Vajon in Translated Hungarian

Abstract

This paper presents an analysis of the structures the discourse marker vajon forms in translated Hungarian fiction. Although translation data has been deployed in the study of discourse markers (Aijmer & Simon- Vandenbergen, 2004), such studies do not account for translation-specific phenomena which can influence the data of their analysis. In addition, translated discourse markers could offer insights into the idiosyncratic properties of translated texts as well as the culturally defined norms of translation that guide the creation of target texts. The analysis presented in this paper extends the cross-linguistic approach beyond contrastive analysis with a detailed investigation of two corpora of translated texts in order to identify patterns which could be a sign of translation or genre norms impacting the target texts. As a result, a distinct, diverging pattern emerges between the two corpora: patterns of explicit polarity show a marked difference. However, further research is needed to clarify whether these are due to language, genre, or translation norms.

Full text

Vajon in Translated Hungarian
Diverging Patterns in Two Fiction Genres
Andrea GÖTZ
Eötvös Loránd University
Translation Studies PhD Programme
gotzandrea@caesar.elte.hu
Abstract. This paper presents an analysis of the structures the discourse
marker vajon forms in translated Hungarian ction. Although translation
data has been deployed in the study of discourse markers (Aijmer & Simon-
Vandenbergen, 2004), such studies do not account for translation-specic
phenomena which can inuence the data of their analysis. In addition,
translated discourse markers could offer insights into the idiosyncratic
properties of translated texts as well as the culturally dened norms of
translation that guide the creation of target texts. The analysis presented in
this paper extends the cross-linguistic approach beyond contrastive analysis
with a detailed investigation of two corpora of translated texts in order
to identify patterns which could be a sign of translation or genre norms
impacting the target texts. As a result, a distinct, diverging pattern emerges
between the two corpora: patterns of explicit polarity show a marked
difference. However, further research is needed to clarify whether these are
due to language, genre, or translation norms.
Keywords: discourse markers, translating discourse markers, translation
method, language norms, corpus-based translation studies
1. Introduction: translating discourse markers
In recent years, the translation of discourse markers, pragmatic markers, or
discourse particles – as several terms are used for these forms – has increasingly
gained attention. Following the introduction of the so-called “translation
method” by Aijmer & Simon-Vandenbergen (2004), many studies deployed
translation data to investigate the function or meaning of discourse markers
(DMs), as reected by their translations (e.g. Aijmer 2007, Degand 2009, Fischer
2007, Furkó 2015, Mortier & Degand 2009). However, these studies represent a
corpus-driven, contrastive approach which does not account for the properties
of translated or translational texts (Károly 2007) as such. In order to facilitate
ActA UniversitAtis sApientiAe, philologicA, 8, 3 (2016) 31–41
DOI: 10.1515/ausp-2016-0029
Unauthenticated
Download Date | 2/18/17 8:59 PM

