Book Reviews
DOMINIC W. MASSARO, editor
University of California, Santa Cruz
THE PSYCHOLOGIST’S SHORTCUT
TO ACTOR–NETWORK THEORY
Actor–Network Theory in Education
By Tara Fenwick and Richard Edwards. London, England:
­Routledge, 2010. 188 pp. Paper, $41.95; hardcover, $135.
Actor–network theory is “a way to intervene,
not a theory of what to think” (p. 1).
This puzzling title of the first chapter of Tara Fen-
wick and Richard Edwards’s book Actor–Network
Theory in Education reflects very well what the
book is about. The juxtaposition of intervention
and thinking is confusing as it violates categorical
distinctions between action and cognition. And
indeed, messing up categorical naturalness is what
actor–network theory is about. It is about doing
“messy research,” as chapter 10 emphasizes; it is
about abandoning well-known categories accepted
in our discipline, and it is about doing empirical
research that questions just what categories have
become established by the people themselves and
their practices, which we study.
The interdisciplinary philosophy of actor–­network
theory was first developed by the French and British
researchers Bruno Latour, Michel Callon, John Law,
and Madelaine Akrish in the early 1980s. These schol-
ars were neither educational researchers nor psychol-
ogists. They were philosophers, political scientists,
anthropologists, and sociologists, the disciplines in
which actor–network theory is still most widely read
and discussed. Biographies of Bruno Latour already
exist, and several books introducing the interdisci-
American Journal of Psychology
Fall 2013, Vol. 126, No. 3 pp. 369–384
plinary field of science and technology studies have
chapters introducing actor–network theory. However,
Actor–Network Theory in Education is pioneering in
dedicating a whole book to this conceptual scheme.
Not only educational researchers but also psy-
chologists who are curious about actor–network
theory should be grateful to Fenwick and Edwards
for writing an introduction that not only explains the
vocabulary and principles of actor–network theory
but also provides a good overview of ways in which
these can be translated into relevant educational re-
search. This is indeed an accomplishment, as one of
the peculiar characteristics of actor–network theory
is to give up on the human as a well-delimited and
clearly specified ontological entity. It takes its point of
departure in the principle of symmetry, which urges
the researcher to ignore one of psychology’s (and
pedagogy’s) most basic ideas of subjects and objects
as being of fundamentally different kinds.
I believe the principle of symmetry is the reason
why actor–network theory has only marginally been
taken up by psychologists, and at the same time, it is
the core reason why this line of thought is fascinat-
ing and important for psychology (and educational
research). Actor–Network Theory in Education is in-
deed a timely book. Fenwick and Edwards emphasize
that in our contemporary technological world we are
all intimately entangled with “glasses and watches . . .
kettles and iPhones, . . . social networking (e.g., Face-
book, Twitter), information sharing (e.g., Delicious,
Flickr), identity construction and marketing (e.g.,
blogs), immersive ecologies (e.g., Second Life)” and
that in such circumstance “to be human is the exotic
 categorization, purifying humanity of the non-human
elements that make its own existence possible” (p.
72). Not only as a category but also when being en-
acted as distinct from any other phenomenon in the
world, the contemporary “human” may find himself
increasingly estranged, and indeed exotic. He may
feel more at ease when he is not “himself ” as a willful
autonomous individual in control of his emotional
and cognitive processes. Psychological well-being
may come to have more to do with being immersed
in the technological environment and with following
the flow of becoming of which he is a part.
These ideas question not only our psychological
diagnostic system, but even core principles of psy-
chological thinking. It may sound like a postmodern
view of the contemporary condition. However, ac-
tor–network theory adds that there is nothing new
about this because “to be human is to be cyborgs” (p.
84). To understand this, we need to remind ourselves
of the title of the first chapter and the fact of actor–
network theory itself as not being a theory of what
to think. The point is not that we have always been
cyborgs but that treating humans as cyborgs opens
up new perspectives on how to think and practice as
psychologists and as educational researchers.
Cyborgs are creatures made up of heterogeneous
matter, of nonhuman as well as human components. In
actor–network terms, investigating cyborgian psychol-
ogy means inquiring into the coming into being—the
“becoming”—of such creatures, along with the mate-
rial–semiotic becoming of psychological traits, emo-
tions, and cognition. Different from inquiring into
causality, links or paths between stimuli and effects,
actor–network theory investigates networks, as its
name indicates. Anything from the largest universe to
the tiniest microbe, and from the hardest stone to the
most intangible philosophical idea, is seen as a network
by actor–network theorists. Networks are assemblages
of human and nonhuman materials, and as networks
“all things, as well as all persons, knowledge and loca-
tions, are relational effects” (p. 17). Any person, trait,
path, emotion, thought, norm, etcetera is what it is be-
cause of the way in which the materials constituting it
are “brought together and linked through processes
of translation” (p. 12). According to actor–network
theory, phenomena do not exist in themselves and
are influenced by external stimuli, nor do they create
output that may affect other discrete entities. On the
contrary, actor–network theory sees everything as fun-
damentally related and when, for example, a child plays
violent video games ­actor–network theory would not
say that the child is affected by the game. Instead, ac-
tor–network theory conceptualizes the relationship (or
370 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
network) as an emerging one through which specific
effects result that translate both the child and the video
game, in comparison to what they were before they
came together in this specific network. Whatever emo-
tions, cognition, norms, behavior, and so on that may
be a result of the child–game network, actor–network
theory would recognize these as “assemblage effects”
and not as being the result of one influencing the other.
When we investigate an object in terms of net-
works, the boundaries of that object become blurred.
One would never look simply at a child and a violent
video game but would include the variety of relations
these are entangled in through the network of which
they are a part and which constitute them. In order
to understand education and learning, Fenwick and
Edwards draw our attention to the development of
curricula, to educational standards, to teaching and
learning practices, to learning technologies, to ac-
counting practices, to educational policies, and many
more. Actor–network theory indeed challenges psy-
chology to start investigating research objects as results
of material–semiotic relations across time and space.
Actor–network theory is not particularly inter-
ested in historical or cultural analysis. Instead, it
questions the way phenomena are enacted and how
phenomena act in practice. Fenwick has published
widely on cybernetics, and the step from cybernetics
to actor–network theory is very understandable. Both
theories aim to conceptualize complexity, they are
both sensitive to the hybrid nature of phenomena,
and they both emphasize system effects. A core differ-
ence between cybernetics and actor–network theory
is the latter’s orientation toward contingent practices
and multiplicity. Even such fixed things as standards
are uncertain in practice, Fenwick and Edwards em-
phasize (p. 87). In each situated instance standards
become entangled in new networks of things, peo-
ple, and semiosis. More than a banal statement that
people change according to the situation they are in
and the people they encounter, actor–network theory
provides a vocabulary for analyzing how a person—or
a specific psychological trait or some other phenom-
enon—is “becoming” through its specific situated
material–semiotic entanglements. This is never pre-
dictable because it is always contingent.
Although it may seem complex enough to think
about people or psychological traits as results of mate-
rial–semiotic relations across time and space, Fenwick
and Edwards emphasize throughout the book that
according to actor–network theory such phenomena
are “multiple.” Again, this conceptualization resonates
with postmodern thought. However, later develop-
ments of actor–network theory (after-ANT), which
Fenwick and Edwards seem particularly sympathetic
to, argue against the understanding that there may al-
ways be different perspectives on a phenomenon. In-
stead, they argue that a phenomenon is multiple. Social
psychologists develop different models of aggression,
and one may understand these as different perspectives
of the phenomenon aggression. However, after-ANT
criticizes such a conceptualization because it moves
the focus away from aggression and instead looks at
the models as something separate from the phenom-
enon they model. It is not only the phenomena we
investigate that become related in networks. According
to actor–network theory, investigating a phenomenon
means entangling this phenomenon in the network of
our own models and methods. Thus, aggression mod-
els entangle aggressive people, aggressive behavior, and
norms about aggression in different material–semiotic
networks of scientific thought. The models become,
in different ways, part of the phenomenon itself. This
does not create a plurality of “aggressions” but a mul-
tiplicity, after-ANT argues. This means that aggression
is one single phenomenon, but it comes in multiple
versions, depending on the material–semiotic network
in which it is entangled. Tracing the material–semi-
otic becoming of aggression, as actor–network theory
does, would reveal the continuity from one version to
the other and thus the simultaneous singularity and
multiplicity of this phenomenon.
Actor–Network Theory in Education presents the
principles and ideas of actor–network theory along
with a large number of examples of how educational
scholars have applied them in empirical research. This
way of discussing actor–network theory is faithful to
the approach’s emphasis on empirical research accom-
panied by conceptual innovation. However, the long
chain of sketches of empirical studies also makes the
reading tedious at times. Some parts could usefully be
read as references to be consulted when searching for
studies of particular fields or approaches.
The book partly suffers from too much jargon. Ac-
tor–network theory has given birth to a broad range
of theoretical concepts that are often evocative and in-
spiring. When exaggerated—as is unfortunately often
the case in studies using actor–network theory—the
question arises whether there is actually any innovative
thought behind the fancy formulations. Fortunately,
these irritations are minor compared with the extreme-
ly helpful and well-written introduction Actor–Network
Theory in Education provides for understanding this
complicated line of thought, for educational research-
ers as well as for psychologists.
“To conclude is . . . also to begin. To close is also to
open. Therefore, as we reach the final chapter of this
text, we are also aware that this is no ending” (p. 165).
Fenwick and Edwards continue by discussing pos-
sible ways in which their discussion of actor–network
theory in education may be taken up and reassembled.
In the spirit of their book, instead of representing the
book’s arguments I have here tried to reassemble the
arguments with examples from psychology. In com-
parison to the tendency of postmodern and other
theories that introduce new ways of thinking, actor–
network theory is not critical in the sense of provid-
ing competing perspectives from the outside. Fenwick
and Edwards conclude with a quote by Bruno Latour:
“The critic is not the one who debunks but the one
who assembles. The critic is not the one who lifts the
rugs from under the feet of naïve believers, but the one
who offers the participants arenas in which to gather”
(p. 165). This ethic is one of the reasons why I find
it so comfortable to work with actor–network theory.
Actor–network theory is an invitation to intervene. Few
psychologists have taken up this invitation, but I would
like to mention a few who have for colleagues curious
about how to take up such interventions in psychol-
ogy: Steven Brown, Nick Lee, and Paul Stenner.
Estrid Sørensen
Ruhr-University Bochum
Mercator Research Group “Spaces of Anthropological
Knowledge”
Universitätsstrasse 150
44801 Bochum, Germany
E-mail: estrid.sorensen@rub.de
BEYOND LATOUR: RETHINKING KNOWLEDGE
AND EDUCATION THROUGH MATERIALITY
The Materiality of Learning: Technology
and Knowledge in Educational Practice
By Estrid Sørensen. New York, NY: Cambridge University Press,
2011. 210 pp. Hardcover, $42.
In recent decades of educational writings we have
witnessed debates about learning framed by op-
posing metaphors of acquisition and participation,
roughly corresponding to more psychological or
more sociological and anthropological approaches.
Such divisions are oversimplistic, of course, and have
led to some rather simplistic uptakes of notions such
as communities of practice to explain learning pro-
cesses. But conceptions of activity and practice have
become increasingly well developed in recent de-
cades of sociocultural studies of learning, ­eschewing
BOOK REVIEWS • 371
 assumptions that learning is primarily a mentalist,
person-centered or even intersubjective process.
Some have even announced a “practice” turn (Mi-
etinnen, Samra-Fredericks, & Yanow, 2009; Nicolini,
Gherardi, & Yanow, 2003; Schatzki, Knorr Cetina, &
von Savigny, 2001), drawing attention to the impor-
tance of nonhuman as well as human interrelations
in learning and everyday activity. Many have begun
to point to the problematic absence of materiality
in conventional studies of learning: artifacts, tools,
technologies, bodies, environments, and nature.
We have therefore begun to witness increasing
interest in the material in learning that goes beyond
the materialism of forms of Marxism and draws in
theoretical framings from fields ranging from Latour’s
actor–network theory and science and technology
studies to material culture and human geography. In
these understandings, the material is not separate and
distinct from the social and certainly not subordinate.
Instead the central premise is that the social and the
material are entangled and together constitute every-
day life. That is, knowing and learning, identities,
activities, and environments are understood to be
sociomaterial enactments.
It is in this context that Estrid Sørensen’s book
The Materiality of Learning is a focus of much in-
terest. Presenting her own empirical studies of chil-
dren’s practices with computers, she draws heavily
on actor–network theory (ANT), activity theory, and
communities of practice to illuminate the materiality
of that which is enacted. Her focus is on these prac-
tices as learning, but we suggest that actually some
aspects of her theory potentially raise questions about
whether learning is itself other than a categorization,
part of the materiality of education. But let us start
back a few steps: If we begin to give the material its
place in learning, how do we theorize this? What role
do artifacts, tools, and environments play in learning?
Can we even think of them as entities separate from
the human? Does the material embrace both the hu-
man and nonhuman, and with what consequences?
When one thinks about learning practices, it is
difficult not to consider the material, yet often in our
accounts it is overlooked or marginalized. It is per-
haps with the increasing ubiquity of computing tech-
nologies that it becomes impossible to overlook the
material, even as ironically the digital is often posi-
tioned as not really real. Computing as both tool and
environment has enhanced the questioning of such
categorizations, and some educationalists have gone
to actor–network theory and science and technology
studies more broadly to help with the reframing of
372 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
learning in such circumstances. Here, as Sørensen’s
studies clearly show, matter is not simply an inert
background or tool, dependent on human agency,
but an active contributor to such agency. Learning
then is not about acquisition or participation but is
the entangled (un)doings of matter.
This raises fundamental questions about many
assumptions about learning and education and the
strongly mentalist foundations underpinning many
of our understandings. It challenges the centering of
the human who learns as that which matters. It sug-
gests a significant reframing of educational practices,
wherein categories such as the social and technological
are not taken for granted but are themselves seen as
the outcome of material practices, or material effects.
It also challenges what many, including Sørensen, have
critically described as the representationalist knowl-
edge that still dominates formal educational processes.
From the outset, Sørensen argues that the general
omission of materiality in educational studies has cre-
ated serious limitations in our current conceptions of
children’s learning, classroom life, and pedagogic
practices. She complains about “the blindness toward
the question of how educational practice is affected
by materials” (p. 2) and argues that its consequence
is to treat materials as mere instruments to advance
educational performance. In her book-length study
of the materiality of learning, she shows how forms
of knowledge and forms of presence—the dynamics
of everyday performance in education—are critically
shaped through materiality. She makes it clear that the
way forward is not to simply focus on material tools or
other artifacts as separate from human forces. She also
eschews the excessive emphasis on relations, which for
Sørensen continues to reify a notion of well-defined
entities that are linked. Instead, she urges a focus on
spatial formations. Her approach in tracing what we
would call the sociomaterial configurations of learning
and education is to concentrate on the patterns that
emerge and form spaces. In an earlier article (Sørensen,
2007, p. 10), she explains,
I suggest we focus on the pattern of the gather-
ing, hybrid, network or assemblage or whichever
relational notion we may prefer. Thus, we may
understand materiality as the formed pattern in
which a particular entity takes part and which
allows it to relate in particular ways to (an)other
particular entity(ies). With this definition, we can
talk about the materiality of materials as well as
about the materiality of social entities. Materiality
is, notably, not an essential property of an entity,
 but a distributed effect. The materiality of an en-
tity achieves its particularities through the way in
which it is arranged with other entities.
Sørensen’s ethnographic studies proceed not
only to attempt to illustrate why materiality is im-
portant to consider in educational concerns but also
to theorize how materiality actually works in pat-
terns to produce different forms of knowledge and
different learning practices. Working with methods
advocated in part from ANT, she traces the ways
in which disparate elements become enrolled in
heterogeneous (e.g., human, imaginary, nonhu-
man, virtual, textual) assemblages or networks that
order knowledge and identities in particular ways.
Sometimes these assemblages become stabilized or
blackboxed as taken-for-granted routines or as par-
ticular concepts. Many regions of pedagogic activity,
for example, are organized to perform representa-
tionalist knowledge, such as Sørensen’s example of
children learning mathematics through enrollment
in concepts of standard measures embedded in rul-
ers, textbooks, and jumping demonstrations.
However, as Sørensen shows, these enrolling dy-
namics are more fluid and multiple than the blackbox
story might suggest. Her study of children building
virtual villages in the virtual world of “Femtedit”
shows the difficulties of actually sustaining children’s
engagement in a new assemblage of learning while
reconciling multiple school participants, information
and communication technology lab access, local ver-
sus online communications, children’s experimenta-
tions, technological difficulties, interests of research-
ers from different countries, and so forth. Although
the technology as a system of components needed
to be sufficiently stabilized to support all the activi-
ties for the duration of the project, it also needed to
remain sufficiently flexible to adapt to the changing
interactions and knowledge of those it was enroll-
ing. In such analyses, Sørensen finds the metaphor
of “network” from ANT too limiting to capture such
inherently diverse regional textures that she uncovers
in classroom activities and their relationship to forms
of knowledge produced. The network technology
along with all the other human and nonhuman ma-
terial elements enrolled in the Femtedit pedagogy is
stabilized and mobilized but also enables a liquid as-
semblage of playful, nonrepresentational knowledge:
“Liquid knowledge of the virtual environment is not a
knowledge that maps the practice of the environment
‘on the scale of a mile to a mile.’ Liquid knowledge is
not a map. It is not regional” (p. 128). She contrasts
this with the “communal knowledge,” in Sørensen’s
terms, that she traces in shared classroom experi-
ences such as celebration of a completed collabora-
tive project. Communal knowledge spreads among
its participants even after the experience, depending
on a sort of resonance space. It may gather around a
center, but it is not knowledge about this center.
Perhaps one of the most original contributions in
Sørensen’s analysis is to conceptualize materiality and
material regions temporally. As she justifiably points
out, with particular reference to Latour’s conceptions,
ANT’s relational views are temporally flat and there-
fore lack a certain analytic power. In her earlier article
reporting the studies presented in this book (Sørensen,
2007, p. 10), she develops this point explicitly:
Moving from an understanding of material-
ity as associated with entities in a temporally
stable context to a relational comprehension,
the temporality of materiality is no longer pre-
given; neither as stable nor as ephemeral. The
question of the materiality as well as the time of
materiality of an entity thus becomes an empiri-
cal matter. As we shall see, time indeed becomes
a defining aspect of materiality.
This is a particularly important observation in
applying ANT to analyze learning processes, which
are inherently temporal phenomena. Sørensen works
with ANT to examine three educational occasions,
showing how the pattern of material relations in each
varies according to the temporality of each. A black-
board in a classroom activity, for instance, exercises
a regional materiality that is atemporal. In contrast, a
distinct temporal horizon enacts a networked materi-
ality in her example of a school celebration of a bed
loft built in the classroom. The celebration brings to-
gether parents and headmaster with children, snacks,
and a song in one event, which is also networked
with the substances and activities constructing the
bed loft and the children’s activities on the loft after
the celebration. The materiality is networked, claims
Sørensen, but with temporality linked to the lifetime
of the bed loft (as long as no representations of it
circulate beyond this materiality, which would con-
stitute a different pattern of material relations). In her
third example, learning activities within an online
virtual environment have a fluid materiality, mediated
by ongoing transformation in a temporality that was
extended and continuous. In all of these, humans are
understood to be performances of human presence
within these material patterns. These sociomaterial
dynamics can be examined in terms of the openings
BOOK REVIEWS • 373
 and support they enable for human participation. In
her final analysis, different spatial–temporal patterns
of materiality produce different kinds of knowledge,
only some of which involve learning.
Liquid knowledge is inseparable from learning.
Representational knowledge can be stagnant. It
is not dependent on learning taking place, and
it is indifferent to whether or not learning takes
place. Communal knowledge can endure with-
out learning, but, when learning happens, the
communal knowledge is affected. The ongoing
mutation that characterizes liquid knowledge is
the epitome of learning. (p. 131)
Sørensen’s book is based on empirical studies
of the participations, performances, and imaginings
of two initiatives to introduce online environments
into Danish and Swedish classrooms. As she states,
“The logical meaning and coherence of the concepts
we use is less important: what is crucial is how they
help us do empirical studies and analyses and the
kinds of studies and analyses in which they result”
(p. 12). Yet one of the curiosities of this book is that
although it ostensibly focuses on the materiality of
learning, the empirical matter of the study tends to
become a background with which there are some oc-
casional entanglings. The matter at hand seems more
centrally focused on how to theorize the materiality
of learning more appropriately. In this, the book il-
lustrates some of the potentialities of an exploration
of materiality but also certain important current theo-
retical and empirical limitations. At heart, this offers
a highly stimulating work that brings many differ-
ent theoretical insights to the framing of children’s
practices but—and this is the case for most existing
studies—struggles to establish a fully materialized
and materializing discourse of the empirical data on
which she draws. In an approach that is trying to
show (to perform the materiality of learning) and not
tell (write about the materiality of learning), we end
up with a bit more of the latter than the former.
Sørensen concludes her study with five core sen-
sitivities: to materiality as performed, to the variations
in the uptakes of technologies, to spatial patterns of
relations, to multiplicity, and to particularity. These are
valuable categories pointing to broad methodological
orientations that would be familiar for those working
in ANT, STS, and new anthropologies. Such sensitivi-
ties provide broad methodological orientations. How-
ever, in the detail of Sørensen’s own study, we question
certain theoretical issues. The first is her adoption of
activity theory and communities of practice as part of
374 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
her analytical framing. Although each of these has re-
lated concepts and focus, they do not sit comfortably
with ANT in their framings of materiality. And ANT
is itself a much-divided territory with complexities
that perhaps escape the categories of “classic ANT”
and “after-ANT” used by Sørensen, the latter term
arguably arising in some quarters when “post-post”
fashions were assailing social sciences. In particular
the works of Bruno Latour, John Law, and Anne-Marie
Mol, on which Sørensen draws, resist binary categories
and have followed diverse, nuanced trajectories. Sec-
ond, and perhaps more fundamentally, one is left with
the question of what the learning was at the heart of
this study. Learning itself seemed to be somewhat taken
for granted as a concept and a process (i.e., if chil-
dren are in schools, then they are learning). How the
practices in which these children were engaged come
to be framed as learning remains a central question.
And what is it to be not learning? Where is struggle, or
capacity, in learning, and how is materiality configured
in these dynamics? Sørensen rightly calls for a different
vocabulary or discourse of learning. While significantly
opening up some possibilities for such a discourse, her
book also shows how far we have yet to travel.
Tara Fenwick
School of Education
University of Stirling
FK9 4LA
United Kingdom
E-mail: tara.fenwick@stir.ac.uk
Richard Edwards
School of Education
University of Stirling
FK9 4LA
United Kingdom
E-mail: r.g.edwards@stir.ac.uk
REFERENCES
Mietinnen, R., Samra-Fredericks, D., & Yanow, D. (2009).
Re-turn to practice: An introductory essay. Organization
Studies, 30(12), 1309–1327.
Nicolini, D., Gherardi, S., & Yanow, D. (2003). Knowing in
organisations: A practice-based approach. Armonk, NY:
M.E. Sharpe.
Schatzki, T., Knorr Cetina, K., & von Savigny, E. (Eds.).
(2001). The practice turn in contemporary theory. Lon-
don, England: Routledge.
Sørensen, E. (2007). The time of materiality. Forum: Qualita-
tive Social Research/Sozialforschung, 8(2). Retrieved from
http://www.qualitative-research.net/index.php/fqs/issue/
view/6
A LONG EXCITING JOURNEY
IN BILINGUALISM RESEARCH
Language and Cognition in Bilinguals
and Monolinguals: An Introduction
By Annette M. B. de Groot. New York, NY: Psychology Press,
2011. xiii + 514 pp. Hardcover, £35.
Once upon a time, psycholinguistics was a mono-
lingual affair. Nowadays, bilingualism is seen as
a normal condition for human beings, and it is an
omnipresent phenomenon in societies where most
psycholinguistic research is produced. Today’s psy-
cholinguists ignore bilingualism at their own peril.
The times are therefore ripe for an introduction that
can present the rapidly growing field of the psycho-
linguistics of bilingualism to researchers and students
from various backgrounds.
Language and Cognition in Bilinguals and Mono-
linguals: An Introduction is much more than an in-
troduction. The volume makes a major contribution
to the field, and in spite of its accessibility it is a seri-
ous read for interested researchers and postgradu-
ate students from different backgrounds. Annette de
Groot, professor of experimental psycholinguistics
at the University of Amsterdam, is not only a major
figure in bilingualism research who masters her field
and has made major contributions to it. She also has
long experience in making psycholinguistics research
on bilingualism accessible to graduate students and
researchers, having previously co-edited two widely
used reference tools, Tutorials in Bilingualism (de
Groot & Kroll, 1997) and the Handbook of Bilingual-
ism (Kroll & de Groot, 2005). She is therefore the
ideal person to write an introduction to the psycho-
linguistics of bilingualism. The resulting book is not
the only introduction on the market (for example, see
the notable Introduction to Bilingualism: Principles
and Processes [Altarriba & Heredia, 2008]), but it
surely is the only single-authored introductory vol-
ume that spans more than 500 pages. Producing such
a long and authoritative text requires knowledge,
experience, and dedication, and the field should be
grateful to Professor de Groot for tackling and com-
pleting such a tour de force.
The main attractions of this volume are its exten-
sive, updated, and authoritative coverage and its great
clarity. The volume is chockablock with information,
clearly and expertly presented. At first glance, it can
appear daunting, with its 500 pages in a tightly packed
two-column format, with page after page covered in
dense text, very long chapters (50–70 pages each),
and a weight of more than two kilograms (hardcover).
There are no cute pictures of bilingual babies or color-
ful graphs, everything is black, white, or gray, and all
pictures are informative rather than decorative, in a for-
mat that will be familiar to those who learned or taught
from previous editions of Harley’s The Psychology of
Language (2001). The book cites a large number of
studies, presenting both seminal papers from the past
and recent research and providing a good historical
perspective from the early studies to 2009 (the book
was published in 2010, despite its 2011 copyright). The
35 pages of references will keep students busy reading
for quite a while. However, de Groot does not simply
list facts; she puts them in historical context, explains
which topics have been widely studied and which ones
have only become possible because of which techni-
cal advances, and she points readers to what is very
new, what is currently used, and what is definitely
passé—a very important approach for students who
often struggle to see things in perspective. The volume
presents a number of research topics that have become
prominent only in recent times, from third-language
acquisition to simultaneous interpreting, from sign lan-
guage to bilingual thought. Theory is well presented,
with a clear view of current trends and discussions
of major debates that nicely present both sides of the
fence. de Groot is obviously well placed to present her
own copious contributions in various areas of bilin-
gualism research. Judging from the author’s index, her
name is the most frequently cited in the book, closely
followed by the name of her collaborator, Judith Kroll,
and readers are lucky to be able to hear all these stories
straight from the horse’s mouth.
The other strong point, and probably the stron-
gest selling point of the book, is clarity. The text-
book reveals the experienced teacher and educational
writer who has learnt not to assume too much knowl-
edge on the part of her audience. de Groot happily
drops in explanations of even the most obvious (to
lecturers) acronyms and terms. As someone guilty of
absent-mindedly filling her PowerPoint slides with
mysterious acronyms, I admired de Groot’s thought-
fulness in clarifying the meaning of “C” and “V” in
“CV syllable,” not to mention explanations of other
“self-evident” conventions, explaining that a Span-
ish-Catalan bilingual has probably acquired Spanish
as a chronological first language, for example. This
is not as self-evident as it may appear.
The text does an excellent job of explaining meth-
ods and tasks. These sections are extremely clear and
useful, and I advise lecturers to point their students
to these sections even if they adopt a different text-
BOOK REVIEWS • 375
 book. Methods and tasks are described in detail,
visualized in pictures, and summarized in tables.
The author discusses different ways to implement
the same technique and different names to call it. In
case readers still have doubts, methods and tasks are
then explained again in sufficient detail in the Glos-
sary at the end of the volume. de Groot is very good
at explaining the rationale of studies. Students some-
times understand how a study was done but not why
it was done this way. The author does an admirable
job of explaining both how and why. She explains
how a specific finding answers the research ques-
tion and what opposite results would have meant. For
instance, in describing a study of speech perception
that used novel syllables, de Groot first italicizes the
word novel and then, just in case someone missed
that, explains that results were not due to syllable
familiarity because all syllables were new (p. 35).
Everything is clarified in equally painstaking detail.
Results from main studies are often described in the
text and also presented with table and graphs, either
taken from the original paper or created for this book.
These will be very useful both for students who want
to read the original paper and for those who do not.
One reason for the book’s clarity is that all chapters
follow the same structure. Each chapter contains an
introduction of a couple of pages, which presents the
main concepts, terminology, history, and importance
of the topic. For instance, the chapter on cognitive
consequences of bilingualism opens with two 1920s
articles, describes the relevance of the topic outside
academia for policy makers and educators, and links
it to research reviewed in previous chapters. Intro-
ductions are followed by a section on methods and
tasks, which ranges from a couple of pages for more
basic research areas to more than 10 pages for research
on the bilingual brain. This section is detailed and
clear, and it does a very good job of explaining how
methods have evolved. For instance, in describing the
“language-switching paradigm” (p. 281), the author
describes the “blocked” design used in early stud-
ies and provides four references to such early stud-
ies, then moves on to more recent studies. Another
very useful feature is that bilingual research is always
preceded by a preliminary discussion of monolingual
findings. This helps the reader situate bilingual re-
search, as it often draws on or stems from monolingual
research in its context, and it is particularly useful to
orient readers who do not have a background in psy-
cholinguistics. Each chapter then presents a series
of topics, each one discussed in a section, and each
section ends with a conclusion that sums up the sec-
376 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
tion. Finally, each chapter ends with a Summary, a
couple of pages of bullet lists that summarize the main
points, with each bullet point consisting of three or
fewer sentences, like a dense PowerPoint slide. These
lists will help readers revise their understanding of the
main points of the chapter. The detailed explanation
of methods and results, the coverage of monolingual
psycholinguistics literature, and the extreme attention
to detail make the book suitable for students without
a background in psycholinguistics.
The book’s stated target audience consists primar-
ily of graduate students and researchers in applied
linguistics, cognitive psychology, and linguistics. Per-
sonally, I would add education to this list. Given the
ever-increasing number of school children who are
bilingual, and given that most school children around
the world are learning at least one foreign language in
school, education specialists need to know much more
about bilingualism and its effects on language use and
cognition. I am using the volume as recommended
reading for an MA module in bilingualism in a depart-
ment of education, and reactions have been positive.
The book consists of eight long chapters. These
cover what one would expect from a psycholinguis-
tics textbook, from language acquisition to compre-
hension and speech production, but with a focus on
the bilingual. There are also chapters that would not
feature in a book about monolingual psycholinguis-
tics, such as a chapter on language control and one
on the cognitive consequences of bilingualism.
The first chapter, “Introduction and Preview,” very
briefly presents psycholinguistics and bilingualism.
This very short chapter does no justice to the rest of
the book. It introduces bilingualism in just over three
pages, including all terminology and main concepts.
I disliked the definition of bilinguals as “people who
know and use two languages” (p. 1), which was not
discussed or justified further. Readers who are not fa-
miliar with the main concepts of bilingualism research
will need additional sources to get some grounding
in bilingualism. Unfortunately, the chapter does not
point readers to suitable sources, so if lecturers want
students to be familiar with issues of definition, ter-
minology, and measurement of bilingualism, they will
have to point their students to one of the many good
introductions to bilingualism currently on the market.
With the second chapter, “Early Bilingualism and
Age of Acquisition Effects on (First and) Second
Language Learning,” de Groot begins her impres-
sive journey. The chapter opens with an introduction
and a clear description of methods and tasks such as
the high-amplitude sucking paradigm and the pref-
erential looking technique. It then covers two topics:
early bilingualism, which focuses on phonological and
lexical development in monolingual and bilingual in-
fants, and age effects in first-language acquisition and
in adults’ second-language acquisition, including a
discussion of the critical period hypothesis. Chapter
3 looks at “Late Foreign Vocabulary Learning and
Lexical Representation.” It explains the central role
of vocabulary acquisition in foreign language learning;
evaluates vocabulary instruction methods; discusses
factors that affect vocabulary acquisition, including
characteristics of words such as concreteness and form
similarity to L1 words and characteristics of the learner
such as prior knowledge and phonological memory;
evaluates the revised hierarchical model; and discusses
vocabulary teaching in the classroom. Chapters 4 and 5
deal mostly with lexical access in bilinguals, the former
covering word recognition, and the latter word pro-
duction. Chapter 4 discusses bilinguals’ processing of
interlexical homophones and homographs, illustrates
the bilingual interactive activation model of lexical ac-
cess, and then reviews the scant literature on bilinguals’
sentence processing. Chapter 5 discusses bilinguals’
speech production, referring to research on picture
naming, the Stroop effect, and word translation, and
then examines foreign accent. Chapter 6 looks at lan-
guage control in bilinguals, with more than a third of
the chapter devoted to language control in simultane-
ous interpreters. Chapter 7, “Cognitive Consequences
of Bilingualism and Multilingualism,” covers various
aspects of cross-linguistic influence on language use,
including third-language use and first- and second-lan-
guage loss, and then moves on to effects of bilingualism
on nonverbal cognition, including cognitive control
and thought. The final chapter covers the study of the
bilingual brain and is followed by a glossary, references,
and author and subject indices.
Being an overview of a field, the volume reflects
the limitations of the field. I can imagine readers com-
plaining that, despite the title, the volume does not
cover multilingualism well enough, and those with a
linguistics background may complain about the use
of slashes rather than square brackets in the chapter
on phonology, or about the lack of references to any
linguistics sources that may describe the aspects of
language studied by psycholinguists, for instance,
when phonemes are introduced. These are short-
comings not of this volume but of the field itself,
which are faithfully reflected in this volume.
The book has been accurately produced with the
impeccable editing and proofreading we have come
to expect from Psychology Press. I was slightly an-
noyed by the fact that text on one side of the page
was visible on the other, and by the darkness of some
gray backgrounds in tables and pictures, although the
latter seems to vary across chapters. I was also slightly
disappointed by the subject index, in particular by
the absence of entries for languages in this index.
Furthermore, I would have liked some suggestions
for readings at the end of each chapter. I hope these
points will be taken into account for the next edition.
Despite the very minor drawbacks discussed in
the two previous paragraphs, I believe that this vol-
ume is a godsend for students and lecturers alike.
Students will be surprised that so much complex in-
formation can be made so easily digestible. Lecturers
will be delighted that so much work has been done
for them, and although they may want to supplement
the book with other sources (an introduction to bi-
lingualism, linguistics sources), they will be able to
rely confidently on this extensive and authoritative
volume for their teaching.
In the last sentence of the volume (p. 446), de
Groot describes her book as a “long journey through
the multifaceted study of language and cognition in
bilinguals and multilinguals.” One can only agree
with this description and hope that many researchers
and postgraduate students will be willing to embark
on this long journey, which will provide them with
a very strong grounding in a rapidly evolving and
exciting field.
Benedetta Bassetti
Centre for Language Learning Research
Department of Education
Derwent College
University of York
York YO10 5DD
United Kingdom
E-mail: benedetta.bassetti@york.ac.uk
References
Altarriba, J., & Heredia, R. R. (Eds.). (2008). An introduction
to bilingualism: Principles and processes. London, Eng-
land: Erlbaum.
de Groot, A. M. B., & Kroll, J. F. (Eds.). (1997). Tutorials in
bilingualism: Psycholinguistic perspectives. Mahwah, NJ:
Erlbaum.
Harley, T. (2001). The psychology of language: From data to
theory (2nd ed.). Hove, England: Psychology Press.
Kroll, J. F., & de Groot, A. M. B. (Eds.). (2005). Handbook of
bilingualism: Psycholinguistic perspectives. Oxford, Eng-
land: Oxford University Press.
BOOK REVIEWS • 377
 ABOUT DOUBLESPEAK AND DOUBLETHINK
Language and Bilingual Cognition
Edited by Vivian Cook and Benedetta Bassetti. New York, NY:
Psychology Press, 2011. 591 pp. Hardcover, £52.99.
The strikingly similar titles of Cook and Bassetti’s
Language and Bilingual Cognition (LBC) and mine,
Language and Cognition in Bilinguals and Multi-
linguals: An Introduction (LCBMI), suggest that
these two volumes have a very similar coverage. But
contrary to this suggestion, they clearly comple-
ment rather than duplicate one another, the fleeting
misperception of duplication being caused by the
reader’s spontaneous assignment of the same mean-
ing to the noun language and the conjunction and
shared by the two titles while implicitly different
senses are emphasized.
The larger part of LBC deals with the time-
honored question of linguistic relativity, that is, the
question of whether our thinking is influenced by the
structural characteristics of the language we speak.
In addressing this question, LBC is particularly con-
cerned with the implications of linguistic relativity for
bilinguals, whose two languages can differ in the way
their grammar and lexicon reflect particular aspects
of reality and, consequently, may exert a different in-
fluence on thought. So language in LBC’s title refers
primarily to the structure of (specific) languages, the
language system, and and is used in a relational sense,
the book trying to elucidate the relation between lan-
guage structure and thought. Although language in
the sense of language use is the focus in a couple of
LBC’s chapters, in the majority of them it is present
implicitly only in the role of mediator, the languages’
specific structural characteristics under examination
exerting their potential effects on thinking through
habitual language use (and, initially, language acquisi-
tion). Cognition in LBC’s title is used as synonymous
to thought, mental life, and in the myriad studies re-
viewed in LBC particular aspects of cognition (e.g.,
the mental representation of objects, colors, time,
space, or motion) are typically examined by means
of nonverbal tasks (e.g., choosing the most similar
objects or color chips from a set; memorizing spa-
tial arrangements and reproducing them after having
changed orientation in space).
Instead of focusing on language structure, lan-
guage in LCBMI emphasizes language use, verbal
behavior, in its various forms (especially speaking,
listening, and reading) and the mental processes in-
volved. In most of the experimental work discussed
378 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
in LCBMI bilingual participants performed some
verbal task, such as reading words or sentences or
producing words, and the influence of the contex-
tually inappropriate language on performance was
determined. The conjunction and in the volume’s
title is used in a loose coordinate sense, intending to
convey the information that the volume does not deal
exclusively with language use, verbal manifestations
of bilingual cognition, but that it has something to
say about nonverbal manifestations of it as well. In
fact, the sections of LCBMI that deal with nonverbal
bilingual cognition address the same issues as LBC
does, but the two volumes cover them in very dif-
ferent proportions. For instance, LCBMI has a sec-
tion on linguistic relativity and how it plays out in
bilingualism, but this section covers only 2.5% of the
volume while, as mentioned, the larger part of LBC
deals with this topic. Conversely, both volumes dis-
cuss the relation between bilingualism and nonverbal
executive control, but this is a more central topic in
LCBMI than in LBC.
There are further salient differences between the
two volumes: LCBMI is a monograph intended as an
introduction in psycholinguistics from the viewpoint
of the bilingual language user and is meant to be ac-
cessible to undergraduates (in cognitive psychology
and applied linguistics). Instead, LBC is an edited
volume that aims at a more advanced and specialized
audience, its 25 chapters written by experts in their
respective fields: applied linguistics, psychology,
anthropology, and philosophy. A further salient dif-
ference is the degree of linguistic sophistication that
speaks from the two volumes, this being much larger
in LBC than in LCBMI. This difference obviously fol-
lows from LBC’s focus on language diversity and how
it relates to cognitive diversity. In turn, this specific
focus is likely to be responsible for a final clear differ-
ence between the volumes: The bilinguals examined
in LCBMI most often master a pair of typologically
close languages (e.g., two Germanic languages), and
those in LBC more often master two distant languages
(e.g., Chinese and English). Plausibly, the larger a
particular structural difference between a pair of lan-
guages, the larger the chance a differential effect on
cognition materializes, hence the focus on distant-
language bilingualism in LBC.
LBC consists of three parts. Part A, containing
six chapters, pays tribute to the founding fathers of
the linguistic relativity research field: Edward Sapir,
Benjamin Lee Whorf, Franz Boas, and Wilhelm von
Humboldt. In addition, it provides an informative
overview of the themes, current methods, and find-
ings in the field, doing so from the perspective of the
different contributing disciplines (linguistics, psychol-
ogy, anthropology, and philosophy) and focusing on
cross-language studies of monolingual speakers of dif-
ferent languages. It thus lays the foundation for Part B,
which contains 15 chapters and deals with LBC’s core
issue: language diversity and cognition in bilinguals.
Finally, Part C, containing four chapters, considers im-
plications of the insights gained from research on bi-
lingual cognition for various settings, such as language
teaching and marketing. The book concludes with an
entertaining epilogue on how linguistic relativity has
inspired the science fiction literature.
Parts A and B both start with an introductory
chapter in which the editors prepare their audience
in two ways for what comes. First, they introduce
the cognitive domains that have been examined most
in the study of linguistic relativity (e.g., motion and
time) and briefly point out how each of these is as-
sumed to relate to a specific structural difference be-
tween languages (e.g., differences between languages
in the way they encode the manner and path of mo-
tion and in their use of grammatical tense). Second,
they introduce a couple of important distinctions
(to be explained later) that greatly help the reader
to mentally organize the diverse studies in this field
and to see how all the volume’s chapters indeed heap
together under its title, Language and Bilingual Cog-
nition, even though on first glance there appear to be
some strangers in their midst (not all chapters deal
with linguistic relativity).
In the introductory chapter of Part A, Cook fur-
thermore poses the preliminary questions of whether
differences in thinking between people exist at all
and, if so, whether such differences are indeed cor-
related with differences between the languages they
speak (instead of being correlated with the different
cultures and environments they live in). On the basis
of a succinct presentation of the available evidence,
Cook gingerly provides affirmative answers to both
these questions, concluding that “at least some aspects
of human cognition are not universal” (p. 6) and that
“some aspects of cognition seem to go with particular
aspects of language in a measurable way” (p. 9).
Cook then takes the next logical step, wonder-
ing whether the correlation between language and
cognition implies causality and considering both
possible directions of causation, from cognition to
language and from language to cognition. Of course,
the latter direction of causality constitutes the tenet
of linguistic relativity theory, bringing Cook to pose
a final question: whether changing people’s language
actually alters their (nonverbal) cognition. This being
so is the central idea in the aforementioned epilogue
on linguistic relativity as it features in science fic-
tion (think of George Orwell’s 1984 [1949] and Jack
Vance’s The Languages of Pao [1958/1974]). But more
to the point in view of LBC’s main subject, the wide-
spread phenomenon of second-language (L2) learn-
ing provides a natural playground to examine effects
of language change on people’s cognition (after all,
adding a new language to one’s linguistic repertoire
involves language change). Furthermore, acquiring
two languages from the start (simultaneous bilingual-
ism) should result in a cognitive disposition that dif-
fers from cognition in monolingual language users.
Indeed, it is due to the growing understanding that
L2 learning and bilingualism may have consequenc-
es for cognition that linguistic relativity researchers
have started to embrace L2 learners and bilinguals as
suitable, perhaps even ideal, participants in their ex-
periments (see Pavlenko, 2005, for a discussion). Not
only LBC but also a second recent edited volume on
language and thought in bilinguals (Pavlenko, 2011)
and a special issue of Bilingualism: Language and
Cognition (Jarvis, 2011) attest to this.
As mentioned, Cook and Bassetti present a couple
of contrasts that enable the reader to help structure
LBC’s content and to see the connections between
the chapters. Two correlated contrasts concern one
between transient and lasting effects of language on
cognition and a second one that distinguishes be-
tween effects of linguistic diversity on online thought
processes during language use on one hand and those
on nonlinguistic cognition as manifested in nonverbal
tasks on the other hand. The thinking that takes place
during language use was coined “thinking-for-speak-
ing” by Dan Slobin (1987), who gathered a substantial
amount of evidence showing that such thinking is
language specific, that is, that the thought processes
that are mobilized to express a particular mental
content in one language differ from those mobilized
when the same content is to be expressed in another
language. The reason these thought processes are
language specific is that they must be must be molded
to map onto the available linguistic means, that is, the
grammatical and lexical categories that are encoded
by the language to use (and that differ between lan-
guages). This thinking-for-speaking is presumably
synonymous with conceptualization, the cognitive
processing that in models of speech production (e.g.,
Levelt, 1989) is assumed to be the first stage in the
production process. (Note, however, that Slobin uses
thinking-for-speaking as an umbrella term that covers
BOOK REVIEWS • 379
 not only the thought processes involved in speech but
also those involved in reading, listening, writing, and
other forms of language use.)
In LBC thinking-for-speaking is presented as a
weak version of the hypothesis that language influences
cognition. This version is cursorily alluded to in many
chapters, and it is the main object of study in some.
The strong version, though, the one assuming a last-
ing influence of language structure that is manifest in
cognitive tasks that do not involve the use of language,
is the focus of LBC’s attention. Note, however, that the
two versions are closely related, the lasting effects of
language-specific structures on nonverbal cognition
originating from the fact that during language use the
available structures time and again enforce a particular
mode of thinking. Dealing primarily with bilingualism,
LBC specifically aims to discover whether bilinguals’
thinking-for-speaking in two languages causes their
nonverbal cognition to differ from that of monolin-
guals. The reported evidence suggests that such is the
case and that this holds not only for bilinguals who
have grown up with two languages but also for those
who acquired a second language later in life. A conclu-
sion that language structure exerts lasting effects on
nonverbal cognition therefore should not be taken to
mean that such effects are nonmalleable but only that
they extend beyond actual language use into nonverbal
areas of cognition.
A further relevant contrast pointed out by the
editors distinguishes between two fundamentally
different ways in which bilingualism can exert an ef-
fect on cognition: at a micro level and a macro level.
Micro-level effects result from a specific linguistic
contrast between a bilingual’s two languages. For
instance, French and English differ in the way they
express motion, and, consequently, monolingual
speakers of English may think about motion in a
different way from monolingual speakers of French.
One of the possible effects of this cross-linguistic
difference on French–English bilinguals may then
be that their thinking about motion involves some
merger of French-specific and English-specific mo-
tion cognition. In the majority of LBC’s chapters such
micro-level effects of bilingualism on cognition are
examined, and these are the ones that reflect the lin-
guistic relativity research tradition.
Cognition may also be influenced by bilingual-
ism per se, the fact that the people in question know
and use two languages, irrespective of the structural
contrasts that exist between these two languages.
Such effects of bilingualism are called macro-level
effects by the editors. As I argued in LCBMI, studies
380 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
that look at such effects are relatively easy to design
because they are not constrained by the availability
of specific linguistic contrasts that may affect cogni-
tion in a specific way. Furthermore, in studies of this
type it is not imperative that all bilingual participants
master the same pair of languages so that participant
recruitment is relatively easy.
Two of LBC’s chapters clearly deal with macro-
level effects of bilingualism on cognition. One of
them discusses the effect of knowing two languages
on theory of mind (TOM), defined as the human
faculty that “permits us to reason about the mental
states of others—their beliefs, desires, and inten-
tions—and to understand and anticipate how these
can differ from our own and from reality” (p. 431).
The little evidence there is suggests that bilingual
children are better at TOM tasks than monolingual
children, as is, for instance, suggested by bilingual
children’s superior ability to recognize appropriate
responses in a conversation and violations of conver-
sational maxims such as Grice’s “maxim of manner”
(which prescribes that a speaker should avoid ambi-
guity, confusion, and obscurity). The second chapter
on macro-level effects of bilingualism on cognition
focuses on the beneficial effects of bilingualism on
nonverbal executive control, a topical issue in the
study of bilingualism that has been popularized by,
especially, Ellen Bialystok and the present chapter’s
author (David Green). The underlying idea is that bi-
linguals must incessantly control their two languages
and that this has turned them into experts in execu-
tive control in general. That expertise in language
control transfers to executive control as required
in nonverbal behavior has been demonstrated in
multiple ways, for instance by showing a bilingual
advantage when the performance of bilinguals and
monolinguals is compared on visual-motor tasks that
require the inhibition of conflicting visual informa-
tion (e.g., the Simon task).
In passing, these two chapters on macro-level ef-
fects of bilingualism on cognition exemplify a further
relevant contrast in this field of study, also mentioned
by the editors: It has to be established whether a par-
ticular effect of bilingualism on cognition is caused
by knowing or by using two languages. An effect of
bilingualism on TOM is plausibly due to knowing two
languages, more precisely, to a higher level of metalin-
guistic awareness that knowing two languages as com-
pared with knowing just one brings about. Conversely,
an effect of bilingualism on general executive control
is probably caused by extensive practice in using the
two languages appropriately in each specific context,
which forces the general executive control system to
incessantly monitor that performance is in accordance
with the contextual requirements and to adapt it the
moment it deviates from these requirements.
To get back to LBC’s core theme of linguistic rela-
tivity, the volume’s joint chapters provide a kaleido-
scopic picture of the great structural diversity among
natural languages. In addition, they suggest that the
cross-language differences in question affect nonverbal
cognition in subtle ways and that these differences can
play out differently in bilinguals and monolinguals.
There are languages other than English that require
their speakers to specify in the verb whether an event
they are talking about was actually witnessed by them
or whether they know about it through hearsay. Con-
versely, English marks verbs for grammatical tense, re-
quiring that the time of happening of an event (in the
past, present, or future) is coded in the verb, whereas
in other languages (e.g., Mandarin) time coding in
linguistic expressions is optional, and if time is coded
at all, it is by means other than verb marking. Lan-
guages differ in aspect marking, that is, in whether or
not action verbs are inflected to indicate that the ac-
tion being talked about is completed or still going on
(the perfective–imperfective distinction in English).
They also differ in how they deal with grammatical
number, plural marking of nouns referring to inanimate
objects being obligatory in some languages (e.g., in
English glass must be pluralized as glasses) but forbid-
den in others (e.g., Japanese or Yucatec). These latter
languages talk about inanimate objects the same way
the former talk about substances (from the English
perspective, treating glass as if it were a mass noun
like sugar or sand). Furthermore, languages differ in
grammatical gender marking and the degree to which
grammatical gender and natural gender may be cor-
related, in how they code the location of objects in
space (e.g., in relative or absolute terms) or describe
movements through space, in the direction of their
writing system and whether the written symbols map
onto meaning directly or via phonology, in how many
words they have to describe the color spectrum, and
in what means they have to describe emotional states.
Languages even appear to differ in whether verbs that
describe the handling of objects specify the shape the
objects take during the handling action: In Navajo “a
blanket laid out flat requires a different verb stem from
a rolled-up blanket” (p. 221).
In LBC, all these cross-language differences and
their potential effects on nonverbal monolingual and
bilingual cognition come up one way or the other, in
reviews of the literature or in original experimental
work. Space limitations prevent me from detailing how
each of them may affect nonverbal cognition. I will
therefore confine myself to a detailed illustration of the
effect of language diversity on motion cognition, the
cognitive domain that is represented most prominently
in LBC and perhaps also in linguistic relativity research
in general (Slobin, 2003, called it his “parade-case” in
developing his thinking-for-­speaking theory). In addi-
tion, a brief illustration regarding time cognition will be
presented, just to demonstrate that language structure
indeed influences cognition across a more varied range
of cognitive domains.
Because motion is a universal human experience, all
languages have means to describe motion events, but
the means to do so differ between them. Two features
can be distinguished in all motion events: their path
(the location change) and manner (the way the loca-
tion change is effectuated, e.g., by running, crawling,
creeping, jogging, strolling, or whatever other way we
can move from A to B). Considering these two aspects
of motion, Talmy (2000) classified natural languages in
two categories. “Satellite framed” (S) languages, such
as the Germanic and Slavic languages, encode paths by
means of prepositions and particles (the eponymous
“satellites,” e.g., in, out, under) while they convey
manner in verbs (e.g., crawl, trod, jump, run, walk,
fly, sprint). In contrast, “verb-framed” (V) languages,
such as the Romance and Semitic languages, encode
path in verbs (e.g., Spanish entrar, “enter,” and cruzar,
“go across”), whereas the encoding of manner in these
languages is optional, and, if expressed at all, it is done
by means of adverbial phrases. A correlated difference
is that S-languages have more verbs expressing man-
ner of motion than V-languages do. As a consequence
of these two different patterns of expressing motion,
speakers of S-languages are thought to attend to man-
ner more, and are consequently more sensitive to it,
than speakers of V-languages, whereas speakers of V-
languages are more attentive and sensitive to the path of
motion events (Czechowska & Ewert in LBC; Slobin,
2003). Consequently, in addition to expressing them-
selves differently in verbal tasks that require them to
describe motion events, speakers of S- and V-languages
may also behave differently in nonverbal tasks that re-
flect motion cognition (e.g., memory, mental imagery,
or visual perception tasks).
One possible consequence of these differences
between S- and V-languages for L2 learning is that
the way motion events are verbalized changes when
native speakers of an S-language are learning a V-lan-
guage (or vice versa). A second is that the changes in
verbal expression to be observed become reflected in
BOOK REVIEWS • 381
 nonverbal task performance as well, suggesting cogni-
tive change. In LBC Hendriks and Hickmann exam-
ine the verbal descriptions that L1 English learners of
L2 French (and English and French monolinguals)
provided when they were presented with animated
cartoons showing motion events and asked to de-
scribe them in French. Looking at language use, this
study thus exemplifies a test of the weak thinking-
for-speaking version of the linguistic relativity hy-
pothesis. On the other hand, Czechowska and Ewert
test its strong version by examining how differences
between Polish and English in the expression of mo-
tion events affect nonverbal visual perception.
Even though they both qualify as S-languages, the
path aspect of motion is lexicalized more prominently
in English than in Polish. Czechowska and ­Ewert
therefore hypothesized that speakers of English gen-
erally pay more attention to the path of motion than
speakers of Polish during language use and that this
might lead to differences in the perception of mo-
tion events between speakers of Polish and English.
In addition, motion perception in Polish learners of
English may change such that path of motion gradu-
ally becomes more salient to them.
In one experiment the researchers presented
English and Polish monolinguals and three groups
of L2 English speakers, all with Polish as their L1
but differing in their level of English proficiency,
with pairs of photographs depicting motion events.
The depicted events differed in manner or path of
motion (e.g., one and the same woman walking or
jumping into a room or walking into or out of it), and
the participants were instructed to rate the similarity
of each pair. In a further experiment, triads of these
pictures were presented (e.g., the woman walking
in vs. walking out vs. jumping in), and the partici-
pants were asked to select the two they regarded most
similar. The performance of the two monolingual
groups showed clear differences that could be relat-
ed directly to their specific language experience and
confirmed the hypothesis that English monolinguals
focus more on path than Polish monolinguals (e.g.,
by showing higher similarity ratings for same-path
picture pairs). The results furthermore confirmed
the hypothesis that with increasing levels of English
proficiency path of motion would become gradually
more salient for the L2 English learners. Interestingly,
the similarity ratings for same-path picture pairs of
the most proficient learner group were even higher
than those of the English monolinguals, suggesting
that path of motion was more salient for the former
group. The joint results suggested that in the least
382 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
proficient L2 English speakers the conceptualization
of motion had shifted to that akin to English, whereas
in the two groups with a higher level of proficiency
in L2 English a restructuring of motion cognition
had taken place such that their conception of motion
differed qualitatively from the way Polish and English
monolinguals conceive of motion.
In addition to thus demonstrating two forms of
cognitive change that can take place during L2 learn-
ing (conceptual shift and restructuring), Czechowska
and Ewert stress a crucial point that has remained
implicit so far, namely, that attention allocation plays
a pivotal role in the development of conceptual rep-
resentations from linguistic experience: The spe-
cific linguistic expressions perceived and produced
by a language user (and learner) guide attention to
specific aspects of the situations that embed these
expressions, and what is attended to most becomes
stored in the ensuing memory representations most
prominently. This idea of attention as the underlying
cognitive mechanism determining the content of con-
ceptual representations, the changes of this content as
a consequence of becoming bilingual, and the differ-
ences in conceptual content between monolinguals
and bilinguals are three more general themes in LBC.
