translation, TT, hedges, the translator, source text, av, modal auxiliary verbs, translational, academic text, target, interpretation, target text, Hedging, Swedish texts, vissa, TT Sava och Oisin, English expression, Word-forword Sava, modal auxiliary verb, translation process, literal translation, free translation, transferred, text, English to Swedish, discussion, translated
Translating Hedges A Study of the Translation of Hedges from English to Swedish in an Academic Text Author
: Linda Axelsson Supervisor: Magnus Levin Examiner: Ibolya Maricic Semester: Spring 2013 Subject: English Level: advanced course
Abstract This paper discusses the translation of hedges from English to Swedish in an academic text. Hedging is defined as the use of expressions which modify the level of strength and/or precision in a statement. The aim of the study is to investigate the frequency of different categories of hedges in an academic text and to discuss the challenges encountered in the translation process. The primary source used in this study is a popular science textbook. By translating a number of pages from this text, the translation of hedging expressions is exemplified and analysed. Hedges belonging to categories such as adverbs, lexical verbs, and modal auxiliary verbs are brought up for discussion. The analysis is supported by translation theory as well as corpora and parallel texts in Swedish. The research method
is a combination of a quantitative and a qualitative approach. quantitative data
presenting the overall frequency of hedges and the categorisation of hedges in the source text, is complemented with a qualitative discussion about translational options and challenges. Hedging is identified as a common feature of the current source text. The most frequently used hedges are the adverb perhaps and the auxiliary may, together constituting nearly ј of all hedges in the source text. The discussion illustrates the translator's process in determining the most appropriate translation in different contexts. The vast majority of hedges were transferred to the target text
through literal or word-for-word translation to maintain the stylistic traits of an academic text, but a few examples of omission are also analysed. Although most hedges were transferred, the discussion also shows that the most important factor to consider in the translation process is that the target text preserves the meaning of the source text and that a freer, or sense-for-sense translation, may also be motivated on occasion. Keywords: academic discourse, equivalence, hedging, sense-for-sense, word-for-word. i
Contents 1. Introduction..................................................................................................................1 1.1 Aim......................................................................................................................... 2 1.2 Material and method............................................................................................... 2 2. Background.................................................................................................................. 3 2.1 Hedging.................................................................................................................. 3 2.2 translation strategies
.............................................................................................. 5 3. Analysis......................................................................................................................... 7 3.1 Hedging strategies and their translations................................................................8 3.1.1 Adverbs........................................................................................................... 8 188.8.131.52 Perhaps and maybe.................................................................................9 184.108.40.206 Other adverbs........................................................................................13 3.1.2 Lexical verbs.................................................................................................16 3.1.3 Modal auxiliary verbs................................................................................... 18 220.127.116.11 May........................................................................................................19 18.104.22.168 Must.......................................................................................................20 3.1.4 Phrasal hedges.............................................................................................. 23 3.1.5 Easily overlooked and false hedges..............................................................24 4. Conclusion.................................................................................................................. 27 5. References...................................................................................................................28 ii
1. Introduction The translation of a text from one language into another is a fascinating process. However straightforward the work may seem to an observer, there are many pitfalls in the translation process that a skilled translator has to be aware of, and be able to deal with. One issue for the translator is deciding whether to use a semantic approach, defined by Ingo (2007:87) as dealing with the specific elements in a text which carry meaning, or a pragmatic approach, which Ingo (2007:126) describes as focusing on the communicative function of a text. The translator's dilemma is that a translation should never come across as being exactly that, a translation, but at the same time it needs to stay true to the source text in terms of aim, meaning, style and content. Translators must be in command of an array of different research tools and methods available to make necessary background research and fact checking. They must also possess great knowledge of their languages, both source and target, and be closely familiar with the source and target cultures in order to make adequate adaptations when necessary, as illustrated by the following extract from the translation used in this study:
(1) [p. 165]
Animal-goddesses depicted on jewellery appear elsewhere in early Celtic art. The recent great find of torcs from Snettisham in Norfolk includes one decorated with a goddess of beasts. Scenes like these must represent myths which are entirely lost to us but which were meaningful symbols within the context of religion in Iron Age Europe.
Djurgudinnor avbildade pе smycken dyker upp pе andra stдllen i den tidiga keltiska konsten. Det storslagna fyndet av keltiska halsringar som gjordes i Snettisham, England, i mitten och senare delen av 1900-talet, omfattar en halsring prydd av en djurens hдrskarinna. Scener som denna torde fцrestдlla myter som gеtt helt fцrlorade fцr oss, men som var meningsfulla religiцsa symboler i Europa under jдrnеldern.
This example brings attention to translational issues such as how to translate, clarify or explain specialist terminology (torcs), and instances where it is necessary to make different types of adaptation (early Celtic art, recent, Snettisham in Norfolk, Iron Age Europe). More importantly, (1) includes an example of hedging, which is the topic of this paper. According to Mauranen (1997:115), a hedge can be identified as an expression which modifies the level of strength and/or precision in a statement. In (1), it is the modal auxiliary verb must which functions as a hedge. By adding must to the statement, the level of certainty is weakened. This example, along with other types of hedges, will be discussed further in the analysis. Hedging expressions can pose a challenge to the translator. The identification of hedges is not always straightforward, and what words are perceived as hedges in any
given context is ultimately subjective. When dealing with hedges, the translator needs to take the style and meaning of the source text into consideration while still making sure that the target text fits naturally into the target culture. Hedges are worth studying from a translational perspective as they may be treated differently, and used more or less frequently, in different languages. 1.1 Aim The aim of this study is to investigate the frequency of different categories of hedges in an academic text and to discuss the challenges encountered in their translation. The hedges that will be discussed belong to the following categories: adverbs, lexical verbs, modal auxiliary verbs, phrases stating limitation, "easily overlooked hedges", and false hedges. Is hedging used frequently enough for it to be identified as a genre-specific trait of academic discourse in English as well as in Swedish? Are hedges generally transferred from the source text to the target text, and if so, what translation strategies are most appropriate in doing so? What types of hedges are most common in academic writing? 1.2 Material and method The source text (henceforth ST) and primary material for this translational study is a popular science textbook Celtic Goddesses. Warriors, Virgins and Mothers by Miranda Green, published in 1995 by British Museum Press. Dr Green is Professor of Archaeology at Cardiff University
, and an authority on Celtic religion, mythology, and iconography (Green, 1995:back sleeve). The main aim of a textbook is to inform its reader on a particular topic. It has what Ingo (2007:127) calls an informative function, which means that the pragmatic aspect is put first: the translator's main task is to ensure that the communication with target text (henceforth TT) readers works as well as it did with ST readers by conveying the necessary information. According to Hellspong & Ledin (1997:207f, 237f), the style of an informative text is usually rather formal, with formal vocabulary and without the use of informal stylistic traits such as contractions. The author is distanced from his/her readers and not visible in the text in the first person. An informative text is usually also perceived as objective, as there is no visible author whose voice comes through in the 2
text. The use of hedging is one way for the author to convey this objectivity, by distancing his/her claims from a personal to a more general level. Instead of saying "I think" or "I believe", more formal and impersonal phrases like "it has been claimed that" or "it appears that" are used. These defining traits are true of the ST used in this study. The targeted ST and TT readers are English and Swedish speaking university students
