EDITORS: Angela Grimm, Anja Müller, Cornelia Hamann, Esther Ruigendijk
TITLE: Production-Comprehension Asymmetries in Child Language
SERIES TITLE: Studies on Language Acquisition [SOLA] 43
PUBLISHER: De Gruyter Mouton
Ursula Kania, English Department, University of Leipzig (Germany)
In language acquisition, comprehension usually precedes production -- for
example, children comprehend more words than they are able to produce (e.g.
Fenson et al. 1993 for English). The reverse pattern, i.e. the observation that
in some cases adult-like production seems to precede comprehension is a much
more recent and still underresearched phenomenon.
This edited volume contains ten selected papers from the DGfS (=Deutsche
Gesellschaft für Sprachwissenschaft/’German Linguistic Society’) workshop
''Production-Comprehension-Asymmetries in Child Language,'' which was held at the
University of Osnabrück in March 2009, presenting research on cases where
“production outperforms comprehension in the same linguistic domain” (p. 1). A
particular focus lies on the acquisition and use of pronouns, since the
asymmetry is particularly well-attested in this area (e.g. Bloom et al. 1994;
Chien & Wexler 1990).
The book is organized into two sections (A and B), which correspond to the two
major aims of the volume:
The first aim is to present cross-linguistic research on
production-comprehension asymmetries, including contributions on lesser-studied
languages such as Bulgarian and Tamil (Section A). The second aim is to discuss
possible sources of production-comprehension asymmetries. Based on the
assumption that both comprehension and production rely on a single grammar, the
seven papers in Section B mainly explore 1. in how far the observed differences
can be attributed to methodological decisions (i.e. performance or task effects)
and 2. “what a grammatical explanation of the observed asymmetries could look
like” (p. 2).
In her contribution ''Testing the Aspect Hypothesis in child Tamil,'' the first
chapter in Section A, Lavanya Sankaran presents experimental findings on the
influence of verb semantics on the production and comprehension of the aspectual
markers ‘kondiru’ (consisting of the participle of ‘kol,' ‘to hold,' and the
perfect auxiliary ‘iru’; Lehmann 1993, p. 207) and ‘vidu’(derived from the
lexical verb for ‘leave, let’; Schiffman 1999, p. 85) in Tamil, a lesser-studied
Dravidian language. The results from an elicited production task carried out
with two child groups (mean age = 3;0 and 4;1) and one adult control group
suggest that the perfective marker ‘vidu’ is acquired before the imperfective
marker ‘kondiru’ and used in an adult-like way in production from early on.
However, a sentence-picture matching task used to measure comprehension showed
the reverse pattern, i.e. a significantly better performance for ‘kondiru’ than
for ‘vidu’. The author suggests that this two-way asymmetry is caused by the
fact that ‘vidu’ has a dual function (as a perfective and an inceptive marker),
which might lead to a disadvantage in comprehension.
The second paper, ''An asymmetry in the acquisition of accusative clitics in
child Romanian,'' by Martine Coene and Laris Avram, discusses the use of
accusative clitics and reflexives in the spontaneous production data of two
children (1;9 and 3;0). Based on Uriagereka (1995) they argue that 1st/2nd
person clitics and reflexives belong to a different class than 3rd person
clitics, leading to different developmental paths in acquisition. The production
data show that the former types of clitics were used in an adult-like way
earlier than the latter. Even though the study focused on production only, the
findings call into question the well-established observation that comprehension
precedes production for Romance pronominals. It is argued that the observed
asymmetry stems from the fact that previous research considered 3rd person
clitics only. The paper thus suggests that the observed production-comprehension
asymmetry may disappear when the whole paradigm of accusative clitics is
The third and last paper with a focus on cross-linguistic evidence is
''Comprehension and imitated production of personal pronouns across languages'' by
Dagmar Bittner, Milena Kuehnast, and Natalia Gagarina. In order to examine the
effects of the cues structural prominence and animacy on the interpretation and
use of personal pronouns in subject position and to compare the cue pattern in
comprehension and production, German, Russian, and Bulgarian children in two age
groups (3 and 5 years) were tested with a question-after-story design. Even
though the production task did not provide enough data for statistical analyses,
the results indicate that the children relied on the same cues in both
production and comprehension, suggesting symmetric processing. However, the
authors point out that the results do not “mirror the complete anaphoric
capacity of [personal pronouns] in the children’s grammar” (p. 91) since only
two cue types and one pronoun type were investigated.
