See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/323720408

Translingual Practice as Spatial Repertoires: Expanding the Paradigm beyond

Structuralist Orientations

Article  in  Applied Linguistics · February 2018

DOI: 10.1093/applin/amx041

CITATIONS READS

285 2,192

1 author:

Suresh Canagarajah
Pennsylvania State University
175 PUBLICATIONS   12,675 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

https://psu-us.academia.edu/SureshCanagarajah View project

Linguistic Shame and shaming in Language Education View project

All content following this page was uploaded by Suresh Canagarajah on 04 August 2018.

The user has requested enhancement of the downloaded file.

 Applied Linguistics 2018: 39/1: 31–54 ß Oxford University Press 2017
doi:10.1093/applin/amx041 Advance Access published on 13 November 2017

Translingual Practice as Spatial

Repertoires: Expanding the Paradigm
beyond Structuralist Orientations

SURESH CANAGARAJAH
Department of Applied Linguistics, 303 Sparks Building, Penn State University, PA
16803, USA
E-mail: asc16@psu.edu

The expanding orientations to translingualism are motivated by a gradual shift

from the structuralist paradigm that has been treated as foundational in modern
linguistics. Structuralism encouraged scholars to consider language, like other
social constructs, as organized as a self-defining and closed structure, set apart
from spatiotemporal ‘context’ (which included diverse considerations such as his-
tory, geography, politics, and society). Translingualism calls for a shift from these
structuralist assumptions to consider more mobile, expansive, situated, and hol-
istic practices. In this article, I articulate how a poststructuralist paradigm might
help us theorize and practice translingualism according to a spatial orientation
that embeds communication in space and time, considering all resources as work-
ing together as an assemblage in shaping meaning. I illustrate from my ongoing
research with international STEM scholars in a Midwestern American university
to theorize how translingualism will redefine the role of constructs such as lan-
guage, non-verbal artifacts, and context in communicative proficiency.

Translingual practice—in its many guises as translanguaging, plurilingualism,

or metrolingualism—has contributed significantly to our understanding of
verbal resources in communicative practice (see Canagarajah 2013 for an ex-
tended discussion of these labels). Challenging traditional understandings of
language relationships in multilingualism, which postulates languages main-
taining their separate structures and identities even in contact, translingualism
looks at verbal resources as interacting synergistically to generate new gram-
mars and meanings, beyond their separate structures. According to this defin-
ition, the prefix ‘trans’ indexes a way of looking at communicative practices as
transcending autonomous languages. In recent years, scholars have expanded
the ramifications of this definition. An emergent second definition focuses on
the need to transcend verbal resources and consider how other semiotic re-
sources and modalities also participate in communication. Block (2014: 54), for
example, has critiqued what he calls the ‘lingual bias’ that predisposes linguists
to treat only words as primary in communication or deserving analytical focus.
Yet another approach focuses on transcending the text/context distinction and
analyzing how diverse semiotic features previously relegated to spatiotemporal

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 32 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

context actively participate in communication (Blommaert 2013; Pennycook

and Otsuji 2015). A fourth meaning that is becoming significant is how semi-
otic resources transform social structures (Li and Zhu 2013). In this case, ‘trans’
indexes ‘transformation’ and challenges understandings of language as regu-
lated or determined by existing contexts of power relations.
One way to understand these expanding orientations is that they are moti-
vated by a gradual shift from the structuralist paradigm that has been treated as
foundational in modern linguistics (see special topic issue of this journal on
poststructuralism, vol.33/5, 2012). Though structuralism is an expansive
movement, I refer here to the way it has been taken up in linguistics, following
the seminal work of de Saussure (see Harland 2010). Structuralism motivated
linguists to consider language as organized as a self-defining and closed struc-
ture. From this perspective, other modalities of communication were separated
from language, maintaining their own structures. Furthermore, linguistic
structure was set apart from spatiotemporal ‘context’ (which included diverse
considerations such as history, geography, politics, culture, and society). As
Hymes (1971) has observed, Chomsky took structuralism further in a cognitive
and individualized direction. The language structure was provided a mental
locus, treating the grammar as internalized, and providing a representational
system of meaning-making for the speakers. Though such approaches define
language as value-free and abstract, certain ideologies subtly enter through the
unproblematized ‘context’. In dominant approaches, context was treated as a
container of language, framed as domains such as speech community or
nation-state. These constructs territorialized and essentialized language, pro-
viding ownership to certain groups of speakers and/or their lands.
The meanings of ‘trans’ that I have reviewed above call for a shift from the
above assumptions to consider more mobile, expansive, situated, and holistic
practices. However, the connection between structuralism and translingualism
needs to be explored further to theorize the analytical benefits of the new
paradigm. This examination would help us identify new possibilities inherent
in translingualism. I articulate below how moving beyond structuralism might
help us theorize and practice translingualism differently. Though such an
orientation is implicit in earlier theorizations and analyses of translingualism,
it has not been sufficiently taken up for critical examination.

THE CASE FOR SPACE

Challenging the structuralist paradigm, scholars are becoming more sensitive
to space as a more expansive framework for explaining communicative and
social life. Developing from the findings in theoretical physics (Barad 2007;
Coole and Frost 2010) on the agentive and vitalist potential of physical nature,
the spatial orientation is gaining thoughtful uptake in other disciplines as well.
Scholars in applied linguistics, such as those in posthumanism (Pennycook
2016), mobility studies (Blommaert 2010), linguistic landscapes (Shohamy
and Gorter 2008), and literacy (Kell 2010), have been influenced by a spatial

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 33

orientation. Spatiality is everything that a structuralist orientation has tried to

avoid, as theorization of scholars in human geography (see Soja 2011; Massey
2005) points out. Situating communicative interactions in space and time ac-
commodates diversity and unpredictability. Conceiving of language and other
human activities as abstract and autonomous structures, however, tends to
favor homogeneity, normativity, and control. Structures are abstracted from
the messiness of material life and social practice. In making structures funda-
mental and generative, structuralism imposes order and control over material
life. When structures are interpreted as located in the mind (as Chomsky did),
they also feed into the Cartesian bias of mind over matter. Treating spatiality as
significant means understanding every practice as situated, holistic, net-
worked, mediated, and ecological, thus integrated with diverse conditions,
resources, and participants. Spatiality does not mean that we abandon all con-
siderations of order, pattern, or norms, but reformulate them beyond abstract,
homogeneous, and closed structures.
Since space was understood traditionally as dead matter to be shaped by human
cognition and language, we have to redefine space according to emerging under-
standings. Though structuralist linguistics did acknowledge space, materiality, and
environment, it treated them as passive, inert, static, and pliant. A spatial orien-
tation treats them as active, generative, and agentive. Space is emerging as a
holistic construct that includes geography, history, and society. We can summar-
ize the extensive theoretical discourses for our purposes as follows (see Thrift
2007; Barad 2007; Massey 2005; Soja 2011; Coole and Frost 2010):
 Space is vitalist, not dead. In this sense, it is self-generating and self-regulating,
with things shaping each other and other beings, including humans.
 While spatiality shapes social life, people shape the material environment.
 If place is space ascribed with social meaning and shaping, as in bounded
constructs, such as nations, communities, and cities, we must also hold
places in dynamic tension with space as an expansive material construct,
providing possibilities for reconstruction.
 Space includes time, with both acting together, as in Bakhtin’s (1986)
notion of chronotopes, which encourages us to consider them as interact-
ing, layered, and dynamic.
 As human exceptionalism is questioned, spatiality adopts a ‘flat ontology’
(Marston et al 2005) that appreciates the ecological interconnection of all
things and beings. It does not predefine any particular category as more
dominant and significant before situated analysis. Spatiality helps consider
how multiple resources mediate and co-construct activities as an ‘assem-
blage’ (Deleuze and Guattari 1987; Latour 2005).
 As the power of mind over matter/nature is questioned, cognition is
understood as distributed across bodies, objects, and social networks, call-
ing for distributed practice in thinking and communicating (Thrift 2007;
Wilson and Golonka 2013).
Such an orientation to communication as an assemblage has been anticipated
in applied linguistics. Early scholars in the field have questioned the relegation

