Research Article
Open Linguistics 2017; 3: 1–17
Open Access
Daniel Rodríguez-Vergara*
A Systemic Functional Approach to the Passive
Voice in English into Spanish Translation:
Thematic Development in a Medical Research
Article
DOI 10.1515/opli-2017-0001
Received May 11, 2016; accepted November 28, 2016
Abstract: The purpose of this study was to explore, from the perspective of Systemic Functional Grammar,
how passive clauses in a medical research article were translated into Spanish, specifically if they were
kept in the passive voice, were changed into the active voice, or were turned into some other structure, and
if voice change in the translated version affected the original thematic development. The medical paper
chosen for this study was originally written in English and published in an Anglophone journal; it was then
translated into Spanish and published in a Mexican journal. Both the original and the translated article
were analyzed in terms of Theme and Rheme; all of the instances of passive and active voice were quantified
and compared. The results show that in some cases the original thematic patterns were modified in the
translation due to the use of the reflexive passive in Spanish, which results in the fronting of the verb in
the sentences, thereby causing a change of Themes in the paragraphs with respect to the original structure.
This study contributes to our understanding of the function of passive constructions in English and Spanish
and its relationship with thematic progression.
Keywords: Theme-Rheme, research writing, systemic functional grammar
1 Introduction
Currently, writing advice for research articles regarding personal and impersonal grammatical forms, such
as the passive voice or first person constructions, is a highly debatable issue (Bennett 2009, Salazar et al.
2013). For instance, Hyland (2001) reports that some style manuals and textbooks advise using the first
person with caution, or even abstaining from using it altogether. Nevertheless, Hyland (2001) found that,
in a corpus of 240 Research Articles (henceforth RAs) from eight different disciplines including both the
so-called ‘hard’ and ‘soft’ fields, there was an average of 22.7 first person pronouns per paper.
The use of the passive voice in RAs is controversial as well. In order for an RA to be clear and precise
for the reader, some scientific writing experts recommend either not overusing passive clauses (Canter &
Fairbairn 2006), or using active clauses unless there is good reason not to (Matthews & Matthews 2014)
in order to avoid obscurity and dullness. Other experts advise not going beyond 10% of passive voice
use (Amdur et al. 2010), and some others recommend even abandoning the practice altogether (Day &
Sakaduski 2011). However, some studies have shown that there is a trend for anglophone RA writers from
different fields of study to choose the passive form for around 30% of their clauses (Martínez 2001, Alvin
*Corresponding author: Daniel Rodríguez-Vergara, Applied Linguistics Department, Foreign Language Center, National
Autonomous University of Mexico, Mexico City, Mexico, E-mail: [email protected]
© 2017 Daniel Rodríguez-Vergara, published by De Gruyter Open.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.
Unauthenticated
Download Date | 6/15/17 1:17 AM
2 D. Rodríguez-Vergara
2014), and that the percentages go far beyond when it comes to the Methods sections (Swales 1990, Nwogu
1997, Biber & Conrad 2009).
Medical RAs have received particular attention since it is in this area that authors are claimed to make
excessive use of the passive voice (Taylor 2005, Goodman & Edwards 2006, Amdur et al. 2010). According
to Amdur et al. (2010: 102), medical writers are likely to use passive constructions with a high frequency
because they might want “to reflect objectivity, to avoid first person pronouns, to appear scholarly and
sophisticated, to avoid responsibility, and to conform to established writing style.” Indeed, objectivity and
impersonality play an important role in the use of the passive voice; however, few studies have attempted to
explore how the choice between active and passive voice might affect the flow of discourse in medical RAs,
or how passive clauses in an Anglophone paper are rendered in another language.
It has been suggested that word order in English is more grammatically governed than in Spanish, the
latter being more pragmatically controlled (Gómez & García 2014). In addition, unlike English, Spanish
is a pro-drop language, which entails that pronouns may be omitted when they are deducible from verb
morphology. For those reasons, it is important to investigate the extent to which those language-specific
differences have an effect on thematic choices at the level of both clause and discourse. The purpose of
this study was to find out how passive clauses in a medical RA were translated from English into Spanish:
whether they were kept in the passive voice, changed into the active voice or into some other structure,
or if a voice change in the translated version affected the original thematic development. Although the
analysis of a minimal sample of texts will not allow for conclusive generalizations, our intention will be to
shed light on the relationship between grammatical constructions and flow of discourse, thereby providing
guidelines for future large-scale studies. For the purposes set out in this study, we used the Systemic
Functional Linguistics (henceforth SFL) framework (Halliday 1985, 1994), which will be described in the
following section.
2 Theoretical background
Although several studies have contributed to the description of medical writing (cf. Salager-Meyer 1994,
Gotti & Salager-Meyer 2006, Atkinson & Valle 2012), few have approached it from the perspective of SFL.
SFL is a social semiotic theory, developed mainly by Halliday (1985, 1994), that emphasizes the meaningmaking property of language; its main interest is to explain the relationship between text and context.
It is functional, as opposed to formal (Matthiessen et al. 2010), in the sense that language is described
principally in terms of how it is used in cultures and for which purposes, and not in terms of how language
knowledge is stored in the speaker’s mind. Additionally, it is systemic, as opposed to structural, in that it
gives priority to the notion of ‘paradigm’ over that of ‘syntagm’, i.e., it is concerned mostly with the linguistic
choices that are available to speakers when they create and communicate meaning.
SFL postulates three types of meaning (or ‘metafunctions’) that language has evolved to perform
through which its organization is reflected: ideational, interpersonal, and textual. Ideational meaning
is used to construe experience by representing, linguistically, what goes on in the world; interpersonal
meaning is used to interact with other speakers and to take a stance towards people and things; textual
meaning is used to create discourse flow by presenting ideational and interpersonal meaning as a coherent
whole.
