the target language pattern i.e. that of the
L signer group. At T the learners exhibit a greater number of pointing signs and
a somewhat greater number of depicting signs. In comparison to conversation
data, depicting signs are used more frequently in the L group. The narrative style
of “The Plank” stimulus obviously elicits the use of depicting signs. L signers
may also be aware of the use of depicting signs in narratives as an eect of their
education. Depicting signs seem to increase with time from T to T. Still, the L
group exhibited greater use of depicting signs. A Welch’s was performed
to compare the eects of groups on normalized frequencies of sign categories,
as well as on raw frequencies of total signs. The results revealed marginally
signicant dierences between at least two groups in the categories depicting
signs (F(, .) = [.], p=.), and lexical signs (F(, .)=[.],
p=.) and a signicant dierence in total signs (F(,.)= [.],
p=.). For multiple comparisons, A Games-Howell post-hoc test shows that
there was a statistically signicant dierence between L and L-T (p=.),
CI (., .) in depicting signs and between L-T and L (p=.,
CI [., .]) in lexical signs. Furthermore, there was a signicant dierence
between L and L-T in total signs (p=., CI [., .]). Although
we see dierences in normalized frequencies of the sign categories, statistical
evidence is limited, probably as a result of a small sample size. However, there
seems to be a tendency for dierences in the categories depicting signs and lexical
signs. This will be explored further in the next sections, through more qualitative
approaches to the data.
. Distribution of the subcategories of depicting signs
In order to investigate the variation in the use of handshape units in the depicting
signs we present the distribution of subcategories of depicting signs across the
groups: (entity handshapes), (handle handshapes), and
(size and shape descriptive handshapes). Proportions in the use of
these depicting sign categories are shown in Table . As depicting signs occur
rarely in the conversation data, we only consider the retelling data here. Overall,
there are similarities between the groups, i.e. in both L and L, is the
most frequent category followed by and . There seems to be
a tendential dierence between L signers and L signers, in particular in the use
of and handshapes, but due to small numerical dierences this
could not be conrmed statistically.
Krister Schönström and Johanna Mesch
Table . Average frequencies (number/participant) and proportion for depicting sign
categories , and at group level (Retelling data)
Depicting signs T T T L
. (.) . (.) . (.) . (.)
. (.) . (.) . (.) . (.)
. (.) . (.) . (.) . (.)
Total . (.) . (.) . (.) . (.)
. Exploring the use of depicting signs
Below we describe some qualitative observations on L learners’ production in
the retelling data. We have found some structures common in L depicting sign
production: () using handle handshapes in contexts where entity handshapes
should be the target form, () choice of handshapes, and () depicting signs as
anchors. Thus, we provide information on some learner variation in the produc-
tion of depicting signs in particular contexts that were not observed in the L
group. It should be emphasized that the examples are analyzed according to their
use in specic contexts, and these examples represent our observations in the
data.
In the gures, Swedish Sign Language (STS) signs are shown using their sign
gloss in small capital letters. All examples are taken from the STS L corpus with
a code for L or L, a unique code for a recording session, an identier for the
participant, for example S, and a start-time for the sequence, for example
SSL_L__S, ::..
.. Handle handshapes instead of entity handshapes
In our distribution data above, we saw a tendency for small dierences between
L and L signers with respect to handshape choices regarding and
handshapes. In our qualitative analysis of the data, we noted that the L
signers made some choices that dier from the L signers with respect to partic-
ular handshape choices. We observed examples of the learners using handshapes
(and movements) typically used for handling (‘open a window’) to express entity
contexts (‘an open window’), see Figure . In the stimulus data (“The Plank”)
there was no such handling context with somebody opening the window. L
describes the open window with the entity handshape at hand, Figure .
In another context (Figure ) the participant intends to produce (as a
noun) using a handle handshape to represent the plank. This is emphasized by
the repeated movement in the shoulders, as if she is carrying the plank. L signers
tend to use entity handshapes in such contexts as shown in Figure .
Second language acquisition of depicting signs
Figure . The T participant describing an ‘open window’ using a handle handshape.
(SSL_L__S, ::.)
Figure . The L participant describing an ‘open window’ using an entity handshape.
(SSL_L_, ::.)
Figure . The T participant expressing the plank (as noun) using handle handshapes
(grip hand). (SSL_L__S, ::.)
Figure . The L participant expressing the movement of the plank using the entity
handshape (at hand). (SSL_L__S, ...)