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 34 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

of space and time into insignificant ‘context’ and encouraged a more expansive
orientation to semiotic resources, though their models were somewhat differ-
ent from that developed in this article. Highlighting what he calls the ‘the
neglected situation’, Goffman (1964: 134) has urged linguists to problematize
context and unveil the features that mediate communication. Context gets
treated in ‘the most happy-go-lucky way’, in an ‘opportunistic’ fashion; ‘an
implication is that social situations do not have properties and a structure of
their own but merely mark, as it were, the geometric intersection of actors
making talk and actors bearing particular social attributes’. Silverstein (1985:
220) has similarly encouraged analysts to attend to ‘the total linguistic fact, the
datum for a science of language is irreducibly dialectic in nature. It is an un-
stable mutual interaction of meaningful sign forms, contextualised to situ-
ations of interested human use and mediated by the fact of cultural
ideology’. For the spatial orientation, the ‘total linguistic fact’ involves sign
forms beyond verbal resources, as articulated in the expansive definitions of
translanguaging. Ideally, such an approach would mean that we reexamine
the text/context distinction and consider how features we may have treated as
part of context may constitute an assemblage that is integral to meanings and
communication.
In this article, I discuss how translingual practice might be understood ac-
cording to a spatial orientation. Though this is primarily a conceptual article
that outlines the changes in our understanding of language, non-verbal semi-
otic resources, context, and competence, I illustrate from my ongoing research
with international Science, Technology, Engineering, and Mathematics
(STEM) scholars in a Midwestern American university. The data consist of
the following: interviews on the language attitudes and communicative prac-
tices of 24 Chinese scholars; research practices and communication of a South
Korean postdoctoral research fellow in Molecular Biology (whom I will call
Jihun1); literacy practices of a Turkish doctoral student in Entomology
(Gunter); and video recordings of four 1-hour episodes of classroom instruc-
tion of two Chinese Math teaching assistants.2 The communicative practices of
the Korean and Turkish scholars have been studied through video observation,
discourse-based interviews, and photographic documentation, accompanied
by a textual analysis of multiple drafts of published articles and other written
genres. These studies aimed at understanding the participants’ multilingual
and polysemiotic practices, and ongoing learning/socialization into profes-
sional communication. An attempt was made to understand their communi-
cation and proficiency in their situated—that is social and spatial—contexts
without unduly imposing the disciplinary frameworks of applied linguists.

BEYOND LANGUAGE TO SPATIAL REPERTOIRES

As we shift to a spatial orientation, we have to abandon the traditional notion
of separately structured languages. Words are mobile signifiers located in space
and time. How they gain meaning and grammatical status is explained by the

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 35

processes of indexicality (Agha 2003). This depends on how people put words
to use in situated activity in specific locations. Indexicality is a spatiotemporal
process, as meanings sediment over time to develop grammatical status and
norms. However, these norms have to be kept open to change as words par-
ticipate with other semiotic assemblages to construct meaning. This orientation
has many implications for language proficiency.
To begin with, one might not have proficiency in the whole of a language
(including what we might consider an advanced or even a basic proficiency in
the ‘underlying’ grammatical structure) and still be able to perform meaningful
activities using that language. Though my interviewees acknowledged that
they would not claim grammatical proficiency in English, they were confident
in communicating in professional contexts. Many of them mentioned that
while they were fluent in their academic and professional communication, it
was in nonacademic genres and contexts, such as casual conversations in bars
or campus corridors, that they found the most difficulty. Consider the claim of
a scholar in astronomy:
 CM: The good thing is that- it’s- astronomy is a very narrow field, only-
not so many words, totally that thousand of words is enough. Like black
holes, everybody know that. @@, the mass, the units. So when you talk
about something, even if it’s not clear enough, we know each other. And
also, because you are writing papers, people already know your ideas, or
your point, so even before you talking with each other, I can guess what
the people- what the other people are going to say. @@. So when we
discuss, we always have to say ‘well, what’s your current/prediction for
this object? What’s your current theory? What’s any—is there any pro-
gress in your model?’ So it’s—the thing is the conference in English is
always good enough to do something. Ehm, yeah. But sometimes the
difficulty maybe come from the daily life, for example, when I have to
find a lawyer or find something, a lot of words I don’t know, I don’t
understand, even never heard about that. So that’s simply the most dif-
ficult thing. [Chunguang Mao; Male; Postdoc in Astronomy and
Astrophysics]3
Note that Chunguang attributes his confidence to the mastery of his disciplin-
ary register, despite his difficulty in having a conversation with a lawyer on an
everyday need. This claim was confirmed in my observation and recording of
the Korean research associate. Jihun demonstrated idiomatic problems in
casual conversation, such as when he said on multiple occasions that he was
‘caught by a cold’ when he could not show up for an interview. However, as I
will demonstrate below, his writing for disciplinary purposes was very
advanced. He was the lead author of a publication whose early drafts were
grammatically well formed.
This inconsistency can be explained in many ways. One possibility is that
these STEM scholars have participated in shaping and sharing the indexicality
of words relevant for their field, though they have not engaged extensively in
casual interactions to develop the repertoires of everyday genres. I found that

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 36 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

Jihun had mastered the English for scientific purposes already in Korea, before
he migrated to the USA. Because that language is shared by scholars from
diverse nationalities for disciplinary communication, Jihun had engaged in
developing the indexicals required for his work. He had developed this profi-
ciency through professional activity in Korea and the USA, not formal learn-
ing. STEM scholars are able to use these indexicals decoupled from the rest of
the language structure because they find their coherence in spatial locations in
particular activities. Blommaert (2010: 23) labels this ability to use selective
words and grammars from languages for specific activities as ‘truncated multi-
lingualism’. (Note that this term is biased towards considering whole language
as the norm. The construct I introduce below, ‘spatial repertoires’, will accom-
modate the possibility of selective verbal resources being sufficient for com-
municative purposes in situated interactions, in combination with other
semiotic resources.)
Similarly, participants were able to use words from diverse languages for ac-
complishing their communicative purposes. Jihun mentioned that his profes-
sional activity involved shuttling between Korean and English constantly. He
mentally planned his research articles in Korean and wrote them in English. He
discussed some English academic publications and experimental findings with a
Korean colleague in Korea, and wrote about them in English. Even if the final
draft of Jihun’s published article is in ‘standard written English’, he has shuttled
between Korean and English in various interactions and stages of the writing
process in shaping this final product. What the spatial orientation means is that
the labeled language structures do not necessarily constrain one’s situated pro-
fessional activity. They combine in making activities possible. This was con-
firmed by my interviewees. Li Sun [Visiting scholar in Industrial Science]
mentioned that she wrote certain drafts first in Chinese before translating
them into English closer to submission. Chao Li [Postdoc in Biochemistry and
Molecular Biology] mentioned that he took notes on readings and discussions in
Chinese or a mix of languages though the source was in English.
Bits and pieces of words and grammatical structures from diverse languages
work together for these participants because these communicative resources
find coherence in terms of the spatial ecology, not necessarily in terms of the
grammatical structure. Since activity is the frame of reference, not the
bounded grammatical structure, it is possible to understand how diverse lan-
guages in the spatial ecology facilitate communication. As we will see in the
next section, multimodal resources also combined with verbal resources to
facilitate situated communication. From this perspective, language works
with an assemblage of semiotic resources, artifacts, and environmental affor-
dances in specific settings to facilitate communicative success. To accommodate
these configurations of communicative resources that go together in particular
activities, some scholars have adopted the term ‘spatial repertoires’. Pennycook
and Otsuji (2015: 83) define spatial repertoires as: ‘link[ing] the repertoires
formed through individual life trajectories to the particular places in which
these linguistic resources are deployed’ (emphasis added). I would modify this