In the present study we will focus on textual meaning, which is manifested in the clause as a periodic
structure that includes Theme and Rheme. According to Halliday (1994: 37): “The Theme is the element
which serves as the point of departure of the message; it is that with which the clause is concerned. The
remainder of the message, the part in which the Theme is developed, is called in Prague School terminology
the Rheme.” From the textual point of view, clauses are considered messages, a quantum of information
where Theme is that which appears in first position. Halliday (1994) maintains that there are three types of
Theme, each of which reflects one of the three metafunctions referred to above. First, ‘experiential Theme’
refers to any of the configurative functions within the clause: Participant, Circumstance, and Process,
respectively, as shown in (1) through (3).
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 3
(1) Intravenous furosemide [Exp. T] is commonly administered to patients with acute heart failure to relieve
pulmonary congestion through diuresis.
(2) At the end of each treatment period [Exp. T], patients were studied using forearm venous occlusion
plethysmography.
(3) Fue rechazada [Exp. T] la participación de pacientes con antecedentes recientes de infarto
miocárdico o accidente cerebrovascular, y con antecedentes de trastornos clínicamente importantes
de origen hepático, gastrointestinal, hematológico, endocrinológico, renal, pulmonar o neurológico.
[Patients were not studied if they had a history of a recent myocardial infarction or cerebrovascular
event, or a history of a clinically important hepatic, gastrointestinal, hematologic, endocrine, renal,
pulmonary or neurologic disorder.]
As in all of the examples of this paper, sentences (1-3) are taken from our parallel texts. In (1) the Theme is
Intravenous furosemide, which is the main Participant of the material Process realized by the verbal group
is administered. In (2) the Theme is At the end of each treatment period, which is a time Circumstance in
the mental clause that has were studied as its Process (verbal group). Finally, a Process can also be in
thematic position; in English, thematic Processes are mainly found in imperative clauses, but in Spanish
they are commonly used in declarative clauses such as (3). This happens because the inflectional verbal
morphology of Spanish indicates grammatical person (first, second, third) and number (singular, plural).
In other words, Spanish is a pro-drop language.
The other types of Theme are interpersonal and textual. As opposed to experiential Theme, interpersonal
and textual Themes are not obligatory, i.e., every clause has an experiential Theme but only some have
interpersonal and textual ones:
(4) It is believed [Int. T], therefore [Tex. T], that venodilation [Exp. T] is brought about through the release
of local prostaglandins.
In (4) there are three Themes. First, It is believed is a projecting mental clause that functions as an
interpersonal Theme. In declarative mood, interpersonal Themes typically codify probability, possibility
and frequency not only through projecting clauses but also through modal Adjuncts (probably, maybe,
often). Second, therefore is a conjunctive Adjunct that functions as a textual Theme. Textual Themes are
cohesive elements that relate clauses to previous discourse. Finally, the clause obligatorily contains an
experiential Theme, which in this case is venodilation.
Although so far we have characterized Theme as an element of the clause, the choice of Theme (as well
as the choice of grammatical voice) plays a role in the thematic development of larger units of discourse. As
Eggins (2004: 342) puts it, “a very significant contribution that Theme makes to the cohesion and coherence
of a text has to do with how thematic elements succeed each other.” Following Daneš (1974) and Fries (1983),
Eggins (2004) recognizes three main types of method of Theme development. First, ‘Theme reiteration’,
through which the same Theme is repeated in a fragment of discourse:
(5a) Nine of these patients [T] had participated in the furosemide arm of the study; the method of diagnosis
and the severity of heart failure were the same as described previously.
(5b) Patients [T] were not studied if they had a history of a recent myocardial infarction or cerebrovascular
event, or a history of a clinically important hepatic, gastrointestinal, hematologic, endocrine, renal, pulmonary
or neurologic disorder.
(5c) Patients with uncontrolled or insulin-dependent diabetes mellitus or uncontrolled hypertension
[T] were also excluded.
Unauthenticated
Download Date | 6/15/17 1:17 AM
4 D. Rodríguez-Vergara
In these three sentences the topical element always has to do with ‘patients’. Although the noun groups that
realize the Themes in (5a-5c) are not identical, the head of the noun groups is always patients. The second
method of development is called ‘the zig-zag pattern’, in which the elements of Rheme become elements of
Theme in subsequent clauses:
(6a) They reported that secretion [R] of prostacyclin, measured indirectly by 6-ketoPGF51∞ production,
increased within 5 min of exposure of the culture to furosemide.
(6b) Secretion [T] was maximal at ~15 min 2 and was maintained for 20 to 30 min [R].
(6c) This time scale for the release of prostacyclin [T] fits well with our finding, as well as that of others, that
the venous dilator effect of furosemide occurs within 15 to 20 min of drug administration in vivo.
As can be observed, here the rhematic element secretion in (6a) becomes the Theme of (6b). Similarly, the
rhematic element 20 to 30 min in (6b) becomes the Theme in (6c), albeit with a different wording (This
time scale for the release of prostacyclin). The third method of Theme development is ‘the multiple-Rheme
pattern’, where the Rheme of a clause is fragmented into a number of subsequent Themes:
(7a) Baseline FVC did not differ between treatments [R].
(7b) In the compound symmetry mixed model [T], the effects of period (p = 0.53) and time (p = 0.84) and
the interaction of time with period (p = 0.34) were nonsignificant.
(7c) In the placebo period [T], the mean (95% CI) response to furosemide over 20 min was a 2.2% (-0.9% to
5.2%) increase in FVC.
Here both the Theme in (7b) and that of (7c) refer to the treatments mentioned in the rhematic part of (7a).