Krister Schönström and Johanna Mesch
.. Choice of handshapes
Next, we analyzed the use of dierent handshapes and constructions in the corpus
data. The list with the number and distribution of handshapes is included in the
supplementary les. We observed some dierences between the L and L groups
in the choice of handshape as well as in the construction of depicting signs, and
will account for some of these dierences here. First, we observed a dierence
in usage with respect to two depicting sign types: (representing a human
being) and . Table shows the variations used by the L learners as well as
L signers and how the entity handshape for ‘human being’ is distributed between
the L learners as well as when compared to L signers. Table also shows the use
of the non-dominant hand in every signer group, where the signers use the non-
dominant hand independently (apart from the dominant hand, which is the right
hand for right-handed signers). The L group uses the handshape index nger/s
(Band BB) for entities representing human beings, e.g. ‘two men walk along’. Such
handshape and depicting sign types are not found in T-T for human beings.
Instead, the handshape Yis used for the entities representing human beings in all
L groups; however this was not observed in the L group. This handshape is also
used for the lexical sign . We also found examples of deviations in the choice
of handshape. For example, the use of the deviating handshape in the T group
when the signer attempts to depict a person. This could be due to the learners’
diculty in forming and producing a signed utterance in a short time.
The handshape 6representing occurs strongly in the L group for
both dominant (DH) and non-dominant hands (NonDH), e.g. in the context
when the signer depicts how a man in the movie holds a beer glass (using one
hand (e.g. the dominant hand) when reading a newspaper with the other hand
(e.g. the non-dominant hand). The L group seems to prefer the use of a hand
when describing holding a glass. This hand is also used in the lexical sign for
.
For the description of carrying the plank, the L signers seem to alternate
between dierent handshapes and seem to use a 6hand to express carrying the
plank. Using that handshape indicates handling an entity that is round and thin
in shape. We suggest that this is inuenced by the lexical signs or possibly
, which both use a 6hand.
Learners use also entity handshapes with a handshape or orientation that
represents a dierent meaning. Figure shows a participant expressing the plank
(as an entity) using two at hands with upright orientation. This could be inter-
preted as emphasizing a specic open upper surface or handshape as in “give
anything”. Compare this to the L signer’s way of describing the path of the plank
into the window, Figure .
Second language acquisition of depicting signs
Table . Examples of usage of depicting signs for (human being) and
T_DH
T_NonDH
T_DH
T_NonDH
T_DH
T_NonDH
L_DH
L_NonDH
(human being) (one-handed)
(human being) (two-handed)
(human being)
(human being)
(one-handed)
(two-handed)
Figure . The T participant trying to express using a depicting sign and upright
orientation of at hand. (SSL_L__S, ::.)
Some learners seem to nd using a depicting sign as a target construction
challenging. We have observed that these learners can be somewhat ambivalent in
using/choosing either a lexical sign or a depicting sign. The end result looks like a
blend between a lexical sign and a depicting one, as described above, for example,
with ‘walk’ and + ‘a person is walking’.
Krister Schönström and Johanna Mesch
Figure . The L participant describes how a plank moves into the window.
(SSL_L__S, ::.)
We have also observed further examples of the inuence of lexical sign forms
on the production of depicting signs. For example, in Figure we see a learner
using a sign with formal properties from the STS lexical sign for ‘wall’ but
modied in the movement part as a depiction of the characteristics of a ‘wall’.
At the same time, the learner is mouthing the Swedish word [vägg] ‘wall’ which
is associated with the lexical sign ‘wall’. In our data we annotated this as a
depicting sign following the target form principle. Interestingly, the L signers do
not use any depicting sign for ‘wall’, only a lexical sign ‘window’ when
the plank is carried through the window.
Figure . The T participant describing a ‘wall’ using a blend of a lexical form and a
depicting form. (SSL_L__S, ::.)
It seems to be dicult for the learners to drop the use of lexical signs. There
are also many cases in which the learners alternate between trying to express
depicting signs but add lexical signs as well, doubling the information. As
Figure illustrates, the learner (T) shows doubt while trying to express how the
men in the movie moved the plank from the window (using a handle handshape).