of theology, archaeology or history, or people with a specific interest in Celtic religion and culture. These people will read the book to gain information about the Celts and their cultural and religious tradition
s. Secondary material was also used in this study. It first and foremost contains texts about hedging, but also about translation methods. The ST has not been translated using one specific translational approach, but dictionaries, Swedish and English corpora
, Swedish texts about the Celts, and literature about translational method have been used as reference material and support for translational choices. The translation of pp. 160173 of Green's book forms the basis for the discussion about hedging. The examples included in this paper were chosen to represent the most common and/or notable examples of hedging found in the ST. The analysis combines a quantitative and a qualitative method, presenting numbers of occurrence of specific hedges as well as analysing specific translational options. The quantitative discussion illustrates the general frequency of hedges as well as the frequency of particular hedges in the ST. It also presents numbers on how many hedges were transferred to the TT and how many were omitted. As the material has a limited range, results cannot be generalised on a larger scale. However, they are complemented by a qualitative discussion which comments on specific translational choices and the way in which the hedges influence the style of the text. In this discussion, Munday's (2008:19f) concepts of word-for-word (literal) and sense-for-sense (free) translation have been used, as well as two of Vinay & Darbelnet's (Munday, 2008:57f) oblique (free) translation procedures, namely transposition and equivalence. 2. Background 2.1 Hedging According to Markkanen & Schrцder (1997:V) and Clemen (1997:235), hedging was 3
first identified as a linguistic phenomenon in the mid-1960s. However, it was linguist George Lakoff (1972:195) who first defined hedges more specifically as "words whose meaning implicitly involves fuzziness words whose job is to make things fuzzier or less fuzzy." In this paper, a hedge is defined as a word or phrase which expresses caution, hesitation or uncertainty regarding the truth value
or level of precision in a statement. According to Schrцder & Zimmer (1997:253), hedging research has gone from a focus on semantics in the 1970s, to a pragmatic focus in the 1980s. Initially, hedging seems to have been regarded as a primarily negative phenomenon, expressing imprecision and fuzziness, but over time it has come to be perceived as a means of expression which authors use purposefully to express politeness and/or uncertainty. Historically, research on hedging has dealt mainly with two aspects, according to Mauranen (1997:115) and Hyland (1998:9), namely writers' reasons for using it, and the different means of expressing hedging. The numerous ways of expressing hedging will be discussed in detail further on in the analysis. Hedges originate in the author's intentions with his/her text. The problem in the identification and translation of hedges is that the author's intentions are rarely known to the translator. One may speculate, however, that Green, the author of the ST, needs to use hedging as a precautionary measure, as the ST deals with the belief systems of the ancient Celtic civilizations in Central Europe
and on the British Isles
more than 2,000 years ago. Any conclusions drawn from archaeological finds, myths, or texts written by contemporary Greek or Roman authors
, are speculative and subject to interpretation. These conclusions can therefore not be said to be definite or unquestionable, and it would be hazardous for Green to present them as such. Another reason for Green to use hedging may be that it is customary in this kind of academic text. There is wide recognition of hedging as a feature of academic and scientific writing
: Hedging is central to academic writing as it expresses possibility rather than certainty and collegiality rather than presumption. Scientific claims are rarely made without interpretive statements and these involve both assessments of probability and judgements concerning the impact of linguistic choices on readers. (Hyland, 1998:viii) Apart from Hyland, researchers such as Mauranen (1997:116), Hellspong & Ledin (1997:189f), and Varttala (2002:143) acknowledge hedges as a genre-specific trait in academic discourse. The field of scientific research
needs to use hedges due to its very 4
nature: predictions are uncertain, results preliminary, and findings modified over time. As Hyland (1998:90) points out, researchers cannot make claims which are inadequate or false. Hence, they use hedges to reduce the risk of losing face if proven wrong. Hedges appear in almost 54 per cent of all sentences in the ST. This finding is supported by Hyland (1998:53), who states that hedges appear in between and Ѕ of all sentences in academic texts. Thus, the many hedges in the ST could be seen simply as Green's way of adhering to the style of the academic text genre. However, due to the nature of the research presented in the ST, it probably has more to do with the uncertainty of interpretations of different archaeological findings, in combination with an unwillingness or inability to make assertive claims. To make definite claims of truth when dealing with the interpretation of ancient cultural and religious practices seems hazardous to say the least, and the difficulty in interpreting archaeological evidence can be seen throughout the ST. Using Hyland's (1998:253f) conclusion, hedges allow writers to put forward their propositions with a certain degree of reservation about content, or concerning the extent of personal commitment they wish to invest in them. They also allow writers to express a particular attitude to their readers in order to attract their interest and meet readers' expectations of modesty and negotiation. Hyland (1998:79) states: [...] hedges are rhetorical means for projecting due caution, modesty and humility when making statements, and their removal is a major linguistic means of conferring greater certainty on propositions. These strategies are therefore central to the whole enterprise of science, [...]. 2.2 Translation strategies For a translator of non-fictional texts there are numerous aspects to consider when dealing with hedges. First of all, hedges need to be properly identified in the ST before one can move on to consider whether they should be transferred to the TT. Hedging can be expressed through a vast array of linguistic means: modal auxiliary verbs (may), adjectives (possible), adverbs (perhaps), modal lexical verbs (appear to), nouns (a claim), if-clauses (if goddess she is), and compounds (seems probable). According to Hyland (1998:3), words or expressions from these categories function as hedges when they are used to bring attention to the reduced strength of an utterance. However, no words or expressions are inherently hedges. Instead, as Hyland (1998:244) points out, whether or not a word or expression should be considered a hedge is ultimately 5
determined by context. Once hedges have been properly identified as such, the Next Step
in the process is determining how to transfer them to the TT, and whether or not it is necessary to do so in the first place. Translation theory provides some hints regarding this. However, there seem to be no clear rules on how hedges should be handled in the translation process. Munday (2008:19f) uses the terms word-for-word and sense-for-sense to represent literal and Free Translation
. The boundary between the two is not clearly defined, but in this paper a literal translation is defined as a translation which follows the ST both structurally (sentence structure
), lexically (choice of words) and pragmatically (communicative function), while the primary concern of a free translation is to convey the message of the ST, regardless of style and form. Hedges and other linguistic features are treated differently depending on which translational approach is used. In a word-for-word translation, all elements in the ST are transferred to the TT, while this may not be the case in texts which allow for greater translational freedom, i.e. sense-for-sense. In this context it is relevant to bring up Ingo's (2007:154) pragmatic discussion. It is centred around the concept of bruksmotsvarighet, which is a synonym to Vinay & Darbelnet's concept of equivalence, as cited in Munday (2008:58). In using this translational approach, a translator should first and foremost transfer the meaning or message of the ST to the TT, and not necessarily make direct translations of all semantic components. According to Ingo (2007:133), a translator using a pragmatic approach should process the text and make necessary adaptations to make sure that the TT fits the target culture with regards to both form and content. From this pragmatic perspective, it would be possible to omit hedges as they are usually not the words which carry the primary meaning of a text. As reader acceptance is an important factor in the academic accreditation process, scientific writers must consider not only the plausibility of their claims, but also their effect on readers. But how does the translator achieve the same effect on TT readers as the original author did on ST readers? This aspect is discussed by Cassirer (2003:33), who states that as it is virtually impossible to know what effect authors aim to achieve when writing their texts, it is equally impossible to compare the effect on ST readers with the effect on TT readers. The only way of doing that is to conduct a readerresponse study where readers get to study both the ST and the TT and compare them. As Cassirer (2003:125f) points out, how written words, especially those that like hedges can be considered vague in nature, are interpreted, is ultimately dependent on the frame 6
of reference of the readers. Thus, the translator needs to consider and try to replicate not only the effect the author aimed to achieve with his/her text, but also the way in which TT readers will understand and receive it. The main challenge in the translation of the current ST was how to deal with the large number of hedges. Should they all be transferred to the TT and if so, how are they best translated? Based on the previous discussion one might argue that when translating an academic text, all its linguistic features, including hedges, should be transferred to the TT in order to maintain its genre-specific traits. This will be discussed further in the analysis.