There is thus a fragmentation of the rhematic element treatments into multiple subsequent Themes.
Using one of the cohesive methods of development exemplified above makes the information in a
text flow coherently, and one important consequence is that readers will follow with more ease the ideas
presented. However, because of language-specific phenomena (such as the pro-drop nature of Spanish),
when rendering one text into another language, methods of development presented in a source text (ST)
might be affected by all the syntactic adjustments that must be made in a translated text (TT). For that
reason, we decided to investigate the effects of grammatical voice shift in the thematic structure of an
English-Spanish translated article. We present the study’s methodology below.
3 Methods
3.1 Selection of parallel texts
For the study, we interviewed an expert informant, who provided us with a medical paper used as a required
reading for undergraduate students of the School of Medicine of the Benemérita Universidad Autónoma de
Puebla, located in the city of Puebla, east-central Mexico. The informant was a professor of the School
who, at the time, was teaching Pharmacology to senior students. The key reading was a Spanish-translated
article from the Mexican Spanish edition of the Journal of the American College of Cardiology1. We thus
looked for its English counterpart in its corresponding Anglophone journal (refer to the bibliographic data
1 Unfortunately, since the Mexican journal does not include information about the translator of the article, we were not able
to provide her/his language background (command/training in both Spanish and English) here. This information could have
helped us understand some of the language choices, and decide whether language influence played a role in them.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 5
in the Appendix). The article is concerned with how aspirin inhibits the acute venodilator response to
furosemide in patients with chronic heart failure, and contains the typical ‘IMRD’ (Introduction, Methods,
Results, Discussion) (Swales 1990) structure of a medical research paper (Salager-Meyer 1994, Nwogu 1997).
For the analysis, we only included the main body parts (Introduction, Methods, Results and Discussion)
and excluded the abstract, the title, headings, footnotes, tables, figures, captions, and references. In the ST,
the IMRD sections contained 2,508 words, 160 non-embedded clauses, and 105 sentences, whereas in the
TT, there were 3,100 words, 155 non-embedded clauses, and 104 sentences.
Despite the limited sample, we intended to carry out a ‘comprehensive analysis’ (Matthiessen 2014) in
order to study text as process (Halliday, 1985), i.e., as a linear and syntagmatic actual instance of language
use. As Matthiessen puts it:
Analysis always involves choices on the part of analysts; and in making such analytic choices, analysts must weight
different objectives against one another: analysts are always faced with trade-offs, e.g. between the coverage of the analysis
of text and the volume of text analyzed. Such trade-offs are particularly important to consider in relation to ADA, Appliable
Discourse Analysis, because the choices analysts make obviously affect the appliability of the results of the analysis.
(Matthiessen 2014: 187-188)
Thus, the aim of our analysis is to reveal what Matthiessen (2006: 105) calls ‘logogenetic patterns’ in our
texts, that is, “instantial patterns in the creation of meaning through the instantiation of the system, and
also the accumulation of patterns of texts as product.” We shall describe our analysis procedures below.
3.2 Analysis
As suggested by Fries (1994), our unit of analysis was the T-unit, which is an independent clause that
includes the clauses that depend on it. Thus, when a coordinated conjunction was used to link two clauses,
they were taken as two units of analysis; but when a subordinating conjunction was used to bind two
clauses, they were taken as a single unit. When the dependent clause follows the main clause, the former
is simply considered part of the Rheme of the latter; but when the dependent clause precedes a main one,
then in the systemic functional tradition the former is treated as the Theme of the unit (Halliday 1994).
However, given that our purpose was to trace the corresponding thematic development in the parallel
Spanish text, we found it useful to consider the first experiential element of the main clause as Theme. For
example, consider the following sentences:
(8a) All patients [T] gave written, informed consent.
(8b) To study the effects of aspirin on the response to furosemide, 11 patients [T] with chronic heart failure
(CHF) were studied.
(8c) Patients [T] had New York Heart Association class II or III CHF caused by left ventricular systolic
dysfunction, confirmed by a left ventricular ejection fraction <40% on echocardiography.
Note that sentence (8b) starts with the dependent clause. According to Halliday (1994), the Theme of the
sentence should be the entire dependent clause, whereas the Rheme should be the entire main clause.
However, as noticed by Thompson (1996), a more delicate analysis would suggest that the main clause has
its own Theme, which, in the case of (8b), is 11 patients. If we treat this noun group as the Theme of unit
(8b), then the result is a Theme-reiteration method of development where (8a-8c) share the same thematic
element (patients).
Because grammatical voice is used to switch the order of several experiential elements of the clause,
our analysis was limited to experiential Themes, excluding interpersonal and textual ones. Thus, having
segmented the texts into units, we marked all experiential Themes and registered whether every T-unit was
passive or active. In the case of Spanish, the two passive types, periphrastic and reflexive, were considered.
Unauthenticated
Download Date | 6/15/17 1:17 AM
6 D. Rodríguez-Vergara
The former is similar to the English passive construction in that it is formed with the auxiliary verb ser (be)
plus the main verb in the past participle:
(9) Esta respuesta venodilatadora frente a la furosemida es inhibida por el inhibidor de la ciclooxigenasa
indometacina. [This venodilator response to furosemide is inhibited by the cyclooxygenase inhibitor
indomethacin.]
In contrast, the reflexive passive construction is formed with the reflexive clitic se plus a verb that is
in agreement with the subject/patient (see Hidalgo (1994) for a comprehensive account of reflexive
constructions in Spanish, and also Arús (2006), who discusses multiple uses of se in Spanish from a
systemic functional perspective):
(10) El estudio se realizó con la aprobación del Local Ethics Committee del West Glasgow University Hospital
NHS Trust. [The study was performed with the approval of the Local Ethics Committee of the West Glasgow
University Hospital NHS Trust.]