She repeats her depicted sign but adds a mouthing based on the Swedish word
[ytta] ‘move’ and then the lexical sign for ‘move’, again while mouthing
[ytta] ‘move’. The handshapes for these signs are similar but the lexical sign
has a strict phonological construction with rightward and leward palm orienta-
Second language acquisition of depicting signs
tion. The total use of the lexical sign for the L group is zero, whereas we
found it in six of the L signers. We found similar examples of lexical sign use for
other lexical entries as well, such as [gå] ‘to walk/to go’, where L learners
show cases in comparison to L signers with six. We believe this is a kind of
lexical transfer from their L Swedish, as lexical signs are the closest linguistic
device to Swedish words, at least conceptually.
Figure . Sequence in which the T participant alternates between depicting signs and
lexical signs. (SSL_L__S, ..)
.. Simultaneous use of two depicting signs with two hands
A prominent part of sign languages is simultaneity: using both hands at the same
time to express dierent linguistic information, see Figure .
However, a qualitative look beyond the number of depicting signs using the
NonDH reveals that using both hands and dierent handshapes as two depicting
signs or in combination with lexical signs are complex constructions for L
learners, rarely observed in our data. First, it is probably in part associated with
modality-dependent skills: it is dicult to combine the dominant and non-
dominant hands in the production of two depicting signs (cf. Boers-Visker,
and Ferrara & Nilsson, ). Second, non-dominant hand handshapes are oen
mirrored by dominant hand handshapes or are dierent from target handshapes.
For example, Figure illustrates an anchor handshape holding a newspaper
using a handshape (as in holding a thick entity) (target language form: Ior
at hand as in Figure and ), at the same time as the dominant hand is also
using a hand (as in drinking beer, holding the glass). L signers clearly use a 6
hand for holding the beer glass and a xat hand representing the newspaper. The
use of constructed action adds spatiality and complexity to the signing process, a
point which has been conrmed in earlier studies (Ferrara & Nilsson, ; see
also Gulamani et al., ).
Krister Schönström and Johanna Mesch
Figure . The L participant’s simultaneous use of DH and NonDH depicting holding a
beer glass and a newspaper in the last picture. (SSL_L__S, :::)
Figure . The T participant uses the dominant hand depicting holding a beer glass and
the non-dominant hand holding a newspaper. (SSL_L__S, ::.)
Figure . The L participant uses the dominant hand depicting holding a beer glass and
the non-dominant hand representing a newspaper. (SSL_L__S, ::.)
Second language acquisition of depicting signs
. Discussion
Learning a sign language with its modality for the rst time as an L can be a
dicult task for many sign language students. Worth mentioning too is that L
signers oen undergo a relatively formal training rather than informal learning;
they rarely learn to sign and develop their sign language outside the classroom,
for practical reasons (there being few L signers available). Through our corpus
data, we have provided a window into L learners’ production of signs. By means
of our longitudinal corpus, we have shown quantitative outcomes in distribution
and frequency of acquisition of the sign lexicon. Specically, we have provided
an account of the use of depicting signs in the learners’ interlanguage at certain
points in their development, as well as by means of a qualitative description of the
use of depicting signs.
Regarding the rst question in the study on frequencies and patterns of sign
type distribution and how it changes over time in the L group, we identied
some interesting results pertaining to how variation in sign type distribution
changes over time. In conversation data, where the learners were at a more novice
stage (T), they adopted a strategy of using ngerspelling, due to a limited sign
vocabulary. We also noted a higher frequency of rst-person pronoun ()
among L signers in earlier stages (T and T). It is a possible result of crosslin-
guistic inuence from L Swedish, as Swedish in comparison to STS has a stricter
obligation of subject marking in clauses. However, this needs to be further
explored in detail in another study, which would include a syntactic analysis.
Furthermore, even if depicting signs occur rarely in the conversation data, it is
notable that L learners at T did not use depicting signs at all. In retelling data, we
observed a change mainly in the use of the categories depicting signs and lexical
signs, as well as pointing. L learners at T used lexical signs to a greater extent,
and depicting signs and pointing to a lesser extent. However, the frequencies of
depicting signs and pointing increased at T and T as well, while at the same
time the use of lexical signs decreased. Concerning depicting signs and lexical
signs, the dierences within the L group were not statistically signicant, and
only marginally between the L learners at T and L signers. Taken together, we
would suggest that although statistical evidence is limited, based on this study’s
data L learners seem to develop their use of the sign lexicon with time, and the
patterns in the lexicon approach the patterns of the target language as used by
L signers. Our interpretation is that the L learners stick at rst to a “one word-
one sign” pattern, more similar to their L Swedish, through the use of lexical
signs and ngerspelling, rather than using partly-lexical signs such as pointing
and especially depicting signs, a prominent and more sophisticated part of sign
language lexicon. For example, we observed in our data that they tend to use
Krister Schönström and Johanna Mesch
multi-word constructions as [‘put on’], using a verb + preposition phrase
to express movement and placement which is prevalent in L Swedish (i.e. “lägga
på”), but that in STS would typically consist of a single depicting sign used as a
verb. Although there are equivalent signs for the verb (‘put’) and the preposition
(‘on’), they are not used in the same manner, which understandably causes di-
culty for the learner. We would suggest that this may be closely related to the inter-
lexical factor of vocabulary diculty, i.e. the conceptual similarities between an
L word and the new L word (sign) do not necessarily overlap semantically and
in use, which causes diculty for the L learner (cf. Laufer, ).
