1 – THE NOVEL AND ITS TRANSLATIONS IN CONTEXT 1.a – Methodological Introduction I Promessi Sposi (henceforward abbreviated IPS ) is a novel by Italian writer Alessandro
Manzoni first published in 1827, but considerably revised for a new, now authoritative,
1840 edition. Manzoni introduced the historical novel, a literary genre inaugurated by
Sir Walter Scott, to the Italian reading public. The first English translation of IPS (1827)
is by the Rev. Charles Swan, issued in Pisa as early as 1828 as The Betrothed Lovers .
Ever since many translations have followed, each one under the common title of The
Betrothed (abbreviated TB )
1
. The most accomplished translations are the 20
th century
ones, by D. J. Connor (1924), A. Colquhoun (1951) and B. Penman (1972). Less
successful were the 19
th century renditions, culturally biased and abridged.
The complexities associated with the translation of works of fiction have always
drawn the attention of translation professionals and amateurs. Theoretical issues such as
the cultural constraints of different tongues and the difficulty of preserving the author’s
particular style have been raised. As for the first point, IPS is a novel bound up with
Italian traditions not so familiar to an average English reader, as can be the case with the
Sunday morning, which in the mind of a local community may generate shared
associations like the merry call of the bells for Mass or the ensuing gathering of the
family for lunch, taking delight in good company if the day is pleasant and charming.
The possibility to convey these ideas rests not only on linguistic but also on correlated
historical, geographical and sociological notions which is part of the translator’s task to
teach. As remarked by Taylor: ‘It is this phenomenon of cultural specificity that does so
much to impede cross-cultural examination’
2
. For this reason the English translator
should beware of the risk of precluding real-life fidelity with easy stereotypes on
Catholicism, the Italian peasantry and their strong family values; such risk was
accurately avoided not only by the Italian Manzoni, but also by Scott and Maria
Edgeworth in their historical and regional novels from the other side of the Channel.
1
The English translators may have owed their choice of a shorter title to the presence of a homonymous
novel by Scott himself ( The Betrothed , 1825)
2
Christopher Taylor, Language to Language: A practical and theoretical guide for Italian/English
translators (Cambridge: University Press, 1998) p. 103.
3
The problem of culture-bound knowledge was famously discussed by the
anthropologist Malinowski in the 1923 paper ‘The problem of meaning in primitive
languages’. Confronted with the problem of rendering all the facets of meaning
contained in a linguistic specimen by a Trobriandese, he opted for translations provided
with commentaries, a sort of extended footnotes. He stressed the necessity to describe
the ‘context of situation’, a useful notion resumed years later by the linguist M. A. K.
Halliday who applied it to grammar studies before it was imported into translation
studies as well. The latest handbooks on translation are unanimous in stressing the
importance of a contextual analysis of the text under scrutiny before the actual process
of translation begins, as acknowledged by landmark translation theorist Eugene Nida:
‘For a number of years I have been increasingly interested in the role of contexts in
translating, because failure to consider the context of a text is largely responsible for the
most serious mistakes in comprehending and reproducing the meaning of a discourse’
3
.
Traditionally, at their outset, handbooks on translation recur to the age-long debate
between free versus literal translations, it is something which divides but on which
everybody seems to agree: translators must negotiate the two seemingly opposite
requirements of source text (ST) fidelity and target text (TT) understandability. But
Eugene Nida and Peter Newmark have proposed more refined distinctions such as those
between formal/dynamic equivalence and semantic/communicative translation
respectively. This move was meant to produce ‘types of equivalence appropriate to
particular circumstances’
4
. A novel like IPS has different stylistic conventions (form and
content) as well as a different readership and requires different translation strategies
from less thought-out literary products such as newspaper articles or advertising leaflets.