After coding all Themes in the texts and registering if they were part of active or passive clauses in English,
or in the case of Spanish, active, periphrastic passive, or reflexive passive ones, we used RSTTool version 3.45
(O’Donnell 2004) in order to build rhetorical diagrams that showed the thematic development of the texts’
sections. Although the software was originally created to analyze texts according to Rhetorical Structure Theory
(Mann and Thompson 1988), we found it useful to create thematic pattern diagrams. For the Introduction and
Results categories, only one diagram per section was built, but, because of their length, for Methods and
Discussion, one diagram was built for each of their subsections (four for Methods and six for Discussion). To
illustrate this, in Figure 1 we present the rhetorical diagram of the Introduction in English.
Figure 1. Rhetorical diagram of the English Introduction
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 7
Worth noting is that because of the nature of the study, of the three methods of development mentioned
by Eggins (2004), we only took into account Theme reiteration, since it is the strategy that was used the
most by the author of the ST, and also the one where potential disruptions in the Spanish translation can
be observed most easily (although, as we will see later, a new method of cohesive development was found).
The Introduction in Figure 1 has eight T-units; all of the Themes are marked in bold. In the figure there are
two graphical means to associate units with their corresponding previous referents, i.e., diagonal straight
lines and a curved arrow. The diagonal lines indicate that the associated units share the same Theme, e.g.,
in units 2 and 3 the Themes are lexically different but referentially the same (symptomatic relief and This
effect); and in units 4 and 5 the Themes are This venodilator response to furosemide and venodilation, where
there is a shared meaning (venodilator/venodilation). In this respect, when there was at least one shared
meaning expressed through one word in two units, the Themes were considered to be reiterated. The other
means of unit association in the figure, the arrow, indicates that the Theme is not reiterated.
After generating all the rhetorical diagrams, we compared and contrasted Themes of the ST with Themes
of the TT to detect if there were differences in their thematic structure, and, when there were differences,
we sought to determine whether or not it was due to grammatical voice shift. Since each of the sections of
the articles showed important singularities, we decided to present the results separately for every section.
4 Results and discussion
4.1 Introduction
This is the shortest section, with eight T-units in both ST and TT (see Figures 1 and 2). Of the eight units, six
are in the passive voice in the ST, while in the TT, only two are in the passive (one is periphrastic and one
reflexive), which means that the majority of the passive units in English were made active in Spanish. For
those changes, the original Theme reiteration pattern was disrupted, and a new one was created (compare
Figures 1 and 2).
Figure 2. Theme pattern in the Spanish Introduction
Unauthenticated
Download Date | 6/15/17 1:17 AM
8 D. Rodríguez-Vergara
Firstly, the two Theme reiteration patterns found in units 2-3 and 4-5 of the ST are disrupted because of
the voice activization of units 3 and 5 in the TT (here we follow Teich’s (1999: 111) terminology of ‘activization’
and ‘passivization’ within the voice system). As a consequence of activization, the ST Theme of unit 3 (This
effect) is changed into Suponemos (we suppose) in the TT, and the ST Theme of unit 5 (venodilation) is
changed into suponemos in the TT. As can be observed, the TT shows a preference for the active version of
the clauses using the first person plural form of the verbs. Consequently, a new Theme reiteration pattern is
formed with Themes where first person constructions are found (Suponemos (we suppose) in units 3 and 5,
nuestra (our) in unit 6, and hemos (we have) in unit 8).
In general, the ST is characterized for an impersonal style that is achieved through the use of passive voice
in the majority of the sentences. In contrast, the TT has a more personal character, which is achieved through
the use of first person constructions. Thus, although in the process of translation the textual meaning of the
Introduction has been re-worked so that it shows a new but still fluent information pattern, the interpersonal
meaning has been changed. The Spanish translation now shows a clearer authorial visibility.
Authorial presence has already been investigated by Perales and Swales (2011) in the context of research
article abstracts translated from Spanish into English and vice versa. Although the abstracts they analyzed
belong to a different discipline (applied linguistics), one of their findings was that, in the Spanish abstracts,
there was greater authorial presence with respect to their English counterparts. This suggests that the use
of first person constructions in our TT might reflect a general trend in Spanish rather than the translator’s
individual style.
4.2 Methods
This is the largest section of the article and is divided into four subsections: 1) Patients, 2) Study design, 3)
Venuous occlusion plethysmography protocol, and 4) Data analysis. The ST has 50 T-units, 36 of which are
in the passive voice (72%); the TT has 49 T-units, 33 of which are in the passive voice (67%), and of those 33,
21 are reflexive and 12 are periphrastic passives.
The first subsection, Patients, shows a Theme reiteration pattern that includes eight units (see Figure 3,
where we show the relevant part of the diagram to only include Themes for each unit). As expected, because
this is the part where the subjects of study are described, the reiterated Theme in the pattern is patients.
When we compare this diagram with its TT counterpart, we see that the latter contains the same pattern;
however, the TT pattern has only five units where pacientes (patients) is part of the Theme. This time, the
thematic change in the three other units does not have to do with a change from passive into active voice,
since they were all maintained passive (periphrastic). The units changed Theme in the TT because, instead
of placing the grammatical subjects at the beginning of the sentence, the translator opted for placing the
verb in initial position:
(12a) ST: 11 patients (10 men and 1 woman; mean [±SD] age 70 ± 7 years) with chronic heart failure (CHF)
were studied.
(12b) TT: fueron estudiados 11 pacientes (10 varones y una mujer; edad media [± DE], 70 ± 7 años) con
insuficiencia cardíaca crónica (ICC).
