Grounding the Translation: Intertwining analysis and translation in cross-language grounded theory research
Svetlana Shklarov, MD, RSW, PhD Candidate
Abstract
Grounded theory research in cross-language, cross-cultural
context is associated with the challenges of linguistic sensitivity
of conceptualization. The author, a bilingual researcher, offers
reflection on her experience of doing grounded theory research,
assuming a dual role of a theorist and a translator. The reflection
is based on self-observations. Grounding the translation is shown
to be acheived through the strategy of intertwining the activities
of translation and conceptual analysis, performed by one person.
The two activities are inseparable in time and take place along
with constant comparison across language boundaries.
Intertwining requires that theoretical sensitivity of the
researcher be enriched with the sensitivity to linguistic and
cultural meanings. Intertwining, through revealing differences
between linguistic meanings or language structures, facilitates
the emergence of concepts and theoretical categories from the
very act of translation. Combining the functions of translation
and analysis and using the strategy of intertwining worked
effectively for this researcher.
Introduction
Translation has been so much a part of qualitative research
in multicultural settings that we rarely give heed to the depth of
processes involved in cross-language data collection and analysis.
Certain aspects are better understood than others. The most
common, traditional concern is the accuracy and equivalency of
information transferred from one language to another – the
quality and ethics of translation (e.g., Houbert, 1998; Hunt &
Bhopal, 2004). More recently researchers began to analyze the
challenges of representation across languages, multiple
interpretations, reflexivity, and the integral role of the translator
(e.g., Friedrich, 1992; Mc Laughlin & Sall, 2001; Muula, 2005;
Temple & Edwards, 2002).
In this article I present a reflection on my experience of
conducting grounded theory research, as a sole bilingual
researcher with monolingual participants. Through analyzing my
self-observations in this project, which I was conducting for my
PhD, I examined the aspects of translation in cross-language
grounded theory study. In this study I explored life stories
narrated by Russian-speaking Holocaust survivors, recent
émigrés from the Former Soviet Union. My research design
involved a full combination of translation and analysis, in which I
assumed the position of a bilingual investigator who performed
both functions. In such setting, a theory emerges from the data
written (or spoken) in the language of monolingual participants
that is unknown to the audience (the source language). Research
results are presented in the language of the audience (the target
language).
There are no specific prescribed procedures for translation in
the context of grounded theory research. Therefore, I
experimented in my study with some patterns of working in
cross-language area, using my previous experiences in
translation. I applied more systematically those patterns that
worked for me, and observed how these patterns fit into the
analysis. In my research, I have found that doing cross-language
grounded theory involves strategies that differ from those
involved in traditionally understood translation. My
experimenting led me to the discovery of a strategy that emerged
naturally in my work, namely, the intertwining of the activities of
translation and conceptual analysis. Both translation and
conceptual analysis were activities, or acts, which I performed as
a bilingual person (in that, these were both my functions).
Intertwining these activities was the strategy that I used to
achieve better grounding of my translation in cross-language
data, while discovering a grounded theory.
In this article I analyze some properties of the strategy of
intertwining, and reflect on the relevant features of language
translation in this context. In my attempts to reflect on my
experiences doing this research, I intend to ground my
conclusions in thorough self-observations, but also recognize the
limitations of these conclusions, which are based on one person’s
practices. Therefore, I do not aim at developing an integrated
theory on the basis of my limited reflections. The purpose of the
hypotheses presented in this article is only to begin to understand
the complex patterns that are associated with cross-language
grounded theory research, and to invite further exploration of
this area.
Conceptual Analysis and Translation – The
Intertwining
Conventionally, any form of cross-language exchange implies
the need for text translation, which is most naturally understood
as conveying the full textual message verbatim from one
language to anther. By text I understand a unit of analysis, any
set of verbal qualitative materials, whether written or spoken.
Traditionally, the original text is expected to be presented in the
target language, before the analysis begins and before the
findings can be reported (Glicksman & Van Haitsma, 2002;
Temple, 2006). However, the expectation of full, verbatim
translation and its separation from the analysis does not always
match the grounded theory research objectives, when both
functions are performed by one bilingual person. The interplay
between the two activities – translation and conceptual analysis –
involves the strategy of intertwining, which includes analytical
efforts that take place before, in parallel to, or independent of the
technical textual translation. Intertwining the two activities
(which are the two functions of one researcher) can also yield
additional emerging concepts that would be missed otherwise. To
explore the strategy of intertwining and try to understand what it
involves, we might begin with comparing some strategies
involved in language translation and discovering a theory.
Translation and GT: The Parallels
Generating conceptual theory from empirical data, as a
cognitive act, has some similarities with language translation.