The second question of the study was how partly-lexical signs such as
depicting signs are used and developed in relation to the other sign types in
the STS lexicon. Overall, in the retellings, L signers seem to relatively quickly
approach the target pattern of L signers with respect to sign type distribution.
L signers use depicting signs relatively frequently and according to a distribution
similar to that of L signers, especially aer one year of instruction. This seems
to be in line with previous ndings for L NGT (Boers-Visker, ). A possible
explanation for this may be that depicting signs are conventionally “weaker” and
closer to the formational patterns of gestures. In addition, there may be a matter
of methodological issues related to the glossing of the signs in the annotation
process; the line between depicting signs and gestures can be somewhat blurred,
especially when annotating, as it can be hard to identify to what type some signs
belong, especially signs that involve handling contexts, i.e. “acting”. For example,
some of the L forms seem to be more of a gesture than an actual depicting sign,
due to their vague phonological constructions, involving indistinct handshapes
and movements, as well as dierent orientation. This is comparable with earlier
studies on signers and gesturers in the use of handshapes (Brentari et al., ).
One example is when one participant used a spread hand with indistinct move-
ment instead of a proper at hand or holding hand for ‘carrying the plank’ with
clear movement, and we had to decide whether to annotate it as a depicting sign
or a gesture. In this particular case, we annotated it as a depicting sign following
the target language denition. This may have contributed to higher quantities of
the depicting signs in the L group, especially for the early learners. Nevertheless,
our ndings show that L signers use depicting signs to a lesser extent, especially
at the beginning, possibly because they struggle to learn these signs.
In addition, our qualitative analyses revealed some dierences in formational
parameters of depicting signs between L and L signers with regard to the use
of depicting signs. The use of depicting signs among L signers is characterized
by a more varied palette of handshapes and movements used for dierent mean-
ings in comparison with the L control group, who are more consistent in their
use. This is oen associated with specic sequences in the movie “The Plank” and
Second language acquisition of depicting signs
the ways in which events are expressed. Among other things, the L group varied
more in depicting signs in the choice of handshapes when compared to the L
group. seems to be one of the types which is easier to acquire and use. An
explanation for this was given in the qualitative analysis: the L signers use handle
handshapes in contexts where entity handshapes would have been more appro-
priate. This may have some connection to previous research on iconicity and
gestures. In a study on silent gestures produced by non-signers, Ortega & Özyürek
() found that among four dierent types of iconicity strategies in gestures
(acting, representing, drawing and personication (aer Müller, and Hwang
et al., )), gesturers have a high prevalence for the acting type of iconicity
in their silent gesture production. The acting type is similar to depicting signs
representing handling using handshapes (see also Ortega & Özyürek,
, p.). Therefore, we would suggest that our nding that L signers’ use of
depicting signs and the high prevalence of handshapes is inuenced by
their gestural knowledge and based on an acting strategy.
Furthermore, we observed that L signers tend to blend lexical signs and
depicting signs that are formally similar. This could be interpreted as an intralex-
ical factor of a vocabulary diculty as well (cf. Laufer, ). The formal similar-
ities between the lexical sign and the depicting signs for entities describing
‘human beings’ makes it dicult for the learner. This aects both the choice of
handshapes and the structure of signing.
We also observed that using depicting signs together with mouthing was
common in L signers. We believe it is a cognitive eect based on their L Swedish
that has been transferred to STS production. The L learners are familiar with
the concept of conventionalized lexicon and lexicogrammar with words (and
phrases) in L Swedish, which are based on a higher degree of conventionaliza-
tion. The learners thus rely on this word concept, which inuences the produc-
tion of signs. We speculate that it may be cognitively more challenging to use
partly-lexical signs i.e. depicting signs whose formational properties depend to
a degree on the context, and which are therefore more productive as a category.