(13a) ST: Patients were not studied if they had a history of a recent myocardial infarction or cerebrovascular
event...
(13b) TT: Fue rechazada la participación de pacientes con antecedentes recientes de infarto miocárdico o
accidente cerebrovascular...
(14a) ST: Patients with uncontrolled or insulin-dependent diabetes mellitus or uncontrolled hypertension
were also excluded.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 9
(14b) TT: También fueron excluidos los pacientes con diabetes mellitus no controlada o en tratamiento con
insulina, así como aquellos con hipertensión no controlada.
Figure 3. Theme pattern in the first subsection of Methods (elliptical Themes were recovered and written in upper case and a
circumflex symbol)
In English, declarative clauses usually start with the grammatical subject or, more markedly, with a
Circumstance, but almost never with the verb. As stated earlier, using verbs at the beginning of a clause
is normally an indication that the clause is in the imperative. Nevertheless, because Spanish is a pro-drop
language, where verb morphology normally indicates grammatical subject and number, and also because it
has a more flexible syntax than English, the choice of starting a declarative clause with a verb is unmarked
in Spanish. Spanish units (12b-14b), albeit passive, change the order of sentential elements with respect to
those of the ST; for that reason the thematic pattern of the original text is partially broken. Notice that even
in unit (13b) the whole syntax is changed in the translated text (the literal translation of (13a) would be Los
pacientes no fueron estudiados si...).
The ST’s second subsection, Study design, has a series of reiterated Themes where the element in
common is patients again (see Figure 4). The Themes are kept in the TT, save one exception, where the
reflexive passive is used, and again the verb is fronted:
(15a) ST: they were asked to withhold the last dose that was scheduled before attending the study.
(15b) TT: se pidió a los pacientes que eliminasen la última dosis programada antes de la realización del
estudio.
Figure 4. Theme pattern in the second subsection of Methods
In example (15b) we see that the flexibility with which the verb is fronted in Spanish is possible not only in
periphrastic passive constructions—as observed in (12b-14b)—but also in reflexive ones.
The third subsection, Venuous occlusion plethysmography protocol, has a thematic pattern of four units
where there are two elements in common: blood flow and venous capacitance (see Figure 5). We decided
Unauthenticated
Download Date | 6/15/17 1:17 AM
10 D. Rodríguez-Vergara
to take these two elements as part of the same thematic pattern because unit 6 contains both elements
(Forearm blood flow and forearm venous capacitance). This complex Theme is then fragmented into two,
i.e., the Theme in unit 7 (Blood flow) and in unit 8 (Venous capacitance). Finally, unit 10 integrates the
two into a single Theme again (Baseline recordings of forearm blood flow and forearm venous capacitance).
This method of development is similar to the cohesive structure described by Eggins (2004), ‘the multipleRheme pattern’, where one Rheme is fragmented into a number of Themes, the difference being that here it
is not a Rheme that is fragmented, but rather a Theme.
Figure 5. Theme pattern in the third subsection of Methods
In the TT, this pattern is broken since, although units 7 and 8 keep the same Theme and voice, units 6 and
10 are rendered as reflexive passives with fronted verbs:
(16a) ST: Forearm blood flow and forearm venous capacitance (FVC) were measured by venous occlusion
plethysmography...
(16b) TT: Se evaluaron el flujo sanguíneo y la capacidad venosa del antebrazo (CVA) mediante pletismografía
con oclusión venosa...
(17a) ST: Baseline recordings of forearm blood flow and forearm venous capacitance (FVC) in both
arms were made every 5 min for 20 min.
(17b) TT: Se efectuaron registros del flujo sanguíneo del antebrazo y de la capacidad venosa del antebrazo
(CVA) en ambos brazos, cada 5 min hasta un total de 20 min.
It seems in these cases that the complexity of the Themes in English could have led to the decision of using
reflexive passive constructions in the translation, in which case the heavy noun groups are placed after
the Spanish verbs. Such a decision would follow what has been called ‘the End-weight principle’, which
predicts that complex groups will be placed at the end of clauses (Quirk et al. 1972).
The final subsection of Methods, Data analysis, contains a thematic development similar to the one
observed in the previous subsection (see Figure 6). Here, unit 5 contains, in the Theme, the item parallel
groups, which later becomes part of the Theme in unit 6, and comparison and placebo, which later become
part of the Theme in unit 7.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 11
Figure 6. Theme pattern in the fourth subsection of Methods
In the ST, the same thematic pattern is observed. Unit 7 is maintained in the periphrastic passive with the
same order of elements. Nevertheless, units 5 and 6 are rendered as reflexive passives, and thus verbs and
Participants shift positions; despite the shift, the Circumstances remain thematic:
(18a) ST: For the parallel groups’ comparison of NTG versus furosemide with placebo, a covariance
pattern, mixed model...
(18b) TT: Para las comparaciones de los grupos paralelos de NTG y furosemida con el placebo se
estableció un patrón de covariancia sobre un modelo mixto...
(19a) ST: For both the crossover and parallel groups’ models, when no treatment by time interaction was
found, the main effects model was used.
(19b) TT: Con respecto a los modelos de los grupos transversales y paralelos, se utilizó el modelo de
efectos principales en aquellos casos en los que no se detectó interacción del tratamiento con el tiempo.
Generally speaking, Methods is the section where most passive clauses are found, a fact that was already
noted in other studies (Swales 1990, Nwogu 1997, Biber & Conrad 2009). According to Biber and Conrad
(2009: 130), this distribution is due to this section centering upon the research procedures, data and
techniques, rather than to the people who carried out the study: “…it makes no difference if it was an
individual or the entire research group who performed an action. The important point is to understand how
the experiment was conducted. As a result, passive voice is the norm in this section.”