Section B starts out with a study on the ''Comprehension and production of
subject pronouns in child Dutch'' by Charlotte Koster, Jan Hoeks, and Petra
Hendriks. The explanatory framework used is Optimality Theory (e.g. Prince &
Smolensky 2004), a constraint-based system in which language production and
comprehension are viewed as processes aiming at an optimal input-output
relationship. The fact that production and comprehension proceed in opposite
directions (production going from meaning to form, comprehension from form to
meaning) leads to a potential asymmetry between the two modalities (Asymmetric
Grammar Hypothesis). Adult speakers overcome this asymmetry by taking into
account both the listener’s and the speaker’s perspective (bidirectional
optimization, Blutner 2000). Based on the assumption that children are not yet
able to do so, the authors hypothesized that this would have consequences for
children’s production and comprehension of anaphoric subject pronouns and NPs in
discourse. In the study, 31 Dutch children (mean age = 5;6) and an adult control
group were given a picture storybook production task and a question-after-story
task to test comprehension. The children preferred to produce subject pronouns
rather than NPs even after a topic shift (thus being overly economical).
Furthermore, they failed to interpret the referents of subject pronouns
correctly in the comprehension task, presumably because they did not interpret a
preceding NP as a topic shift signal (i.e. failing to take into account the
speaker’s perspective). The authors argue that these asymmetries will vanish
once the children have optimized bidirectionality.
The fifth paper is ''Asymmetries in the processing of object relatives in child
Hebrew and Italian'' by Irena Botwinik. Based on the well-attested observation
that the production of object relatives precedes their comprehension, this study
lays out a possible explanation for this asymmetry by reanalyzing experimental
data on the comprehension of object relatives in Hebrew (Günzberg et al. 2008)
and Italian (Arosio et al. 2006). It is suggested that the correct parsing and
hence interpretation of object relatives is similar to that of other garden path
sentences involving local ambiguities, leading to processing difficulties which
are encountered in comprehension but not in production.
The sixth contribution, ''A comprehension delay of subject-object [S-O] word
order in Dutch preschoolers,'' by Gisi Cannizzaro, presents experimental findings
on the comprehension and production of S-O word order in children (mean age =
3;6) and adults. The two groups were tested on the same sentences in both a
picture-selection and a picture-description task, for the latter of which
eye-tracking data was obtained as an additional online-measure. The children,
but not the adults, performed significantly worse on the comprehension than on
the production task. The author bases her explanation in the framework of
Optimality Theory and suggests that the observed delay in comprehension may
arise from a different ranking of constraints in the two modalities which the
children still have to overcome. However, since the prediction that animacy
(which is related to the PROMINENCE-constraint, Hendriks et al. 2005) would
influence comprehension strategies was not borne out by the data, the author
concludes with suggestions for further research in order to investigate the
possible sources of the asymmetry in more detail.
In the seventh paper, ''Asymmetries in children’s language performance within and
across modalities,'' Oda-Christina Brandt-Kobele and Barbara Höhle first
summarize findings from previous studies on production-comprehension asymmetries
as well as selected proposals on how to account for the observed patterns. They
then consider methodological explanations in more detail, reexamining their own
experimental data on the comprehension of verb-inflections in German 3-4
year-olds (Brandt-Kobele & Höhle 2010). Participants were tested in 1. a
preferential looking task and 2. a picture-selection task. Since the 3rd-person
singular female pronoun and the 3rd-person plural pronoun are homophones in
German, only these two pronouns were used in the test sentences in order to
force children to obtain information on the number of participants from the verb
inflection. While eye-tracking data from the first experiment provides evidence
for the children’s ability to do so, this is not the case for the second
experiment -- the authors suggest that pointing increases the processing load,
leading to poorer performance. It is argued that different methods may not only
be responsible for the observed within-modality asymmetry (i.e. within
comprehension) but also for attested cross-modal asymmetries (i.e. between
comprehension and production).