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 37

slightly to move spatial repertoires beyond the methodological individualism,

human agency, and verbal resources the definition favors. Spatial repertoires
may not be brought already to the activity by the individual but assembled in
situ, and in collaboration with others, in the manner of distributed practice.
These repertoires may not be part of one’s existing proficiency. I would expand
the repertoires beyond the linguistic to include all possible semioticized re-
sources. I would also spatialize these repertoires more completely by treating
them as embedded in the material ecology and facilitated by social networks.
Spatial repertoires are an alternative to grammar in analyzing meaning making
and communicative success. I will demonstrate below the diverse resources
that serve my participants in different communicative activities.
All this is not to deny that there are languages called Korean or English. This
possibility is also explained by indexicality. At a limited scale of consideration,
certain words index certain places and communities, and develop identities as
distinctly labeled or territorialized languages. Indexicals sediment over time to
gain an identity as belonging to one language or the other, with a specific
grammatical status in that language. Language ideologies further give identity
to a collection of words as belonging to Korean or English. However, such
structures or labels do not constrain people from drawing from all of them
to accomplish their activities in practice, giving new meanings and identities to
these words, as translingual scholars theorize.

FROM MULTIMODALITY TO ASSEMBLAGE

One of the reasons my research participants were confident about professional
communication, despite their limited proficiency in English grammar is that
their communicative activity involved diverse other semiotic resources beyond
words in their communication. They cited using gestures, visuals, ‘body lan-
guage’, and modalities such as PowerPoint and blackboard that mediated their
words. The significant role of visual and bodily resources was confirmed in a
video recording of Chinese Math TA Tan’s classroom instruction. As we applied
linguists viewed the teaching, we found his speech difficult to understand. The
following is an example of his speech:
[00: 14: 36.11] TAN: i think that fluids might be a good example. if
we consider a pipe and we have some- some fluids inside. so we have
a (velocity) field. so if we call this (velocity) field as <PAUSING TO
WRITE> conservative vector field or irrotational vector field, that means,
if we consider- (a regular) section, then, there would be no (.) rotation
of this field, so this- th- this th- of the- of this fluid. this fluid will go
along the tide and there is no- no turn shows this fluid will go rotation
inside. ok, (that’s) just a glance of this irrotational vector field. we will
talk about that in the section about this Stoke’s theorem (later).
The transcripts confirmed that Tan did not have an extensive English vocabu-
lary or complex syntactic structures as part of his personal repertoire.

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 38 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

There were also many pauses and hesitations in his speech. What he used
mostly were deictics such as here/there and this/that (as underlined above).
A count of all the transcripts from his recordings also proved the preponder-
ance of these deictics. In addition to the deictics, he used technical words such
as irrotational, vector, and velocity, as italicized above. However, the students in
the class did not seem to have any problems following the lecture, as our
second video camera for student uptake proved. Their periodic interjections
to clarify matters, note-taking, and rapport suggested that they were following
the instruction very well.
A discourse-based interview with a more experienced native-speaker TA,
Adam, suggested the reason for Tan’s communicative success, despite what
we perceived as his grammatical limitations. Adam’s immediate response as
the video clip started playing was: ‘This guy’s boardwork is excellent. Some
instructors might talk more clearly and fluently. But if their boardwork is poor,
the talk doesn’t help at all. One of our Chinese professors doesn’t speak much,
but because her boardwork is good, we like her teaching a lot’ (12 January
2016). What the notion of boardwork implied was that Math instructors and
students were not focusing on speech in isolation from the other spatial rep-
ertoires, as we applied linguists were doing in our analysis. What mattered to
them was the embodied activity of boardwork. In fact, language can get in the
way of teaching.
A consideration of the whole embodied activity suggested why Tan’s teach-
ing was successful and effective. In teaching Stoke’s Theorem, and attempting
to illustrate the curl, Tan first draws a pipe, with arrows for liquid flowing
through. Then he draws a cross-section to examine if the water is flowing
straight or in curls. He also waves his hands and fingers to demonstrate
water flowing without and with curls to clarify the point about conservative
vectors. As he makes the point, he also reaches further left on the board to
underline the words curl, irrotational, and conservative that he had introduced
earlier. Thus he draws the attention of students and makes connections to
points he had made previously. Since the visuals and gestures are communi-
catively very functional, Tan did not have to verbally describe many of these
concepts and processes. The preponderant deictics played a crucial role in
pointing to relevant information.
There were other semiotic resources that facilitated this effective teaching. In
contrastively facing the board or facing the students as he taught, Tan indexed
teaching math and doing math, respectively. In the former, he was focused on
demonstrating how he thought through problems and developed solutions as a
math professional. When he faced the students, he made himself available for
questions and explicated concepts. His posture suggested shifting between
identities and activities. Analyzing videos of the same TA from a CA perspec-
tive, Looney et al (2017) demonstrate that words such as ‘okay’ and ‘so’ are
self-talk that indexed the TAs’ activity of doing math by thinking to themselves
and organizing their activity. In fact, often writing on the board preceded
talking about it. For example, Tan wrote a matrix first, ruminated on it silently

