EDITORS: Mittelberg, Irene; Coulson, Seana; Gonzalez-Marquez, Monica; Spivey,
Michael J.
TITLE: Methods in Cognitive Linguistics
SERIES: Human Cognitive Processing 18
PUBLISHER: John Benjamins
YEAR: 2007
Bram Vandekerckhove, Center for Psycholinguistics, University of Antwerp, Belgium
SUMMARY
The book under review is a methodological handbook for linguists who are
interested in the empirical study of theoretical issues in Cognitive
Linguistics. It should be ''a resource of basic facts a linguist needs to know
before tackling a new methodology and [...] a useful reference later on'' (p.
XXIII). The volume contains 17 papers in total, preceded by a short introduction
by the editors, with a summary of each paper. Leonard Talmy provides the
foreword, in which he claims that all known methodologies have a necessary
contribution to make to the study of language, given their unique strengths and
limitations.
The volume is organized into five parts whose titles more or less speak for
themselves: ''Methods and Motivations'', ''Corpus and Discourse Analysis'', ''Sign
Language and Gesture'', ''Behavioral Research'', and ''Neural Approaches''.
Part I: Methods and Motivations
In the first chapter of Part I, Raymond Gibbs suggests ways for cognitive
linguistics to engage in a more collaborative relationship with experimental
psychology. He argues that cognitive linguists should take more care to
explicate the methods they use to arrive at their theoretical claims and to
frame their theoretical analyses in such a way that they can be tested
experimentally. The author illustrates his arguments by giving some examples
from his own experimental research on conceptual metaphors.
Mittelberg et al. introduce the reader to discourse analysis and corpus
research. In the first part of the chapter the authors give an overview of
different cognitive-functionalist approaches to discourse that can function as
frameworks for empirical qualitative and quantitative discourse analysis. They
also survey some empirical work on written discourse. The second part of the
chapter focuses on corpus-based quantitative research. This part also contains a
summary of some common terms in corpus research, and an annotated list of
popular corpora.
Gonzalez-Marquez et al. discuss experimental methodology in the context of
language research. They go through each part of a research article and explain
why it is there and what it should contain. After a brief discussion of what
constitutes ''the scientific method'', they explain the different steps involved
in the scientific research process. They discuss how to deal with factors such
as cultural differences and linguistic ability, and point out some common
interpretative pitfalls caused by biased thinking on the part of the researcher.
Rafael Núñez gives an introduction to inferential statistics for empirical
cognitive linguistics. The author explains what the difference is between
descriptive and inferential statistics, what variables are, and what role they
play in the design of an experiment. There is a short but very clear discussion
of how theoretical probability distributions relate to the logic of hypothesis
testing. Two parametric statistical tests, the t-test for two independent
samples and the Analysis of Variance, and one non-parametric test, the
Chi-square test, are introduced.
Part II: Corpus and Discourse Analysis
Waugh et al. present three case studies of research on small spoken discourse
corpora. These serve to illustrate the benefits of a cognitive-functionalist
approach to discourse analysis that integrates a variety of different empirical
approaches and data sources in the study of language usage data. Bonnie
Fonseca-Greber and Linda R. Waugh investigated the forms, meanings and uses of
subject pronouns in spoken French, Caroline Vickers studied the different ways
in which native and non-native speakers of English in an US academic community
accommodate to each other, and Betil Eröz explored the influence of cultural
norms on the classroom discourse behavior of Chinese students in a US academy.
Starting from the observation that cognitive linguists in general show ''a
relative lack of enthusiasm'' (p. 149) for corpus-based research. Grondelaers et
al. provide some arguments in favor of corpus research for cognitive
linguistics. They present a methodologically very sound case study on the use of
the Dutch particle 'er' in adjunct-initial presentative constructions. In the
conclusion they argue that the adoption of this kind of corpus methodology will
make linguistics a more collaborative, more cumulative, and slower enterprise.
Part III: Sign Language and Gesture
In the first chapter of Part III, Sherman Wilcox and Jill P. Morford make a case
for the empirical study of signed languages in the context of cognitive
linguistic research, since both approaches can offer each other very interesting
research opportunities. They present some examples of cognitive linguistics
research in signed languages on iconicity, metaphor, mental spaces, gesture and
grammaticization, and language evolution. The authors point out some
methodological issues concerning the use of video data and the transcription of
the data. The chapter ends with a survey of some specific empirical approaches
to signed language research.
