This paper aims to investigate the styles of the two German translators of F. Scott Fitzgerald’s novel The Beautiful and Damned – Renate Orth-Guttmann and Hans-Christian Oeser. Orth-Guttmann and Oeser both published translations of Fitzgerald’s novel in 1998 under the title Die Schönen und Verdammten, thus presenting the researcher with a unique opportunity to compare the styles of two translators while keeping constant many other factors in the translation process, namely source text and author, time of publication of original and translation and source and target language. For present purposes style is understood as ‘preferred or recurring patterns of linguistic behaviour, rather than individual or one-off instances of intervention’ (Baker 2000: 245). An obvious type of recurring patterns is repetition at the lexical level. In this paper I focus on repetition – and variation – in the translators’ use of speech-act report verbs. Speech acts and speech-act report verbs are explained in section Two below. Section Three describes the methodology applied in the study and section Four presents the analyses and results. The paper concludes with possible interpretations of the findings (section Five) and final remarks on these results (section Six).

A speech act is ‘an utterance that constitutes some act in addition to the mere act of uttering’ (Collins 1994) and a performative is defined as ‘denoting an utterance that constitutes some act, esp. the act described by the verb’ (Collins 1994). Austin (1976 [1962]), generally considered the father of speech-act theory, introduced the term ‘performative verbs.’ These verbs appear in the utterance itself and have either the form of: (i) first person singular, indicative, active, present (e.g. ‘I warn you.’); or (ii) second or third person, singular or plural, indicative, passive, present (e.g. ‘You/passengers are warned to…’). The utterances in which these verbs occur perform an action, over and above the actual utterance (e.g. the act of warning). For further detailed discussion of happy and unhappy performative utterances, as well as explicit as opposed to primary performative verbs see Austin (1976 [1962]; I-VII).

According to Austin (1976 [1962]), all these performative utterances have an illocutionary force. This means that the performed act (illocution) has a function (force). While the locutionary act merely accounts for the act of uttering, the illocutionary act includes the action. However, usually any speech act has been uttered for some reason and therefore has some communicative function. This means that most locutionary acts are also illocutionary acts and they basically become one category. The other category is the perlocutionary act which accounts for the consequential effects of the speech act.

Among the locutionary acts Austin distinguishes between phonetic acts, merely accounting for ‘the act of uttering certain noises’ (Austin 1976 [1962]: 95), rhetic acts (indirect speech acts) and phatic acts. According to Austin, phatic acts are mimicable and reproducible. One can mimic not merely the statement in quotation marks, but also the more complex way in which the statement was made (Austin 1976 [1962]: 95). Austin describes phatic acts as ‘the ‘inverted commas’ use of ‘said’ as we get it in novels: every utterance can be just reproduced in inverted commas, or in inverted commas with ‘said he’ or, more often, ‘said she’[1], &c., after it’ (Austin 1976 [1962]: 96).

In Short et al’s (1996) speech classification these phatic acts are called direct speech (acts). Short distinguishes between speech and thought, direct and indirect speech/thought, and the narrator’s representation of speech/thought acts. Direct speech is divided into the categories of free direct speech/thought, i.e. speech/thought without a reporting clause, and what I will call reported direct speech/thought, defined as appearing with a reporting clause.[2] The present paper is concerned with reported direct speech acts. As narrator’s representation is only marginally concerned with speech, it is not discussed any further here. In addition, all classes of thought acts as well as indirect speech acts are omitted, as they can have different illocutionary or perlocutionary forces to those associated with direct speech, such as the effect of distancing the reader from the speaker.

Reported direct speech acts consist of a reported clause, which appears in inverted commas, and a reporting clause. The reporting clause, which contains the relevant verb, can occur in initial, mid or end-position. In order to exclude borderline cases and ensure a consistent selection of verbs, reporting clauses in this study must end with a colon, when they occur in initial position.