Both activities are rooted in discovering and conveying conceptual
meanings: the former from descriptive data into general patterns
and a theory, and the latter – across texts written in two different
languages. The direction of translation cannot be presented as a
one-way vector. The constant search for a suitable word involves
the reiterative comparison between words and textual contexts
that flows in both directions to balance the equivalency of
meanings. By analogy, in search for working concepts, the
procedures of grounded theory require continuous comparison
that is carried out across the data and the emerging concepts.
A given concept is often signified in two languages by words
that have similar meanings but bear different subtle nuances and
cultural connotations (Hunt & Bhopal, 2004; Tsai et al., 2004; see
also Schopenhauer, 1800/1992). These differences have to be
captured in the translation. Often it is impossible to express a
complex concept in different languages with precise equivalency,
and the translator has to settle for the most effective compromise.
The settling for a compromise involves elements of theorizing.
Translating is challenging when complex, contextually
foreign words are under consideration. For example, in my
research, it was difficult to find a Russian equivalent to the
English word resilience. I tried to translate this word into
Russian in the beginning of my data collection, because I
intended to present the concept of resilience to my participants in
our interview conversations.
In English-language psychology, the meaning of resilience is
metaphoric: “Resilience, according to the dictionary, means
recoiling or springing back to the original shape after bending,
stretching, or compression. Psychological resilience implies a
similar springing back after having been subject to severe
stressors.” (Valent, 1998, p. 517). There are a number of Russian
words that can express the concept, but none have this
metaphoric meaning, and each of the translation options has a
connotation that is somewhat different from other versions
(Muller, 1990). Occasionally, authors use the metaphor of
elasticity (эластичность) (Pearce, 1997), but this translation is an
exception. Most common translation choices in the available
Russian-language academic literature have a meaning close to
hardiness or [life-] steadiness. (In an English presentation, I can
only give the closest literal translation of the Russian words.)
The contextual connotation of the concept is also
controversial: in western academic discourse, it is often
associated with the psychology of trauma, vulnerability, and
posttraumatic sequelae. This connection is very weak when the
concept is used in the Russian cultural context, because
historically the notion of psychic trauma has been suppressed in
Soviet academic and public discourse until the recent decade.
The word that I used in my conversations with the Russianspeakers
was not a literal reflection of the English metaphoric
concept. Therefore, I often employed more than a single word. I
added background explanations and frequently invited my
interview partners’ input by asking about their understanding
and interpretations of the phenomena.
In this example, a routine translation, through analysis and
comparison, became a micro-theory development, with an
outcome expressed in a set of working and fitting words, together
with a context-specific interview strategy and data collection
technique. According to my observations, translation is associated
with constant comparing of the concepts, the words’ meanings,
and their properties in the contexts of different cultures.
Translation and GT – The Intertwining
An ability to conceptualize and a commitment to constant
comparison are paramount for good translation, to the same
extent as these qualities are important for building a grounded
theory. Therefore, combining the functions of a researcher and a
translator is most natural for a bilingual theorist.
Because one bilingual person carries out both translation
and conceptual analysis, these two functions become intertwined.
According to my self-observations, the intertwining involves the
following properties. First, the two functions (the activities of
translation and analysis) are inseparable in time and happen
simultaneously. Second, constant comparison, which is an
essential tool of analysis, takes place across language boundaries,
transcending the technical stage of isolated translation. Third,
theoretical sensitivity of the researcher needs to be enriched with
the sensitivity to differences in language meanings and its
implications for the emerging theoretical concepts. The analyst
takes an active role engaging in the interplay, reiteration
between the two activities. And finally, data for conceptual
analysis can be collected from the very act of translation, and the
differences between meanings or language structures can become
a source of important concepts and theoretical categories. I will
now review and illustrate these properties.
Inseparability in time.
The first property (the simultaneousness of translation and
analysis) is implied by the intrinsic similarity of the two activities
and interplay between them, which prevents one from separating
them in time. My experience has demonstrated that when I work
with texts across languages, the acts of translating and analyzing
literally become intertwined in one cognitive act of
conceptualizing, regardless of cross-language textual boundaries,
or any intentions of scheduling time for translation separately. It
would be a simplification to expect that the two functions of
translation and analysis be performed by the researcher in
sequence (e.g., translation first, coding and analysis next),
because by translating the text one will inevitably engage in
conceptualizing the ideas that emerge from its content.
I have noticed that analysis, coding, and memoing cannot be
delayed until the full translation is completed. Most grounded
theory researchers are familiar with the emergence of ideas
during the interviews or shortly after, before the formal, technical
and written implementation of coding or other procedures. In a
similar way, my cross-language mindset became tuned into the
subtleties of cultural meanings early in the data collection stage.
I could not separate translation from analysis.