32 Andrea GÖTZ
an analysis that can accommodate translation phenomena, the overarching
patterns found in translation corpora need to be investigated. These patterns can
be shaped by the culturally diverse and genre-related norms of translation. The
effects of translation norms are especially relevant for the translation of DMs, as
the socio-pragmatic functions of DMs are well established (see Schiffrin 1987,
Foolen 2012). However, variation in translation data regarding DMs can be
heavily inuenced by genre norms as well (Niemegeers 2009).
Translated texts do not solely reect the properties of the source texts but are
increasingly seen as the products of a particular type of textual composition that
includes both re-productive and creative processes (Károly 2014). The textual
properties of the source text, due to cross-linguistic and, frequently, genre
differences, cannot always be simply re-created in the target text. Translated texts
will exhibit their own patterns of cohesion and coherence, which are of particular
interest to the study of translated discourse markers since DMs are also thought to
contribute to discourse cohesion and coherence (Fraser 1999, Schiffrin 1987). It
is becoming increasingly clear that contrastive approaches need to be combined
with translation-studies-specic considerations in order to distinguish between
cross-linguistic and translation-specic phenomena. Indeed, studying cohesive
markers (grammatical or lexical forms) in translation has proved to be useful for
both elds (Behrens 2005, Becher 2011).
The extent to which the translator is seen as an executor of norms and, indeed,
how profoundly the controversial phenomena known as translation universals
inuence the target text remain a much-contested issue (Meylaerts 2008). Since
the emergence of corpus-based translation studies (Baker 1993, 1996), the
idiosyncratic patterns of translated texts (e.g. universals such as explicitation),
which differentiate them from non-translated, or authentic texts, and the effect
language norms have become widely researched topics. Patterns specic to
translated texts are usually accommodated within the framework of translation
universals, i.e. a higher count of explicit grammatical features in translated texts
are attributed to the translation universal of explicitation (Olohan & Baker 2000),
which is thought to affect all translated texts, hence its universal nature.
This study addresses these issues by presenting an analysis of translation
data, which expands contrastive analysis with a corpus-based investigation. The
present paper investigates the translation data of the Hungarian DM vajon in two
ction corpora composed of English source texts and their Hungarian translations.
Vajon-structures in translated Hungarian are examined for their patterns, which
could point either to the inuence of specic translation phenomena and genre
norms or to cross-linguistic differences. The marking of explicit polarity is found
to show a distinct difference between the two corpora. The variation in the two
corpora in terms of marked polarity is examined in detail.
Unauthenticated
Download Date | 2/18/17 8:59 PM

33Vajon in Translated Hungarian...
2. Classication of structures
In the following, the categories for classifying the various vajon-structures present
in the Hungarian corpora (see 4.1.) are introduced and notions from functional
grammar (Halliday & Matthiessen 2014) are applied to their description. Since
vajon occurs in a diverse set of structures, a brief overview of these is in order.
Vajon, traditionally understood as an interrogative particle with modal
properties (Keszler 2000), appears in both interrogative sentences and subordinate
interrogative clauses. As part of a hypotactic projected clause, in a subordinate
construction, it can be preceded by the complementizer hogy. In the case of a
hypotactic structure, various cognitive verbs (referred to as c in 4.1.), such as
wonder, think, know, contemplate, can appear in the projecting clause. This
is relevant since, in some cases, an English projecting clause complete with a
cognitive verb can be translated as an interrogative sentence, not as a hypotactic
clause, as in (1). Source contexts are marked “a”, translations “b” in all examples,
the abbreviation of the corpus and the text of the example are enclosed between
square brackets. In (1), the rst part of the clause nexus to the sign || represents
the projecting clause, and the latter half the projected clause. The Hungarian
translation does not follow this structure, and the cognitive verb is not retained.
(1)a I wonder || if she lives alone except this little girl; (…)
b Vajon kettesben él-e a kislánnyal?
dm [two of them] together lives-pol.pArt with the little
girl? [B/Je]
In other cases, cognitive verbs are retained in the translation, as in (2). In (2),
the cognitive verb wonder is substituted for ‘tűnődik’ in the target context, and
the polar interrogative subordinate clause is retained. The complementizer hogy
may or may not appear in the projected clause in the position whether occupies
– in (2), it is not present.
(2)a I often wonder, Shirley, || whether most men resemble
b Gyakran tűnődöm, Shirley, vajon más férak is olyanok-e
I often wonder Shirley dm other men as well are
like [B/Sh]
In a high number of cases, vajon cannot be attributed the presence of a
linguistic form in the source context. In the literature, such instances of “added”
target forms that do not correspond to a source form are referred to as additions,
insertions, zero forms, or zero equivalents. In this paper, these are referred to as
zero forms. In (3), we can see such an example.
Unauthenticated
Download Date | 2/18/17 8:59 PM