Table 1: Number of hedges in ST, categorization, transference frequency, and translation strategy used
Number of words in ST Number of hedges in ST Adverb hedges Lexical verb hedges Modal auxiliary verb hedges Phrasal hedges Adjective hedges If-clause hedges Compound hedges Pronoun hedges
5,644 171 67 34 30 18 10 8 3 1
Translated word- Translated sense-
transferred to TT for-word
The translation of a selection of these hedges will be further discussed in the analysis. The examples were chosen from the categories with the highest frequency of hedges, namely adverbs, lexical verbs, modal auxiliary verbs and phrases stating limitation. Adjectives and if-clauses are discussed in the context of "easily overlooked" or false hedges. Compounds and pronouns were consciously left out of the analysis as they are used on only four occasions in the ST and therefore not to be considered representative or typical of the ST.
The author's frequent use of hedging becomes obvious immediately when reading the text:
(2) [p. 169]
Snakes are carnivorous; they are hunters, Ormar дr kцttдtare; de jagar och dцdar pе
and they kill violently, by a poisonous bite ett vеldsamt sдtt genom giftbett eller genom
or by constriction. Their predatory nature att krama ihjдl bytet. Deras rovlystna natur
perhaps gave rise to negative, fearful
gav kanske upphov till en negativ,
perceptions of the snake, particularly if the skrдmmande uppfattning av ormen, sдrskilt
danger to humans were fully recognised. om dess farlighet fцr mдnniskan
The snake is essentially an earthbound
uppmдrksammades till fullo. Ormen дr en i
creature, gliding close to the ground with a huvudsak jordbunden varelse som glider tдtt
rippling, water-like motion, able to
intill marken med en vеglik rцrelse, kapabel
insinuate itself in and out of minute cracks att nдstla sig in och ut genom minimala
or holes in rocks. Maybe this behaviour sprickor och hеl i klipporna. Kanske ledde
created a link in people's minds between detta beteende till att mдnniskorna
the snake and infernal symbolism. Fertility- fцrknippade ormen med underjorden. Ormen
imagery was also associated with the
fцrknippades ocksе med fruktbarhet: honan
serpent: the female gives birth to a large fцder mеnga ungar samtidigt, hanen har tvе
number of young at once; the male has a hemipenisar, och givetvis kan reptilens form
multiple penis; and of course the shape of ha givit den en fallisk symbolik. Slutligen
the reptile may have endowed it with
betraktades ormens skinnцmsande som en
phallic symbolism. Finally, the practice of allegori fцr fцrnyelse och pеnyttfцdelse.
skin-shedding was seen as an allegory of Dцden, underjorden, fruktbarhet, lдkande och
regeneration and rebirth. So death, the
fцrnyelse kunde alltsе alla symboliseras av
underworld, fertility, healing and renewal bilden av detta enda djur.
could all be symbolised by the image of
Fцrbindelsen mellan ormar och gudinnor
this one beast.
verkar frдmst ha handlat om symbolik med
The association between serpents and fruktbarhet, lдkande och fцrnyelse.
the goddesses seems to have been
principally concerned with the symbolism
of fecundity, healing and regeneration.
Seven hedges are found in (2), which should be regarded as a high number in a text extract this short, containing only 179 words. With this example, I want to illustrate that it is impossible to read the ST without noticing the many hedges. It also shows the variety of hedging expressions used in the ST. We will now take a closer look at some of these hedges and the strategies used in their translation.
3.1 Hedging strategies and their translations 3.1.1 Adverbs There are many different ways for a writer to express hedging. Hyland (1998:130f,135) states that of the major grammatical classes, adjectives, adverbs and nouns are used to express hedging in more than 50 per cent of cases in scientific research articles
. The use of hedging nouns is non-existent in the ST, and only 10 adjective hedges have been found, but adverbs are used to express hedging all the more frequently. A total of 67 adverb hedges were identified in the ST. All of these were transferred to the TT and all 8
but two were translated with corresponding Swedish adverbs. The two exceptions are discussed in (15) and (16) below. 22.214.171.124 Perhaps and maybe The adverb perhaps appears 17 times as a hedge in the ST, but the synonymous adverb maybe appears on only one occasion. A few examples of how perhaps/maybe were translated are presented below:
(3) [p. 162] (4) [p. 169]
[...] she holds the great dish towards the sky, as if she is perhaps receiving her supplicants' offering of blood or wine.
[...] hon hеller den stora skеlen mot himlen, som om hon kanske mottar offergеvor i form av blod eller vin frеn de bedjande.
Maybe this behaviour created a link Kanske ledde detta beteende till att
in people's minds between the snake mдnniskorna fцrknippade ormen med
and infernal symbolism.
In (3) and (4), perhaps and maybe are translated kanske, which was chosen as the most neutral and most frequent of the possible solutions. Kanske renders more than 976,000 hits in the Swedish corpus Korp, and must therefore be considered the most natural Swedish equivalent as synonyms such as eventuellt receives only 97,232 hits. The concept of equivalence is defined as a linguistic key term in translational method by Jakobson, cited in Munday (2008:37), where he discusses how equivalence in meaning can be achieved in different languages. According to Jakobson, there is no full equivalence between words (or so-called code-units) in different languages, so the translator's task is to substitute the source language message with an equivalent message in the target language. This pragmatic approach has been kept in mind when translating the many perhaps in the ST. The key issue is not whether or not the expressions chosen as translations of perhaps are its full equivalents, but that the message of the ST is fully transferred to the TT. According to synonymer.se [www], other possible translational options for perhaps, some of which have been used elsewhere in the text although not as frequently as kanske, are mцjligen, mцjligtvis, eventuellt, mеhдnda, and kanhдnda:
(5) [p. 165] (6) [p. 167]
[...] his size perhaps suggestive of his inferior, human status [...]
[...] eventuellt antyder hans storlek hans underlдgsna, mдnskliga status [...]
The beast stands beneath a tree which Djuret stеr under ett trдd som
is perhaps symbolic of the animal's mцjligtvis symboliserar dess naturliga
However, none of these synonyms are as common as kanske. As stated above, the one that comes closest is eventuellt with approximately 97,000 hits. The main reason for using some of these synonyms occasionally was to lessen the feeling of repetitiveness due to the high frequency of perhaps in the ST. Another option which was considered is the adverb nog, also common in Swedish and rendering more than 666,000 hits in Korp. However, judging from the examples presented in Korp, nog is more often used in informal contexts, thus considered less appropriate in this kind of academic text. The translations in (3)(6) are literal, but with slight differences in connotations between the chosen words. Connotation is defined by Ingo (2007:109) as associative content, and complements denotation, i.e. the basic meaning of a word. In his discussion of how words change in frequency over time, Lindquist (2009:59ff) presents statistics regarding the use of perhaps and maybe. According to his study of the Corpus of Contemporary American English
(COCA), perhaps has a clear advantage over maybe in academic writing, the ratio being 262 to 28 tokens per million words. This finding is reflected in the frequency of perhaps and maybe in the ST, perhaps being used 17 times and maybe only once. A reason why academic writing favours perhaps to maybe could be, according to Lindquist (2009:62), that the modal auxiliary verb may in combination with the verb be, is used instead of maybe to express epistemic meaning, i.e. to what extent something is likely. May + be is used more frequently in academic writing than in other subcorpora. The high frequency of hedges in the ST, particularly perhaps, signals a certain hesitation or reluctance to state absolute truths or facts. This becomes especially evident in passages where numerous hedges are found:
(7) [p. 171]
The light and dark aspects of the snake's Ormens ljusa och mцrka symboliska aspekter
symbolism are perhaps manifest also in mдrks kanske дven dдr den fцrekommer med de
its appearance with the Romano-Celtic romersk-keltiska modergudinnorna, sдrskilt i
mother-goddesses, especially in the
det tyska Rhenlandet. Dessa keltisk-germanska
Rhineland. These Celto-Germanic
moderliga tretal fцrknippas ibland med bilden
maternal triads are sometimes associated av en orm som slingrar sig om ett trдd,
with the image of a snake curled round a symbolik som pеminner om ormen som Adam
tree, symbolism which resembles that of och Eva mцter vid kunskapens trдd i Edens
the serpent and Tree of Knowledge
lustgеrd i Gamla testamentet. Ett altare till
encountered by Adam and Eve
Aufaniae som var Rhenlandets lokala
Garden of Eden of the Old Testament. On modergudinnor, har pе baksidan en bild av ett
the reverse of one Rhenish altar to the stort trдd, eventuellt en ek eller ett piltrдd, som
local mother-goddesses, the Aufaniae. a strдcker sina grenar mot himlen med en orm
great tree, perhaps an oak or willow,
ringlande om dess stam. Bilden avbildar
stretches its branches to the sky, its trunk mцjligen livets trдd som vaktas av ormen och
encircled by a snake. The image may be еterspeglar kanske kopplingen mellan den цvre
that of the Tree of Life, guarded by the vдrlden och underjorden: [...]