Such productive word categories do not exist in Swedish, at least not the same
way and to the same degree as in STS. We think learners are aware of the impor-
tance of using depicting signs, but they are acquired and mastered at later stages
in acquisition.
Finally, our ndings show that simultaneous use of dominant hand and non-
dominant hand in order to express two separate references or events by using
two depicting signs seems to be complex for L learners regardless of acquisition
stage. This conrms previous ndings by Ferrara and Nilsson () for L NTS,
showing that L learners have diculties with the coordination of the depicting
signs in the signing space and coordination of the hands in relation to each other.
Krister Schönström and Johanna Mesch
For example, with respect to in our data, the simultaneous use of two
separate handshape units was more common in L signers. We believe this is
also associated both with modality-dependent skills, i.e. expressing two dierent
handshapes at once, and with the linguistic complexity in expressing two separate
events simultaneously.
To conclude, this study was a rst step in the description of L acquisition of
signs with a particular focus on depicting signs. In future work, exploring how the
signs are used in utterances (especially as verbs) would add a syntactic perspective
to the study of the signs and would oer further insights. We have observed some
issues related to syntactic transfer from the L signers’ L Swedish, i.e. a prefer-
ence for using prepositional phrases (including use of lexical signs) to constitute
adverbial constructions, and this issue seems to inuence or overlap with the use
of depicting signs. This needs to be studied further in future research.
. Conclusion
Sign language data are time-consuming to construct, annotate and analyze, but
provide a rich set of naturalistic and semi-spontaneous linguistic data for further
studies. In this study, we have focused on the frequency and distribution of sign
types, specically by providing analysis of the use of depicting signs in L signers.
In addition, we have compared the use of the signs of L signers with L signers.
The ndings presented here demonstrate that beginner L signers use depicting
signs to a lesser extent and lexical signs to a greater extent, especially compared
to L signers. However, with time, L signers seem to approach a target language
pattern in the distribution of signs. Furthermore, our qualitative analyses show
that L signers have some diculties in the formation of depicting signs, which
may be related to the simultaneity of sign languages as well as sources based on
their L Swedish and gestural knowledge.
Funding
This article is a result of the project “From Speech to Sign – learning Swedish Sign Language
as a second language” funded by the Bank of Sweden Tercentenary Foundation (Riksbankens
Jubileumsfond) (P-:).
Second language acquisition of depicting signs
Acknowledgements
We thank our incredible informants for their willingness to participate. We also would like to
thank our fantastic crew of research assistants for their work with the data: Ylva Larsson, Josen
Bark and Moa Gärdenfors.
References
Bardel, C., Gudmundson, A., & Lindqvist, C. (). Aspects of lexical sophistication in
advanced learners’ oral production. Vocabulary acquisition and use in L French and
Italian. Studies in Second Language Acquisition,34(), –.
https://doi.org/10.1017/S0272263112000058
Bochner, J.H., Christie, K., Hauser, P. C., & Searls, J.M. (). When is a dierence really
dierent? Learners’ discrimination of linguistic contrasts in American Sign Language.
Language Learning,61(), –. https://doi.org/10.1111/j.1467-9922.2011.00671.x
Boers-Visker, E. (). Learning to use space: A study into the SL2 acquisition process of adult
learners of Sign Language of the Netherlands. Doctoral thesis. Amsterdam: LOT.
Boers-Visker, E., & van den Bogaerde, B. (). Learning to use space in the L acquisition of
a signed language: Two case studies. Sign Language Studies,19(), –.
https://doi.org/10.1353/sls.2019.0003
Brentari, D., Coppola, M., Mazzoni, L., & Goldin-Meadow, S. (). When does a system
become phonological? Handshape production in gesturers, signers, and homesigners.