The eighth paper is ''Adults’ on-line comprehension of object pronouns in
discourse,'' by Petra Hendriks, Arina Banga, Jacolien von Rij, Gisi Cannizzaro,
and John Hoeks. Studies on English have shown that six-year-olds correctly
produce but still often misinterpret object pronouns (but not reflexives), a
pattern supposedly resulting from children’s inability to observe Principle B of
Binding Theory (the so-called Delay of Principle B-Effect). It has been proposed
that this could be due experimental artifacts (Conroy et al., 2009) or the fact
that the correct interpretation relies on contextual factors (Spenader, Smits, &
Hendriks 2009). Results from a picture-verification task the authors conducted
with 25 Dutch adults showed that the accuracy of off-line-responses was not
influenced by discourse context. However, reaction times were longer when the
discourse topic was not established unambiguously in the very beginning. It is
argued that this influence of discourse context on adults’ online behavior and
children’s off-line interpretations speaks against the Delay of Principle
B-Effect resulting (only) from experimental artifacts.
In the ninth contribution, ''Production and comprehension of sentence negation in
child German,'' Magdalena Wojtecka, Corinna Koch, Angela Grimm, and Petra Schulz
investigate whether there is a developmental asymmetry between the two
modalities. 34 German pre-schoolers were tested on the sentence-negator ‘nicht’
(‘not’) in an elicited production task and a truth value judgment task, with an
interval of 6 months between two test rounds. While children showed mastery of
‘nicht’ in production already in the first test round (mean age = 3;7),
comprehension was still non-adult-like in the second test round (mean age =
4;2). It is suggested that further research should use a variety of methods to
test the same children in both modalities in order to shed further light on the
The tenth and last paper is ''Principle B delays as a processing problem:
Evidence from task effects,'' by Sergio Baauw, Shalom Zuckerman, Esther
Ruigendijk, and Sergey Avrutin. Similar to Hendriks, Banga, von Rij, Cannizzaro,
and Hoeks (this volume) they argue that findings on children’s mistakes in the
interpretation of object pronouns are not due to experimental artifacts but stem
from processing problems. They compare findings from experiments on
comprehension involving both picture-selection and truth-value judgment tasks
conducted with Dutch and Spanish children and Spanish agrammatic Broca’s
aphasics. The fact that latter type of task requires more processing resources
(since the acceptability of a particular reading has to be considered) should
result in poorer performance. In line with this prediction, performance was
found to be significantly worse in the truth-value judgment task throughout,
suggesting that the Pronoun Interpretation Problem stems from processing
Considering that all papers focus on or at least include findings from languages
other than English, the first aim of the volume -- i.e. to present
cross-linguistic research on production-comprehension asymmetries -- has
certainly been met. Especially the papers on lesser-studied languages such as
Tamil (Sankaran) or Hebrew (Botwinik) are a welcome and much-needed addition to
previous research since they widen and evaluate the potential scope of the
existing explanatory frameworks.
The second and certainly very ambitious aim was “to shed light on the source of
the production-comprehension asymmetries” (p. 4). The studies which really
succeed in this respect are the ones fulfilling one of the following two
criteria: first, in order to rule out the possibility that the observed
asymmetry is due to differences between (groups of) participants, it is
necessary to test the same children in both modalities. Furthermore,
within-modality asymmetries have to be explored by comparing the results
obtained via different experimental methods and measures (e.g. offline vs.
online). In total, eight of the ten papers in this volume satisfy either the
first (e.g. Sankaran; Bittner, Kuehnast & Gagarina) or the second criterion
(e.g. Brandt-Kobele & Höhle), thus providing valuable converging evidence on the
observed asymmetries both across and within modalities.
Since the volume takes a generative grammar perspective, it will at first glance
mainly be of interest to nativist language acquisition researchers and
generative linguists concerned with the phenomena discussed (most notably
pronouns, but also negation and aspect marking).