Reporting verbs are those that appear outside the inverted commas in written utterances presenting direct speech. Burgess (2000) calls these verbs ‘words introducing direct speech.’ Ballmer and Brennenstuhl (1981) speak of ‘speech act verbs’ using the term ‘in the more general sense’ and defining it ‘to specify the content or means of expression of the linguistic activity’ in inverted commas. I find the designation ‘speech act verb’ for the reporting verb, i.e. the verb outside the inverted commas, ambiguous, especially given Austin’s (1976 [1962]) treatment of the subject. Therefore, I follow Bourne (2002) who names these verbs ‘speech-act report verbs.’ Regarding the speech acts proper I follow Short et al’s (1996) terminology and will refer to the utterance in inverted commas as the ‘reported clause,’ which contrasts with the ‘reporting clause’ which introduces or follows the direct speech.

As opposed to Austin’s performative verbs, Ballmer and Brennenstuhl consider as speech-act report verbs all verbs designating any kind of (aspect of) speech activity (Ballmer and Brennenstuhl 1981: 16). This means that they account for all verbs which can possibly appear in the reporting clause surrounding direct speech. These are the verbs relevant for the present paper. Therefore, I follow Ballmer and Brennenstuhl’s methodology for extracting speech-act report verbs. If a verb is inserted in the sentence frame below and if the sentence then makes sense, the verb is counted as a speech-act report verb.

Sentence (1) makes sense, therefore said is a speech-act report verb whereas sentence (2) is not possible, thus seemed is not a speech-act report verb.

The aligned parallel corpus consists of F. Scott Fitzgerald’s novel The Beautiful and Damned and its two German translations, both entitled Die Schönen und Verdammten. Wordsmith Tools (Scott 1998) and Multiconcord (Woolls 1997) were used to facilitate corpus analysis. A frequency wordlist was created with Wordsmith Tools for The Beautiful and Damned. The speech-act report verbs were manually extracted from the wordlist according to the methodology suggested by Ballmer and Brennenstuhl (1981) and speech-act report verbs that occurred a minimum of 15 times were considered.

In some instances, the selected speech-act report verbs did not actually introduce or follow an utterance; these occurrences were excluded. Thus only the occurrences where the verbs appeared in reported direct speech acts, according to Short’s classification (see section Two), were selected. These selection criteria brought the frequency of occurrence down to one or two for some verbs. As the present study attempts to investigate potential patterns of translators’ habits, rather than unique uses of linguistic features, only the speech-act report verbs occurring in a reported direct speech act a minimum of ten times were considered. The selected speech-act report verbs and their frequencies in The Beautiful and Damned are listed in Table 1.

The speech-act report verbs that the two translators used were analyzed from different angles. First, the number of different translations for each of the speech-act report verbs in Table 1 was investigated. Second, the verbs which each translator used only once for each of the speech-act report verbs were examined. Third, the verbs which each translator used most frequently as a translation of the speech-act report verbs under investigation were analyzed. The analysis of these most frequently used speech-act report verbs in the German translations led to a qualitative analysis of these cases, investigating whether the options chosen were also the most obvious. Finally, the instances where the translators did not opt for the obvious translation were examined with regard to the semantic meaning of the alternative verb used.

Table 2 shows the most frequent speech-act report verbs and their total numbers of occurrence in The Beautiful and Damned and the number of different verbs each of the translators uses to translate the English verbs. Orth-Guttmann uses a wider variety of verbs than Oeser for all but one verb. The exception is the verb whispered, which Oeser translates with three different verbs, flüsterte (9 times), wisperte (twice) and murmelte (once), while Orth-Guttmann chooses only flüsterte (11 times) and hauchte (once). The most striking example is the verb said, which Orth-Guttmann translates with 27 different verbs, whereas Oeser uses only seven. In the case of cried, Orth-Guttmann chooses six times as many verbs as Oeser and makes twelve different choices, while Oeser uses only two different verbs. There are four cases in which Oeser sticks to only one verb, while Orth-Guttmann uses a variety of verbs. The verbs translated with least variety by Orth-Guttmann are asked, called and whispered (2 options), and demanded and inquired (3 options). In most cases Orth-Guttmann uses more than five alternatives, whereas Oeser uses more than five different verbs in only two cases (said and suggested). In total, Oeser translates with fewer than half as many different verbs as Orth-Guttmann.