34 Andrea GÖTZ
(3)a What can it be?
b Vajon mi történhetett?
dm what could have happened? [B/Se]
In some cases, the source context cannot be identied as a clause nexus
composed of a projecting and a projected clause. Instances such as these are
labelled as “other structures”. This category is illustrated by (4).
(4)a (…) how I longed to follow it farther!
b (…) vajon hová vezet?
dm where does it lead? [B/Je]
3. Corpus and methods
The corpus of this study consists of eight novels translated from English into
Hungarian, four from the young adult genre (corpus A) and the other four are
romantic novels (corpus B) dating from the 19th century (see Sources). The
young adult corpus represents a part of Robin’s revisional corpus (Robin 2014)
of translated Hungarian ction, who gave permission for its use. The texts were
selected for consistency within the corpora, similarities in the genres, and some
marked differences between the two in register and style. These differences are
assumed to be reected in the Hungarian translation of the novels as well. From
each translated novel, the rst 25 contexts featuring vajon were selected, together
with their source language contexts. Thus, the eight novels yielded 200 source-
and 200 target-language contexts in total, with 100–100 occurrences in each
corpus. Although the data collection itself represents a corpus-driven approach,
the data are analysed from a more corpus-based point of view.
4. Results and discussion
4.1. Quantitative results: target and source forms
The two ction corpora show an overall similarity and some signicant as well
as marginal differences. Table 1 presents the various target structures containing
the DM vajon in the two corpora.
Unauthenticated
Download Date | 2/18/17 8:59 PM

35Vajon in Translated Hungarian...
Table 1. Target-language vajon-structures
Corpus A (YA) Corpus B (R)
vajon 45 38
vajon + -e 7 32
c || vajon 7 4
c || vajon + -e 8 23
c || hogy + vajon 18 2
c || hogy + vajon + -e 12 1
|| hogy + vajon 3 0
Both datasets comprise similar numbers of vajon; however, there is a striking
difference in the prevalence of vajon + -e structures between the two genres.
In total, across all structures, corpus B displays 56 target forms with the
interrogative particle -e in contrast to the 27 occurrences of -e in corpus A. At the
same time, projected clauses displaying an explicit complementizer hogy were
more frequent in corpus A, including hogy-clauses with the particle -e. Corpus
B contained a signicantly lower number of clauses with the complementizer
hogy being present. From all 48 subordinate clauses in corpus A, 33 contain the
complementizer hogy, which number for corpus B is three, in relation to a total
of 30 subordinate clauses. This difference could point to a diverging norm of
translation, although it could also be affected by the variation of source forms
regarding the two corpora.
In summary, corpus B displays a greater tendency to use the particle -e and
insert vajon + -e structures, and is less prone to use the complementizer hogy in
vajon-structures. As these variations, in addition to translational norms, could
also be inuenced and motivated by a variation in the source structures, it is
necessary to examine their source forms in order to investigate the reasons for the
discrepancies in question.
Table 2 demonstrates the English source structures that were translated with
vajon-structures shown in Table 1. A few differences are immediately evident.
Table 2. Source-language structures
Corpus A (YA) Corpus B (R)
zero 34 44
wonder 20 7
wonder || whether 4 6
wonder || if 22 1
c9 7
c || if 2 3
c || whether 5 17
Unauthenticated
Download Date | 2/18/17 8:59 PM