serpent and reflective, perhaps, of the link
between the upper and lower worlds: [...]
All hedges in (7) were kept and transferred to the TT based upon arguments regarding the nature of academic and scientific texts as discussed in section 2. However, to lessen the feeling of repetitiveness while keeping the expression of hesitation as to the interpretation of the symbols discussed, perhaps was translated kanske as well as eventuellt. These instances where there are numerous hedges in one paragraph, give the text a feeling of indecisiveness and uncertainty. Nothing can be said to be fact, but everything is questionable and subject to one's personal interpretation. That is naturally the case when dealing with ancient belief systems, but it brings the question of omission to the fore. It seems possible to create a somewhat smoother, less repetitive TT if some, but far from all, hedges are simply omitted. Removing a few hedges would not alter the tone or style of the ST, and would still allow the TT to signal the same uncertainty as the ST. Is the wish to create a less circumstantial and easier to read TT a strong enough argument against considerations as to the nature and genre-specificity of the ST and the level of uncertainty in the statements presented? As stated in section 2, hedges are considered an important feature of academic writing and should from this perspective be kept to maintain the style of the text. However, Kranich (2011:77) has conducted a study which shows that hedges are used much less frequently in German than in English popular science texts. Kranich claims that there are different communicative preferences in English and German: English texts are more indirect and reader-oriented, while German texts are direct and content-oriented. Thus hedges are used more frequently in English. Kranich (2011:78) illustrates her claim by a number of examples; for instance
she shows how the English expression may solve is translated simply lцst (solves) in German, thus representing the indirect vs. direct approaches she considers typical of English and German. As Swedish and German are similar in many aspects and belong to the same language family, this difference in the use of hedging may be true of English vs. Swedish as well. Based upon this argument, hedges can be omitted from the TT to better follow the writing conventions in the target language culture. On the other hand, without the use of hedging researchers need to provide strong evidence for their claims, which can be difficult as pointed out above. When discussing genre-specific traits, it is useful to compare the ST to Swedish texts on the same subject, but these are hard to find. Although there are several similar texts about the Celts available in Swedish, most of these have been translated from other languages, most notably French or English. A good example is Druidens dцd. En arkeologisk thriller by Ross & Robins. When comparing the chapters dealing with Celtic religion and archaeology, similar hedges as the ones found in Green's book are easily identified:
(8) [p. 136) (9) [p. 137]
Den hade fallit i bitar, men elegansen hos den enda bevarade handen tydde pе en hцg social stдllning hos offret. Andra tecken visar pе en offerdцd. Dessa kan ha avlдgsnats under nеgot skede av offerriten, antagligen efter det att han fеtt sin coup de grвce.
However, hedges found or missing in texts translated to Swedish from another language may be a result of literal translation just as well as pointing to the genre-specificity of Swedish academic texts. Access to the originals would be necessary to make comparisons in order to draw any conclusions about genre-specific traits. I have had two Swedish original texts at my disposal to compare the ST to: Harrison's article about the Celts in the Swedish online encyclopedia Nationalencyklopedin [www], and Gцrman's book Nordiskt och keltiskt. Sydskandinavisk religion under yngre bronsеlder och keltisk jдrnеlder. These texts use hedging in much the same situations as the ST, i.e. interpretations of archaeological finds or speculations about Celtic rituals and society:
(10) [NE p. 5]
Gravfynd tyder pе att styrande funktioner innehades av jordдgande stormдn.
Av dessa fцremеl framgеr att djuren ifrеga vanligen hцr samman med kult av solen.
[Gцrman p. 31]
As Hyland (1998:90) states, hedges are an integral component of scientific writing, and judging from (10) and (11), this seems to be true of these types of texts in both English and Swedish. It may well be as Kranich (2011:77) claims, that English uses hedging to a greater extent than other languages, but as the examples above show, the use of hedging is frequent in Swedish academic texts as well, although the percentage is possibly not as high as the 54 per cent found in the ST. This would, however, support a translational approach where most, but not all hedges are transferred to the TT.
126.96.36.199 Other adverbs Apart from perhaps/maybe, a number of other adverbs are used as hedges in the ST, most notably probably and sometimes, appearing 5 and 10 times respectively:
(12) [p. 165] (13) [p. 171]
Word-for- [...] and the date is now regarded as [...] sannolikt tillverkad under andra
probably second or first century BC. eller fцrsta еrhundradet f. Kr.
These Celto-Germanic maternal triads Dessa keltisk-germanska moderliga
are sometimes associated with the
tretal fцrknippas ibland med bilden
image of a snake curled round a tree, av en orm som slingrar sig om ett
The translation of the hedges in (12) and (13) is literal as they have suitable and conventional equivalents in Swedish. Sometimes is translated with the equally neutral ibland, which contains the connotation "more seldom than often". It receives more than 323,000 hits in Korp and must therefore be considered common and conventional. Probably is translated sannolikt, which implies a fairly strong possibility and generates more than 57,000 hits in Korp. This is further supported by NE.se, which uses sannolikt in the description of exactly the same object as the ST, namely the Gundestrup cauldron:
(14) [p. 4]
Silverplatta frеn den s.k. Gundestrupkittlen, sannolikt tillverkad under bцrjan av fцrsta еrhundradet f.Kr.
According to Hauck (2009:479), probably can be paraphrased as there is the possibility that and suggests a chance greater than 50 per cent, which seems reasonable to me. A common synonym to sannolikt is fцrmodligen, which generates more than 79,500 hits in Korp. The choice of sannolikt instead of fцrmodligen in (12) was based on its use in the parallel text and the fact that sannolikt has fewer syllables and therefore creates a
slightly smoother TT sentence. Several other adverbs are used as hedges less frequently in the ST. One example is especially interesting:
(15) [p. 168]
Translational ST approach
So the meeting between Finn and Sava may not have been entirely fortuitous, although clearly the hero had no part in her enchantment.
TT Sе mцtet mellan Finn och Sava kanske inte var en ren tillfдllighet, дven om hjдlten uppenbart inte var delaktig i hennes fцrtrollning.