There is one particular variation which Orth-Guttmann uses frequently, while it occurs only twice in Oeser’s translation, and that is the translation of a reported direct speech act with a free direct speech act. In Orth-Guttmann’s translation there are 41 occurrences without a reporting clause in total, distributed across 13 verbs. The highest number of free direct speech acts can be found for the verbs said (14 instances), cried (7 instances) and answered (6 instances), which are among the most frequent speech-act report verbs in The Beautiful and Damned. This could be interpreted as another means by which Orth-Guttmann avoids repetition.

Having examined the most frequently used speech-act report verbs in the translations, the question arises whether they are also the most ‘obvious’ choices available to the translators. To investigate this question, the main lexical equivalents, taking their function as speech-act report verbs into account, of each of the English verbs were extracted from a bilingual dictionary (Pons 2001) and defined as ‘obvious’ translations. They are shown in Table 5 along with frequencies in Die Schönen und Verdammten.

After looking at the obvious options, the alternative choices of verbs were analyzed. They were examined with regard to their semantic differences in relation to the obvious verbs, which were considered to be equivalent to the English source-text verbs, i.e. literal in recreating the semantic meaning of the source language word in the target text. Some of the alternative translations, such as wisperte for whispered, are equivalent to the obvious options (flüsterte for whispered). Other choices take up a secondary meaning of the English verb, such as empfahl (recommended), legte nahe (advised) and unterstellte (implied/insinuated) for suggested. There are also cases of change in register. For example, in three instances, began is translated as the higher register verb hob an. The most interesting observation is, however, that many of the choices the translators made are different from the source-text verb regarding their semantic meaning. Examples of this would be translations of remarked with staunte (to be amazed), murmured with bestätigte (confirmed) or cried with wiederholte (repeated). Each of these examples shows a shift in the semantic meaning of the source-text verb. It is mainly Orth-Guttmann who uses such verbs with a different semantic element, and this she does on 87 occasions. She uses 33 different verbs which appear as translations for 15 out of the 19 verbs analyzed. Oeser only adds a semantic component in six instances when translating the verbs answered and said. He uses fragte (asked) three times and versetzte (retorted) once for said (sagte). For answered he uses versetzte and beschied (rejected/rebuffed; officialese). In one instance he translates repeated as rief (cried), which slightly changes the semantic meaning. In three instances he translates the verbs cried, added and agreed as sagte (said), which all lack an element of the original semantic meaning. Verbs which lack a semantic element from the source text occur in Orth-Guttmann’s translation in 33 instances, distributed across seven verbs, most of them translated as sagte and erklärte (said). These data indicate that Oeser stays closer to the source text and repeats more, while Orth-Guttmann deviates more from the source text and uses a greater variety of verbs.

In examining the translations of the speech-act report verb murmured, it is striking that Oeser uses only murmelte, which supports the above hypothesis that he stays closer to the source text. Orth-Guttmann uses the word murmelte only once. In most cases she translates murmured as sagte (said, 13 times), which misses out on the component of speaking in a low voice. However, she compensates for this element by using adverbial modification, adding halblaut (half aloud, 9 times) or leise (quietly, 3 times). Another example with modification is the verb agreed, which Orth-Guttmann translates as meinte (said) in three instances; in two of these she adds the adverb zustimmend (in agreement). In the third instance, meinte is not modified; this could be interpreted as a case of implicitation. For the verb repeated, Orth-Guttmann uses jammerte (whined) in one occurrence, which she complements with immer wieder (again and again); the element of pity comes from the source text which reads ‘she repeated piteously.’ The semantic meaning is kept in this example and it does not show explicitation or implicitation. However, there is a slight shift in the focus. In Orth-Guttmann’s translation, the person in question seems to lament more than in the original.

There is no instance of adverbial or other modification, or motivation from the source text in the cases where Oeser uses a basic speech-act report verb instead of the obvious option, or where he uses verbs with an additional semantic element. These occurrences could be interpreted as cases of explicitation or implicitation, respectively, but there are too few instances to conclude that it would be Oeser’s style to implicitate and even less to explicitate.