Natural Language and Linguistic Theory,30, –. https://doi.org/10.1007/s11049-011-9145-1
Brentari, D., & Padden, C.A. (). Native and foreign vocabulary in American Sign
Language: A lexicon with multiple origins. In D. Brentari (Ed.), Foreign vocabulary in
sign languages: A cross-linguistic investigation of word formation (Lawrence E, Issue May
, pp. –). https://doi.org/10.4324/9781410601513-10
Chen Pichler, D. (). Sources of handshape error in rst-time signers of ASL. In G. Mathur
& D.J. Napoli (Eds.), Deaf around the world: The impact of language (pp. –). Oxford
University Press. https://doi.org/10.1093/acprof:oso/9780199732548.003.0005
Chen Pichler, D., & Koulidobrova, E. (). Acquisition of sign language as a second
language. In M. Marschark & P. E. Spencer (Eds.), The Oxford handbook of Deaf studies in
language (pp. –). Oxford University Press.
Cormier, K., Quinto-Pozos, D., Sevcikova, Z., & Schembri, A . (). Lexicalisation and de-
lexicalisation processes in sign languages: Comparing depicting constructions and
viewpoint gestures. Language and Communication,32(), –.
https://doi.org/10.1016/j.langcom.2012.09.004
ELAN . [Computer soware], (). https://archive.mpi.nl/tla/elan
Emmorey, K. (Ed.). (). Perspectives on classier constructions in sign languages. Lawrence
Erlbaum Associates. https://doi.org/10.4324/9781410607447
Ferrara, L., & Nilsson, A.-L. (). Describing spatial layouts as an LM signed language
learner. Sign Language & Linguistics,20(), –. https://doi.org/10.1075/sll.20.1.01fer
Granger, S. (). Contrastive interlanguage analysis: A reappraisal. International Journal of
Learner Corpus Research,1(), –. https://doi.org/10.1075/ijlcr.1.1.01gra
Krister Schönström and Johanna Mesch
Gulamani, S., Marshall, C., & Morgan, G. (). The challenges of viewpoint-taking when
learning a sign language: Data from the ‘frog story’ in British Sign Language. Second
Language Research,38, –. https://doi.org/10.1177/0267658320906855
Hodge, G., & Johnston, T. (). Points, depictions, gestures and enactment: partly-lexical
and non-lexical sign as core elements of single clause-like units in Auslan (Australian
Sign Language). Australian Journal of Linguistics,34(), –.
https://doi.org/10.1080/07268602.2014.887408
Hwang, S.-O., Tomita, N., Morgan, H., Ergin, R., Ilkbasaran, D., Seegers, S., … Padden, C.
(). Of the body and the hands: Patterned iconicity for semantic categories. Language
and Cognition,9, –. https://doi.org/10.1017/langcog.2016.28
Jarvis, S. (). The scope of transfer research. In L. Yu & T. Odlin (Eds.), New perspectives on
transfer in second language learning (pp. –). Multilingual Matters.
https://doi.org/10.21832/9781783094349-004
Johnston, T.A., & Ferrara, L. (). Lexicalization in signed languages: When is an idiom not
an idiom? Selected Papers from the 3rd UK Cognitive Linguistics Conference,1, –.
Johnston, T.A., & Schembri, A. (). Variation, lexicalization and grammaticalization in
signed languages. Langage et Société,131(), –. https://doi.org/10.3917/ls.131.0019
Kita, S., Gijn, I. van, & Hulst, H. van der (). The non-linguistic status of the Symmetry
Condition in signed languages: Evidence from a comparison of signs and speech-
accompanying representational gestures. Sign Language & Linguistics,17(), –.
https://doi.org/10.1075/sll.17.2.04kit
Kurz, K.B., Mullaney, K., & Occhino, C. (). Constructed action in American Sign
Language: A look at second language learners in a second modality. Languages,4().
https://doi.org/10.3390/languages4040090
Laufer, B. (). Words you know: How they aect the words you learn. In Fisiak, J. (Ed.),
Further insights into contrastive analysis (pp. –). John Benjamins.
https://doi.org/10.1075/llsee.30.35lau
Laufer, B., & Nation, P. (). Vocabulary size and use: Lexical richness in L written
production. Applied Linguistics,16(), –. https://doi.org/10.1093/applin/16.3.307
Leeson, L., Sheridan, S., Cannon, K., Murphy, T., Newman, H., & Veldheer, H. (). Hands
in motion: Learning to ngerspell in Irish Sign Language. Teanga, 11, (Special Issue),
–. https://doi.org/10.35903/teanga.v11i1.199
Lepic, R. (). A usage-based alternative to “lexicalization” in sign language linguistics.