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 39

(and perhaps compelled the students to think along with him), before he ex-
plained or commented on them. It is possible that writing and visuals help in
thinking, as a form of ‘thinking with your hands’ (Van Compernolle and
Williams 2011), where gestures facilitate thinking. Furthermore, Tan shuttled
between the pronouns ‘we’ and ‘you’. ‘We’ indexed dong math together, and
‘you’ indexed making space for student questions or feedback.
What this example illustrates is that the non-verbal resources are not a
supplement to talk or thinking. They mediate and shape language use.
Applied linguists who address these resources as ‘multimodality’ have trad-
itionally treated gestures, visuals, and other paralinguistic resources as com-
pensatory—that is resources that help when language is not adequate for the
purpose. Consider the orientation of the Douglas Fir Group (2016: 29):
‘Nonlinguistic, multimodal semiotic resources are used to make the coupling
of a form and a meaning socially available during unfolding interactions.
They are not peripheral or complementary to language learning. Instead,
they provide crucial social cues to grammar’. Though the authors give con-
siderable importance to non-verbal resources, it is evident that they are trea-
ted as providing ‘social cues’ to grammar, which can presumably work on its
own. What the practice of the math TAs suggests is that the diverse semiotic
resources work together as an assemblage, without the possibility of separat-
ing them. Language is considered inefficient and insufficient by itself for the
successful outcome of the activity. We might even say that the preponder-
ance of deixis suggests that in certain activities the TAs treat language as
indexing the meanings of other non-verbal resources to facilitate
understanding.
The notion of assemblage helps to consider how diverse semiotic resources
play a collaborative role as a spatial repertoire in accounting for the success of
this activity, when language is not predefined as the sole, superior, or separate
medium of consideration. Assemblage corrects the orientation to non-verbal
resources in scholars addressing ‘multimodality’. From the perspective of as-
semblage, semiotic resources are not organized into separate modes. To think
so is to fall into structuralist thinking. According to assemblage, all modalities,
including language, work together and shape each other in communication.
Some sociolinguists and conversation analysts have made remarkable strides in
adopting such a perspective to interactional analysis—see Mondada (2014) and
Streek (2013), for example. However, in focusing on the resources evident in
the immediate setting, they differ from the spatial orientation which accom-
modates meanings and influences from expansive spatiotemporal scales, as I
will demonstrate below.
There are other differences from studies in multimodal analysis. Semiotic
resources can be agentive, shaping human cognition and communication.
Traditional multimodal approaches have treated objects and the human
body as orchestrated by the human mind. Spatial approaches encourage a
rethinking of human agency. Consider how an artifact, a visual model of

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 40 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

over wintering of bees in an article by Gunter (a doctoral candidate in

Entomology, from Turkey), shaped his thinking and communication in dra-
matic ways. Gunter started his manuscript as a review article to examine how
existing studies explained why bee stocks depleted in successive winters,
which affected brood numbers for pollination. Though he ends up reading
around 100 articles, prodded on by his advisor who is also a co-author, track-
ing new publications and questions as he writes, his conclusion is somewhat
broad. A very advanced draft states:
Overwintering in honey bees is a complex process, which integrates
multiple environmental cues, social cues and interactions within
the colony, and physiological and molecular changes in individual
bees. Using the available information, we have developed a model
which explains how the entry, maintenance and exit from over-
wintering may be regulated by these factors, but further studies are
necessary to comprehensively test this model by uncoupling and
individually testing these factors, many of which are closely corre-
lated (2 June 2015).
At this point, what he terms a ‘model’ is only verbally articulated in the
preceding pages. However, his adviser suggests that he develop a visual rep-
resentation of his findings. Gunter’s attempts illustrate the limits of human
agency and mental representation. I have examined seven renditions of his
visualization. Though he has formed a mental picture from his reading, rep-
resenting it visually becomes a challenge. Initially, his visuals do not seem to
index his mental representation well. Gunter realizes that this is not a cog-
nitive problem but a prosthetic one. That is, he lacks the physical abilities and
material resources to render his knowledge in a satisfactory model. He sees
the need to complement his mind and body with instruments that might
become an extension of him to represent his knowledge. Consider an early
representation in Figure 1 below:

Figure 1: Overwintering bees early visual

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 41

He mentions his dissatisfaction with his attempt in an interview: ‘I couldn’t

figure out how to make curved arrows with normal lines. Thus, used these
terrible things instead’ (INT, 16 April 2016).
He eventually gets help from a technician in his laboratory who has better
skills and computer resources for modeling. What emerges is not just an ele-
gant model, but one that projects a different cause/effect relationship than
Gunter had originally envisioned. Consider the final representation included
in the publication in Figure 2 below:

Figure 2: Overwintering bees final visual

Note that this model distinguishes developments inside and outside the bee
nest. It projects not just a temporal sequence of processes but a cause/effect
relationship. The visual projects two pheromones—that is reduced brood
pheromone and increased forager pheromone—as leading to the slowed mat-
uration of winter bees. The empirical studies he reviewed do not make this
causal connection or provide an integration of the various factors and stages in
the overwintering process of bees. Also, the visual projects its own represen-
tation that differs from Gunter’s earlier mental picture. It forces Gunter to
revise his understanding of the overwintering process. Gunter says in a dis-
course-based interview:
Drawing this thing made me rethink my explanation. The problem
was I was trying to put it in a mechanical explanation while draw-
ing it. Those arrows are not just arrows, they are directional. They
are talking about inductions and inhibitions of things. You can see

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 42 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

reduced brood pheromones is causing reduced foraging. These are

all causal relationships and not just temporal relationships. So once
you start thinking about those [. . .] I was like, well does it, and how
does it? . . . There were a couple of times we took things out and
changed them a little. Because it stopped making sense once you
put it on the figure. You see that it doesn’t make sense. It made
sense when I said it, but once you see it all together it doesn’t, so
you change things. (INT, 16 April 2016)
In addition to the agency of the visual object, the process of drawing it and
seeing it on the page changed Gunter’s thinking. In this sense, this communi-
cative activity is performative, not representational. As he proceeds with his
research, this artifact begins to play a central role in his scholarship. Future
writings, research, and presentations are shaped by this visual, as it provides
him a hypothesis that he chooses to explore for his doctorate. In one of his
poster presentations in a conference, Gunter circles the two pheromones in red
to show their centrality. Based on this hypothesis, he also subsequently writes
a successful grant application to test the role of these two pheromones. What
we find is that the visual model does not simply convey preconstructed ideas or
supplement words but is itself agentive in shaping thinking and communica-
tion. Other applied linguists studying scientific interactions have also pointed
to the significance of material and social networks that mediate language and
thinking (see Ochs et al 1996; McNamara 1997).
To summarize the implications of the spatial orientation, consider the fol-
lowing differences from dominant approaches to multimodality:
 Multimodality studies treat human agents as having the power to deploy
semiotic resources as they will for their purposes; spatial orientation the-
orizes that objects in the environment also shape human actors.
 In multimodality studies, there is a tendency to invoke and interpret the
predefined values, meanings, and structures of each modality separately;
spatial orientation treats meaning as emergent in relation to the diverse
assemblages as an ‘ensemble’ (Kress 2009) that shapes each other.
 Multimodal studies explain semiotic resources as indexing and motivated
by cognitive representations; spatial orientation considers resources as
performative, generating meanings in activity.
From this perspective, though there are significant developments in applied
linguistics to accommodate more expansive semiotic resources (as cited ear-
lier), we have to be alert to the differences between the spatial orientation and
alternate approaches.