Another observation that can be made is that whereas in English passive constructions seem to allow
complex nouns at the beginning of sentences, the Spanish translation shows a preference for reflexive
passives where complex nouns are placed in the rhematic portion of units. In addition, we observed that
although the TT had basically the same thematic patterns as the ST, some of the units changed their Theme
mainly because some verbs were fronted.
Unauthenticated
Download Date | 6/15/17 1:17 AM
12 D. Rodríguez-Vergara
4.3 Results
This is a short section, with 15 T-units in both the ST and the TT. In contrast to the previous sections, this
one exhibits a constant use of active voice in all but one unit in the ST. However, although the TT has the
same number of units, five of them are reflexive passive. This means that, contrary to the expectation that
the direction of voice shift in the translation would be from passive to active (Gutiérrez 1997), the opposite
holds true in this section.
In this section the ST has four Theme reiteration patterns (see Figure 7). In the first (units 1-3), we
observe the fragmentation of a Theme whose parts are Themes of other units. However, in contrast to the
Theme fragmentation that we observed in the previous section, the pattern formed in units 1-3 shows the
reversed order, i.e., the Themes that are fragmented from unit 3 (Baseline mean arterial pressure and heart
rate) appear earlier than the Theme of which they are fragments (Mean arterial pressure and heart rate). In
the TT, this three-unit pattern is broken because unit 2 is rendered as a reflexive passive with a fronted verb
(se observaron) as well as with an adverb (tampoco):
(20a) ST: Similarly, heart rate did not differ between treatments.
(20b) TT: De la misma manera, tampoco se observaron diferencias en la frecuencia cardíaca entre estos
grupos.
Figure 7. Theme pattern in the English Results
The second thematic pattern in the ST consists of three units (4-6), with forearm as the item in common. This
pattern is kept in the TT despite the fact that unit 5 was passivized. In turn, the third thematic pattern in the
ST has four units (7-10), but worth noting is that there are no items in common. We nevertheless considered
them part of the same thematic pattern because all start with a Circumstance. It is a fragment that describes
how results were obtained under specific conditions, and can thus be regarded as a unitary sequence. In the
TT, although units 9-10 were passivized, the thematic pattern was kept since the Circumstances remained
fronted:
(21a) ST: After pretreatment with 300 mg of aspirin, there was a 3.7% (-6.8% to -0.7%) fall in FVC...
(21b) TT: Tras el pretratamiento con 300 mg de aspirina se observó una disminución del 3,7% (-6,8 a-O,
7%) en la CVA...
(22a) ST: After 75 mg aspirin, there was a 1.1% (-4.2% to 1.9%) fall in FVC.
(22b) TT: Tras el pretratamiento con 75 mg de aspirina se observó una disminución del 1,1% (-4,2 a 1,9%)
en la CVA.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 13
The fourth thematic pattern in the ST has three units (11-13), where the item in common is response to
furosemide. In the TT, this pattern is maintained with no voice changes.
Generally speaking, the Results section has an active voice character because, in it, writers do not
generally describe actions using transitive verbs. Instead, they talk about how conditions were like in the
study, thereby assessing the findings of the experiments as significant, not significant, similar, different,
etc. For this reason, the most frequent verb in this section is (there) be; therefore, the possibility of using the
passive voice is almost null.
4.4 Discussion
This final section is relatively large and is divided into six subsections: 1) an unnamed introductory
subsection, 2) Effect of aspirin and role of prostaglandins in furosemide-mediated venodilation, 3) Possible
direct effect of aspirin on furosemide action in veins, 4) Possible venoconstrictor effect of furosemide in
presence of aspirin, 5) Study limitations, and 6) Conclusions. In both the ST and the TT there are 42 T-units,
nine (21%) of which are in the passive voice in the ST, and six (14%) of which are in the passive voice in the
TT; of those six, four are reflexive and two are periphrastic.
Although this is a large section, in the ST there is only one thematic pattern with more than two units: a
three-unit pattern found in the second subsection, Effect of aspirin and role of prostaglandins in furosemidemediated venodilation. Not surprisingly, the thematic item in common is aspirin, the medication that is
investigated in the paper. In the TT, the pattern is maintained with no voice changes.
There are an additional seven patterns composed of only two units each, only one of them showing a
possible thematic disruption in the ST:
(23a) ST: These authors studied the effects of therapeutic concentrations of furosemide on cultures of human
umbilical vein endothelium.
(23b) TT: Estos autores estudiaron los efectos de las concentraciones terapéuticas de furosemida sobre
cultivos de endotelio de vena umbilical humana.
(24a) ST: They reported that secretion of prostacyclin, measured indirectly by 6-ketoPGF518 production,
increased within 5 min of exposure of the culture to furosemide.
(24b) TT: Observaron que la secreción de prostaciclina, medida de manera indirecta a través de la producción
de 6-cetoPGF518 aumentaba a los 5 min de exposición del cultivo a la furosemida.
The possible disruption occurs in (24a-b), where, due to the pro-drop nature of Spanish, the translated
sentence has the verb (in the active voice) as its Theme, and no pronoun is used. We characterized this as a
‘possible’ disruption since it might as well be argued that the grammatical person indicated by the English
pronoun (They) is realized not lexically but morphologically in (24b), thus reflecting the same Theme
reiteration pattern.
Even though the Discussion section does not have complex Theme-reiteration patterns, it could have
other patterns with different methods of cohesive development. Perhaps the fact that there are few passive
constructions in this section has an effect on the low number of Theme-reiteration patterns. Further research
is needed to confirm this assertion.