From the above examples it can be concluded that the instances where Orth-Guttmann’s choices of speech-act report verbs deviate from the source text cannot as yet be interpreted as instances of explicitation or implicitation. A detailed analysis of the data regarding the source-text influence, as well as possible modifications Orth-Guttmann might have used to compensate for the different semantic meanings of her chosen verbs, would be necessary to draw such conclusions.

To illustrate a first step in this further detailed analysis, see the instances of insisted in Table 6.

The example of the speech-act report verb insisted shows that Oeser stays closer to the source text and uses more repetition than Orth-Guttmann. Orth-Guttmann uses a lot of variation and moves much further from the source text than Oeser. From the examination of her translations of insisted, a possible hypothesis might be that she tends to explicitate or even interpret the source text.

On the basis of the analyses described above, I would like to propose three tentative conclusions: (i) Oeser uses lexical repetition more than Orth-Guttmann, who tends to avoid repetition; (ii) Oeser stays closer to the source text than Orth-Guttmann; and (iii) Orth-Guttmann tends to explicitate.

(i) There is a significant difference in the number of instances of repetition between the two translators (see Tables 2-4). Oeser repeats verbs he chooses as translations of a particular speech-act report verb to a considerably greater extent than Orth-Guttmann. She uses a significantly wider variety of verbs to translate the individual speech-act report verbs and consequently repeats these verbs to a lesser extent. She also frequently omits the reporting clause before or after an utterance, in order to avoid repeating the particular verb. Orth-Guttmann may be motivated by a desire to avoid what is seen by some commentators as bad style. Ben-Ari (1998) says about repetition:

[…] it seems that avoiding repetitions has to do with a deep-rooted need to display richness of vocabulary, passed down by generations of normative stylistic do’s and don’ts and [is] extremely difficult to root out. […] It seems that, from kindergarten on, pupils have been complimented on their rich vocabulary and reprimanded for repetitiveness.

Ben-Ari 1998: 70

(ii) Oeser uses verbs repeatedly according to the instances of repetitions of speech-act report verbs in the source text. He stays close to the source text and in the majority of cases chooses verbs which produce quite literal translations of the speech-act report verb. Orth-Guttmann, on the other hand, does not keep up the repetitions she encounters in the source text, but tries to avoid them by variation and omissions of reporting clauses.

(iii) Orth-Guttmann uses a greater variety of speech-act report verbs to translate the 19 verbs under investigation. A considerable number of these verbs deviate from the semantic meaning of the source-text verb. In some instances she chooses a verb less specific than that in the source text. However, in the majority of instances, she chooses a verb with an additional semantic feature. Like the translators studied by Bourne (2002) who found ‘considerable diversity in the choice of Spanish verbs used to render ‘said’ in the context of impositive directive speech acts’ (Bourne 2002: 241) Orth-Guttmann intervenes ‘to provide variety, to convey precision and to specify the illocutionary force of an utterance’ (Bourne 2002: 252). Orth-Guttmann also appears to use speech-act report verbs to emphasize characters’ emotional states to the extent that she influences the reader’s attitude to the characters in a manner consistent with that described by Burgess (2000) in his discussion of Goethe’s Die Wahlverwandtschaften.

The analyses carried out show significant differences in the use of speech-act report verbs between the two translators. Hans-Christian Oeser repeats the speech-act report verbs in congruence with the source text. It is likely that he was aware of the repetitions in the source text and retained them deliberately. Renate Orth-Guttmann also noticed the source text repetitions. It is often argued that translators avoid repetition (Laviosa-Braithwaite 1998). Orth-Guttmann was aware of the accepted wisdom that repeated use of speech-act report verbs, particularly sagte (said), is considered to be bad style, or as Orth-Guttmann (personal communication, July 2005) put it, that German readers do not like the repetition of sagte. Thus, Orth-Guttmann’s strategy conforms to perceived target-language conventions by avoiding too much repetition. Oeser uses repetition as a stylistic device which runs counter to current conventions.

Orth-Guttmann’s extensive use of variation goes together with her choices of verbs with an additional semantic feature, with a potentially different illocutionary force. Thus she highlights certain traits of the characters (see Table 6, Nos. 4-9). The question remains, however, as to whether she does this systematically and to such an extent that the readers of her translation would perceive the characters differently from those of Oeser’s translation.