ArticlePDF Available

The Critical Period Hypothesis reconsidered: Successful adult learners of Hungarian and English

Authors:

Abstract and Figures

The strong Version ofthe critical period hypothesis (CPH) encouraged the research on successful adult learners of Hungarian and English at the English Department of Janus Pannonius University, Pecs. Altogether 33 successful learners were interviewed: in Study l 20 learners ofvarious mother tongues have been acquinng Hungarian in the host environment, whereas in Study 2 13 Hungarians have been learning English äs aforeign language and have stayed in the host environmentfor a relatively shortperiod. Two tapes were developed with short samplesfrom the interviewees and native Speakers ofthe target languages. These tapes were administered to three groups of native Speakers in judgement tasks with a follow-up task eliciting clues judges used in deciding whether Speakers were native or non-native. The findings of this study challenge the strong version of the CPH. Other outcomes include typical clues applied by native judges and some insights into ways how these successful learners have developed native proficiency.
Content may be subject to copyright.
The
Critical Period
Hypothesis
reconsidered:
Successful
adult
learners
of
Hungarian
and
English
MARIANNE
NIKOLOV
Abstract
The
strong
Version
ofthe
critical period hypothesis
(CPH)
encouraged
the re-
search
on
successful
adult learners
of
Hungarian
and
English
at the
English
Department
of
Janus Pannonius
University,
Pecs.
Altogether
33
successful
learners
were
interviewed:
in
Study
l 20
learners
ofvarious
mother tongues
have been
acquinng
Hungarian
in the
host environment,
whereas
in
Study
2 13
Hungarians have been
learning
English
äs
aforeign
language
and
have stayed
in
the
host
environmentfor
a
relatively
shortperiod.
Two
tapes were developed
with
short
samplesfrom
the
interviewees
and
native
Speakers
ofthe
target
lan-
guages. These tapes were
administered
to
three groups
of
native
Speakers
in
judgement
tasks
with
a
follow-up
task eliciting
clues
judges used
in
deciding
whether
Speakers
were native
or
non-native.
The
findings of
this study
challenge
the
strong version
of the
CPH.
Other
outcomes
include
typical
clues applied
by
native judges
and
some
insights into
ways
how
these successful learners have developed native
proficiency.
...
when
they
say,
T
m
amazed
that
you
sound
just
like
a
native!'
they
are
really
saying
something
like
'You
speak
my
language
brilliantly
-
especially
for
a
foreigner!"
(Scovel
1988: 177)
Introduction
The
critical
period
hypothesis
(CPH) Claims that
there
is a
period
during
which
learners
can
acquire
a
second
language
easily
and
achieve
native-speaker
com-
petence,
but
after
this
period
second
language
acquisition
(SLA)
becomes
more
difficult
and is
rarely
entirely
successful
(Lenneberg
1967).
Researchers
differ
over
when
this
critical
period
comes
to an
end.
The
strong
version
of the CPH
Claims
that
an
authentic
accent
is not
available
unless
SLA
begins
before
the
critical
age
(e.g.:
Scovel
1988; 1995;
DeKeyser
forthcoming).
Long
(1990)
IRAL
38
(2000),
109-124
0019042X/2000/038-00109
©Walter
de
Gruyter
110
Marianne
Nikolov
suggests that
the
acquisition
of
native-like
accent
is not
possible
by
learners
who
begin
learning
the
target
language
after
the age of
six; Scovel (1988)
ar-
gues
that
the
critical
period
for
pronunciation
is
around
puberty,
whereas
in
Krashen's
(1985)
view, acquisition
is
always
available
to
adults
äs
they have
continued access
to
Universal
Grammar.
As for the
pedagogical implications,
several authors
argue
for the
teachability
of
phonology
in
adulthood
(e.g.:
Pen-
nington
1998).
Recently,
extreme positions have also been
put
forward
(Pinker 1994; Ske-
han
1998)
relying
on
evidence
not
only
from
SLA but
also
evolutionary
biol-
ogy.
Skehan's
summary
(1998: 232),
illustrates
the
widening
of the
scope
of
these discussions:
1.
The
evidence
for a
critical period
is
strong.
It
is
based
on
multiple sources:
second language acquisition,
the
acquisition
of
ASL, acquisition
by the
hearing-impaired
whose problems have been
corrected,
and the
problems
that
feral
children
encounter
who
are
exposed
to
language
after
the
onset
of
puberty.
None
of
these
sources
of
evidence
has to
bear
the
explanatory
bürden
ahne.
2. The
evidence
is
consistent with
a
gradual
decline
in a
language capacity
which
is
complete
by the
onset
of
puberty.
3.
Explanations
for the
onset
and
close
of the
critical period based
on
evolu-
tionary
biology
are the
most
convincing,
since they suggest reasons
why a
language learning
ability
would
benefit
survival
chances.
In
contrast,
in
Singleton's
Interpretation,
"...
the
evidence does
not
con-
sistently
support
the
hypothesis that younger learners
are
inevitably
more