Eve Sweetser advocates a multimodal study of language. She argues that co-speech
gesture is a crucial data source for research in cognitive linguistics, since it
can reveal much about embodied conceptual structure that is less obvious in
spoken language. She discusses some of the findings in this research area and
shows how these help forward the study of the relation between language and
cognition. She especially argues that Mental Spaces theory should be a very
productive approach.
The chapter by Irene Mittelberg provides the reader with an overview of the
practical steps involved in the study of spontaneous speech and the accompanying
co-speech gestures. She provides some starting points for research and discusses
some issues concerning recording, assessing and editing video data. Considerable
attention is paid to the transcription of discourse and gestures.
Part IV: Behavioral Research
Laura A. Carlson and Patrick Hill discuss the experimental study of how people
map linguistic descriptions on the spatial arrangements of objects. The methods
they discuss are acceptability judgments, speeded verification tasks, placement
tasks, space-parsing tasks, and production tasks. For each method they give a
general description and some research examples, and they tackle some design
issues and the strengths and weaknesses of the method. They underline the
importance of choosing the right method for the specific research question, and
argue to use a combination of methods.
Simulation semantics hypothesizes that understanding a piece of language
involves the mental simulation of its contents. Benjamin Bergen surveys methods
that can be used to investigate the claims of simulation semantics. On the basis
of research examples, he discusses four types of methods: visual and motor
compatibility and interference experiments, experiments investigating simulation
time effects, and neural imaging methods. The chapter ends with some questions
for future research.
Uri Hasson and Rachel Giora review experimental methods that are used in
psycholinguistic studies of language comprehension. They tackle lexical decision
and naming tasks, free recall and recognition tasks, item verification measures,
reading times, feature-listing and feature-choosing tasks (self-report
measures), and experiments in which the effects of language comprehension on the
performance of a subsequent task is investigated. With each method they provide
its motivations, research examples relevant to cognitive linguistics, and some
further considerations like its strengths, weaknesses and practical issues
concerning its use.
Due to their sensitivity, speed and involuntary nature, eye movements give very
good insights into cognitive processing. Richardson et al. briefly discuss the
role of eye movements in the visual system and survey what eye-tracking research
in experimental psychology has revealed about the cognitive processes underlying
both 'online' perception and action and 'offline' remembering, imagining (e.g.
during narrative comprehension) and reasoning. They then discuss some results of
eyetracking research on sentence processing and metaphor comprehension. The last
section of the chapter addresses the practical side of eye-tracking research.
Brandone et al. present two methods they use at their laboratories to
investigate the cognitive foundations of language, the effects of language
input, and their coordination in preverbal infants: the habituation paradigm and
the intermodal preferential looking paradigm. With each method they describe the
rationale behind it, the procedure, and its relevance for cognitive linguistics.
After the presentation of the two methods, their usefulness is demonstrated
through a case study on the process of verb learning.
In the last paper of Part IV, Kira Gor discusses two experimental paradigms in
the context of research on the processing of inflectional morphology in English,
Italian and Russian by different populations of speakers: a verb elicitation
task with real and nonce verbs, and a lexical decision task. Various frequency
effects and effects of phonological similarity are investigated. The results are
interpreted in favor of a probabilistic rule model of inflectional morphology
which positions itself in between the well-known single-route en dual-route
approaches.
Part V: Neural Approaches
Seana Coulson introduces the use of event-related potentials (ERPs) in the study
of language processing. The chapter starts with a general introduction to the
methodology. Next, an overview is given of the most important language-sensitive
ERP components. This is followed by a brief review of ERP studies that are of
particular interest for cognitive linguists. Coulson discusses studies on the
psychological reality of semantic frames, the processing of figurative language
and the effect of temporal iconicity. She points out some methodological
constraints, and ends the chapter with some suggestions for future
electrophysiological research in cognitive linguistics.
The book ends with Shimon Edelman's exploration of the possibility to integrate
linguistics into a general cognitive science. He discusses some basic
general-purpose computational principles or mechanisms that seem to underlie the
whole of cognition, and presents a theory of language in line with these general
principles. An implemented model of this theory is presented, which learns
language structure in an unsupervised manner from raw text corpora. Edelman
points to some questions that are still open to empirical research.
EVALUATION
I think this handbook could indeed be a very good starting point for every
language researcher who wants to engage in an empirical study of an issue or
claim in cognitive linguistics but does not yet know what methodological options
are available or who wants to know more about the methods he or she considers
using. Most papers in this volume provide very clear and, considering the space
limits, quite comprehensive introductions to empirical methodology, especially
those on experimental methods.
However, the book is not an all-round winner. Generally, I have the impression
that the handbook format is somewhat forced upon the collection of papers that
the editors had at their disposal. The different subject areas that the book
deals with are not equally represented. There are as much as six chapters on
behavioral methods, yet only two on corpus research - one of which is actually
only partly devoted to quantitative corpus research as we know it, and one of
which is almost entirely dedicated to a case study. The last chapter of the book
discusses a computational model of language acquisition and language processing,
but there is actually no general introduction to computational modeling for
cognitive linguistics.
Furthermore, not every paper makes an equally valuable contribution to the book.
I do not know that much of discourse analysis or its methods, but I was not
really impressed by the two chapters on that subject (Mittelberg et al. and
Waugh et al.). Maybe the reason is that this field of research is still
searching for a set of established methods, but in those chapters I missed the
thorough and focused emphasis on methodology that characterizes most of the
other papers in the book. To give just one example of what I mean, on page 121,
Waugh et al. mention that the analyst who is present or participates at a
discourse event he or she wants to study should be careful ''not to bias what the
participants say and how they say it'', but neither of the two chapters seems to
contain any concrete guidelines on how to minimize this researcher bias. It also
seemed to me that the case studies presented in Waugh et al. often miss some
methodological rigor, or at least the methodology is not reported in a very
sound manner. I will take the contribution of Betil Eröz on the classroom
interactions of Chinese students in the US as an example. According to what she
reports, she first observed the classroom behavior of a group of international
students. She mentions that she then selected the subgroup of Chinese students
from that bigger group of five nationalities, without providing any motivation
for that choice. If the investigation followed the order of the report, she only
did a literature review on Chinese classroom behavior and cultural norms after
having noticed some characteristic behavior in the chosen subgroup of Chinese
students. This seems to be a perfect example of an ad-hoc explanation. I think
her empirical investigation should have started from a concrete prediction about
the classroom behavior of the Chinese students in the US based on the
literature, and not the other way around.
I also do not think that a methodological handbook should function as a vehicle
to reporting one's own research if it does not function as illustrating material
in a more general didactic discussion on methodology. Therefore I wonder if some
contributions really belongs in this book, however interesting the research in
itself may be (the best example being the paper on inflectional morphology by
Kira Gor).
Apart from these issues, there are some minor considerations I would like to
address. I do not know if a glossary at the end of the book would have been
doable, but it is always nice to see one in a reference work. I would also have
liked to see a more standardized chapter structure for the papers in parts II to
V, so that the reader knows where to find some specific kind of information.
This is more or less the case for the papers on experimental methods, but I
think it is possible to impose a similar general structure on the other papers
too. In relation to this, I also missed a specific ''Further reading'' section at
the end of each chapter. References to introductory literature are mostly spread
throughout each chapter.
I would like to stress that most of the papers in this book are very good
introductions to empirical methods that could be of interest to cognitive
linguistics. However, I think the book would have been even more of a handbook
than it already is if the different topics for the papers to cover were chosen
and delimited in advance, and if the topics were more evenly distributed among
the papers. In that way, a more balanced and comprehensive coverage of the
different methodological approaches would have been achieved, with each chapter
earning its rightful place among the others. As a methodological reference work,
the book would have benefited from an adherence of the different papers to a
unified set of structure and content guidelines.
ABOUT THE REVIEWER
Bram Vandekerckhove is currently a PhD student in linguistics at the University
of Antwerp, Belgium. His research focuses on the exemplar-based modeling of
human sentence processing. He is also interested in morphological productivity,
which was the subject of his Master's thesis.
|