In (15), the ST adverb hedge has not been translated with an adverb in Swedish. Instead, entirely has been translated with the adjective ren, because it is a nice solution to translate the whole ST expression entirely fortuitous with the collocation en ren tillfдllighet. "Ren tillfдllighet" receives 495 hits in Korp, and the corpus shows the expression being used in contexts similar to the one in (15). The meaning of the ST sentence, that the event described was not a sheer coincidence or something that happened by accident, is transferred to the target language and expressed through equivalence in Vinay & Darbelnet's use of the term. According to Vinay & Darbelnet as cited in Munday (2008:58), equivalence is used when the same situation is described in different languages by different stylistic or structural means, and this method is particularly useful when translating fixed expressions such as idioms and proverbs, as seen in the example above. To illustrate the variety of adverb hedges in the ST, a few additional examples should be commented on:
(16) [p. 162]
Translational ST approach
Hunting, sacrifice, war and fertility are all seemingly present.
TT Jakt, offer, krig och fruktbarhet tycks alla vara representerade.
Seemingly in (16) is the other of the two adverbs which was not translated with another adverb, but with the verb tycks. This can be seen as an example of Vinay & Darbelnet's translation procedure transposition, as presented in Munday (2008:57). Transposition is the process of exchanging one part of a sentence for another without changing the sense. Transposition is a common translational change, and works in many different ways:
verb to noun, adverb to verb, etc. The translational choice is supported by the article about the Celts in NE.se, where tycks occurs frequently (see further in 3.1.2).
(17) [p. 167] (18) [p. 169]
In addition, there are rare instances of Dessutom finns nеgra fе bilder av en
images which depict goddesses
gudinna med horn, trots att detta
wearing antlers, although this is
vanligtvis дr ett privilegium
usually the prerogative of male
fцrbehеllet manliga gudomar.
This is complex and contains
Denna дr komplex och innehеller
messages which are both positive and budskap som дr sеvдl positiva som
negative, based partly on observation negativa, delvis baserade pе
of the behaviour of the snake in the iakttagelser av vilda ormars beteende.
Usually in (17) is translated vanligtvis to express a relatively frequent occurrence. Partly is translated delvis in (18), and has the connotation "only in part, not the majority of".
(19) [p. 171f] (20) [p. 173]
Translational ST approach
In the Celtic system, however, particular birds were considered sacred because they somehow epitomised the personae of the divinities with whom they were associated.
[...] presumably brought to the temple by hopeful or grateful pilgrims [...]
TT I det keltiska systemet betraktades dock vissa fеglar som heliga eftersom de pе nеgot sдtt personifierade karaktдrerna hos de gudomar de fцrknippades med. [...] fцrmodligen fцrda till templet av hoppfulla eller tacksamma pilgrimer [...]
Somehow in (19) expresses uncertainty regarding method or manner and is translated pе nеgot sдtt. Presumably in (20) is translated fцrmodligen, expressing a relatively high level of probability. The translation of the hedges in (17)(20) is literal, and the hedges have been translated with expressions containing the same connotations of (un)likelihood or (in)frequency and chosen to be the best equivalents, or in Ingo's (2007:154) words, bruksmotsvarigheter, of each term, based on their frequency in Korp. This section has illustrated the frequency and variety of adverb hedges in the ST, and it has exemplified and discussed the translational choices of some of these. Another important group of hedges is lexical verbs, which will be discussed in the next section.
3.1.2 Lexical verbs Lexical verbs is a group which covers all verbs except the auxiliaries (discussed in 3.1.3). Hyland (1998:119) states that lexical verbs such as suggest, indicate, and predict are the most common means of expressing mitigation. The ST contains 34 lexical verb hedges (suggest, appear to, perceive, regard, imply, associate, indicate, seem to, see as, evoke, resemble, consider). Of these, 32 were transferred to the TT. All but one were translated with verbs, and the only exception is discussed in (26). The modal lexical verbs appear to and seem to occur as hedges 10 and 5 times respectively:
(21) Word-for[p. 161] word
The Classical attitude to both women Den klassiska synen pе bеde kvinnor
and animals appears to have been och djur verkar ha varit annorlunda
different from that of the Celts.
(22) Word-for[p. 170] word
In a shrine whose main power seems to have been to heal eye-afflictions, the torch may reflect light after the darkness of disease and clear vision in place of defective sight.
I en helgedom vars frдmsta styrka tycks ha varit att bota цgonеkommor kan facklan ha representerat ljuset efter sjukdomens mцrker, och tydligt seende som ersдtter dеlig syn.
Tycks and verkar are the translations chosen in (21) and (22). It would also have been possible to use fцrefaller. However, verka is chosen as it receives almost 360,000 hits in Korp, a big difference from the 22,595 generated by tyckas, and the 26,978 by fцrefalla. The choice of tycks is supported by the article in NE.se which shows tycks to be a conventional hedge in Swedish:
(23) [p. 1] (24) [p. 2]
[...] delvis p.g.a. att den дldsta latиnetiden (400-talet f.Kr.) uppvisar en kultur vars sociala bas tycks ha liknat hallstattkulturens. [...] kelter frеn vдster tycks ha erцvrat Bцhmen [...]
The expressions appear to and seem to present a hesitant possible interpretation on behalf of the author. Symbols and practices of ancient peoples are to be interpreted through our limited understanding of their culture and religion, so a hesitant guess seems to be a suitable way of doing it. As Hyland (1998:253f) states, researchers use so called content-oriented hedges to express reservations concerning the content of their propositions, to protect themselves from possible criticism and to help them save face if their claim is refuted. This can be compared to reader-oriented hedges, which are used
with the reader in mind, expressing politeness or inviting the reader to a discussion. As Kranich (2011:78) states, English tends to use reader-oriented hedges, while German (and possibly Swedish) prefers content-oriented hedges. According to Hauck (2009:480), the use of expressions of uncertainty or probability makes a person more reliable and in possession of greater persuasive power over time compared to people who tend to express certainties and are then proven wrong. It proves that one is not inclined to draw hasty conclusions and express certainty prematurely. This aspect is further developed by Meyer (1997:21), who draws attention to the paradox of hedging strategies. In speech, hedging signals weakness, but in academic discourse, it can do the opposite and actually work to strengthen an argument. In line with this reasoning, most of the hedges in the ST were transferred to the TT. Other lexical verbs are used as hedges in the ST:
(25) [p. 161f]
The stags may be the victims of a Hjortarna kan vara offer fцr en jakt
hunt or a sacrifice (or both): the eller en offerritual (eller bеda delar):
presence of armed men implies nдrvaron av bevдpnade mдn antyder
the former, the presiding goddess det fцrstnдmnda, den nдrvarande
gudinnan det sistnдmnda.
In (25), implies is translated literally as antyder. There are no other translational options that express the same hesitant conclusion. The whole sentence has a strong feeling
of uncertainty due to the fact that it contains three hedging expressions: may, or both, and implies. The use of may is discussed in detail in section 188.8.131.52, and the expression or both is an obvious hedge, which protects the author from being accused of making a false statement.
(26) [p. 164]
Both the torc and one of the
Bеde halsringen och en av
bracelets are decorated with
armringarna pryds av bildkonst med
iconography which is suggestive of kopplingar till en fеgelgudinna: [...]
imagery associated with a goddess
of birds: [...]
In (26), associated with is translated with the prepositional phrase med kopplingar till. This is the only instance where a lexical verb hedge has not been translated using a corresponding verb in Swedish. This approach is another example of Vinay and Darbelnet's procedure transposition (compare ex. 16), cited in Munday (2008:57),
where a word/expression from one category is replaced by a word/expression from another category in the translation (i.e. verb to noun, adverb to verb, etc). The sentence has been restructured and made less circumstantial by omission of the adjective hedging suggestive of, also present in the ST sentence. As the sentence already contains a verb hedge, the adjective hedge was omitted. This strategy can be seen as an example of Ingo's (2007:124) concept implicitation (implicitgцrande), where words or expressions are consciously left out of the translation if the semantic content can be read between the lines.