When I first attempted translating my transcribed
interviews in the early days of my research, concepts began to
emerge as soon as I started translation, even prior to creating the
written target-language text. I decided to change the tactics. I
transcribed the interviews fully in Russian, and then proceeded
with open coding and writing my memos in English, without prior
full translation of the source texts. I skipped the stage of
translating my transcripts verbatim. This strategy proved useful,
and allowed me to avoid the distortion of the original sourcelanguage
words before their conceptual meaning became evident
through the analysis. I realized that premature language
translation could have influenced the consequent coding, and
decided to preserve all my raw data in Russian.
Constant comparison across language boundaries.
In my research, the simultaneous undertaking of translation
and analysis led to the simultaneous work with texts in both
languages. The second property of intertwining, which is the
constant comparison across the languages, helps ensure the link
between data and emerging concepts. For example, during
selective and theoretical coding, I continued working with
concepts signified by English words, constantly comparing the
meaning of English words with the meanings embedded in source
language data. Thus, the comparison between concepts, data
incidents, and emerging theoretical hypotheses, expressed in
theoretical codes and memos, occurred and was constantly
reiterated across languages, transcending cross-language
boundaries.
I translated selected excerpts of the interviews into English.
This verbatim translation usually occurred at the time of writing
memos that were grounded in the particular data excerpts – the
indicators of the conceptual categories that were central in these
memos. Writing such memos, accompanied with translating
related data excerpts, enriches the memos with concepts
emerging as a result of deeper focusing on source data through
translation.
Working across languages is reiterative, and is not limited to
one separate procedure, such as open, selective, or theoretical
coding. Constant comparison requires returning to the source
language, if the need for it emerges, at any time of the analysis,
for example, when writing a theoretical memo or at the stage of
writing up the theory. Consistent with general grounded theory
procedures, in the early stages of research comparative working
across languages happens at the level of data incidents, or
between data incidents and emerging concepts. In later stages of
theoretical coding and writing, language comparison is applied to
ideas expressed in the two languages, rather than only data.
I found that, with the intertwining of translation and
analysis, reiterative returning to the source language does not
always mean referring to source data texts, but often also
pertains to working with concepts and theoretical ideas, through
fitting them in both languages. For example, the researcher can
create source-language memos in parallel with the ones in
English, or check emerging concepts for their fit in both
languages upon their emergence. The purpose of such activities is
not verification by back translation, but rather deepening the
analysis through cross-language enrichment of the emerging
categories.
Association between language ability and theoretical
sensitivity.
While intertwined in time and in constant comparison across
the texts in both languages, the activities of translation and
analysis remain the functions of one person: the bilingual
theorist. The qualities of this person, such as her ability to
conceptualize and her theoretical sensitivity, are essential in
research. According to Glaser (2002b), the researcher’s position,
in general, is “a vital variable to weave into the constant
comparative analysis” (para. 11). The researcher’s language
knowledge is, thus, an important part of such a variable in crosslanguage
setting. This leads us to the third property of
intertwining, which I observed in my study: the significance of
the researcher’s sensitivity to differences in language meanings
for her overall theoretical sensitivity, and the use of these
associated qualities in performing both functions.
In grounded theory research, the discovered theory is
expected to carry features of the individual researcher’s
theoretical sensitivity and creative ability to conceptualize. As a
researcher, I was constantly aware that my theoretical sensitivity
was partially dependent on my ability to conceptualize the subtle
linguistic and cultural differences. Having been involved in
multiple projects that required both translation and analysis, I
was not new to language translation. In more than ten years of
my work with multicultural projects prior to initiating this study,
I had extensive experience of oral and written translation in
different settings and disciplines, ranging from client information
in services to seniors, to international teaching situations, and to
professional publications in mental health and psychiatry. These
experiences have sharpened both my sensitivity to language
differences and my skills necessary for achieving cultural
relevancy in translation. I discovered that these skills
significantly enriched my theoretical sensitivity and ability to
conceptualize. In addition, remembering Glaser’s (1978)
recommendation of “reliance on the social psychology of the
analyst” (p. 2), I could rely on my intimate connection with
Russian as my first language, using this connection and my
knowledge of the culture within the constant comparison mode.
Emergence of concepts from the act of translation.
In my research, the intertwining allowed capturing the
differences between linguistic meanings, which facilitated
discovering patterns implied by these differences. In doing so, I
discovered the fourth and final property of intertwining: the
emergence of concepts from data collected as a result of the very
act of translation. Important ideas can originate from capturing
the differences between meanings or language structures.
A single word and its context in a participant’s utterance can
provide data for discovering a significant category or a number of
interconnected categories. An analysis of the following interview
episode illustrates my statement. One of my interview partners
referred to herself in our conversation as a “victim of the
Holocaust.” My first reaction was to ask a probing question, “Do
you consider yourself a victim?” The interviewee’s answer was,
“Yes, I am a victim.” No further explanations followed, and she
continued her story as if uninterrupted, without giving much
notice to the issue. I understood that for her, this was not a
question worth discussion (L.Y. transcript, 2006).