36 Andrea GÖTZ
Corpus A (YA) Corpus B (R)
other structures 4 12
|| as to 0 2
|| whether 0 1
In corpus A, we nd 34 zero forms, ten per cent less than in corpus B. Although
the difference is not signicant, it might point to a more pronounced tendency
on the part of the translators or editors to insert vajon into the target text. As
we have seen above, target structures in corpus B contained a higher number
of the particle -e. This nding is surprising, as the source contexts of corpus B
contain fewer instances of explicit markers of polarity, i.e. grammatical clues
that could trigger the use of forms that expressly signal polarity. In corpus A, 33
occurrences of whether or if are observed, which could motivate the use of -e in
the translation, whereas corpus B comprises 28.
Corpus A also has a much higher number of wonder-structures as source forms
(46) than corpus B (14), taking all structures featuring wonder into account.
Wonder is treated separately from the other cognitive verbs due to its frequency.
In corpus A, we nd 20 wonder-structures, which could introduce wh-question
clauses, and 26 wonder-clauses which contain a marker of polarity: whether or if.
In corpus B, on the other hand, from the 14 wonder-structures, seven introduce
wh-question clauses, and seven if-/whether-clauses. Another point of difference
is the number of other structures rendered as vajon-structures in the target texts.
Corpus B contains 12, three times as many as corpus A, and two occurrences of
as to. This could point to a greater willingness on the part of the translators of
corpus B to deploy creative solutions. However, to gain a clearer understanding
of the possible norms guiding translation, the relation of source and target forms
needs to be studied further.
4.2. Qualitative results: the presence of the polar particle -e
As seen before, there is a clear difference in marked polarity between the two sets
of target texts, with corpus B showing a higher percentage of polarity markers
(56) despite exhibiting fewer markers of polarity in the source texts (28). In the
following, we examine the results relevant for the translation of polarity, as
presented in Table 3.
Unauthenticated
Download Date | 2/18/17 8:59 PM

37Vajon in Translated Hungarian...
Table 3. The translation of polarity
Corpus A
(YA) Corpus B
(R)
Marked polarity in the target texts
e.g. structures with -e27 56
Marked polarity in the source texts
e.g. structures with whether, if 33 28
Polar source structures translated in the target texts
without the explicit polar marker -e13 2
e.g. I wonder || whether it is so. → Vajon csakugyan így
esett[-e]? (‘Did it really happen like this?’) [B/Se]
Source wh-structures translated with -e in the target texts
3 2
e.g. I always wondered || when she would notice that…
→ Mindig érdekelt, hogy vajon feltűnt-e neki, hogy… (‘I
always wanted to know whether she noticed that…’) [A/
Sh]
Target structures with -e from zero source forms
3 18
e.g. Will she ever come back? → Vajon hazatér-e valaha?
(‘Will she ever return home?’)
Target structures without -e from polar zero source forms
15 7
e.g. Will he leave it again soon? → Vajon rövidesen
megint útra kel[-e]? (‘Is he going to leave again soon?’)
Polar target forms with markedly polar source forms
20 25
e.g. I wonder || if she lives alone except this little girl;
→ Vajon kettesben él-e a kislánnyal? (‘Does she live
together with the little girl?’) [B/Je]
Other source structures rendered with -e in the target text
1 9
e.g. a doubt sometimes entered her mind of their being
really engaged → Elinor elméjébe olykor már-már kétely
lopózott: vajon csakugyan jegyesek-e (‘a doubt snuck
into Elinor’s mind: whether they were really engaged’)
[B/Se]
The ndings indicate that there is no complete “conversion” between source
and target structures marked for polarity. A closer look at the structures reveals
that in corpus A 13 instances of marked polarity in the English source texts did not
become manifestly polar in the Hungarian target texts. This number for corpus B
is two. The number of evidently wh-structures in the source texts translated into
Hungarian as polar structures is low for both corpora, with three in corpus A and
two in B, although zero forms and other structures, which might be interpreted
as polar, were not included in this gure.
Zero forms served as source structures for polar structures in the target texts
in three cases in respect to corpus A and in 18 in corpus B, which means that
Unauthenticated
Download Date | 2/18/17 8:59 PM