However, the neglect of certain aspects (inevitably) resulting from this
theoretical decision could also serve as a starting point for a debate on
production-comprehension asymmetries between generativists and researchers
working within the competing usage-based paradigm (e.g. Tomasello 2003). For
example, the potential role of item-based effects is not considered in this
volume. One implication resulting from a usage-based perspective is that one
should be careful about crediting children with adult-like productivity, since
(spontaneous) productivity has been found to be more item-based and thus more
limited than is often assumed in generative approaches (e.g. Dabrowska & Lieven
2005 for questions in English). It is thus to be hoped that this volume
stimulates future research (within both generative and usage-based frameworks)
in order to shed further light on the observed patterns.
Furthermore, on a more general level, the inclusion of studies using both off-
and online-measures (e.g. Cannizzaro) and the resulting discussion about which
parts of language processing the different methods and measures actually tap
into make this volume a stimulating read for everyone interested in the design
and evaluation of psycholinguistic experiments.
Arosio, Fabrizio, Flavia Adani & Maria Teresia Guasti. 2006. Children’s
processing of subject and object relatives in Italian. In Adriana Belletti,
Elisa Bennati, Christiano Chesi & Ida Ferrari (eds.), Language Acquisition and
Development, 15-27. Cambridge Scholars Press.
Bloom, Paul, Andrew Baars, Laura Conway & Janet Nicol. 1994. Children’s
knowledge of binding and coreference. Evidence from spontaneous speech. Language
Blutner, Reinhard. 2000. Some aspects of optimality in natural language
interpretation. Journal of Semantics 17. 189-216.
Brandt-Kobele, Oda-Christina & Barbara Höhle. 2010. What asymmetries within
comprehension reveal about asymmetries between comprehension and production: The
case of verbal inflection in language acquisition. Lingua 120. 1910-1925.
Chien, Yu-Chin & Kenneth Wexler. 1990. Children’s knowledge of locality
conditions on binding as evidence for the modularity of syntax and pragmatics.
Language Acquisition 1. 225-295.
Conroy, Anastasia, Eri Takahashi, Jeffrey Lidz & Colin Phillips. 2009. Equal
treatment for all antecedents. How children succeed with Principle B. Linguistic
Inquiry 40. 446-486.
Dabrowska, Ewa, & Elena Lieven. 2005. Towards a lexically specific grammar of
children's question constructions. Cognitive Linguistics 16(3). 437--474.
Fenson, Larry, Philip S. Dale, J. Steven Reznick, Donna Thal, Elizabeth Bates,
Jeffrey P. Hartung, Steve Pethick & Judy S. Reilly. 1993. The MacArthur
Communicative Development Inventories: User’s guide and technical manual. San
Diego: Paul H Brookes.
Günzberg-Kerbel, Noa, Lilach Shvimer & Naama Friedman. 2008. “Take the hen that
the cow kissed the hen”: Comprehension and production of various relative
clauses by Hebrew speaking children. Language and Brain 7. 23-43 (in Hebrew).
Hendriks, Petra, Helen de Hoop & Monique Lamers. 2005. Asymmetries in language
use reveal asymmetries in the grammar. In Paul Dekke & Michael Franke (eds.),
Proceedings of the 15th Amsterdam Colloquium, 113-118. Amsterdam: Institute for
Logic, Language, and Computation.
Lehmann, Thomas. 1993. A grammar of modern Tamil. Pondicherry Institute of
Prince, Alan & Paul Smolensky. 2004. Optimality Theory: Constraint interaction
in generative grammar. Malden, MA: Blackwell.
Schiffman, Harold. 1999. A Reference Grammar of Spoken Tamil. Cambridge:
Cambridge University Press.
Spenader, Jennifer, Erik-Jan Smits & Petra Hendriks. 2009. Coherent discourse
solves the Pronoun Interpretation Problem. Journal of Child Language 36. 23-52.
Tomasello, Michael. 2003. Constructing a language: A usage-based theory of
language acquisition. Cambridge, Mass.: Harvard University Press.
Uriagereka, Juan. 1995. The syntax of clitic placement in Western Romance.
Linguistic Inquiry 26(1). 79-123.