I knew the difference between the cultural connotations of
the word victim in the two languages, and sensed the potential
discrepancy. In English, and in particular in the context of
traditional conversations with the Holocaust survivors, the word
victim bears a somewhat negative, inferior connotation that
makes it relatively uncommon in the contemporary vocabulary of
western-educated survivors. The connotation relates to the
western discourse, in which this word is paired with the word
survivor. The common victim-survivor dichotomy implies the
victorious nature of survivorship, and the triumph of the human
spirit over life adversities. Within this binary opposition, victim
would be the negative polarity, and survivor – the positive one
(for reference on binary opposition in social contraction and
language, see Gergen, 1999). It is possible that a Holocaust
survivor who is used to western listeners would have recognized
the prompt in my question (“Do you consider yourself a victim?”)
and responded to it differently. Conversely, for my research
participant, in her language context of a former Soviet citizen and
a Riussian-speaker, there was no conflict between the two
categories. My probing question and the conceptual connection I
was trying to imply appeared irrelevant.
The Russian word victim, although a precise equivalent of
the English word, does not always bear the same contextual
nuances. In many contexts it has a somewhat heroic connotation
(it also has a meaning of sacrifice that is stronger than in
English). Conversely, a precise structural and grammatical
equivalent of the word survivor does not exist in the Russian
language. This makes it difficult to find a literal and
grammatically accurate translation of the common word
combination Holocaust survivor. In Russian, one would use such
words as victim, or [former] inmate, or a combination of several
words in an awkward grammatical form.
It is worth noting that historically, Soviet social and political
attitudes towards the Holocaust have been ambiguous, rooted in
denial, almost taboo associated with the Holocaust memory
(Altman, 2005). One can draw comparative parallels between
historical and social processes in Russia, the conventional
language related to the Holocaust, and the meanings that people
attribute to their experiences. Formulating and integrating my
hypothesis required additional data and further analysis, but the
categories that emerged from this episode sensitized me and
concurred with my other data. Analyzing the nuances of this
word’s meaning had direct relevance to my emerging theory.
Having discussed the properties of intertwining, it is
important to explore the features of the act of translation, when,
together with conceptual analysis, it becomes an integral part of
discovering a grounded theory.
Transforming Translation Paths: Interferences and
Adjustments
When research is performed by a single bilingual theorist,
translation as a pure, isolated process is not relevant. In turn,
conceptual analysis does not exist in isolation from language
translation. Therefore, although the title of this article includes
the word translation, I believe that this term, in its classical
meaning, does not ideally describe the activity that is so closely
intertwined with conceptual analysis in cross-language grounded
theory work. When translation becomes such an inseparable part
of research, the act of translation loses some of its traditional
properties and acquires other features. Pure, isolated traditional
translation can potentially become a source of interferences, lead
to a clashing of strategies, and create a misfit between analytical
goals and language relevance.
What are the interferences between pure translation and
conceptual analysis that require such change? What are the
adjustments that can enhance cross-language analytical
strategies? In this section, I review the properties of translation
that do not fit into the context of discovering a grounded theory,
and therefore have to be adjusted. These adjustments have to be
made to the act of translation, to balance the interferences
between its traditional properties and the demands of grounded
theory research.
Conceptualization versus Description
Text analysis in grounded theory is based on
conceptualizing, as opposed to full and exhaustive portrayal of
the area under study. Grounded theory ascends to the abstract
level of concepts, which are grounded in the empirical
descriptions – the text.
Conversely, conventional translation aims at complete, full,
and accurate equivalence between the source and the target
texts, so that they convey the same details of the message. As
opposed to a traditional devoted translator, a grounded theorist is
not concerned about the exact and accurate presentation of the
source text, and therefore can transcend an accurate verbatim
translation. The descriptive fixation on textual details can
become interference to conceptualization, if translation is used in
a traditional way. Therefore, in grounded theory research, the
adjustment is made to the conventional act of translation:
Translation loses its worrisome fixation on literal details or
concrete words, but rather aims at conveying the conceptual,
abstract meaning.
Technically, this means that while coding, the analyst can
draw conceptual categories from the source text and formulate
them in the target language, providing that the discovered
concepts and theoretical patterns emerge directly from the data
and fit the area under study. The researcher can also refer to
partial full translation for illustration or other purposes defined
by the objectives of the analysis.
To ensure that the findings are grounded in data, the
researcher needs to remain faithful to conveying conceptual
meanings emerging from the text, the meanings expressed by
participants. One of the major properties of grounded theory,
constant comparison, is paramount for such grounding, and takes
the form of continuous cross-language comparison, as it happens
with the intertwining of translation and analysis. The relevancy
of translation, therefore, requires the abandonment of the quest
for descriptive accuracy, but remains rooted in the principles of
constant comparison and conceptual equivalency.