(27) [p. 173]
[...] its black colour and its
[...] dess svarta fдrg och pеstеdda
reputed cruelty towards other birds grymhet gentemot andra fеglar [...]
In the last example in this section, reputed is translated pеstеdda. Here, the idiom ha rykte om sig, which carries the same connotations as the ST expression, was considered. However, the idiom did not fit smoothly into the sentence, so pеstеdda was chosen instead. Pеstеdda implies, in the same way as reputed, that this is something that people believe or claim to be true. To some extent, it draws the attention away from the author and signals that this is something that others believe. Lexical verbs used as hedges appear frequently in the ST which has been indicated by the examples in this section. In the next section, the modal auxiliary verb hedges may and must will be discussed.
3.1.3 Modal auxiliary verbs The term modality is defined by Holmes (1982:10) as the speaker's attitude towards what is being said: it expresses ability, likelihood, permission and obligation. Especially interesting is the subcategory epistemic modality, which deals with strategies for expressing different degrees of certainty. According to Hyland (1998:105f), modal auxiliary verbs are frequently used as hedges. Modal auxiliary verbs are verbs such as may, might, can, could, should, would, ought, must, will, and shall. A total of 30 modal auxiliary verb hedges were identified in the ST. Of these, 23 were translated with corresponding Swedish auxiliaries, one was omitted which is discussed in (36), and the remaining six were translated with adverbs or compound expressions. 18
184.108.40.206 May The use of one particular modal auxiliary verb, may, is frequent in the ST:
(28) [p. 161] (29) [p. 162]
The stags may be the victims of a hunt or a sacrifice (or both): [...]
Hjortarna kan vara offer fцr en jakt eller en offerritual (eller bеda delar): [...]
The imagery of fertility is enhanced by the phallic symbolism of the male attendant or suppliant, who may even have represented the dead man himself: [...]
Bilderna av fruktbarhet fцrstдrks av den falliska symboliken hos den manliga fцljeslagaren eller tjдnaren, som till och med kan ha representerat den dцde mannen: [...]
May is, without comparison, the most frequently used modal auxiliary verb hedge in the ST, and it is one of the most frequently used hedges overall, occurring a total of 25 times. This finding is supported by Hyland (1998:116), who states that may and might are the most commonly used modal auxiliaries and considered prototypical hedges, appearing equally frequently in all domains of scientific research writing. Hyland (1998:106) goes on to say that may and might can be used interchangeably to indicate a fifty-fifty assessment of possibilities and can be paraphrased I believe/perhaps. However, in my opinion, might seems to signal weaker possibility than may. In both examples above, may has been translated literally by using the corresponding modal auxiliary kan. However, kan is also translated with the more obviously similar modal auxiliary can. According to Meyer (1997:37), can and may are very similar, but not entirely synonymous. In his opinion, may can always be replaced by can, but not vice versa. May implies alternativity, while can has a more general meaning. As suggested by Hyland, replacing may with might as well as I believe and perhaps, is possible in the examples above:
The stags might be the
The stags, I believe, are victims The stags are perhaps the
victims of a hunt or a
of a hunt or a sacrifice (or
victims of a hunt or a
sacrifice (or both): [...]
sacrifice (or both): [...]
The imagery of fertility is The imagery of fertility is
The imagery of fertility is
enhanced by the phallic enhanced by the phallic
enhanced by the phallic
symbolism of the male
symbolism of the male
symbolism of the male
attendant or suppliant, who attendant or suppliant, who, I attendant or suppliant, who
might even have
believe, even represented the perhaps even represented
represented the dead man dead man himself: [...]
the dead man himself: [...]
However, I disagree with Hylands's conclusion that may, might, I believe and perhaps all indicate a fifty-fifty possibility. May and perhaps are neutral, but in my opinion, I believe indicates a higher level of certainty than the other three, a possibility greater than 50 per cent. I believe, in contrast to the other three expressions, includes a personal pronoun which serves to strengthen the argument: this is something I consider to be true. I also believe that might signals greater hesitation than may, as stated above. This is supported by Hyland (see below). In spite of the fact that may and might are considered synonymous, may appears more than twice as often as might in academic writing according to Hyland (1998:106). This is supported by the frequent occurrence of may in the ST, whereas might is not used at all. May is also the only modal auxiliary verb which appears more frequently in academic research writing than in other genres. When working with the translation of the ST, I considered if there was in fact any difference between how one should translate may and might. The neutral may most often became the equally neutral kan in the Swedish translation. According to Hyland (1998:117), might can be regarded as a more "remote" form of may which expresses a higher degree of conditionality or tentativeness. According to Hyland (1998:118), this is further indicated by the fact that might often occurs in combination with other tentative verbs: might speculate, might be suggested. I did not, however, find any example of this in the ST. In all places where may appears as a hedge in the ST, it has been transferred to the TT. One of the main reasons for choosing to keep the numerous hedges expressed through may was that hedges belong to this kind of academic research writing, an aspect which has been discussed previously in this paper (see Section
2). Hence, a decision to omit hedges in the TT would alter the style of the text and make it less genre-specific, something which did not seem right as it is important for translations to stay true to the style of the ST. Even though Kranich (2011:77) argues that English uses hedging more frequently than German, and possibly Swedish, the Swedish parallel texts support the frequent use of hedges. 220.127.116.11 Must Interestingly, the ST contains two examples of the modal auxiliary verb must used as a hedge. They will be presented and discussed in this section. 20
(32) [p. 160]
This led inevitably to a respect and Detta medfцrde ofrеnkomligen en reverence for the creatures inhabiting respekt och vцrdnad fцr varelserna what must have been perceived as a som bebodde det heliga landskapet. sacred landscape.
According to Meyer (1997:33), must is an unusual hedge since there is a more cautious way of expressing the necessity implied by must, and that is by using the modal auxiliary should. However, no examples of should as a hedge were found in the ST. In (32), the hedge must (have been perceived) has been omitted in the translation. This is actually one of only six instances where the ST hedge has not been transferred to the TT. In my opinion, there are no strong enough reasons to transfer the hedge in (32) to the TT. Even without the hedge, the reader will understand that the landscape is not sacred in its very nature, but that it was perceived as such by the Celts. This is what the sentence would look like without the hedge:
This led inevitably to a respect and reverence for the creatures inhabiting the sacred
I would argue that THE UNDERSTANDING
of the message is not impaired if the hedge is removed. The unhedged statement comes across as very strong and confident, but the message is still communicated to the reader. Omitting 6 of 171 hedges to promote the flow and legibility of the TT does not alter the style of the text or make it less typical to its genre. In these instances where the ST sentences are unnecessarily tedious, and nothing important is lost if the hedges are left out, it is possible to omit them and still remain true to the style of Swedish popular science texts. The degrees of tentativeness or assertiveness which the ST author wishes to convey by using hedges are still signalled in the text as the majority of hedges were transferred to the TT. The applied translational approach is discussed by Ingo (2007:124, 286f) in terms of semantic omission and implicitation (implicitgцrande). Semantic omission, i. e. to leave out meaningful components in the ST in the translation, almost always results in a loss of information. However, in the process of implicitation, as discussed in (26), words or expressions can be omitted when the message is otherwise implied in the sentence.
(34) [p. 165]
Translational ST approach
Scenes like these must represent myths which are entirely lost to us but which were meaningful symbols within the context of religion in Iron Age Europe.
TT Scener som denna torde fцrestдlla myter som gеtt helt fцrlorade fцr oss, men som var meningsfulla religiцsa symboler i Europa under jдrnеldern.