To sum up, of all the T-units that were analyzed in all the sections, the ST and the TT had roughly the
same proportion of passive constructions: 56% were passive in the ST and 56.06% in the TT. In turn, 64.86%
of all passive constructions in the TT were reflexive. The limited use of the periphrastic passive in Spanish
vis-à-vis the reflexive construction is thought to be due, in part, to its complex aspectual restrictions,
as attested by Takagaki (2005) in his registerially diverse corpus study. Takagaki (2005: 304) notes that
“this may be in accord with the characterization of the passive with ser as representative of a singular and
Unauthenticated
Download Date | 6/15/17 1:17 AM
14 D. Rodríguez-Vergara
dynamic event and the passive with se as typical of a repetitive and habitual event. The high ratio of the
imperfect tenses in the passive with se can be said to demonstrate their tendency to be used in general and
generic statements.” Thus, it seems that because in medical research articles writers try to generalize and
universalize their findings (Saleger-Meyer 1994), they rely more on reflexive constructions rather than on
periphrastic ones.
5 Final considerations
In this paper we have presented a systemic functional analysis of thematic structure in an English medical
research article vis-à-vis its translation into Spanish. We explored how passive clauses in the ST are
translated into Spanish and how voice changes modified the original Theme patterns. The results showed
that most of the modifications of the original thematic patterns in the translation were due to the use of the
reflexive passive form in Spanish. In reflexive passives, it is common to front the verb in the corresponding
sentences, causing a change of Themes with respect to the ST’s sentences. However, we also identified some
periphrastic passive constructions where the verb was fronted.
Overall, the majority of the ST’s clauses in the Introduction and Methods sections were passive,
whereas the opposite was true for the Results and Discussion sections. In the TT, only the Methods section
contained a majority of clauses in the passive voice, where the frequency of the reflexive passive variant
almost doubled the periphrastic construction. Furthermore, some active ST clauses were passivized in
the TT using the reflexive form; when English Themes tended to be complex, they were usually moved to
rhematic position in the translation, a phenomenon that the End-weight principle actually predicts (Quirk
et al. 1972).
The sections where we observed the most changes in thematic progression were Methods and Results.
In these, original thematic patterns were partially disrupted in the translation, since some of the TT
units presented new fronted elements in relation to the ST units. The Discussion section of the ST and TT
presented basically the same thematic progression, but in the Introduction we observed that a brand new
Theme-reiteration pattern was created in the ST. As a result, the tenor of the Introduction was switched from
an impersonal one in the ST, which was reflected by the use of passive sentences, to a personal one in the
TT through the use of active sentences and first person constructions.
Although the scope of this study is rather limited insofar as only one pair of texts was analyzed, we have
tried to provide guidelines for further studies that can potentially contribute to an understanding of the
function of grammatical voice in English and Spanish, as well as its relationship with thematic progression.
In addition, regarding the writing advice controversy on the use of personal and impersonal grammatical
forms in research articles, this study confirms what had already been noted in other studies (e.g., SalagerMeyer 1994, Biber & Conrad 2009): since every section has a different communicative goal, each has its own
specific grammatical and rhetorical characteristics. Therefore, instead of partially or completely banning
the use of passive voice or first person constructions, writing advice should acknowledge the heterogeneity
of RAs, as well as the correspondence between grammatical forms and communicative functions.
Another limitation of this study was that instances of self-mention were not computed and compared
between the texts, and even though some authors have already studied the issue (Tarone et al. 1998,
Hyland 2001, Perales & Swales 2011), the role of self-mention remains to be studied further in the context
of translation in the medical sciences. Furthermore, since we identified a method of cohesive development
that we had not expected (fragmentation of Theme), more research is indeed needed not only on Theme
reiteration, but also on Theme fragmentation patterns and other devices discussed in Eggins (2004), i.e.,
the zig-zag pattern and the multiple-Rheme pattern. Finally, because from the analysis of two texts we
cannot determine whether the thematic changes observed in the ST are significant, the same research
objectives should be applied to more data, ideally to a large corpus of medical texts. A final direction that
future studies can take is to evaluate the extent to which thematic changes affect reading comprehension.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 15
Abbreviations
R: Rheme
SFL: Systemic Functional Linguistics
ST: Source Text
T: Theme
TT: Translated Text
References
Alvin, Leong P. 2014. The passive voice in scientific writing: The current norm in science journals. Journal of Science
Communication 13 (1), 1-16.
Amdur, Robert J., Jessica Kirwan, Christopher G. Morris. 2010. Use of the passive voice in medical journal articles. American
Medical Writers Association Journal 25 (3), 98-110.
Arús, Jorge. 2006. Perspectiva sistémico-funcional de los usos de ‘se’ en Español [Systemic functional perspective on the uses
of ‘se’ in Spanish]. Signos 61, 131-159.
Atkinson, Dwight, Ellen Valle. 2012. Corpus analysis of scientific and medical writing across time. In Chapelle, Carol A. (ed.),
The Encyclopedia of Applied Linguistics. Malden: Blackwell, 1347–1353.
Bennett, Karen. 2009. English academic style manuals: A survey. Journal of English for Academic Purposes 8 (1), 43–54.
Biber, Douglas, Susan Conrad. 2009. Register, genre, and style. Cambridge: Cambridge University Press.
Canter, David, Gavin Fairbairn. 2006. Becoming a author: Advice for academics and other professionals. Berkshire: Open
University Press.
Daneš, Frantisek. 1974. Functional sentence perspective and the organization of the text. In Daneš, Frantisek (ed.), Papers on
functional sentence perspective. Prague: Academia, 106-128.
Day, Robert A., Nancy Sakaduski. 2011. Scientific English: A guide for scientists and other professionals (3rd. edn.). Santa
Barbara: Greenwood.