Emergence versus Verification
Traditional translation is widely used in multicultural
settings of conventional quantitative research. Most of the
commonly accepted standards and criteria of the quality of
translation were developed in traditional, verificational research
(Hunt & Bhopal, 2004; Roberts, Kent, Prys, & Lewis, 2003; Yu,
Lee, & Woo, 2004). In such research, the initial hypothesis needs
to be verified through developing operational indicators and
measuring them (i.e., using deductive logic). Accordingly, in crosslanguage
setting of traditional quantitative research, the initial
hypothesis (conceptualized in English) is routinely
operationalized, and then translated into the source-language
indicators, or measures, which are applied to monolingual
research subjects (e. g., verified translations of questionnaires,
assessment scales, or surveys into the language of participants).
In traditional research, translation commonly requires
verification. Verification can be accomplished through a number
of technical means, such as comparing translations by two or
more independent translators, committees, or analyzing back
translation (a “round-trip” translation technique).
Conversely, in grounded theory, the logical process of
research is reverse, mostly inductive, and so are the translation
processes. Grounded theory is based on emergence and induction,
and is explorative rather than focused on verification (Glaser &
Strauss, 1967; Glaser, 2002a; Stebbins, 2001). The researchertranslator
uses conceptualization and constant comparison
method. My observations led me to believe that when translation
is intertwined with conceptual analysis, and the translated
message emerges from the source data, no verification is needed.
The rigorous procedures of the grounded theory method ensure
that the translated findings are already grounded in the source
text.
When translation is undertaken by a bilingual grounded
theorist, cumbersome verification of such translation can become
interference in research. Target language meanings as a result of
translation, which is intertwined with the conceptual analysis,
are inherently emergent. Trusting the emergence of conceptual
meanings expressed across the languages is an important
adjustment to the act of translation, when it is performed within
the grounded theory method.
Fidelity and Transparency: Shifting the Balance of
Translation Criteria in GT
The properties of translation function become transformed,
when the researcher is using the strategy of its intertwining with
conceptual analysis. Translation loses its fixation on verification
and its descriptive properties. Instead, it becomes focused on
conceptualization and emergence. However, the principles of
conceptual equivalency and constant comparison remain
indispensable. To explore the criteria of good translation, when
its properties are adjusted in such way, I refer to the classic
criteria of translation quality that have naturally “earned its
way” (Glaser, 1978) into my reflection: fidelity and transparency
(Tianmin, 2006). Fidelity is defined as the extent to which the
translation accurately renders the meaning of the source text,
without adding to it or subtracting from it, and without
intensifying or weakening any part of the meaning. Transparency
is the criterion of the extent to which the translation conforms to
the target language’s grammatical, syntactic, and idiomatic
conventions. The text is expected to appear fluent and natural to
the native speakers of the target language.
According to my self-observations, the criteria of fidelity and
transparency in translation used in grounded theory remain
relevant, but the adherence to these criteria is balanced in a way
that is different from conventional translation. As we shall see, in
grounded theory it is necessary to adjust translation criteria, so
the balance is tilted towards the favouring of fidelity over
transparency.
The criteria of fidelity and transparency can be conflicting,
because the meaning can be expressed not only through words,
but also through language structures, style, grammatical form,
underlying context, and culturally rooted metaphors (Lakoff &
Johnson, 1980; Al-Hasnawi, 2007; Whorf, 1956). Maintaining
fidelity is practically impossible in a transparent translation that
conforms to the target language structural elements. Ideal
transparency can be achieved only at the expense of changing
subtle nuances embedded in the natural structures of the source,
which means that “infidelity is built in translation” (Tianmin,
2006). Therefore, the translator traditionally is required to find a
working compromise between fidelity and transparency.
However, in the grounded theory analysis the working
compromise is not acceptable. Fidelity cannot be compromised,
because the theory needs to be grounded in the unchanged source
language data. Transparency cannot be stressed at the expense of
the full equivalency of meanings.
In case of a bilingual speaker conducting both translation
and research, the conflict is almost irrelevant. When I skip the
stage of creating a fluent target language text, I can avoid the
necessity of presenting wordings that are transparent (smoothly
flowing and natural) for a potential reader. By doing so, I have
the advantage of remaining faithful to the source while analysing
my data and bringing it to the higher, abstract level of theoretical
conceptualization. Other bilingual researchers suggested similar
strategies. For example, Temple (2006) recommended allowing
“foreignization of the text” (p. 10; this concept was first used by
Venuti, 1995). The “foreignized” text of translation is not
necessarily fluid in style (which assumes a compromised
transparency), but effectively preserves the meaning of the
original (the maximized fidelity). The strategy of adhering to a
relevant (minimum intelligible) level of transparency and the
maximal level of fidelity can be used effectively in the context of
intertwining translation and conceptual analysis.