ef-
ficient
than
older
learners
in the
phonetic/phonological
domain"
(Singleton
1989: 137).
Discussions
concerning
the
relationship
between
the CPH and
accent
do not
go
beyond pronunciation
and
presume that native-like
proficiency
is
available
for
late
beginners
in
other
areas
of
language
(e.g.:
grammar,
vocabulary
(Sin-
gleton 1999),
use of
idioms
or
writing),
whereas accent
referring
to the
particu-
lar
way of
speaking
will
teil
whether
the
Speaker
is a
native
Speaker.
Tradition-
ally,
research
on
foreign accent
has
been
concerned
with segmental phenom-
ena,
but
recently there
has
been
a
growing interest
in the
role
of
nonsegmental
phenomena
(stress,
Intonation, rhythm, speaking rate
and
voice quality;
e.g.,
Munro
1995).
As
Larsen-Freeman
and
Long
(1991)
point out,
the
issue
of age is an im-
portant
one in SLA
research
for
educational
policy-making
and for
language
pedagogy.
It is a
challenging
one for
any
learner
who
Starts
SLA
after
puberty,
especially with
limited
contact with
the
target language.
The
recent boom
of
interest
in an
early
Start
of
foreign
languages
all
over
the
world
indicates that
despite
the
contraditory
evidence
from
research,
parents
and
stakeholders
tend
to
favour
an
early exposure
to
modern languages (Blondin,
Candelier,
Ede-
The
Critical
Period
Hypothesis
reconsidered
111
lenbos, Johnstone,
Kubanek-German
and
Taeschner
1998;
Curtain
and
Pesola
1994; Dickson
and
Cumming
1998; Edelenbos
and
Johnstone 1996;
Nikolov
and
Curtain
forthcoming).
This
paper focuses
on
exceptional
adult
learners'
SLA,
and
challenges
the
widely
held
view according
to
which
success,
includ-
ing
accentless
proficiency,
is
available
to
early
Starters
only.
Background
to
research
This research
was
designed
to
challenge
the
strong
version
of the CPH and is
based
on
successful
adult learners
at the
English
Department
of
Janus
Pannon-
ius
University,
Pecs
in the
spring
semester
of
1993.
The
aim
of the
study
was to
find
out
if
adults
who
started
SLA
after
puberty
could
achieve
native-like
pro-
ficiency
and
could
be
misidentified
äs
native
Speakers
on a
tape. Besides
this
aim
I
also
wanted
to find out
what motives
and
strategies
these successful learn-
ers
used.
The
case
studies
were conducted
by
two
groups
of
students
in
"The
Critical Period Hypothesis" course
of the
Department
of
English.
Fourth-
and
fifth-year
English
major
students
participating
in the
course were
required
to
identify
adults
in
their
communities
all
over
Hungary
whose second
language
proficiency
was
thought
to be
outstanding.
This research
focus
and
design
is
strikingly
similar
to the
enquiries
described
in
Bongaerts,
van
Summeren, Planken
and
Schils
(1997)
indicating
that
the
synthesis
of the
background
literature
and the
research questions
offer
a
log-
ical
framework
for
enquiries.
In
both
studies
highly
successful learners were
carefully
chosen:
in the
present study
Hungarian
learners
of
English
and
learn-
ers of
Hungarian
with
a
variety
of first
languages,
while
in the
Bongaerts
et
al.
(1997)
study
Dutch
learners
of
English were
targeted.
Similarly
to
these
princi-
ples,
another
study also focused
on
successful adult learners.
loup,
Boustagui,
El
Tigi
and
Moselle
(1994)
investigate
two
adult learners
of
Egytian
Arabic
in
an
untutored
setting.
Participants
In
the
present paper,
altogether
46
case studies were
prepared
but
13
of
these
had
to be
excluded
from
the final
data
either
for the bad
tape
quality
or
because
the
interviewees
started
SLA
between
the
ages
of 12 and 14.
The age of the 33
successful learners ranges
from
20 to 70,
falling
into
two
major
groups:
In
Study
l,
20
learners
of
different
mother
tongues have been
acquiring
Hungarian
in the
host
environment;
whereas
in
Study
2,13
Hungar-
ians
have been
learning
English
äs
a
foreign language
and
have
stayed
in the
host
environment
for a
relatively
short
period
(see
Table
1).
These learners
are
hypothesised
to
form
a
homogeneous sample
from
the
point
of
view
of the CPH
äs
they
all
started
SLA
after
puberty,
at the age of 15
112
Marianne
Nikolov
Table
1.
Interviewees
in
Study
1 and
Study
2
Number
of
study
Study
1
Study
2
Mother
tongue
Hungarian
Russian
American
British
Bulgarian
Spanish
German/English
Laotian
Finnish
Polish
Target
language
English
Hungarian
Hungarian
Hungarian
Hungarian
Hungarian
Hungarian
Hungarian
Hungarian
Hungarian
Number
of
interviewees
13
5
5
4
1
1
1
1
1
1
or
later
äs
adults.
This
paper
reports
on
the
20
learners
of
Hungarian (Study
1)
and
on the 13
case
studies
of
Hungarian learners
of
English (Study
2).
The
measuring
Instrument
After
identifying
successful
learners with
a
good
command
in the
target
lan-
guage
a
structured
interview
was
developed
combining
elicitation
techniques
used
in
other
studies
on the CPH
(e.g.:
Bongaerts
et
al.
1997;
loup
et al.
1994;
Johnson