To provide one further example, in a chapter on
Chinese–English bilinguals’ sensitivity to the temporal
phases of action events, Chen and Su suggested that
long-term experience with a language that contains
explicit markings on the verb for tense and aspect
(such as English) leads to a perceptual system that
becomes sensitive to the temporal aspects of action
events because these explicit markings direct speakers’
attention to time aspects of the event (Has it finished?
Is it still going on? Will it take place in the future?).
Conversely, people who lack such linguistic experi-
ence (e.g., Chinese monolinguals) should lack this
specific perceptual sensitivity. But when such people
start learning a new language containing grammati-
cal tense and aspect marking, they may still gradually
develop this sensitivity in parallel to learning the new
language. These ideas are supported by an earlier
study testing English and Chinese monolinguals and
the current one testing Chinese speakers of L2 English.
The results of the monolingual study suggested that
English monolinguals perceive the present, past, and
future phases of action events encoded in pictures but
that Chinese monolinguals tend to perceive only an
action event’s present phase. The current study (using
a sentence–picture matching task) shows that Chinese–
English bilinguals with a high level of English profi-
ciency behave like the English monolinguals, similarly
manifesting sensitivity to the different temporal aspects
of depicted action events.
LBC’s joint chapters provide more demonstra-
tions of cognitive change resulting from L2 learning,
and the combined evidence suggests that, in addition
to cognitive shift and restructuring, L2 learning and
bilingualism induce other types of cognitive change.
One of them involves the fusing or mingling of an L1
concept and its most closely related L2 concept. L2
learning may also lead to the emergence of totally new
concepts and the doubling of concepts in cases where
a concept belonging to the new language deviates
substantially from the most similar concept in the old
language. In general, the joint chapters add to earlier
evidence (reviewed by Pavlenko, 2005) that bilingual-
ism leads to various types of cognitive change and
enriches the methods to obtain such evidence.
It takes little imagination to see how some of these
types of cognitive change evolving in tandem with
learning a new language might bring about an in-
crease of mental power or, conversely, how reducing
one’s stock of concepts may have the opposite ef-
fect. In George Orwell’s 1984, Oceania’s totalitarian
regime designed Newspeak to curtail the thought of
Oceania’s people, to render all thoughts that diverged
from the principles of Ingsoc (the regime’s ideology)
utterly unthinkable. In constructing Newspeak on
the basis of Oldspeak, standard English, all linguis-
tic means to express politically incorrect ideas were
removed from the latter, and, to generally limit the
variety of ideas that could be expressed, Newspeak’s
vocabulary was brought down to an absolute mini-
mum. This way, Oldthink, the thinking patterns as-
sociated with and enabled by Oldspeak, would no
longer be possible, which was just as well because
Oldthink was considered Crimethink, and you could
end up in jail for it. To borrow and adapt Orwell’s
terminology, Newspeak thus promoted Narrowthink,
a confined mental world.
But as compared with being bilingual, just master-
ing Oldspeak English or any other single language
also implies a form of Narrowthink. This idea is ex-
ploited in Vance’s The Languages of Pao (1958/1974),
as related by Brooke in LBC’s epilogue. Here the
people of Pao, monolingual speakers of a language
that lacks verbs and thus fosters idleness, are oc-
cupied by the enemy. To cut a long story short, the
hero, who happens to be a bilingual, persuades a
part of Pao’s population to learn a second language
(Pastiche) that “allows its speakers to see the world
more completely” (p. 563). This broader view of
the world enables the L2 speakers to develop an ef-
ficient strategy to tackle the enemy, with the effect
that peace is ultimately restored on Pao. In terms of
Orwell’s vocabulary again, Doublespeak encourages
Doublethink, thus lifting mental powers, including
the knowledge of how to take advantage of them.
Fortunately for Singlethinkers, there are ways to
increase one’s cognitive abilities other than becoming
bilingual. One of them is engaging in metacognition.
Dealing with language and thought—presumably
the two quintessential abilities of humankind—LBC
constantly invites metalinguistic and metacognitive
thinking, time and again forcing the reader to step
back from the text and reflect on the essence of lan-
guage and thought. This is likely to lead to valuable
new insights and changed views in all its readers. The
great diversity of cross-language differences covered
by LBC has strengthened my doubts about the ex-
istence of language universals and made me incline
more toward the view that the recurrent patterns in
linguistic organization as manifested among natural
languages arise not from language universals but from
universal constraints on human cognition (Evans &
Levinson, 2009). In addition, I ponder more than be-
fore on the consequences of English having become
the lingua franca among scientists despite the fact that
it includes scientific concepts that have no proper
equivalent in other languages (in one of LBC’s chap-
ters Wierzbicka argues that cognition itself is such a
concept). Another issue to brood on is the ubiquity
of English as the L2 in bilinguals worldwide despite
the fact that the total number of the world’s natural
languages exceeds 3,000. The conclusion must be
that, so far, we have come to know only a minuscule
portion of what there is to know about bilingual cog-
nition and that the task ahead of us is vast. LBC also
makes one see the source of L2 English utterances
that deviate from standard English and that can truly
baffle a naive reader and listener at times. When I
recently checked my e-mail after coming home from a
visit to China, a number of messages from the people
I had met there were awaiting me. They included
phrases such as “It’s really a great honor to be with
you in Beijing and Guanzhou” and “I am so glad to
meet you at Guandong University of Foreign Stud-
ies.” But I was back in the Netherlands! The person
who had accompanied me to the gate at Hong Kong
airport even wrote, “How is your flight?” Knowing
about the lack of grammatical tense in Chinese makes
one realize that such grammatical infelicities result
from transferring Chinese thinking-for-speaking to
L2 English. In turn, such awareness informs the for-
eign language curriculum.
BOOK REVIEWS • 383
 Learning about such themes and being encour-
aged to think more deeply about them is why reading
LBC has been an extremely enriching experience to
me. Anyone who wants to know more about the way
bilingualism affects cognition, Singlespeakers and
Doublespeakers alike, is strongly advised to get hold
of a copy of LBC and get immersed in it. Mentally
enriching Widethink will be the reward.
Annette M. B. de Groot
University of Amsterdam
Department of Psychology
Weesperplein 4
1018 XA Amsterdam
The Netherlands
E-mail: a.m.b.degroot@uva.nl
References
de Groot, A. M. B. (2011). Language and cognition in bilin-
guals and multilinguals: An introduction. New York, NY:
Psychology Press.
Evans, N., & Levinson, S. C. (2009). The myth of language
universals: Language diversity and its importance for
cognitive science. Behavioral and Brain Sciences, 32,
429–492.
384 •  AMERICAN JOURNAL OF PSYCHOLOGY, FALL 2013
Jarvis, S. (Ed.). (2011). Crosslinguistic influence in bilinguals’
concepts and conceptualizations. Special issue of Bilin-
gualism: Language and Cognition, 14.
Levelt, W. J. M. (1989). Speaking: From intention to articula-
tion. Cambridge, MA: MIT Press.
Orwell, G. (1949). 1984. London, England: Secker & War-
burg.
Pavlenko, A. (2005). Bilingualism and thought. In J. F. Kroll
& A. M. B. de Groot (Eds.), Handbook of bilingualism:
Psycholinguistic approaches (pp. 433–453). New York,
NY: Oxford University Press.
Pavlenko, A. (Ed.). (2011). Thinking and speaking in two lan-
guages. Bristol, England: Multilingual Matters.
Slobin, D. I. (1987). Thinking for speaking. Proceedings of
the Thirteenth Annual Meeting of the Berkeley Linguistics
Society, 13, 435–445.
Slobin, D. I. (2003). Language and thought online: Cognitive
consequences of linguistic relativity. In D. Gentner & S.
Goldin-Meadow (Eds.), Language in mind: Advances in
the study of language and thought (pp. 157–192). Cam-
bridge, MA: MIT Press.
Talmy, L. (2000). Toward a cognitive semantics. Cambridge,
MA: MIT Press.
Vance, J. (1974). The languages of Pao. London, England:
Mayflower. (Original work published 1958)