FROM CONTEXT TO SCALES

My research with STEM scholars also problematizes the traditional orientation
to ‘context’ in structuralist linguistics. As other scholars have pointed out (see
Blommaert et al 2017), context is often treated as distinct from grammar/text;
monolithic (i.e. not differentiating diverse spatial, temporal, and social scales);

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 43

and of secondary relevance to linguistic analysis. Some applied linguists have

articulated a more dynamic orientation to context, demonstrating how it is
emergent and co-constructed by language (Duranti and Goodwin 1992;
Goodwin 2000; Hanks 2006). A spatial orientation develops an even more
complex understanding of how semiotic resources work in relation to spatio-
temporal conditions.
To begin with, my findings suggest that STEM scholars treat the boundary
between text and context as permeable. As the example of Jihun’s writing
practice below will show, the monolithic term context is inadequate to accom-
modate the diverse factors that mediate and shape language. Diverse features
traditionally relegated to context are part of text or talk. They constitute spatial
repertoires that people use to accomplish their communicative activities.
Consider the composition of a research article Jihun had published as the
first author. The 30 different drafts I analyzed at different stages of develop-
ment suggested the way in which the published article of Jihun was a complex
assemblage of spatial repertoires. To begin with, a range of participants, multi-
modal resources, and artifacts from different networks and spatial ecologies
went into the construction of the text. At a local spatial scale, his collaborator
was the senior professor of his laboratory and Principal Investigator in the
grant he was working on, Nick. Both came from the Department of
Biochemistry. A second junior faculty member from the Department of
Chemical Engineering, Mohan, was also an active member of this research
group and played an important role in shaping the text. These three scholars
worked closely on the manuscript, with many rounds of readings, comment-
ing, and revisions. Though Jihun took the lead in composing the drafts, the
other two shaped the ideas and words in deep and pervasive ways. The in-text
comments show contributions such as the following: suggesting more relevant
references; interpretation of experiment; designing additional experiments;
providing better wording or phrasing; strategizing style, wording, or presenta-
tion of data for actual and potential reviewers; responding to referees; and
clarifying ideas through conversations in research group meetings (RGMs).
In many cases, the other two collaborators not only posed questions and sug-
gestions but wrote chunks of text that Jihun directly embedded in the draft.
Beyond these three, there were seven others who made additional contri-
butions, though they were not involved in face-to-face interactions during the
writing or in conversations in the RGMs. Beyond these 10 named as authors,
there were 9 others who are acknowledged in the end of the article. Some of
these scholars also provided texts and images that are embedded in the article.
At a more distant spatial and temporal scale of contribution, they conducted
different sections of the experiment, provided raw materials, chemicals and
instruments for the study, analyzed the data, or provided technical assistance.
Many of them are from departments and laboratories that are different from
Jihun’s. Three are from other universities. How their institutional and depart-
mental affiliations relate in this collaboration finds clearer representation in

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 44 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

Figure 3: Jihun’s social networks

Figure 3. Those whose names are bolded and underlined are the named
authors.
As we can see, there are different social networks Jihun is tapping into to put
together this text. Rather than being mere ‘context’, these networks account
for the assemblage that is the text. Such social networks and resources go
beyond simply mediating the text Jihun is constructing. The words, figures,
ideas, research findings, and suggestions of those in the networks are literally
embedded in the evolving text, almost forming a tapestry of many voices. The
term entextualization indexes how social, spatial, and material resources come
together in the assemblage of texts (Kell 2010). Though some of the texts,
images, and figures belong to other research projects, isolated from Jihun’s
context, they get entextualized into a new and coherent whole in Jihun’s
published article.
As we open up context to analyze these social networks and spatial reper-
toires that constitute the text, we have to reconsider authorship. Though Jihun
is the first author, he mentioned that this naming convention does not reflect
authorship as much as signals who took the lead in designing the experiment
and putting together the drafts. That Nick’s name is last indexes his authority
as the Principal Investigator (with Mohan’s name coming in front of him to
signal his critical contribution). That at least six of those named as authors
played no direct role in the writing or revising of the draft also compels us to

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 45

understand authorship differently. The collaboration and critical contribution

by others suggest that this article is an emergence of distributed practice.
Anyone going through the serial drafts will see that it is not Jihun’s thinking,
words, or communication that account for this text. As they pass around the
drafts, the participants are generating new ideas, interpreting findings, con-
ducting new research, and developing the product collaboratively. The ideas
and words emerge between and across the social networks. Jihun is merely
keeping record of everything that is emerging in his drafting process. This
finding undercuts the methodological individualism that informs much linguistic
research, which treats the individual as the locus of proficiency and speech.
There are implications for the cognition and mental representation that in-
forms this textual product. We can postulate that the text is an emergence of
distributed practice that involves the thinking and language of multiple parties,
not to mention the artifacts, texts, and resources that also shape their thinking
and communication.
Though I set up a video camera close to Jihun’s laboratory computer to
record his writing practices, it quickly became apparent that his writing
went beyond this immediate setting. Or perhaps we can say that the text
and its features indexed other social, temporal, and geographical contexts
beyond what was captured by the video camera. For example, an important
site of negotiations of this article was a few rooms away from the laboratory
where Jihun, Mohan, Nick, and a couple of rotating graduate students (joining
for apprenticeship purposes) held their RGMs. Swales (2004) studies RGM as
an important genre of academic communication in its own right. In the
embedded comments in the drafts, Nick or Mohan sometimes say that they
should discuss a point more elaborately before revising the matter. I then
videotaped the RGM’s to see how participants reviewed figures and images
from the experiment on a monitor placed centrally in the room, to interpret
them closely and formulate their arguments. Notes from these conversations
are recorded by Jihun in his notebook and help him revise the draft subse-
quently. These oral conversations are also entextualized in the published art-
icle, though there might not be actual linguistic evidence in the finished
product.
Some texts, images, and artifacts come from miles away. While Jihun was
composing, he takes a phone call to a colleague in another laboratory to clarify
some issues about some results. In another situation, he consults a Korean
colleague in Korea about a reference he needs through Iphone text messages.
This is done through texting him in Korean. The colleague provides informa-
tion on a published English article and discusses relevant findings in Korean. In
such cases, Korean sources get entextualized into an article published in
English. The publication is also shaped by raw materials, chemicals, DNA, in-
struments, and results of related experiments run by others beyond his own
laboratory and outside the experiment and writing managed by Jihun. Some
were prepared or conducted long before Jihun’s own study or writing. What
such expansive and layered spatiotemporal contexts suggest is the limitations

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 46 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

of treating face-to-face interactions and empirically evident contextual features

as the unit of analysis in applied linguistics. This is a lingering influence of a
metaphysics of presence (Derrida 1973) which has been discussed much in phil-
osophy. As we open up the monolithic and passive notion of ‘context’, we see
that it contains not only expansive social, temporal, and spatial networks but
that features from these domains are agentive. They shape the text in signifi-
cant ways, consistent with the notion of an assemblage. As the monolithic and
static notion of context is inadequate to address such dynamic relations between
texts and contexts, some scholars are adopting a more layered, relational, and
multifaceted metaphor of scales for language analysis (see Canagarajah and De
Costa 2016). We also find that these diverse scales of influence are embedded
in the text as a tapestry of layered simultaneity (Blommaert 2005: 237). This
orientation to transcending the text/context binary by accommodating re-
sources across space and time as shaping meaning is the third definition of
translingualism mentioned above.