The hedge in (34) was used to illustrate the whole concept of hedging in section 1. It is not obvious that must is in fact a hedge here. At first, sentences containing must come across as expressing a high level of certainty, as must has strong assertive connotations, and are therefore easily disregarded in the search for hedges. However, the interpretation of must as a hedge becomes more evident when looking at the sentence when must has been removed:
Scenes like these represent myths which Scener som denna fцrestдller myter som gеtt
are entirely lost to us but which were
helt fцrlorade fцr oss, men som var
meaningful symbols within the context of meningsfulla religiцsa symboler i Europa
religion in Iron Age Europe.
By comparing the sentences we can see that must should be interpreted as a hedge in this context, as the level of certainty is unexpectedly higher if it is omitted. Must being used as a hedge has been discussed by Coates (1983:45f), who states that must in its epistemic function is often used in hedging. According to Coates (1983:41), epistemic modal auxiliary verbs such as must are subjective, and express the speaker's/writer's opinion or judgment about a statement. Must expresses confidence, but signals less certainty than an expression without must. The translational choice in (34) is torde, but borde and mеste were also considered. Mеste was ruled out because of its imperative connotations, which did not work well as an expression of hedging. To distinguish differences in meaning between torde and borde is less straightforward. The conclusive information was found in a web search at Institutet fцr de inhemska sprеken [www]. According to this site, both words can be used to express probability, but torde is distinguished by having the additional connotation "should appropriately", which works well in this context. In connection to the discussion of how must is easily overlooked as a hedge, it is interesting to discuss other words with the same function, as well as words which work the other way around and are falsely identified as hedges. This will be discussed in 3.1.5.
This section analysed the uses and translations of the modal auxiliary verb hedges may and must. Hedging can also be expressed through phrases indicating limitations in method or knowledge. This will be the topic of the next section.
3.1.4 Phrasal hedges A slightly different type of hedge than the ones previously discussed in this paper, is pointed out by Hyland (1998:141). These hedges consist of phrases where an author openly states limitations of the method used or admits to having inadequate knowledge. This type of hedging phrase is to be found on 18 occasions in the ST, and they were all transferred to the TT due to their strong hedging quality. A few examples will be analysed below:
(36) [p. 160]
We cannot know whether or not the Vi vet inte om kvinnorna och djuren som
females and their animals
avbildas i den tidiga europeiska
represented in the earlier European bildkonsten var skepnadsskiftare eller
iconography were shape-changers or inte.
In (36), the phrase We cannot know whether or not hedges the entire statement as Green signals having insufficient knowledge to make an actual claim. In the TT, the expression has been slightly modified to Vi vet inte om, instead of the literal Vi kan inte veta om. The chosen translation expresses a slight difference in nuance compared to the ST expression: we do not know vs. we will never be able to know. The reason for making this change is that it is unnecessary to state with full certainty that this is something that will never come to be known.
(37) [p. 161]
Translational ST approach
It is clearly extremely difficult to offer any close interpretation of the scene on the Strettweg wagon, but the imagery does suggest certain themes and beliefs.
TT Det дr sjдlvklart mycket svеrt att gцra en korrekt tolkning av scenen pе Strettwegvagnen, men bildsprеket antyder vissa teman och trosuppfattningar.
In (37), the hedging expression It is clearly extremely difficult is translated literally as Det дr sjдlvklart mycket svеrt. This hedge, although expressing reservations, is not as
strong as the one in (36), but opens up for a possible interpretation, which is also presented further on in the sentence.
(38) [p. 164]
Translational ST approach
The identity of the Reinheim goddess, if goddess she is, can never be known, [...]
TT Reinheimgudinnans identitet, om hon verkligen дr en gudinna, kommer aldrig att bli kдnd, [...]
The hedge in (38), can never be known has been slightly modified and is translated kommer aldrig att bli kдnd. The literal translation kan aldrig bli kдnd signals withholding information, that we know something that we do not want others to know. Kommer aldrig is an equivalent to will never, which simply expresses future tense, while kan aldrig implies a lack of will to share information that is known. The use of the same type of phrasal hedges can be seen in Swedish parallel texts, which signals their conventionality in academic writing, regardless of language:
Relationen mellan denna kultur och keltiska sprеk дr dock oklar: дven om mеnga latиnefolk
[NE, p. 1] var keltisktalande, behцver inte alla ha varit det.
En ursprunglig totemism har sеledes inte varit mцjlig att pеvisa.
This section has discussed the practice of expressing hedging through the use of phrases stating limitations. In the next section, there will be a discussion of "easily overlooked" and falsely identified hedges.
3.1.5 Easily overlooked and false hedges In this last section of the analysis, we will reconnect to the discussion in section 18.104.22.168 about words that are easily overlooked as hedges. One word which similarly to the modal auxiliary verb must is easily missed as a hedge is the adjective certain. Like must, certain has a strong connotation of certainty and is thus easily disregarded in the search for hedges:
(41) [p. 168f]
Translational ST approach
Sava and Oisin, with their continuing deer-associations, bear certain resemblances to other mothers and sons [...]
TT Sava och Oisin med sina fortlцpande associationer med hjortar, bдr vissa likheter med andra mцdrar och sцner [...]
The adjective certain in (41) is translated with the adjective vissa. Possible alternatives according to synonymer.se [www] include the adjectives sдrskilda and speciella. However, the use of vissa in this context is supported by Korp, which produces more than 520,500 hits in a search on the adjective viss. Many of the examples given in Korp include viss(a) used in similar contexts as in (41): sker vissa fцrдndringar, i vissa delar av, i vissa fall, till en viss grad, etc. Sдrskild and speciell generate approximately 109,500 and 214,700 hits respectively in Korp, which makes viss(a) the most conventional of the three. Another possible translational option would be to use the pronoun flera. It produces more than 570,000 hits in Korp. However, there is a slight but important difference in meaning between vissa and flera, which makes vissa a more suitable choice in this context. Vissa can be used as a literal translation of some, meaning "a few; not many". Flera on the other hand, is a literal translation of several, meaning "more than a few; a significant part of". In my interpretation of the sentence in (41), vissa is the word with the most appropriate connotations. The choice of vissa is further supported by Swedish parallel texts. In the article about the Celts in NE.se, vissa appears frequently:
(42) [p. 2] (43) [p. 5]
[...] varvid det keltiska inflytandet цkade genom vissa, av oss hittills okдnda invandringar. Vissa oppida tjдnade som tillflyktsorter, andra var дven boplatser fцr jordbrukare.
I would argue that even though it is not obvious, certain should be considered a hedge in (41). Certain can be replaced with some/a few/a number of, which more clearly shows the hedging function. The level of certainty increases if certain is omitted (compare ex. 35): bear resemblances is a stronger statement than bear certain resemblances. Based upon this observation, certain should be identified as a hedge in the above context as hedges serve to weaken the certainty of statements. An interesting point to discuss in relation to words that are easily overlooked as hedges due to their assertive nature, is that there are also words which work the other way around, i.e. which are easily mistaken to be hedges when they are not. When analysing and studying hedges, it is very easy to become over-zealous, finding hedges where they actually do not exist. This emphasizes Hyland's (1998:244) point, brought up in section 2.2, that no words have an inherent hedging function, but that whether or not they are to be perceived as hedges is due to context.
The word if, mostly identified as a conjunction according to the Oxford English Dictionary
(OED), can easily be mistaken to be a hedge in all instances. This must be due to the hesitant and uncertain connotations of the word. In certain contexts, if does express hedging, but this is not always the case. The word if appears 14 times in the ST, but it has a hedging function in only eight of these instances. Hedging if-clauses consist of several words, and therefore stand out as obvious hedges in the ST, expressing a strong level of reservation or hesitation. Four of the hedging if-clauses contain as if, and the other four only if. Compare the examples below:
(44) [p. 164] (45) [p. 168]
Translational ST approach
The identity of the Reinheim goddess, if goddess she is, can never be known, [...]