38 Andrea GÖTZ
translators and editors working on the texts of corpus B were six times likelier
to construct marked polar structures. In fact, following a closer examination
regarding vajon-structures from zero forms in the source texts, it emerges that
in corpus A in 15 cases polar zero source forms were not rendered as markedly
polar in the target texts, i.e. the target contexts contain a vajon-structure, not
a vajon + -e structure. This is observed in seven cases in corpus B. Corpus B
has also constructed nine markedly polar structures in the target texts from
other structures in the source, more than corpus A, which created only one
such structure. Corpus B also contained two as to structures, one of which was
translated with -e in Hungarian.
Corpus B was thus more likely to translate English source contexts as markedly
polar in Hungarian (20 in corpus A, 25 in corpus B), and also less likely not to
translate polar source structures as not markedly polar. This, however, does not
mean that the remaining structures would not have been rendered as polar, only
that they are not marked as such with the polar particle -e. Since clauses and
sentences without the polar particle -e would still be grammatical – and function
as polar questions – the difference in this pattern between the two corpora could
reveal diverging norms of translation, inuenced by genre norms.
All in all, corpus B shows an altogether more pronounced tendency than corpus
A to use the polar particle -e, but without investigating the source structures as
well it would not be possible to discern whether this tendency is due to the
properties of the source texts or, indeed, whether it could be attributed to other
factors such as translation norms. However, we cannot yet claim that corpus B
displays different translational norms than corpus A as due to the small scale of
the study the incomprehensive analyses cannot substantiate such claims.
Conclusions
This paper examined vajon-structures present in two corpora of translated
Hungarian texts from different genres. Although the two datasets displayed an
overall similarity, in terms of marked polarity structures, a marked difference
was revealed, with texts in the romance genre showing a greater prevalence of the
polar particle -e. In addition to marked polarity, texts in the YA corpus contained
a higher percentage of the complementizer hogy in comparison to texts in corpus
B. Since the analysis concerns small datasets, generalizations regarding genre
and translation norms, norms of translated Hungarian, or translation universals
cannot be drawn. The results, however, delineate potential lines of inquiry for
future research.
The presence of these function words and the levels of grammatical explicitness
should be contrasted with the frequency of these forms and levels of explicitness
Unauthenticated
Download Date | 2/18/17 8:59 PM

39Vajon in Translated Hungarian...
found in authentic Hungarian texts, across genres, as the potential inuence of
genre norms cannot be ignored.
A similar nding, which established a difference between the patterns that
in translated English as opposed to authentic English, has been interpreted as
evidence for the translation universal of explicitation (Olohan & Baker 2000).
Although such claims and comparisons are not made in this study, it is clear that
DMs and the structures they form can offer insights into the properties of translated
texts. However, in order to explore this in greater detail, cross-linguistic, corpus-
driven approaches need to be extended beyond the scope of contrastive analyses
in order to accommodate methodologies more suitable for studying translated
language. These translation-specic characteristics, of course, do exert an effect
on the individual tokens found in the corpora. In conclusion, to fully investigate
the patterns of translated DMs, and their implications for translation studies, a
combined approach is needed.
References
Aijmer, K.–Simon-Vandenbergen A-M. 2004. A model and a methodology for
the study of pragmatic markers: the semantic eld of expectation. Journal of
Pragmatics 36(10): 1781–1805.
Aijmer, K. 2007. The meaning and functions of the Swedish discourse marker
alltså — evidence from translation corpora. Catalan Journal of Linguistics 6:
31–59.
Baker, M. 1993. Corpus linguistics and translation studies: implications and
applications. In: Baker Mona et al. (eds), Text and technology: in honour of
John Sinclair. Philadelphia: John Benjamins, 233–250.
1996. Corpus-based translation studies: The challenges that lie ahead. In:
Sager, Juan C.–Somers, H. L. (eds), Terminology, LSP, and translation studies
in language engineering in honour of Juan C. Sager. Amsterdam, Philadelphia:
John Benjamins, 175–186.
Blakemore, D. 2006. Discourse markers. In: Horn, Laurence R.–Ward, L. Gregory
(eds), The handbook of pragmatics. Oxford: Blackwell, 221–240.
Olohan, M.–Baker, M. 2000. Reporting that in translated English. Evidence for
subconscious processes of explicitation? Across Languages and Cultures 1(2):
141–158.
Becher, V. 2011. When and why do translators add connectives? A corpus-based
study. Target 23(1): 26–47.
Behrens, B. 2005. Cohesive ties in translation: a contrastive study of the Norwegian
connective dermed. Languages in Contrast 5(1): 3–32.
Unauthenticated
Download Date | 2/18/17 8:59 PM