Eggins, Suzanne. 2004. An introduction to systemic functional linguistics (2nd edn.). London: Continuum.
Fries, Peter H. 1983. On the status of theme in English: Arguments from discourse. In Petöfi, Janos S., Emel Sözer (eds.), Micro
and macro connexity of texts. Hamburg: Helmut Buske, 116-152.
Fries, Peter H. 1994. On Theme, Rheme and discourse goals. In Coulthard, Malcolm (ed.), Advances in written text analysis.
London: Routledge, 229-249.
Gómez, María de los Ángeles, Ana P. García. 2014. Discourse-organizational patterns
in English and Spanish: Some notes on the thematic management of news reports. Revista Española de Lingüística Aplicada,
27 (1), 87-117.
Goodman, Neville W., Martin B. Edwards. 2006. Medical writing: A prescription for clarity (3rd edn.). Cambridge: Cambridge
University Press.
Gotti, Maurizio, Françoise Salager-Meyer (eds.). 2006. Advances in medical discourse analysis: Oral and written contexts.
Bern: Peter Lang.
Gutiérrez, Bertha M. 1997. La influencia del inglés sobre nuestro lenguaje médico [The influence of English in our medical
language]. Medicina Clínica, 108 (8), 307-313.
Halliday, Michael A. K. 1985/1994. An introduction to functional grammar (1st, 2nd edn.). London: Arnold.
Hidalgo, Raquel. 1994. The pragmatics of de-transitive voice in Spanish: From passive to inverse? In Givón, Thomas (ed.),
Voice and inversion. Amsterdam: John Benjamins, 169-186.
Hyland, Ken. 2001. Humble servants of the discipline? Self-mention in research articles. English for Specific Purposes 20 (3),
207-226.
Mann, William C., Sandra A. Thompson. 1988. Rhetorical Structure Theory: Toward a functional theory of text organization.
Text 8 (3), 243-281.
Martínez, Iliana A. 2001. Impersonality in the research article as revealed by analysis of the transitivity structure. English for
Specific Purposes 20 (3), 227-247.
Matthews, Janice R., Robert W. Matthews. 2014. Successful scientific writing: A step-by-step guide for the biological and
medical sciences (4th edn.). Cambridge: Cambridge University Press.
Matthiessen, Christian M. I. M. 2006. Frequency profiles of some basic grammatical systems:
An interim report. In Thompson, Geoff, Susan Hunston (eds.), System and corpus: Exploring connections. London: Equinox,
103-142.
Matthiessen, Christian M. I. M. 2014. Appliable Discourse Analysis. In Yan, Fang, Jonathan J. Webster (eds.), Developing
Systemic Functional Linguistics: Theory and application. London: Equinox, 138-208.
Unauthenticated
Download Date | 6/15/17 1:17 AM
16 D. Rodríguez-Vergara
Matthiessen, Christian M. I. M., Kazuhiro Teruya, Marvin Lam. 2010. Key terms in systemic functional linguistics. London:
Continuum.
Nwogu, Kevin N. 1997. The medical research paper: Structure and functions. English for Specific Purposes 16 (2), 119-138.
O’Donnell, Michael. 2004. RSTTool, version 3.45 [Computer software]. South Bridge: WagSoft.
Perales, Moisés, John Swales. 2011. Tracing convergence and divergence in pairs of Spanish and English research article
abstracts: The case of Ibérica. Ibérica 21, 49-70.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, et al. 1972. A grammar of contemporary English. London: Longman.
Salager-Meyer, Françoise. 1994. Hedges and textual communicative function in medical English written discourse. English for
Specific Purposes, 13 (2), 149-171.
Salazar, Danica, Aaron Ventura, Isabel Verdaguer. 2013. A cross-disciplinary analysis of personal and impersonal features in
English and Spanish scientific writing. In Verdaguer, Isabel, Natalia J. Laso, Danica Salazar (eds.), Biomedical English: A
corpus-based approach. Amsterdam: John Benjamins, 121-143.
Swales, John M. 1990. Genre analysis: English in academic and research settings. Cambridge: Cambridge University Press.
Takagaki, Toshihiro. 2005. On the productivity of the Spanish passive constructions. In Takagaki, Toshihiro, Susumu Zaima,
Yoichiro Tsuraga, et al. (eds.), Corpus-based approaches to sentence structures. Amsterdam: John Benjamins, 289-309.
Tarone, Elaine, Sharon Dwyer, Susan Gillette, et al. 1998. On the use of the passive and active voice in astrophysics journal
papers: With extensions to other languages and other fields. English for Specific Purposes 17 (1), 113-132.
Taylor, Robert B. 2005. The clinician’s guide to medical writing. Portland: Springer.
Teich, Elke. 1999. Systemic Functional Grammar in natural language generation: Linguistic description and computational
representation. London: Cassell.
Thompson, Geoffrey. 1996. Introducing functional grammar. London: Arnold.
Unauthenticated
Download Date | 6/15/17 1:17 AM
A Systemic Functional Approach to the Passive Voice in English into Spanish Translation 17
Appendix: References of parallel texts
ST:
Jhund, Pardeep, Andrew P. Davie, John J. V. McMurray. 2001a. Aspirin inhibits the acute venodilator
response to furosemide in patients with chronic heart failure. Journal of the American College of Cardiology
37 (5), 1234-1238.
TT:
Jhund, Pardeep, Andrew P. Davie, John J. V. McMurray. 2001b. La aspirina inhibe la respuesta venodilatadora
inducida por la furosemida en los pacientes con insuficiencia cardíaca crónica. Journal of the American
College of Cardiology (Mexican edition), 10 (6), 257-262.
Unauthenticated
Download Date | 6/15/17 1:17 AM