| |
Abstract:
This study investigates whether different lexicalization
patterns of motion events in English and Spanish have significance
for how speakers of these languages perform on non-linguistic
memory and similarity judgment tasks. We show that performance on
linguistic and non-linguistic tasks are dissociable, unless
language mediates activity during both encoding and performance in
the similarity task. These findings have important implications for
interpreting studies of the linguistic relativity hypothesis.
English and Spanish differ in the type of information that
motion verbs lexicalize. English motion verbs typically express the
manner of motion (e.g., run, walk, stagger, etc.), while the path
of motion is expressed by prepositions (in, out, across). Spanish
motion verbs typically express the path of motion (e.g., entrar,
'enter', salir, 'exit', meter, 'put in', etc.), while manner is
optionally expressed with adverbial phrases. Slobin (1996, 1998)
has suggested that lexicalization patterns influence non-linguistic
performance. On this hypothesis, Spanish speakers should pay more
attention than English speakers to path information in their
perception, memory, and assessment of motion events, because manner
is less frequently mentioned and less codable in this language. To
test this possibility, we compared how speakers of Spanish and
English performed on two non-linguistic tasks, recognition memory
and similarity judgment of 36 video clips of motion events, each
with manner and path variants. To understand the effect of
linguistic processing on non-linguistic representations, we varied
the nature of the encoding before testing for recognition and
similarity.
Participants either encoded the events visually or they encoded
the events linguistically (by providing verbal descriptions). The
linguistic relativity perspective predicted that (a) in recognition
memory, Spanish speakers would make more errors (false alarms) than
English speakers with events that varied in manner but had the same
path, and (b) in the similarity task, Spanish speakers would prefer
events with a similar path to a given target, rather than events
with similar manner.
We found no language effect in the recognition memory task for
both linguistic and non-linguistic encoding. In both conditions,
there were equal numbers of false alarms to events varying on
manner or path in both English and Spanish. There also was no
effect of language in the similarity task after non-linguistic
encoding. In this condition, English and Spanish speakers showed no
difference in their likelihood of choosing events with similar
manner or path to the target event. However, there was a linguistic
effect in the similarity task after linguistic encoding. In this
condition, Spanish speakers preferred events with the same path as
the target significantly more often than English speakers. Thus
initial linguistic processing prompted responses triggered by the
linguistic pattern but only during a task, similarity judgment,
that could be effectively mediated verbally. This suggests two
points: a) The relative order of linguistic and non-linguistic
tasks should be taken into account in studies of the linguistic
relativity hypothesis; and b) linguistic and non-linguistic tasks
were dissociated in most conditions but participants used
linguistic regularities made available in the context as a strategy
in making non-linguistic judgments.
Slobin, D. 1996. From "thought and language" to thinking for
speaking". Gumperz J., Levinson, S. (eds.) Rethinking Linguistic
Relativity, Cambridge University Press, UK.
Slobin, D., 1998. Verbalized events: a dynamic approach to
linguistic relativity and determinism. Working Papers for the LAUD
Symposium, Linguistic Agency University-GH Essen, Essen.
|