[...] the behaviour of a thwarted lover, who tried to ensure that if Sava would not accept him, she should never have a husband.
TT Reinheimgudinnans identitet, om hon verkligen дr en gudinna, kommer aldrig att bli kдnd, [...] [...] en meningslцs hдmndaktion frеn en fцrsmеdd дlskare som ville fцrsдkra sig om att Sava aldrig skulle kunna gifta sig om hon nekade honom.
If in (44) and (45) has been translated literally with the corresponding Swedish om. In (44), if is a hedge, expressing reservations as to the interpretation of the image described. However, in (45), if is not a hedge, but an implication of a stipulation and its result (if Sava...so...). When reading the text quickly or in an active search for hedges, it is easy to mark all occurrences of if as having a hedging function. As mentioned in section 1, Green tends to use hedges when interpreting archaeological finds. In (44), the if-clause hedges the whole content of the statement: If, and only if the proposed conditions are true, which may very well not be the case, then they may mean a certain thing. The hedge signals that the entire proposition is highly uncertain. A cult-image of a woman has indeed been found in Reinheim, this is a fact. But the interpretation of the image is uncertain; it may depict a goddess, but it could just as well be another female, whose identity is unknown. The statement offers an interpretation of a hypothetical scenario, which may be of no relevance at all if the condition of the claim is proven to be false. The decision to translate the ST if-clause if goddess she is with the TT if-clause om hon verkligen дr en gudinna, was made to make the translation slightly more straightforward than the ST.
4. Conclusion The aim of this study was to investigate the frequency of different categories of hedges in an academic text and to discuss the challenges encountered in their translation. By presenting a number of examples from a popular science textbook, different ways of expressing hedging have been discussed from a translational point of view. The discussion was supported by quantitative data and the translational choices were analysed and motivated using corpora, dictionaries and parallel texts. The study shows that adverb hedges are by far the most common ones in the current ST, and lexical verb hedges and modal auxiliary verb hedges also appear frequently. Phrasal hedges, adjective hedges and if-clause hedges are less common, but still appear frequently enough to be regarded as significant traits of the ST. The conclusion is that hedges appear frequently enough to be identified as a genre-specific trait of academic discourse in English as well as in Swedish. The use of hedges in Swedish academic writing is supported by parallel texts. The vast majority of hedges were transferred from the ST to the TT, only six out of a total 171 were omitted. The translation strategy most often used was word-for-word or literal translation. On a few occasions, a sense-for-sense or free translation seemed more appropriate, and Vinay & Darbelnet's procedures transposition and equivalence were then used. The study also reveals that the identification of hedges is ultimately subjective and that some hedges are easily overlooked due to their assertive nature, while other words may be falsely identified as hedges. This motivates extra carefulness by the translator, to make sure that only real hedges are transferred to the TT. As the findings of this study are based on limited material, it is difficult to draw any general conclusions on how hedging should be treated in the translation process. An area of further research might be to conduct larger quantitative studies as to the difference between the use of hedging in English vs Swedish. It would also be helpful with further theoretical work regarding how to deal with hedges generally in the translation process. 27
5. References Primary source Green, Miranda. 1995. Celtic Goddesses. Warriors, Virgins and Mothers. London: British Museum Press. Secondary sources Cassirer, Peter. 2003. Stil, stilistik & stilanalys. Stockholm: Bokfцrlaget Natur och Kultur. Clemen, Gudrun. 1997. The Concept of Hedging: Origins, Approaches and Definitions. In Markkanen, Raija, & Schrцder, Hartmut (eds.). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts:235248. Berlin: Walter de Gruyter & Co. Coates, Jennifer. 1983. The semantics of the modal auxiliaries. Worcester: Billing & Sons Limited. Gцrman, Marianne. 1988. Nordiskt och keltiskt. Sydskandinavisk religion under yngre bronsеlder och keltisk jдrnеlder. Lund: Wallin & Dalholm Boktr. AB. Hauck, Ben (ed.). 2009. How to Be Less Wrong More Often. ETC: A Review of General Semantics 66/4:478482. Hellspong, Lennart & Ledin, Per. 1997. Vдgar genom texten. Handbok i bruktextanalys. Lund: Studentlitteratur AB. Holmes, Janet. 1982. Expressing doubt and certainty in English. RELC Journal 13/2:9 28. Hyland, Ken. 1998. Hedging in Scientific Research Articles. Amsterdam: John Benjamins
Publishing Co. Ingo, Rune. 2007. Konsten att цversдtta. Цversдttandets praktik och didaktik. Lund: Studentlitteratur. Kranich, Svenja. 2011. To hedge or not to hedge: the use of epistemic modal expressions in popular science in English texts, EnglishGerman translations, and German original texts. Text & Talk 31/1:7799. Lakoff, George. 1972. Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts. In Peranteau, Paul M., Levi, Judith N. & Phares, Gloria C. (eds.). Papers from the Eighth Regional Meeting of Chicago Linguistic Society:183228. Chicago: Chicago University Press. Lindquist, Hans. 2009. Corpus Linguistics and the Description of English. Edinburgh: Edinburgh University
Markkanen, Raija & Schrцder, Hartmut. 1997. Hedging: A Challenge for Pragmatics and discourse analysis
. In Markkanen, Raija, & Schrцder, Hartmut (eds). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts:318. Berlin: Walter de Gruyter & Co. Mauranen, Anna. 1997. Hedging in Language Revisers' Hands. In Markkanen, Raija, & Schrцder, Hartmut (eds). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts:115133. Berlin: Walter de Gruyter & Co. Meyer, Paul Georg. 1997. Hedging Strategies in Written Academic Discourse: Strengthening the Argument by Weakening the Claim. In Markkanen, Raija, & Schrцder, Hartmut (eds). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts:2141. Berlin: Walter de Gruyter & Co. Munday, Jeremy. 2008. Introducing Translation Studies
. Theories and applications. New York: Routledge. Ross, Anne & Robins, Don. 1990. Druidens dцd. En arkeologisk thriller. Stockholm: Bonnier Fakta Bokfцrlag AB. Schrцder, Hartmut & Zimmer, Dagmar. 1997. Hedging Research in Pragmatics: A Bibliographical Research Guide to Hedging. In Markkanen, Raija, & Schrцder, Hartmut (eds). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts:249271. Berlin: Walter de Gruyter & Co. Sprеkrеdet. 2008. Svenska skrivregler. Stockholm: Liber AB. Varttala, Teppo. 2002. Hedging in Scientific Research Articles: A Cross-disciplinary Study. In Cortese, Giuseppina & Riley, Philip (eds.). Domain-specific English: Textual Practices across Communities and Classrooms:141174. Bern: Peter Lang AG. Internet resources Corpus of contemporary america
n English. Accessed on 24 May 2013. http://corpus.byu.edu/coca/ Institutet fцr de inhemska sprеken. Accessed on 9 April 2013. http://www.kotus.fi/index.phtml?l=sv&s=1841 Nationalencyklopedin. Accessed on 15 May 2013. http://www.ne.se.proxy.lnu.se/lang/kelter?i_whole_article=true Oxford English Dictionary. Accessed on 24 May 2013. http://www.oed.com.proxy.lnu.se/ Sprеkbanken, Korp. Accessed on 23 May 2013. http://spraakbanken.gu.se/korp/ Synonymer.se. Accessed 16 May 2013. http://www.synonymer.se/ 29