In grounded theory procedures, the process of foreignization
has an analogue: the “freedom afforded in memo writing” (Glaser,
1978, p. 85). When we write memos, grammar and correctness of
the text are irrelevant, because the priority is given to recording
ideas, “getting them out.” In a similar way, when a bilingual
theorist works to convey the meaning of the source text in
translation, she does not have to follow the criteria of full
transparency, but rather should focus on the conceptual meaning
of the message.
The bilingual theorist can achieve the transparency goal on
the conceptual level, rather than on the descriptive level of
creating a full, accurately transparent data text in the target
language. The intertwining of translation and conceptual analysis
is a naturally occurring and instrumental strategy for balancing
the opposite trends of achieving fidelity and relevant
transparency in the cross-language generation of a theory.
A Theory Back-Translated: Linguistic Fit, Relevance,
Workability, and Modifiability
In the final stages of a grounded theory project, when the
cross-language researcher engages in theoretical writing, the
work concentrates within the target language (English). The
focus on writing in English is natural, because the intended
audience of theory presentation, in most academic cases, is the
English-language audience. How can the researcher make sure
that linguistically, in relation to the source language, the
resulting theory still has sufficient fit, relevance, workability, and
modifiability? For a cross-language theory, how can the general
judging criteria apply to its cross-language nature and linguistic
quality?
In my research, I experimented with some patterns of the
application of grounded theory method to cross-language setting.
One of such experiments was back-translating my theory. The
following conclusions emerged from my experience of theoretical
writing, and in particular, from the incident of back-translation.
At the later stages of my work, I found myself impelled, by
the emerging need of cross-language comparison, to translate a
summary of my theory back into Russian. I intended my Russian
text to be highly transparent, to enable me to present it to
Russian-speakers (partially because the participants in my study
were curious and kept asking me about the results). Fidelity to
my written theoretical text was also very important, to convey the
meaning of the theory. My criteria of transparency ascended to
the level of conceptual constructs, rather than to the data details.
A few questions emerged. Would my theory read as smoothly in
Russian as in English, would the theoretical concepts fit and
work in both languages? Would my theory naturally fit into the
source language structure, lexicon, and grammar, or would
Russian become a foreign language for the concepts initially born
in it?
According to my observations, the reverse translation of a
written English text into the source language can help the
researcher evaluate how easy it is to achieve the fidelity and
transparency in the source-language version of the presented
ideas. I considered this ease as an indication of the first classical
requisite property of grounded theory, the fit, as it appeared
relevant in its linguistic aspect. If the linguistic fit is achieved,
the categories, expressed in back translation, smoothly fit the
source-language conceptual meanings. Such fit is a direct
outcome of the systematic application of the strategy of
intertwining, from the early stages of the research. Because
throughout the investigation, translation was intertwined with
conceptual analysis, the resulting categories fit the natural
structure of the source language and cultural context, smoothly
and automatically flowing into the back translation.
The linguistic workability and relevance, in my experience,
mean that in both languages, the theory should read smoothly
and sound right. It should be understood by monolingual research
participants, as well as by the English-speaking audience. The
theoretical concepts should sound relevant, have the “grab,” and
work in both languages equally. Source-language words have to
“make sense” to monolingual native speakers, without lengthy
explanations or footnotes explaining English-born wordings.
Linguistic workability and relevance are also emergent
properties, in that they are ensured throughout the analysis,
from its initial stages, in which the linguistic expressions of
concepts have initially emerged. Workability and relevance are
guaranteed by the systematic intertwining of translation with
conceptual analysis. For example, the researcher should abstain
from choosing English codes that use too specific, language-based
imagery, or peculiar idiomatic forms that would sound awkward
in the source language. Such issues have to be built into the
analysis, through using the strategy of intertwining and its
properties. It might be too late or too difficult to adjust the
possible linguistic flaws, if the researcher only discovers them in
the final stages of theoretical writing, with an isolated attempt of
back translation. Fit, work, and relevance have to be ensured
through the systematic procedures specific to the cross-language
setting.
If the linguistic indication of modifiability is to be found in
cross-language research, I would interpret it as a property that
pertains to making the texts transparent in both languages, with
its conceptual nature remaining equivalent, while the writer is
able to modify and adjust some grammatical structures and
language peculiarities. For example, there is no such
grammatical form as gerund in the Russian language. Therefore,
some of my one-word English signifiers of categories had to be
creatively transformed into corresponding Russian grammatical
constructs, at the time of their emergence. At times, these
modifications involved creative translation into Russian, and in
other cases, it required changing English constructs, to achieve
better equivalency. It often happened that sociological constructs
born in English needed modification for their expression in the
source language. Conversely, source-language in vivo categories
needed to be evaluated or modified for their fit in translation.
I believe that all experienced translators are familiar with
these or similar challenges. However, in the context of grounded
theory, these challenges apply not only to the pure act of
translation, but also to the emergent analytical ideas. Once
again, the strategy of intertwining continues to be relevant. It is
through the systematic use of intertwining that the bilingual
grounded theorist can achieve both linguistic and conceptual fit,
workability, relevance, and modifiability of the discovered theory.
Discussion
The hypothetical patterns presented in this article are based
on my self-observations in conducting cross-language grounded
theory research. Therefore, presented hypotheses are limited to
one person’s experiences and relate to one particular research
situation.
In practice, the reflections presented in this article most
closely relate to the work of bilingual researchers. The dual
position of a bilingual researcher in cross-language grounded
theory is unique because of the binocular conceptual vision, the
intimate knowledge of both languages and cultural discourses,
and the general familiarity with cross-language experiences.
However, employing a bilingual analyst is not the only
possible strategy of undertaking cross-language research. For
example, effective research is often carried out by a team of
people who are not necessarily fluent in two (or more) languages
relevant to the area of study (Glicksman & Van Haitsma, 2002;
Nguyen et al., 2008). In such cases, complete and seamless
intertwining is not possible, because the analysts receive second
hand, translated data, and the translator mediates all
communications.
The concept of intertwining and its properties might be
relevant to research situations other than the work of a sole
bilingual theorist, but further data collection and research are
needed to understand whether my conclusions can be extended or
generalized to the work of other researchers in other settings. My
concepts might have to be modified and further integrated to fit a
broad range of research situations. Such analysis is beyond the
scope of this article.
Conclusion
Cross-language grounded theory research involves processes
that spread beyond the traditionally understood translation, and
can also include altering the conventional analytical techniques of
the theorist. In my sole work as a bilingual theorist, the strategy
of intertwining translation and analysis proved to be effective and
natural. I performed myself all the functions associated with my
project, without involving a translator, interpreter, or transcriber,
and found it most appropriate. Using the advantage of intimate
sensitivity to the languages that I know well, I could work across
languages on my own, following Glaser’s (1978) notion that
“grounded theory is a do-it-yourself methodology” (p. 116). The
functions of translation and conceptual analysis, thus, became
intertwined in my research, technically and strategically.
According to Glaser & Holton (2004), “Classic GT is simply a
set of integrated conceptual hypotheses systematically generated
to produce an inductive theory about a substantive area.” (para.
7). All the classical principles of grounded theory apply to the
cross-language strategies involved in theory development.
Translation adds complexity to data analysis, but does not change
any essential properties of the method. In fact, in my reflections I
have not discovered anything that has not been conceptualized as
part of the fundamental principles of the grounded theory
method. Rather, I made an attempt to explain, primarily for the
purposes of my own or similar research, some aspects of what is
going on in the substantial situation of cross-language data
analysis with the purpose of discovering a theory. I conclude that
the essential elements of grounded theory, such as emergence,
conceptualization, and constant comparison, naturally fit into the
cross-language setting, in which translation and analysis are
intertwined.
Author
Svetlana Shklarov, MD, RSW, PhD Candidate
Interdisciplinary Graduate Studies Program
University of Calgary, AB, Canada
Email: shklar@shaw.ca
References
Al-Hasnawi, A. R. (2007). Cognitive approach to translating
metaphors. Translation Journal, 11(3). Retrieved on
September 3, 2007 from
http://accurapid.com/journal/41metaphor.htm
Altman, I. (2005). Мемориализация Холокоста в России: история,
современность, перспективы [Memorializing the Holocaust
in Russia: History, modern times, perspectives]
[Electronic Version]. Неприкосновенный запас, 2-3.
Retrieved April 15, 2006, from
http://magazines.russ.ru/nz/2005/2/alt28-pr.html
Friedrich, H. (1992). On the art of translation. In J. Biguenet &
R. Schulte (Eds.), Theories of Translation: An Anthology
of Essays from Dryden to Derrida. Chicago, IL: The
University of Chicago Press, p. 11-16.
Hunt, S. M., & Bhopal, R. (2004). Self report in clinical and
epidemiological studies with non-English speakers: the
challenge of language and culture. Journal of
Epidemiology and Community Health, 58, 618-622.
Retrieved September 15, 2005, from
http://jech.bmjjournals.com/cgi/content/full/58/7/618
Houbert, F. (1998). Translation as a communication process
[Electronic Version]. Translation Journal, 2(3). Retrieved
on September 3, 2007 from
http://accurapid.com/journal/05theory.htm
Gergen, K. J. (1999). An invitation to social construction.
Thousand Oaks, CA: Sage.
Glaser, B. G. (1978). Theoretical sensitivity: Advances in the
methodology of grounded theory. Mill Valley, CA:
Sociology Press.
