Full text of DOI:10.1017/S0305000900004542

J. Child Lang.  (), –. Printed in the United Kingdom
#  Cambridge University Press
The importance of studying filler-producing children
EDY VENEZIANO
Universite Nancy ‚ and Universite Paris V – CNRS
In her note Ann Peters makes a remarkable effort in bringing together data
on fillers reported to occur in different languages. This is a difficult enterprise
since these data have been obtained with different methodological tools and
are anchored in different theoretical approaches. Peters tries to organize them
under a loose but common underlying theoretical framework in which
phonoprosodic, lexical, syntactic and morphological aspects of acquisition
are considered as inherently interrelated and intertwined. Within this
framework she is able to consider as related fillers that seem to stand for
different kinds of language units (free and bound morphemes, but also
amalgams and unidentifiable chunks that seem to stand for content words).
She points out that fillers may be at the crossroad of these different kinds of
traditional layers of language. As such, they may constitute traces of different
kinds of work from the child’s side and may have a role in changing the status
of the child’s knowledge in these domains.
The importance of studying fillers
Although some authors may consider fillers as a minor phenomenon in early
language acquisition, mainly because they are not found in all children’s
production, I would like to support and strengthen Peters’ position that
considers fillers as a highly important domain of inquiry.
First, as is the case of other phenomena (including child-directed speech
and specific kinds of conversational functioning), the fact that fillers may not
be produced by certain children does not imply that they are extraneous to
early language acquisition. They may not be necessary, but they may still
represent one of the many possible entries into the complex and multifaceted
properties of language systems.
Second, given the inter-relatedness among the different components of
language, the study of fillers provide an invaluable window on the child’s
underlying knowledge of linguistic constructs like word classes and con-
stituent phrases, and thus may provide extremely relevant empirical data for
choosing among competing hypotheses. Indeed children who produce fillers
would have the appropriate means to show structural knowledge of the
language – for example, by producing fillers in slots that require grammatical
morphemes (e.g. Dolitsky, ; Peters & Menn, ; Simonsen, ;



Bottari, Cipriani & Chilosi, \), or by producing different fillers
according to the class to which the word they precede belongs (Veneziano &
Sinclair, ). If there is a period in which fillers do not manifest this
structural knowledge, as seems to be the case for some children (e.g. Peters
& Menn, ; Simonsen, ; Veneziano & Sinclair, ; Kilani-Schoch
& Dressler, ), hypotheses that claim early availability of linguistic
constructs are seriously weakened on empirical grounds.
Fillers and the close relation between phonoprosody and morphology
Peters draws our attention to the fact that fillers may be a privileged example
showing that phonology and morphosyntax are closer than is usually
thought. This point is of great relevance for a constructivist account of
language acquisition. Indeed, the relation between phonoprosody and
morphosyntax stressed by Peters allows to formulate specific hypotheses on
how morphosyntactic categories can be constructed out of concerns of a
qualitatively different kind. It may be supposed that starting with phono-
logical and prosodic concerns, children are led to build specific relations, for
example, between words and their immediate and\or recurring verbal
environments, and between words themselves, providing an intermediary
link between a pregrammatical organization and a protogrammatical one that
starts to use principles resembling the categories of a grammatically-based
description of the language, like word classes, grammatical morphemes and
constituent phrases (see Veneziano & Sinclair, , for a proposal along
these lines). From this point of view it appears strange that, in the final part
of the paper, Peters discusses only two main theoretical positions on fillers.
Position I that views fillers as purely phonological elements, discontinuous
relative to grammatical development, and position II (having a nativist and
a constructivist version) that sees fillers as phonological traces of ‘early
awareness of (some) adult functional categories’.
Considering that, at the same time, Peters stresses the fact that fillers seem
to change status during development (from premorphological to mor-
phological), a third position seems missing: protomorphological fillers
presenting (some) characteristics of adult language categories  
     with premorphological fillers, those
reflecting at first mainly phonoprosodic properties of the language. In other
words, a constructivist position does not see any hiatus between phonological
and partly grammatical or protomorphological fillers, provided that a
continuity in the  of construction linking them can be detailed.
The need for even more precise criteria and methods of analysis
However, to be able to make distinctions among fillers and choose among the
possible positions one needs clear criteria and stringent methods of analysis.


    - 
Although Peters goes a long way out in trying to define and set criteria for
identifying fillers and determining their nature, there is still room for even
more precise and well-defined notions, and for testing specific implications
of the statements that are made.
One of the problems concerns the definition itself of fillers. Two charac-
teristics appear to be common: . fillers are unglossable or not well
identifiable as particular lexical items: . they are produced by the child in
place of some undetermined ‘something else’. If these defining criteria fit
well protomorphological fillers they may not necessarily fit fillers that seem
to be phonologically and\or prosodically motivated. The latter may not be
undetermined since the child may aim specific elements and particular
prosodic patterns. Thus the term ‘fillers’ may not be appropriate to refer to
the latter productions. However, a unifying term seems useful for em-
phasizing the underlying developmental relation existing among the different
kinds of fillers. Should we coin a more general term, reserving ‘fillers’ for
those elements that present protomorphological characteristics? Or should
we have a more general definition of fillers that would then be specified for
each of its different attributes (for example, phonoprosodic, protomorpho-
logical and morphological fillers)?
On a different level, Peters rightly emphasizes the need for child-centred
– rather than adult-centred – analyses that consider fillers not as an isolated
phenomenon but as part of an overall working system that needs to be
explored at the same time as fillers are analysed; she also points out that the
ultimate decision about the status of early fillers must be made post hoc.
However, it seems to me unlikely that we can decide upon the status of fillers
by looking at their phenomenological destiny. The rule of the thumb ‘if they
disappear, they were purely phonological; if they evolve continuously into
identifiable morphemes they were or become protomorphemic’ seems
difficult to handle. It is a known fact that children don’t go forever producing
fillers, so fillers do disappear; and in places where fillers are produced we will
find, sooner or later, grammatical morphemes (or identifiable content words).
What seems more important to consider is  fillers evolve in the course of
development – what kind of changes can we observe in their phonological
and structural properties – as well as  they disappear   to
other aspects of the child’s language functioning. Ultimately, the crucial kind
of information comes from probes of the child’s production of fillers and of
other aspects of the child’s language    \  . If
- and   repeated probes indicate that fillers
undergo well specified qualitative changes in the course of development, this
may provide one of the best kinds of evidence that some features of the child’s
knowledge of language are under construction, and that the  of this
construction is developmentally continuous.



REFERENCES
Bottari, P., Cipriani, P. & Chilosi, A. M. (\). Protosyntactic devices in the
acquisition of Italian free morphology. Language Acquisition: A Journal of Developmental
Linguistics , –.
Dolitsky, M. (). The birth of grammatical morphemes. Journal of Psycholinguistic
Research , –.
Kilani-Schoch M. & Dressler, W. U. (). Are fillers as precursors of morphemes relevant
for morphological theory? In W. U. Dressler, O. E. Pfeiffer, M. Pochtrager & J. Rennison
(eds), Morphological analysis in comparison. Current issues in linguistic theory, Vol. .
Amsterdam\Philadelphia: Benjamins.
Peters, A. M. & Menn, L. (). False starts and filler-syllables: ways to learn grammatical
morphemes. Language , –.
Simonsen, H. G. (). Models in the description of phonological acquisition. Sixth
International Congress for the Study of Child Language. Trieste, Italy.
Veneziano, E. & Sinclair, H. (). The changing status of ‘filler syllables’ on the way to
grammatical morphemes. Journal of Child Language , –.