Glossa,4(): . https://doi.org/10.5334/gjgl.840
Liddell, S.K. (). Grammar, gesture, and meaning in American Sign Language. Cambridge
University Press. https://doi.org/10.1017/CBO9780511615054
Marshall, C., Bel, A., Gulamani, S., & Morgan, G. (). How are signed languages learned as
second languages? Language and Linguistics Compass,15(), –.
https://doi.org/10.1111/lnc3.12403
Marshall, C., & Morgan, G. (). From gesture to sign language: Conventionalization of
classier constructions by adult hearing learners of British Sign Language. Topics in
Cognitive Science,7(), –. https://doi.org/10.1111/tops.12118
Mesch, J., & Schönström, K. (). From design and collection to annotation of a learner
corpus of sign language. In M. Bono, E. Ehimiou, S.-E. Fotinea, T. Hanke,
J. Hochgesang, J. Kristoersen, J. Mesch, & Y. Osugi (Eds.), Proceedings of the 8th
Workshop on the representation and processing of sign languages: Involving the language
community [Language Resources and Evaluation Conference (LREC)] (pp. –).
European Language Resources Association (ELR A).
Second language acquisition of depicting signs
Mesch, J., & Schönström, K. (). Use and acquisition of mouth actions in L sign language
learners – A corpus-based study. Sign Language & Linguistics 24(), –.
https://doi.org/10.1075/sll.19003.mes
Mesch, J., & Wallin, L. (). Gloss annotations in the Swedish Sign Language Corpus.
International Journal of Corpus Linguistics,20(), –.
https://doi.org/10.1075/ijcl.20.1.05mes
Mesch, J., & Wallin, L. (). Annoteringskonventioner för teckenspråkstexter. Version 8, maj
2021. [Annotation guidelines for sign language texts] (p. ). Sign Language Section,
Department of Linguistics, Stockholm University.
Morford, J.P., & MacFarlane, J. (). Frequency characteristics of American Sign Language.
Sign Language Studies,3(), –. https://doi.org/10.1353/sls.2003.0003
Mu
ller, C. (). Gestural modes of representation as techniques of depiction. In C. Mu
ller,
A. J. Cienki, E. Fricke, S. H. Ladewig, D. McNeill, & S. Teßendorf (Eds.),
Body–language–communication: An international handbook on multimodality in human
interaction (pp. –). De Gruyter Mouton.
Nation, I.S. (). Learning vocabulary in another language. Cambridge University Press.
https://doi.org/10.1017/CBO9781139524759
Öqvist, Z., Riemer Kankkonen, N., & Mesch, J. (). STS-korpus: A sign language web
corpus tool for teaching and public use. In E. Ehimiou, S.-E. Fotinea, T. Hanke,
J.A . Hochgesang, J. Kristoersen, & J. Mesch (Eds.), Proceedings of the 9th workshop on
the representation and processing of sign languages: Sign language resources in the service
of the language community, technological challenges and application perspectives
[Language Resources and Evaluation Conference (LREC)] (pp. –). Paris: European
Language Resources Association (ELR A).
Ortega, G. (). Acquisition of a signed phonological system by hearing adults: The role of sign
structure and iconicity. Doctoral thesis, University College London.
Ortega, G. (). Iconicity and sign lexical acquisition: A review. Frontiers in Psychology,8.
https://doi.org/10.3389/fpsyg.2017.01280
Ortega, G., & Morgan, G. (a). Phonological development in hearing learners of a sign
language: The inuence of phonological parameters, sign complexity, and iconicity.
Language Learning,65( ), –. https://doi.org/10.1111/lang.12123
Ortega, G., & Morgan, G. (b). The eect of iconicity in the mental lexicon of hearing non-
signers and procient signers: Evidence of cross-modal priming. Language, Cognition
and Neuroscience,30(), –. https://doi.org/10.1080/23273798.2014.959533
Ortega, G., Schiefner, A., & Özyürek, A. (). Hearing non-signers use their gestures to
predict iconic form-meaning mappings at rst exposure to signs. Cognition,191, .
https://doi.org/10.1016/j.cognition.2019.06.008
Ortega, G., & Özyürek, A. (). Systematic mappings between semantic categories and types
of iconic representations in the manual modality: A normed database of silent gesture.
Behavior Research Methods,52(), –. https://doi.org/10.3758/s13428-019-01204-6
Riemer Kankkonen, N., Björkstrand, T., Mesch, J., & Börstell, C. (). Crowdsourcing for the
Swedish Sign Language Dictionary. In M. Bono, E. Ehimiou, S.-E. Fotinea, T. Hanke,
J. Hochgesang, J. Kristoersen, J. Mesch, & Y. Osugi (Eds.), Proceedings of the 8th
Workshop on the representation and processing of sign languages: Involving the language
community [Language resources and evaluation conference (LREC)] (pp. –). Paris:
European Language Resources Association (ELR A).