Glaser, B. G. (2002a). Conceptualization: On theory and
theorizing using grounded theory [Electronic Version].
International Journal of Qualitative Methods, 1(2), Article
3. Retrieved October 1, 2005, from
http://www.ualberta.ca/~ijqm/
Glaser, B. G. (2002b). Constructivist grounded theory? [Electronic
Version]. Forum: Qualitative Social Research, 3(3).
Retrieved October 1, 2005, from http://www.qualitativeresearch.
net/fqs-texte/3-02/3-02glaser-e.htm
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded
theory: Strategies for qualitative research. Chicago, IL:
Aldine.
Glaser, B. G., & Holton, J. (2004). Remodeling grounded theory
[Electronic Version]. Forum: Qualitative Social Research,
5(2), Article 4. Retrieved July 24, 2005, from
http://www.qualitative-research.net/fqs-texte/2-04/2-
04glaser-e.htm
Glicksman, A., & Van Haitsma, K. (2002). The social context of
adaptation to traumatic events: Soviet Jews and the
Holocaust. Journal of Clinical Geropsychology, 8(8), 227-
237.
Lakoff, G. & Johnson, M. (1980). Metaphors we live by. Chicago,
IL: University of Chicago Press.
Mc Laughlin, F., & Sall, T. S. (2001). The give and take of
fieldwork: Noun classes and other concerns in Fatick,
Senegal. In P. Newman & M. Ratliff (Eds.), Linguistic
fieldwork (pp. 189-210). Cambridge, UK: Cambridge
University Press.
Muller, V. K. (1990). English-Russian dictionary. Moscow:
Russky Yazyk.
Muula, A. M. (2005). What should HIV/AIDS be called in Malawi?
[Electronic version]. Nursing Ethics, 12(2), 187-192.
Nguyen, G. T, Barg, F. K., Armstrong, K., Holmes, J. H., &
Hornik, R. C. (2008). Cancer and communication in the
health care setting: Experiences of older Vietnamese
immigrants, a qualitative study. Journal of General
Internal Medicine, 23(1), 45-50.
Pearce, J. (1997). Последствия злоупотребления и пренебрежения
ребенком для его развития [Consequences of abuse and
neglect in child development]. In N. K. Acanova (Ed.),
Руководство по предупреждению насилия над детьми (pp.
91-106). Moscow, Russia: Владос.
Roberts, G., Kent, B., Prys, D., & Lewis, R. (2003). Describing
chronic pain: towards bilingual practice. International
Journal of Nursing Studies, 40(8), 889-902.
Schopenhauer, A. (1992). On language and words (P.
Mollenhauer, Trans.). In J. Biguenet & R. Schulte (Eds.),
Theories of Translation: An Anthology of Essays from
Dryden to Derrida (pp. 32-35). Chicago, IL: The
University of Chicago Press. (Original work published
1800)
Stebbins, R. A. (2001). Exploratory research in the social sciences
(Vol. 48). Thousand Oaks, CA: Sage.
Temple, B., & Edwards, R. (2002). Interpreters/translators and
cross-language research:
Reflexivity and border crossings. International Journal of
Qualitative Methods, 1 (2), Article 1. Retrieved October
15, 2005, from http://www.ualberta.ca/~ijqm/
Temple, B. (2006). Representation across languages: Biographical
sociology meets translation and interpretation studies.
Qualitative Sociology Review, 2(1), 7-21.
Retrieved on June 3, 2007 from
http://www.qualitativesociologyreview.org/ENG/Volume3/
QSR_2_1_Temple.pdf
Tianmin, J. (2006). Translation in context [Electronic Version].
Translation Journal, 10(2). Retrieved on September 3,
2007 from http://accurapid.com/journal/41metaphor.htm
Tsai, J. H.-C., Choe, J. H., Lim, J. M. C., Acorda, E., Chan, N. L.,
Taylor, V., & Tu, S.-P. (2004). Developing culturally
competent health knowledge: Issues of data analysis of
cross-cultural, cross-language qualitative research.
International Journal of Qualitative Methods, 3(4), Article
2. Retrieved October 28, 2005 from
http://www.ualberta.ca/~iiqm/backissues/3_4/html/tsai.ht
ml
Valent, P. (1998). Resilience in child survivors of the Holocaust:
Toward the concept of resilience. Psychoanalytic Review,
29, 517-535.
Venuti, L. (1995). The translator’s invisibility: A history of
translation. London, UK: Routledge.
Whorf, B. (1956). Language, Thought & Reality. Cambridge, MA:
MIT Press.
Yu, D. S. F., Lee, D. T. F., & Woo, J (2004). Issues and challenges
of instrument translation [Electronic version]. Western
Journal of Nursing Research, 26(3), 307-320.