FROM REPRESENTATIONALISM TO PERFORMATIVITY

How do we characterize the process undertaken by Jihun in entextualizing
these expansive spatial repertoires in his article? In interviews, Jihun men-
tioned that he considered himself an ‘organizer’ rather than a ‘communicator’
(INT. 11 June 2016). That is, he considered his writing process as a pragmatic
and physical activity of assembling relevant resources for the article. In this
sense, the writing resembled bricolage. Consider that his first draft simply
constituted the images and figures relating to the experiment. He said that
his typical writing involved shaping the article around the main results of
his study, as captured through photographs and electromagnetic images.
These visuals were mostly cut and pasted from files saved in his computer
earlier or downloaded from those sent by his colleagues. In a discourse-
based interview, Jihun also showed me an article that he said was ‘fundamen-
tal’ to his own article (INT. 11 June 2016), as it reported on extracting cellulose
from a different raw material. As the experiment involved similar procedures,
he said that he used that article as the template and even borrowed phrases
and terms for his own article. Some phrases, section headings, title, and se-
quence closely resemble that article.
Such bricolage in the writing process was also facilitated by a strategic con-
figuring of the space and objects in Jihun’s workspace. The video recording
shows that, as Jihun composes, he switches between screens to borrow texts
from other published articles and research resources. Since research proced-
ures were fairly standardized, in many cases Jihun cut and pasted texts from
other publications that he pulled up in other windows. He mentioned that this
textual borrowing explained why his early drafts had different fonts (presum-
ably influenced by fonts from other texts. This practice of textual borrowing is
now widely documented in STEM literacy research (see, for example,
Flowerdew 2007)). This is not necessarily plagiarism, as the texts are

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 47

entextualized into a new whole and recontextualized for the purpose.

Sometimes Jihun borrowed from print articles. His writing space was set up
strategically with a stack of printed publications next to his computer. He
shuttled back and forth between the computer and the printed articles to
borrow phrases and words. Also, in his windows, he had a bilingual Korean/
English dictionary always available. He toggled between these screens to trans-
late words and phrases as he wrote. Such bricolage calls for a reconsideration
of representationalism in structuralist notions of competence. That is, one does
not start with a picture in the mind or the required words for accurately rep-
resenting those ideas and images on the page. The resources that are assembled
generate the ideas and words for the publication in situated interactions. Such
bricolage is an example of ‘nonrepresentational thinking’ and a performativity
that orientate to doing as generative of thinking and communicating, or processes
rather than ideas as generative of meanings (see further Thrift 2007).
As we can see above, efficient bricolage involved strategically organizing the
spatiotemporal resources, so that Jihun could draw from them effectively. That
these scholars’ thinking and communicating were shaped by material and
spatial resources does not mean that they lacked agency. In other cases, the
participants used verbal resources strategically to construct new empowering
spaces in the workplace, exploiting the layering of place and space. Though they
are working in a US university, which might be considered as treating Anglo
American native speaker varieties of English as the norm, the participants
found spaces to use their first languages or their localized varieties of English
in many places. Chinese scholars mentioned that in laboratories or offices they
would switch to Chinese to indicate that they would like to have the conver-
sation in their first language if they had their co-nationals as a majority:
 S: Uhm, so I don’t know how to define the workplace. So in- like, you
know if uh, it’s like in a public (place), you know we have a few group
meeting, you know, a few students, so then we talk with uh- in English.
But if the individual meeting, you know, then like with the students in
my office, sometimes we just use Chinese. It’s easier to uh understanding.
[Sencun Zhu; Male; Associate Professor of Computer Science and
Engineering]
Sencun finds it difficult to define the ‘workplace’ because it is layered with
diverse spaces for discourses and identities that defy homogeneous norms.
Another scholar mentioned that his supervisors and advisors were themselves
multilingual speakers—that is one was Irish, another Taiwanese, and the third
Chinese. Therefore, they were tolerant in using their local varieties of English
and negotiating meanings with each other by drawing from other spatial rep-
ertoires. In adopting such practices, they were using verbal resources to re-
define the work context into more inclusive spaces where diverse and
alternate language norms were admissible. Since space is expansive, it provides
resources for participants to construct alternate spaces within bounded and
hegemonic places, to suit their interests.

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 48 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

This orientation suggests the transformative potential of spatial repertoires,

as defined in the fourth definition of translingualism. The diverse languages
used in the interactions of STEM scholars are somewhat resistant in the con-
text of the monolingual policies that are dominant in American universities.
These scholars have been tested—sometimes repeatedly—on TOEFL, IELTS,
and local university assessments for their English language proficiency.
Often their appointments in American universities are dependent on these
scores. Such examinations and policies convey the understanding that
English is the language of the workplace and ensures success in the STEM
professions. The translingual practice of the international scholars suggests
that they appreciate the value of language diversity and subtly act against
dominant policies and discourses. Even though the lack of diverse languages
in the publications in high stakes contexts suggests international scholars sat-
isfying dominant norms, we must not forget that the earlier drafts, notes, and
revision interactions involve translingual resources, suggesting their value in
generating the finished products. Though such translingual practices occur in
safe and protected sites, away from surveillance or high stakes interactions, we
must not underestimate their transformative potential to diversifying the
workplace or pluralizing high stake activities in the long term.

FROM COMPETENCE TO EMPLACEMENT

To summarize the discussion so far, in accomplishing communicative activities,
my participants draw from diverse verbal resources (beyond labeled lan-
guages); from semiotic resources that can be agentive in shaping communica-
tion; adopting distributed practice that involves collaborating with a network
of social agents, objects, and bodies in layered spatiotemporal scales; all of
which constitute an assemblage of situated and emergent spatial repertoires.
These repertoires are not representational but performative. Rather than
coming loaded with values and meanings, they generate meanings in situated
use. I can summarize the shifts involved from a structuralist to a spatial orien-
tation to translingualism as follows in Table 1, in the order they have been
introduced in this article:

Table 1: Shifts from structuralist to spatial orientation

Structuralist Spatial

 Predefined meanings Indexicality

 Whole language Truncated multilingualism
 Grammar Spatial repertoires
 Communication Activity
 Agency of humans Agency of things
 Cognition Embodiment

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 49

Structuralist Spatial

 Hierarchy Flat ontology

 Multimodality Assemblage
 Text/context binary Entextualization
 Methodological individualism Distributed practice
 Metaphysics of presence Spatiotemporal scales
 Discrete contexts Layered simultaneity
 Representational thinking Performativity/bricolage
 Contextual shaping Spatial transformation
 Competence Emplacement
 Arboreal development Rhizomatic development

Addressing translingualism from this broader spatial perspective has many

ramifications for how we conceive of a competence to engage in translangua-
ging. I will articulate two major implications deriving from a spatial orienta-
tion—that is competence as emplacement, and proficiency development as
rhizomatic.
From a spatial orientation, communicative proficiency involves the ability to
align diverse semiotic and spatial resources for successful activity. Along with
the flat ontology of assemblage, it holds that all resources have to be brought
together for successful communication. Also, beyond giving primacy to the
mind, it posits that the body and material objects facilitate thinking. The
notion of ‘alignment’ has enjoyed some currency in the sociocognitive orien-
tation in applied linguistics to suggest the mind-body-world connection
involved in meaning construction and language learning. It is defined as:
‘‘‘the complex means by which human beings effect coordinated interaction,
and maintain that interaction in dynamically adaptive ways’’ (Atkinson et al
2007: 169)’. This is a useful metaphor, as it suggests that meaning and cogni-
tive representations are not predefined but emergent through such alignment.
It also suggests that one can never come ready for communicative activity with
all the required grammars and codes. One has to undertake the alignment of
diverse resources and contextual conditions for meaning. In this manner, the
construct has the potential to index the nonrepresentational orientation de-
veloped in this article.
However, in available studies, alignment has been used in ways that differ
from the spatial paradigm articulated above. Though other languages are part
of the context of language learning and shown as mediating the learning pro-
cess, the authors typically focus on competence for one language at a time,
isolating competence to a single language structure (see Atkinson et al 2007;
Nishino and Atkinson 2015). Also, despite the mediation of the body and ob-
jects, cognition is given importance in orchestrating this alignment and indi-
viduals are the frame of reference, falling into the trap of methodological

