LINGUIST List 2.398

Saturday, 10 August 1991

Disc: Manual Babbling: A Response from Petitto

Editor for this issue: <>


  1. INLO000, Manual Babbling

Message 1: Manual Babbling

Date: Fri, 09 Aug 91 16:27:09 EDT
Subject: Manual Babbling
To: Linguist Members
Several requests for further information about a recent study of
manual babbling have been forwarded to me. The reference is...
Petitto, L.A. and Marentette, P.F. (1991) Babbling in the manual
mode: Evidence for the ontogeny of language. Science, vol 251,
pp. 1493-1496.
(1) Re: Babbling in X, where X is any language.
As can be seen from the title, this study is not about babbling
IN any signed language, per se. As with spoken languages,
infants (esp. in the early period) do not babble in specific signed
(2) Re: Vocal babbling in deaf children. Studies of the vocal
productions of profoundly deaf infants have demonstrated that
previous researchers' claims (e.g., Lenneberg) about the
existence of vocal babbling in deaf infants were unfounded.
While I will limit my comments here to syllabic babbling
(definition below), my comments are true of the other
vocalizations that are produced by deaf infants; that is, there
are qualitative differences between hearing and profoundly
deaf infants' vocalizations throughout early development.
Although profoundly deaf infants may occasionally produce a
well-formed syllabic vocalization, these forms are (1)
infrequent relative to their other vocalizations, and (2) occur in
unsystematic ways. Indeed, only those deaf children who can
benefit from the use of hearing aids (i.e., they have some
hearing) will produce syllabic babbling, but, here, too, other
significant developmental anomalies characterize their babbling
due to the absence of audition (e.g., Oller). However, deaf
infants acquiring a natural signed language from birth, freely
produce manual babbling on the identical time course and
sequence observed for vocal babbling. (Both hearing infants
acquiring a signed language, and bilingual hearing infants
acquiring a spoken and a signed language, also achieve all
linguistic milestones (e.g., babbling, first words, first
combinations) within the same time period commonly observed
in human language acquisition (see Petitto & Marentette for all
relevant references.) Interestingly, hearing infants will
occasionally produce well-formed syllabic manual babbling. As
with deaf infants' vocal babbling, the hearing infants' manual
babbling is (i) infrequent relative to their other manual
productions, and (ii) occurs in unsystematic ways. In the Petitto
& Marentette paper I offer an explanation for the existence of
this phenomenon in hearing and deaf infants.
Below is my response to another question that I received from a
member of Linguist; he suggested that my answer may be of
interest to other members of Linguist.
Re: What do people NOW call what used to be called BABBLING?
There is still controversy over definitions, per se. However, I
think there would be general agreement regarding the following
very ROUGH characterization:
Babbling is the general term used to refer to the class of
meaningless, but linguistically-related vocal (or manual)
productions by children, which typically occurs both prior to
and during early language development (more below).
There is still controversy over whether children's early
babbling inventories comprise a universal set of phonetic forms
and controversy over WHY this may/may not be the case; there
is still controversy over whether babbling is continuous with
later later development, and the extent to which babbling is
My own studies of vocal and manual babbling support the
following conclusions: babbling is a maturationally controlled,
specifically-linguistic phenomenon. As in the literature, the
single term babbling is used to refer to the following different,
but related phenomena over time (the following is not intended
to reflect strict stages, per se): In the early months, depending
upon the modality of the language input, infants (4-7 months)
produce a restricted set of phonetic forms (vocally, if exposed to
spoken languages and manually if exposed to signed languages);
the forms of infants acquiring spoken languages appear to be
drawn from the set of possible sounds found in spoken
languages and are typically not language-specific; the identical
phenomenon is observed in children (hearing or deaf) who are
acquiring signed languages. As such, the initial inventory of
forms common to infants' acquiring spoken languages and
infants acquiring signed languages (respectively) can be
overlapping, although infants do exhibit individual form
preferences. Around 7-9 months, infants begin the production
of syllabic (reduplicated or canonical) babbling (e.g., Oller), for
example, dadada, bababa. The child is still not producing actual
words in her target language at this time; nonetheless, these
forms obey the prosodic and phonological lawfulness (e.g.,
Stampe) of human Language. Syllabic (vocal or signed)
babbling is characterized by (1) use of a reduced subset of
possible sounds (phonetic units) found in language, (ii) syllabic
organization (well-formed consonant-vowel clusters), and (iii)
use without apparent meaning or reference; there are several
other properties (e.g., see Petitto & Marentette for relevant
references and other properties). Around ages 9-12 months, I
have observed (like de Boysen-Bardies) that aspects of the
child's babbling forms can take on language-specific
characteristics, a phenomenon that I have observed in four
different languages (2 spoken, 2 signed). Moreover, there is a
continuity of phonetic form and syllabic type within an
individual child's syllabic babbling and their first words.
Around 12-14 months (and beyond in some children), hearing
and deaf children produce jargon babbling (meaningless
babbling sequences that maintain the timing, rhythm and
duration of spoken and signed sentences, respectively). Jargon
babbling can and typically does continue even during hearing
and deaf children's production of first words and signs
Representational structures: I believe only one explanation of
human language ontogeny fully accounts for these and other
similarities in the time course and content of signed and spoken
language acquisition: Humans are born with a predisposition to
discover particular sized units with particular distributional
patterns in the input, guided by innately-specified structural
constraints. At birth, this nascent structure-seeking capacity is
sensitive to the patterned organization of natural language
phonology common to all world languages, be they spoken or
signed (e.g., rhythmic, temporal, and hierarchical organization)
and is particularly sensitive to structures in the input that
correspond to the size and distributional patterns of the syllable
in spoken and signed languages. --This is a mouthful. For
relevant references and further explication, see Petitto, in press,
and Petitto & Marentette, 1991.
I'd be happy to send people reprints. Just send me your
Laura Ann Petitto
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue