Review of Language in Cognition
|AUTHOR: Cedric Boeckx
TITLE: Language in Cognition
SUBTITLE: Uncovering Mental Structures and the Rules Behind Them
Michael Shelton, Assistant Professor of Spanish and Linguistics, Occidental College
Boeckx's ''Language in Cognition'' serves as an introductory text appropriate for
beginning-level linguistics courses with a focus on the biological nature of
language. The text explores linguistics within the broader context of the
cognitive sciences. As opposed to traditional introductory linguistics texts,
this book is unique in that it does not take the reader on a tour through the
many subfields of linguistic inquiry (e.g. phonology, morphology, syntax, etc.),
but rather it challenges the reader to consider larger questions about the
nature of the language faculty, what its biological foundation is, and what our
linguistic system shares with other cognitive processes such as vision and
Boeckx begins with a Prologue in which he states the goals of his book. He
explains that he will focus on the biological view of language with the goal of
awakening curiosity among his readers. The assumed audience of the text are
students with little or no previous knowledge of the subject matter. The
chapters place less emphasis on results but rather focus on the questions that
have proven fruitful in bringing those results within reach. As it is also the
focus of research in the humanities and springs from natural curiosity, the
author views linguistics as a natural ''bridge discipline'' to introduce students
to the sciences. The remainder of the book is divided into four parts with three
Part I -- Ever Since Chomsky
In Chapter 1, the author lays out the goals of linguistic study. He highlights
that linguistics is indeed a young discipline with many basic questions still
unsolved. The distinction between ''E-language'' (social/cultural perspectives)
and ''I-language'' (cognitive perspectives) is discussed with the view that
individual languages are the tools one can use to test what's happening
cognitively underneath. This chapter also explains the difference between
descriptive and prescriptive grammars and explains that part of understanding
the constraints of the linguistic system includes identifying what is
constrained purely within the system itself, and what limitations are shared
with other cognitive abilities. In this chapter Boeckx also outlines the general
organization of the book stating that there are five broad questions of interest
to understand: how to best characterize our knowledge of language, how that
knowledge is acquired, how it is implemented in the brain, how it is put to use,
and how it emerged in the species. He closes the chapter by motivating
linguistic inquiry as a theory of the language faculty, a model to investigate
cognition, and a program to create questions regarding the relationship between
the brain and the mind. A useful analogy he adopts is to understand linguistics
as the ''laserbeam'' view within the ''floodlight'' scope of the cognitive sciences.
Chapter 2 couches modern linguistic theory within the broader historical context
of language studies. The ''cognitive revolution'' of the 1950s is characterized as
a program to develop a computational-representational theory of mind which was
motivated by the revival of 17th and 18th century philosophical views on
language, ethological studies of animal behavior, advances in mathematical
research studying computation, and a break away from the behaviorist paradigm in
psychology. This chapter discusses Chomsky's seminal review of Skinner (1957) in
which he critiques behaviorism and outlines the framework for a
nativist/biological account of language. Chomsky's views of language are then
related to work in other disciplines. The author describes that nativist
perspectives on language drawn from ethological studies which examine how
behavioral programs mature during development in other animals, including
studies on instincts and imprinting. He argues that Chomsky adopted from the
ethologists the central idea that, while experience with the outside environment
may be necessary, this experience requires mental computations tailored to what
is to be learned, a set of task-specific learning organs that know what to
extract from the specific experiences. It is here that Boeckx first introduces
the nature of the Language Acquisition Device (LAD) or Universal Grammar (UG).
The chapter then continues to identify insights that modern linguists may draw
from Cartesian philosophy. Highlighted are certain claims by Descartes that
resonate strongly with our current understanding of language, such as the
importance of our ability to detach the use of language from its immediate
circumstances, the idea that mechanical explanations should be the aim of
scientific understanding, and the emphasis on the creative aspects of language
use and the importance of internal structures. This chapter also introduces
Hobbes' perspective that thinking consists of performing arithmetic-like
operations on these internal structures, which can be likened to the modern
concept of a computational theory of mind. Hobbes also stressed the importance
of type-token distinctions, compositionality, quantification, and recursion as
essential features of human cognition. Lastly, Chapter 2 examines the
contributions of the mathematical nature of information processing that Turing
and Shannon made to modern cognitive science.
Chapter 3 introduces the poverty of the stimulus with discussions of Plato's
''Meno'', creoles, and newly formed sign languages. The author also presents the
problematic nature of the perhaps more intuitive concept that language is
acquired solely through cultural transmission with a review of cultural
differences in child-directed speech, the fallacy of grammar correction by
caretakers, structure-dependency illustrated through examples of AUX inversion
in English, and the acquisition of ''visual'' words (e.g. look, see) by blind
children. The last section of Chapter 3 explores Lenneberg's work on the
biological foundations of language (1967). The reader is introduced to the
arguments in favor of a critical period for first language acquisition as well
as research on critical periods in other areas of cognition, such as Hubel &
Wiesel's (2004) work on the visual system. Elaborating on the critical period
hypothesis (CPH) for language, the author cites the usual studies on feral
children and Genie, in addition to reviewing research techniques with infants
such as habituation and sucking rate studies. Indirect support for the CPH from
second language acquisition among late learners closes Chapter 3 and Part I.
Part II -- Unweaving the Sentence
Chapter 4 introduces the basic tenets of syntactic theory and the importance of
underlying hidden structure. This chapter in particular makes very insightful
analogies to other cognitive domains to exemplify how we automatically structure
what we extract from the input in our environment, listing examples in our
interpretation of colors and shapes, such as in the Kanizsa triangle (1955), and
in language from Lewis Carroll's ''Jabberwocky'' (1872). In its discussion of
syntax, the chapter explores ideas central to Chomsky's ''Syntactic Structures''
(1957), such as how to formalize the productive nature of language (infinite use
of finite means). The reader learns about phrase structure rules after examining
finite-state machines and how they fail to account for many aspects of
linguistic structure, such as dependencies and imbedded clauses. By viewing
crossing dependencies, the reader sees the need for transformations applied to
underlying structures, thus setting the stage for hierarchical structure in
language in the form of generative-transformational grammar. In the subsequent
section, Boeckx examines other ''hidden'' properties such as fractality which
leads to our understanding of headed hierarchies (exemplified in simplified X'
Theory), and how we are able to transform the hierarchical structure of language
into the linear order needed for speech. In this section the author draws more
connections to the visual system in a discussion of our ability to transform 3D
images into 2D signals in the retina. Dimension-reduction always leads to a loss
of information, and the discussion is brought back to linguistics via examples
of syntactic ambiguity. Chapter 4 also introduces concepts such as locality,
which motivates the interpretation of the shortest dependency available, and
conservativity, which motivates traces and copies within transformations.
A clear exposition of parameters and comparative linguistic studies with
illustrative examples and insightful analogies to other sciences forms Chapter
5. Boeckx begins with the apparent paradox that we find in language: one human
cognitive system, but thousands of linguistic varieties in the world. In this
discussion the reader encounters the proposition that perhaps the surface-level
differences are more alike than they may seem. Using examples such as wh-words
in English and Chinese, pro-drop vs. non-pro-drop languages, adverb placement in
English and French, and head direction in English and Japanese, the chapter
motivates the central concepts of Principles and Parameters, illustrated in the
typical manner as a set of switches. Again referencing biology, the author
describes analogous behavior among genes, in which master genes ''turn on and
off'' other genes, to demonstrate that variation is not endless but rather
severely limited by cognitive/genetic constraints. Lastly, Chapter 5 highlights
the importance of comparative linguistics and emphasizes how comparisons between
seemingly unrelated languages (such as examples from transitivity in English and
Basque) can illuminate a general cognitive aim.
Chapter 6 tackles language acquisition. The author admits that parameters show
how a child could in principle solve the language acquisition problem, yet he
does not think that we have found a way to show how it is done in practice. This
leads to a discussion of computational modeling. A staunch nativist perspective
is presented in a critique of Saffran, Aslin & Newport's (1996) study on
statistical learning, arguing that a baby can only track the transitional
properties of syllables in the input if pre-specified mental structures are
already in place, including knowledge of what a syllable is. The chapter
outlines a rule-based account of language acquisition with examples from regular
and irregular verbal forms, which continues with the need for cues which help to
set parameters by cuing certain structures as important to look for in the
input. The crucial role of frequency of forms and structure is also discussed in
a short section. The chapter closes with an exposition of the Continuity
Hypothesis which describes how children's errors may also be revealing of UG in
that the mistakes they make fall within crosslinguistically attested possibilities.
Part III -- The Mental Foundations of Behavior
Chapter 7 offers a discussion of meaning from a strong philosophical
perspective. In the first section, the author discusses logicist approaches and
supports a mind-dependent view of semantics. The author warns against the
temptation to confuse meaning with reference, arguing that knowing what a word
means and knowing how to use it are different things. The interaction effects of
meaning and reference are expounded with examples in which truth-conditions and
ambiguous sentences exemplify context-dependent variability in meaning. A
description of entailment follows with the importance of negative facts (i.e.
one must investigate what words and sentences cannot be as well as what they
can). The licensing of negative polarity items and certain quantifiers (every
vs. some) illustrate this issue and reiterate that meaning relies on properties
that are intrinsic to sentence structure and do not represent the shifts in
contexts of use. The final sections of Chapter 5 explore the meaning of words as
instructions to build concepts in the mind. The reader sees how certain concepts
can exist in the mind despite varying specific references in the real world
(e.g. book, green, triangle).
Chapter 8 examines the relationship between language and thought, given the
striking similarities yet remarkable differences between our cognition and those
of other animals. Here the concept of modularity is introduced as
special-purpose mechanisms that perform specific tasks related to certain types
of processing. The chapter explores how human infants, chicks, and adult monkeys
represent objects, how we track numerosities, how we interpret agency and goals,
the nature of our sense of space, and beginning evidence for a core knowledge
system for social interaction (such as infants' preference for race, gender, and
native language), ending in a discussion of Spelke's (e.g. 2000) and Fodor's
(e.g. 1983) work on modularity. From here the chapter explores ways in which
natural language breaks the bounds of traditional views of modularity in that
language combines information across various modules. The chapter then ends with
an argument for what makes us human: a theory of mind and the flexibility to
combine representations that would otherwise remain isolated.
Chapter 9 introduces competence and performance and explores theoretical and
experimental methods in linguistics and psycholinguistics. Examples of
processing difficulties with center-embedding illustrate how theoretically
grammatical constructs can break down due to performance issues, in this case
working memory. Next the chapter explores the history that led to a divide
between theoretical linguistics and processing studies in language production
and comprehension. A note on acceptability judgments argues that they should be
viewed as behavioral data despite claims that they are too gradient and exhibit
interaction effects. The last section of Chapter 9 involves a detailed
discussion of six arguments that have divided psycholinguists and theoretical
linguists. Among these issues are discrepancies between theoretical grammars and
parsers, issues of ambiguity resolution, differences in timing (parsing and
production are fast, acceptability judgments are slow), disagreement regarding
unified or separate systems for production and comprehension, and the alleged
failure of the derivational theory of complexity.
Part IV -- Missing Links
In the final chapters of his book, Boeckx explores current questions in the
relationship between the mind and the brain and the evolution of mental
faculties. Chapter 10 begins with the classic description of the functional
localization of language in Broca's and Wernicke's areas. Following the typical
discussion found in most introductory textbooks, the author describes what he
considers to be the major flaws of the classic model, including the complex
symptoms of aphasics, the overly simplistic linguistic foundations of the model,
and the inability of the anatomical assertions to hold up in light of subsequent
observations. The author argues that, despite its limitations, this model
remains in the mainstream due to a history starting with Gall's phrenology and
findings that show strong evidence of localization in other sensory and motor
domains, such as vision, hearing and the somatosensory system. A discussion of
the problematic nature of linking neurology with linguistic theory stresses
fundamental problems such as how linguistic theory focuses on finely-grained
distinctions, whereas neurological studies have tried to localize ''syntax'' or
''phonology'' very broadly defined. The chapter concludes with the argument that
what might be localized are the underlying specific computations that cognitive
scientists have motivated on independent grounds.
Chapter 11 is entitled ''Homo Combinans'' and reiterates the uniqueness of the
human ability to combine linguistic units in productive and creative ways. It
starts with a discussion of evolutionary studies that have moved away from
questioning the emergence of language and instead focus on adaptation and
selection. Comparative studies of vervet monkeys, bees, and songbirds identify
the unique qualities of human language. Shifting the question slightly, Boeckx
argues in Chapter 11 that a fruitful approach to comparative research would be
to study in animals the computations and representation that underlie language
rather than to search for language itself. The chapter concludes arguing that
human cognition is special not only because we combine linguistic items in
creative ways, but also due to our ability to combine information (in
computational terms) across various cognitive domains.
The final chapter identifies current research in linguistics and genetics,
stating clearly from the outset that there is no specific gene for language. The
reader learns of the famous case of the KE family and the FOXP2 gene, with a
description of how foxp2 exists in many mammals and has been shown to have an
important role in various areas of the body. This discussion returns to the view
of the mind as a collection of special-purpose machines, or modules. While the
author admits the possibility of a modular mind, he argues that the modules
share cognitive primitives, such as ''combine'', ''linearize'', or ''concatenate'',
and it is these shared combinations that may underlie localization. The chapter
concludes emphasizing the need for a computation-oriented approach to the study
of the mind, linguistics, and cognitive science in general.
An epilogue concludes the book with an extension of the biological arguments for
language to moral competence and to the mental constructs of music.
Overall this text is a clearly written, thought-provoking discussion of the
cognitive sciences through the lens of language. I believe this book does
achieve its goal of piquing the curiosity of its readers regarding the
biological nature and origins of language, and how language fits into the mind
among other cognitive abilities.
Amongst the book's strengths, particularly commendable are the connections made
to other cognitive domains and the biological sciences. They couch language and
linguistics well within our broader understanding of the mind and the body.