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 50 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

individualism. Mind appears to enjoy more power in orchestrating and bene-

fitting from the alignment, though the body and material resources are treated
as mediating the work of the mind. Note in the definition above that human
beings are given agency in ‘effecting’ this coordination. It provides agency for
humans to do this alignment without fully analyzing the agentive work of
others (including artifacts) in the spatial ecology. Similarly, learning is ac-
counted for primarily in terms of grammar and cognition, though they are
mediated by diverse material, social, and bodily affordances. The analysis of
interactions is also influenced by ametaphysics of presence, prioritizing the
influences that are locally visible. The methodology is defined as ‘focus[ing]
on highly contingent, moment-to-moment attunement of writer and ecosocial
environment, yielding a richly detailed analysis of multimodal interaction’
(Nishino and Atkinson 2015: 42). Though sensitive to multimodality, the se-
quential analysis focuses on the face to face and local. In adopting these orien-
tations, the treatment of alignment eventually fails to transcend some of the
limitations of the structuralist paradigm.
Another metaphor that accommodates the spatial orientation better is em-
placement (adopted by scholars in rhetoric: Pigg 2014; Rickert 2013). Indexing
the necessity for individuals to situate themselves in the spatial ecology, not
only to align the diverse resources but also to be shaped by them, emplacement
accommodates the qualified agency of human beings. It gives more importance
to the body in drawing from the ecological affordances for meaning-making
processes. Emplacement is more a physical activity than mental, as compe-
tence has been traditionally theorized. The word ‘emplacement’ is suitably
more passive in formulation, indexing the possibility that communication
and cognition are shaped by the agency of material and spatial features. This
construct is more open to drawing from all spatial repertoires equally for a
translingual ability. However, emplacement is not completely deterministic.
Communication is not a case of people and meanings completely determined
by things and space. One’s emplacement can be strategic, responsive, and
creative, as we found in the case of Tan and Jihun above. They draw from
all the resources in their environment to make up for their grammatical limi-
tations. Spatial repertoires are affordances for resourceful communication
through strategic emplacement. One can also resist the territorialized norms
of bounded places by constructing alternate spaces that accommodate diver-
sity, as claimed by my Chinese interview respondents. Emplacement can ac-
commodate a qualified human agency, while it gives spatial resources and
semiotic repertoires considerable significance in meaning construction.
Conceiving of competence as emplacement also posits a different direction-
ality for language development. Structuralist-influenced Second Language
Acquisition (SLA) research has been motivated by figuring out the trajectory
or sequence of grammar learning in accounting for competence. Much re-
search has involved finding out the basic or elementary grammatical structures
that should precede the acquisition of more advanced structures for cumula-
tive acquisition (see early work on morpheme order studies, for example,

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 S. CANAGARAJAH 51

Dulay et al 1982). Similarly, casual conversation is considered easier to begin

with, compared to writing and specialized/professional/institutionalized com-
munication. These assumptions posit an arboreal metaphor of structural de-
velopment (as theorized by Deleuze and Guattari 1987). It is as if
communication involves forming strong grammatical roots before growing
trunks and branches of pragmatics, literacy, and discourse. The progression
is linear and causal, from roots to the branches and flowers or fruits.
Furthermore, the roots are subterranean, analogous to the invisible mind
(the Chomskyan ‘language acquisition device’) in relation to the visible ma-
terial life. Such assumptions are embedded in the langue/parole or compe-
tence/performance distinctions which treat the underlying cognitive
competence as shaping social and material practice.
The practice and claims of the STEM scholars suggest the possibility of non-
linear language development. They demonstrate proficiency in their specia-
lized disciplinary registers before casual genres; in writing before speaking; and
bits and pieces of verbal resources from different languages before their
‘underlying’ or ‘basic’ grammatical structures. A metaphor that might describe
their nonlinear proficiency development is the poststructuralist metaphor of
rhizomes. Rhizomes have multiple roots; the roots and the branches are indis-
tinguishable; the roots might be above ground; and each part of the plant can
become a root for new growth. A ginger root might be a good example of
rhizomes, as the multiple roots and the body are indistinguishable.
Rhizomes suggest a different starting point for learning and proficiency, com-
plicating the competence/performance distinction. Performance in a specific
register of a labeled language for communicative activity in isolation from the
full or ‘basic’ knowledge of that language is not only possible, it might motivate
additional learning or even development of other verbal resources. Such per-
formance with a small collection of verbal resources from a language does not
have to make one’s language capacity suspect, as these resources work with
other semiotic resources to gain their coherence and meaning. Furthermore,
other features of a spatial repertoire, such as using artifacts or gestures can also
facilitate language development. Complicating the sole or primary role given to
cognition and grammar as in structuralist understanding of ‘competence’, rhi-
zomes diversify the direction and bases of language development. Thus rhi-
zomes favor a performative orientation to meaning-making rather than relying
on preconstructed cognitive or grammatical representation. Meaning and
thinking can emerge at many different points of the rhizome, in the liminal
spaces of body, objects, and spatial resources. This metaphor captures the pro-
ficiency and learning of the STEM scholars much better.

CONCLUSION
To conclude, this article demonstrates that there is more to the paradigm shift
of translingualism than accommodating more diverse verbal resources in

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018
 52 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES

communicative practice or in one’s proficiency. To appreciate these larger im-

plications, we have to move beyond the foundational structuralist orientation
in linguistics. In adopting a spatial orientation, we realize that translingualism
accommodates communicative practices that include more expansive spatial
repertoires that transcend text/context distinctions and transgress social
boundaries. To consider how communication works in this fashion, we have
to also treat meaning making ability as distributed, accommodating the role of
social networks, things, and bodies, beyond mind and grammar, requiring
strategic emplacement.
As we adopt a spatial orientation to translanguaging, we confront new
methodological and analytical questions. To begin with, we have to consider
how to define the unit and focus of analysis when a flat ontology assumes that
everything is connected to everything else (see for a discussion, Canagarajah
2017). Though we have to adopt pragmatic boundaries (or ‘cuts’) on the unit
of analysis and selectively focus on the semiotic resources playing a more sig-
nificant role in a communicative activity, we have to consider how we can
keep other contexts and resources as part of the analysis. Admittedly, the
choice of resources and meanings I have discussed in the case studies above
are selective. For example, I have not discussed the role of time too extensively
in this analysis. I have focused on the spatial resources that emerged as more
significant in each activity. Therefore, we have to make these analytical cuts
based on a strong rationale, generating more complex interpretations. Another
limitation in my approach is that, for practical reasons, my data gathering has
been circumscribed to bounded settings. We might need multi-sited data gath-
ering as we consider the semiotic resources from expansive and layered times
and places that shape communicative activities. Moving from the currently
dominant methodological individualism and metaphysics of presence in
applied linguistics research will require creative new methods for data collec-
tion and analysis. These questions will be explored further, as we continue to
adopt a spatial orientation to translingualism.