40 Andrea GÖTZ
Degand, L. 2009. On describing polysemous discourse markers. What does
translation add to the picture? In: Simon-Vandenbergen, A. M. et al. (eds), From
will to well: studies in linguistics, offered to Anne-Marie Simon-Vandenbergen.
Gent: Academia Press, 173–183.
Fischer, K. 2007. Grounding and common ground: modal particles and their
translation equivalents. In: Fetzer, Anita–Fischer, Kerstin (eds), Lexical
markers of common grounds. Amsterdam: Elsevier, 47–66.
Foolen, Ad. 2012. Pragmatic markers in a sociopragmatic perspective. In:
Andersen, Gisle–Aijmer, Karin (eds), Pragmatics of society. Berlin; Boston: De
Gruyter Mouton, 217–242.
Fraser, B. 1999. What are discourse markers? Journal of Pragmatics 31(7): 931–
952.
Furkó, B. P. 2015. Perspectives on the translation of discourse markers: a case
study of the translation of reformulation markers from English into Hungarian.
Acta Universitatis Sapientiae, Philologica 6(2): 181–196.
Halliday, M. A. K.–Matthiessen M. I. M., Ch. 2014. Halliday’s introduction to
functional grammar. Fourth Edition. Milton Park, Abingdon, Oxon: Routledge.
Károly, K. 2007. Szövegtan és Fordítás. Budapest: Akadémiai Kiadó.
2014. Szövegkoherencia a Fordításban. Budapest: ELTE Eötvös Kiadó.
Keszler, B. (ed.). 2000. Magyar grammatika. Budapest: Nemzeti Tankönyvkiadó.
Meylaerts, R. 2008. Translators and (their) norms: Towards a sociological
construction. In: Toury, Gideon et al. (eds), Beyond descriptive translation
studies investigations in homage to Gideon Toury. Amsterdam, Philadelphia:
John Benjamins, 91–102.
Mortier, L.–Degand L. 2009. Adversative discourse markers in contrast: the need
for a combined corpus approach. International Journal of Corpus Linguistics
14(3): 338–366.
Niemegeers, S. 2009. Dutch modal particles maar and wel and their English
equivalents in different genres. Translation and Interpreting Studies 4(1): 47–
66.
Robin, E. 2014. Fordítási univerzálék a lektorált szövegekben. PhD dissertation.
Budapest: ELTE.
Schiffrin, D. 1987. Discourse markers. Studies in Interactional Sociolinguistics 5.
Cambridge: Cambridge University Press.
Sources:
Corpus A
Stiefvater, M. 2009a. Shiver. New York: Scholastic.
2009b. Shiver–Borzongás. Szeged: Könyvmolyképző Kiadó.
Unauthenticated
Download Date | 2/18/17 8:59 PM

41Vajon in Translated Hungarian...
Penney, Stef. 2006. The Tenderness of Wolves. Birmingham: Quercus.
2010. Gyengéd, mint a farkasok. Szeged: Könyvmolyképző Kiadó.
Diamand, Emily. 2008. Reavers’ Ransom. Frome: Chicken House.
2009. Kalózok nyomában. Budapest: Agave Könyvek.
Cashore, Kristin. 2009. Fire. New York: Dial Books.
2009. Fire–Zsarát. Szeged: Könyvmolyképző Kiadó.
Corpus B
[All texts are available from: http://www.gutenberg.org and http://mek.oszk.hu.]
Brontë, Charlotte. 1847. Jane Eyre. Available from: http://www.gutenberg.org/
ebooks/1260.
1991. Jane Eyre. Budapest: Európa.
1849. Shirley. Available from: http://www.gutenberg.org/ebooks/30486.
1977. Shirley. Budapest: Európa.
Austen, Jane. 1811. Sense and Sensibility. Available from: http://www.gutenberg.
org/ebooks/161.
1976. Értelem és érzelem. Budapest: Európa.
1817. Persuasion. Available from: http://www.gutenberg.org/ebooks/105.
1980. Meggyőző érvek. Budapest: Európa.
Unauthenticated
Download Date | 2/18/17 8:59 PM