Equally laudable is the discussion of the history of language studies from Plato
to Descartes, something I have never seen in such detail in other introductory
texts on linguistics. It gives a well-grounded basis for understanding the
motivation for current perspectives and lines of research.
In general I have few criticisms for the text. The chapters read in a very
conversational yet detailed manner. Included throughout are extensive notes that
point the reader to the sources from which the author derives his examples, and
where one may find more in-depth discussion of the material.
One question that does arise is the best use for such a text. It is clearly a
good choice for a general audience with interest in language and the mind. For
students, however, it is quite different from a typical introduction to
linguistics textbook, with a scope that makes it difficult to place it within a
traditional introduction to linguistics class. The focus is too narrow in its
discussion of linguistics proper, covering syntax and formal semantics
throughout, but with little attention to other subfields such as phonology or
morphology (or related fields such as sociolinguistics). At other times the
focus is quite broad as the author relates language to other cognitive domains
and the biological sciences. The instructor would need to rely quite heavily on
supplementary material to cover basic topics in an introductory class. Perhaps a
special topics course on the biological bases of language, or a course in
cognitive science on language and the mind would be the most appropriate venues
for a book such as this. On a similar note, the author acknowledges in the
prologue that alternative views have been relegated to endnotes (or are often
absent), that the book is ''unabashedly Chomskyan'' (p.xiv), and that as a
syntactician his examples are predominantly of a syntactic nature while other
subfields likely would have served equally well for the purpose of illustration.
As a reference book for those interested in a basic introduction to issues
pertinent to the study of language and the mind, these caveats may be
sufficient. However, as a textbook, one would like to see a more rounded
discussion of topics in which the student does indeed read how other levels of
linguistic inquiry play a role in understanding language's place in the
cognitive sciences, as well as how other approaches beyond generative theories
have informed and encouraged discourse on the nature of language in cognition.
As the book follows the traditional generative approach, the reader also
encounters a description in which, in the simplest of terms, our linguistic
system can be conceived of as a set of lexical items, and grammars which allow
for combination of these items (be it at the phonological, morphological, or
syntactic levels). The syntactic system is discussed and exemplified heavily in
the chapters, yet the nature of the representation of the lexical items is less
developed. The chapters that deal with semantics and the relation between
language and thought do describe various views on meaning, but the novice reader
may be left wondering how this information is represented cognitively, as a
discussion of the lexicon is lacking (topics such as prototypes, schemata, among
others, which often appear in other texts are absent here).
Turning to the supplementary materials at the end of the book, the Guide to
Further Study is excellent. It contains recommendations to read seminal works,
ideas for essays, discussion topics, and references to sources with additional
exercises for practice with data. This section would undoubtedly enhance the
classroom experience for students. As it is a textbook, it is surprising to find
no glossary, given the technical terminology that appears in the chapters. The
reference section is thorough. Included are seminal works on topics discussed in
the book as well as current sources, both those accessible to the novice for
further reading and those more appropriate for the more advanced reader seeking
technical accounts of the material.
In summary, Boeckx's text is a unique introduction to the biological foundations
of language. While perhaps problematic in its scope for a general linguistics
class, the book is well written, with many insightful connections to other areas
of cognition. Anyone interested in exploring the underlying structures that
connect language and the mind will enjoy both the exposition presented in this
text as well as the wealth of additional references it includes.
Carroll, L. (1872). ''Through the looking-glass, and what Alice found there''.
Chomsky, N. (1957). ''Syntactic Structures''. The Hague: Mouton.
Chomsky, N. (1959). Review of Skinner (1957). ''Language'' 35: 26-58.
Fodor, J. A. (1983). ''The modularity of mind: An essay in faculty psychology''.
Cambridge, MA: MIT Press.
Hubel, D. H. & Wiesel, T. N. (2004). ''Brain and visual perception: The story of
a 25-year collaboration''. Oxford: Oxford University Press.
Kanizsa, G. (1955). Margini quasi-percettivi in campi con stimolazione omogenea.
''Revista di Psicologia'', 49 (1): 7--30.
Lenneberg, E. H. (1967). ''Biological foundations of language''. New York: Wiley.
Plato. (1956). ''Protagoras and Meno''. New York: Penguin Classics.
Saffran, J., Aslin, R., & Newport, E. (1996). Statistical learning by 8-month
old infants. ''Science'' 274: 1926-8.
Skinner, B. F. (1957). ''Verbal behavior''. New York: Appleton-Century-Crofts.
Spelke, E. S. (2000). Core knowledge. ''American Psychologist'' 55: 1233-43.
| ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Michael Shelton received his Ph.D. in Hispanic linguistics from The
Pennsylvania State University. He currently teaches general linguistics,
Hispanic linguistics and Spanish language at Occidental College in Los
Angeles, CA, USA. His principle research interests are experimental
approaches to phonology, the cognitive representation of phonological
structure, and language processing.