NOTES
1 All names are pseudonyms. Consent . Falling intonation
was obtained following IRB approval. ? Rising intonation
2 Thanks to Joan Kelly Hall and , continuing intonation
Stephen Looney for lending me the (aa) speech transcriber is
Corpus of English for Academic and unsure of
Professional Purposes, currently in <GESTURE> nonverbal information
development. [. . .] ellipsis
3 The transcription conventions are as (.) micropause
follows: @@ laughter
- Cut off, self-
interruption

Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948

by guest
on 08 February 2018

REFERENCES
Agha, A. 2003. ‘The social life of cultural value,’
Language and Communication 23: 231–73.
Atkinson, D., E. Churchill, T. Nishino, and
H. Okada. 2007. ‘Alignment and interaction in
a sociocognitive approach in second language ac-
quisition,’ Modern Language Journal 91: 169–88.
Bakhtin, M. M. 1986. Speech Genres and Other
Late Essays. V. W. McGee (trans.). University
of Texas Press.
Barad, K. 2007. Meeting the Universe Halfway:
Quantum Physics and the Entanglement of Matter
and Meaning. Duke University Press.
Block, D. 2014. ‘Moving beyond ‘‘lingualism’’:
Multilingual embodiment and multimodality
in SLA,’ in S. May (ed.): The Multilingual
Turn: Implications for SLA, TESOL, and Bilingual
Education. Routledge.
Blommaert, J. 2013. Ethnography, Superdiversity,
and Linguistic Landscapes. Multilingual Matters.
Blommaert, J. 2010. The Sociolinguistics of
Globalization. Cambridge University Press.
Blommaert, J. 2005. Discourse: A Critical
Introduction. Cambridge University Press.
Blommaert, J., J. van der Aa, and M. Spotti.
2017. ‘Complexity, mobility, migration,’ in
S. Canagarajah (ed.): Routledge Handbook on
Language and Migration. Routledge.
Canagarajah, S. 2017. ‘The unit and focus of ana-
lysis in Lingua Franca English interactions: In
search of a method,’ International Journal of
Bilingual Education and Bilingualism, forthcoming
Canagarajah, S. 2013. Translingual Practice: Global
Englishes and Cosmopolitan Relations. Routledge.
Canagarajah, S. and P. De Costa. 2016.
‘Introduction: Scales analysis, and its uses
and prospects in educational linguistics,’
Linguistics and Education 34: 1–10.
Coole, D. and S. Frost. 2010. New Materialisms:
Ontology, Agency, and Politics. Duke University
Press.
Deleuze, G. and F. Guattari. 1987. A Thousand
Plateaus. University of Minnesota Press.
Derrida, J. 1973. Speech and Phenomena and Other
Essays on Husserl’s Theory of Signs. J. Allison
(trans.). Northwestern University Press.
Douglas Fir Group. 2016. ‘A transdisciplinary
framework for SLA in a multilingual world,’
Modern Language Journal 100: 19–47.
Dulay, H., M. Burt, and S. D. Krashen. 1982.
Language Two. Oxford University Press.
Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948
by guest
on 08 February 2018
S. CANAGARAJAH 53
Duranti, A. and C. Goodwin (eds). 1992.
Rethinking Context: Language as an Interactive
Phenomenon. Cambridge University Press.
Flowerdew, J. 2007. ‘The non-Anglophone
scholar on the periphery of scientific commu-
nication,’ AILA Review 20: 14–27.
Goffman, E. 1964. ‘The neglected situation,’
American Anthropologist 66: 133–6.
Goodwin, C. 2000. ‘Action and embodiment
within situated human interaction,’ Journal of
Pragmatics 32: 1489–522.
Hanks, W. 2006. ‘Context, communicative,’ in J.
L. Mey (ed.): Concise Encylopedia of Pragmatics.
Elsevier, pp. 119–32.
Harland, R. 2010. Superstructuralism. Routledge.
Hymes, D. 1971. ‘On linguistic theory, commu-
nicative competence, and the education of dis-
advantaged children,’ in M. L. Wax , S.
A. Diamond, and F. Gearing (eds):
Anthropological Perspectives on Education. Basic
Books, pp. 51–66.
Kell, C. 2010. ‘Literacy practices, text/s and
meaning making across time and space,’ in
M. Baynham and M. Prinsloo (eds): The
Future of Literacy Studies. Palgrave Macmillan,
pp. 75–100.
Kress, G. 2009. Multimodality: A Social Semiotic
Approach to Contemporary Communication. Routledge.
Latour, B. 2005. Reassembling the Social: An
Introduction to Actor-Network-Theory. Oxford
University Press.
Li, W. and H. Zhu, 2013. ‘Translanguaging iden-
tities and ideologies: Creating transnational
space through flexible multilingual practices
amongst Chinese university students in the
UK,’ Applied Linguistics 34: 516–35.
Looney, S. D., D. Jia, and D. Kimura. 2017.
‘Self-directed okay in mathematics lectures,’
Journal of Pragmatics 107: 46–59.
Marston, S., J. Jones, and K. Woodward.
2005. ‘Human geography without scale,’
Transactions of the Institute of British Geographers
30: 416–32.
Massey, D. 2005. For Space. Sage.
McNamara, T. 1997. ‘Interaction’’ in second
language performance assessment: Whose per-
formance?,’ Applied Linguistics 18: 446–66
Mondada, L. 2014. ‘The local constitution of
multimodal resources for social interaction,’
Journal of Pragmatics 65: 137–56.
 54 TRANSLINGUAL PRACTICE AS SPATIAL REPERTOIRES
Nishino, T. and D. Atkinson. 2015. ‘Second
language writing as sociocognitive alignment,’
Journal of Second Language Writing 27: 37–54.
Ochs, E., P. Gonzales, and S. Jacoby. 1996.
‘When I come down I’m in the domain
state: Grammar and graphic representation
in the interpretive activity of physicists,’ in
E. Ochs , E. A. Schegloff, and S. Thompson
(eds): Interaction and Grammar. Cambridge
University Press, pp. 328–69.
Pennycook, A. 2016. ‘Posthumanist applied lin-
guistics,’ Applied Linguistics, in press doi: 10.
1093/applin/amw016.
Pennycook, A. and E. Otsuji, 2015.
Metrolingualism: Language in the City. Routledge.
Pigg, S. 2014. ‘Emplacing mobile composing
habits: A study of academic writing in net-
worked social spaces,’ College English 66: 250–
75.
Rickert, T. 2013. Ambient Rhetoric. University of
Pittsburgh Press.
Downloaded from https://academic.oup.com/applij/article-abstract/39/1/31/4626948
by guest
on 08 February
View2018
publication stats
Shohamy, E. and D. Gorter (eds). 2008.
Linguistic Landscapes: Expanding the Scenery.
Routledge.
Silverstein, M. 1985. ‘Language and the culture
of gender,’ in E. Mertz and R. Parment (eds):
Semiotic mediation. Academic Press, pp. 219–59.
Soja, E. 2011. Postmodern Geographies: The
Reassertion of Space in Critical Social Theory. Verso.
Streek, J. 2013. ‘Interaction and the living
body,’ Journal of Pragmatics 46: 69–90.
Swales, J. 2004. Research Genres. Cambridge
University Press.
Thrift, N. 2007. Non-Representational Theory:
Space, Politics, Affect. Routledge.
Van Compernolle, R. A. and L. Williams.
2011. ‘Thinking with your hands: Speech-ges-
ture activity during an L2 awareness-raising
task,’ Language Awareness 20: 203–19.
Wilson, A. and S. Golonka. 2013. ‘Embodied
cognition is not what you think it is,’
Frontiers in Psychology 4: 1–35.