Krister Schönström and Johanna Mesch
Rosen, R.S. (). Beginning L production errors in ASL lexical phonology: A cognitive
phonology model. Sign Language & Linguistics,7(), –.
https://doi.org/10.1075/sll.7.1.04beg
Saunders, D., & Parisot, A.-M. (). Constructed action in Quebec Sign Language (LSQ)
amongst Deaf rst language and second language users. Poster presented in the TISLR
Conference, Melbourne, Australia.
Schembri, A. (). Rethinking “classiers” in signed languages. In K. Emmorey (Ed.),
Perspectives on classier constructions in sign languages (pp. –). Lawrence Erlbaum.
Schembri, A., Jones, C., & Burnham, D. (). Comparing action gestures and classier verbs
of motion: Evidence from Australian Sign Language, Taiwan Sign Language, and
nonsigners’ gestures without speech. Journal of Deaf Studies and Deaf Education,10(),
–. https://doi.org/10.1093/deafed/eni029
Schick, B. (). The eects of morphosyntactic structure on the acquisition of classier
predicates in ASL. In C. Lucas (ed.), Sign language research: Theoretical issues (pp.
–). Gallaudet University Press.
Schönström, K. (). Sign languages and second language acquisition research – An
introduction. JESLA, The Journal of EuroSLA,5(). https://doi.org/10.22599/jesla.73
Schönström, K., & Mesch, J. (). Dataset. The project From speech to sign – Learning
Swedish Sign Language as a second language. Department of Linguistics, Stockholm
University.
Simper-Allen, P. (). “Cut and Break”-beskrivningar i svenskt teckenspråk: Barns och vuxnas
avbildande verbkonstruktioner. Doctoral thesis, Sign Language, Department of
Linguistics, Stockholm University.
Sykes, E. (Writer, Director), & Penington, J. (Producer). (). The Plank [Film]. Rank Film
Distributors.
Talmy, L. (). Semantics and syntax of motion. In J.P. Kimball (Ed.), Syntax and semantics.
Volume 4 (pp. –). Academic Press.
Taub, S., Galvan, D., Piñar, P., & Mather, S. (). Gesture and ASL L acquisition. In
R. M. de Quadros (Ed.) Sign languages: Spinning and unravelling the past, present and
future. TISLR9, forty ve papers and three posters from the 9th Theoretical Issues in Sign
Language Research Conference, Florianopolis, Brazil, December 2006 (pp. –).
Editora Arara Azul.
Wallin, L. (). Polysynthetic signs in Swedish Sign Language. Doctoral thesis, Department of
Linguistics, Stockholm University.
Résumé
Cet article concerne l’acquisition du lexique chez les apprenants L de la langue des signes
suédoise. Des données de conversation et de narration ont été sélectionnées dans un corpus
longitudinal de apprenants adultes signeurs L. Ces données ont été analysées et comparées
aux données de neuf signeurs L. L’utilisation de trois grandes catégories de signes a été
analysée : les signes lexicaux, les signes partiellement lexicaux (c’est-à-dire les depicting signs)
et les signes non lexicaux. Les résultats ont révélé certaines diérences entre les signeurs L
et L, en particulier en ce qui concerne les signes descriptifs. La quantité de signes descriptifs
utilisés par les apprenants L a augmenté avec le temps, se rapprochant de leur utilisation dans
Second language acquisition of depicting signs
la langue cible. D’un point de vue qualitatif, nous avons observé quelques diérences entre les
signeurs L et L dans leur utilisation des signes descriptifs, selon le choix de la congura-
tion manuelle et selon la construction des signes. Nous discutons ces résultats à la lumière de
recherches antérieures liées au lexique en L ainsi qu’au rôle de la connaissance des gestes dans
l’acquisition d’une L gestuelle.
Authors’ addresses
Krister Schönström
Department of Linguistics
Stockholm University
Stockholm
Sweden
schonstrom@ling.su.se
Johanna Mesch
Department of Linguistics
Stockholm University
Stockholm
Sweden
johanna.mesch@ling.su.se
Publication history
Date received: January
Date accepted: June
Krister Schönström and Johanna Mesch