3
E. A. Nida, Contexts in Translating (Amsterdam; Philadelphia: J. Benjamins, 2001) p. ix.
4
B. Hatim and I. Mason, Discourse and the Translator (London; New York: Longman, 1990) p. 7.
4
On a first look, in the case of IPS /TB Nida’s formal equivalence and Newmark’s
semantic translation are to be preferred in that as argued by Newmark ‘the more
important the language of a text, the more closely it should be translated. This is valid at
every rank of the text’
5
. This is a valid theoretical starting point, albeit too abstract to be
of any practical use. In as ‘important’ and ‘expressive’ works as IPS , Newmark’s
emphasis in translation is on the contextual meaning of the original, namely on the
‘truth’ of the message, as he argues that ‘There seems no good reason not to reproduce
the truth, even when the truth is not particularly important’
6
. He is aware that author-
centred texts are more often than not culture-connoted and a strict literal adherence
would entail loss of artistic value in the target text, as well as loss in
communicativeness. He lists the few cases where a kind of indirect – ‘ludic’ –
equivalence in translation could be advocated to arouse a response in the target reader
similar to that intended for the source reader, some appear to be particularly relevant to
IPS /TB :
1. Cultural words – objects or activities with connotations that are specific to one
community 2.Transcultural words with similar referents and different connotation – the ‘classical’ example
are the staples: bread, rice, wine, etc.
3.Concept words with different emphases in different communities (‘liberalism’, ‘liberty’,
‘obedience’, ‘bureaucracy’).
[...]
5. Cultural metaphors, idioms, proverbs, puns, neologisms. They may have to be spelt out in
the TL – concision, force, nuances of meaning are lost or compensated.
7
As far as these cases are concerned, the translator may decide to preserve the ‘truth’ of
the ST and recur to expansion/footnotes for clarification (see Malinowski), or he may
cautiously opt for a more expressive rendition, if literal adherence makes no sense at all
(see Newmark). A case for case choice on the translator’s part would seem unavoidable,
and yet careful analysis of stylistic – motivated – variants within the work under
scrutiny, aimed at disclosing the author’s pragmatic intentions, may help establish a
divide between cultural items requiring literal translation and reader-centred items
5
P. Newmark, About Translation (Clevedon: Multilingual Matters, 1991) p. 1.
6
Newmark 1991, p. 2.
7
Ibid., p. 8.
5
requiring a reader-centred adaptation. So, additional information/footnotes would be the
case for general concept words and notions, while a response-based equivalence in the
TT is in order for the many puns and proverbs present in IPS .
Whatever decision the translator makes, either the literal (cultural) or the pragmatic
(communicative) facet of meaning is bound to go lost. This would lead to the
pessimistic conclusion of Edward Sapir and B. L. Whorf that a strictly semantic
translation for culture-bound items is impossible, since different tongues have different
ways of producing cultural values which constrain not only our world views but our
way of thinking. However, mindful of Malinowski’s resourceful example and
Newmark’s emphasis on ‘contextual meaning’, up-to-date translation theorists have not
refrained from drafting more practical models, which, drawing from linguistics, still
seek to combine general cultural constraints with particular stylistic considerations.
It is basically a matter of hierarchy and discourse/narrative structure, whereby content
be linked to form, matter to manner, and culture to style, and once analysed and
matched in ST transposed satisfyingly in TT. However difficult to achieve, this is what
Hatim and Mason’s hierarchical model of context and Juliane House’s situational model
for translation quality assessment suggest the translator should do. In the case of IPS
this could be a tenable goal: Manzoni’s emphasis on plain style in narrative accords well
with the preference of the English language for a fluent prose, so that a reader-centred
translation for stylistic elements is to be advocated, quite the contrary for cultural values
which need be retained and faithfully conveyed in the target version. Finally, rethorical
devices like puns and proverbs belong to a sort zone between culture and style where
the translator is more at liberty to find the best personal solution.
In Discourse and the Translator , authors Basil Hatim and Ian Mason revise past
linguistic theories which in their view come to bear on the translation process. Particular
attention is devoted to the implications of Speech Act theory, developed by Austin and
Searle, for the work of the translator. Pointing to the socio-communicative function of
language, Austin and Searle have shown how every meaningful utterance constitutes a
form of action which can be analysed in terms of locution, illocution and perlocution.
The locution of an utterance is the verbal action performed by the oral apparatus and
entails phonological competence and lexico-grammatical rules. The illocution is the
action or intention behind the utterance, as can be apparent from the use of verbs like
6
‘promise’, ‘warn’, etc. The perlocution is the effect achieved on the hearer/reader in
conversation/narrative, i.e. ‘the extent to which the receiver’s state of
mind/knowledge/attitude is altered by the utterance in question’
8
.
After Austin and Searle’s ternary division of every speech act, Hatim and Mason
propose a roughly complementary text/narrative-view made up of three dimensions: the
communicative (concerned with register analysis in SL and TL) the pragmatic
(concerned with matching intention or illocutionary force in ST and TT) and the
semiotic dimension (concerned with the problem to translate signs across cultural
boundaries). In particular, the semiotic dimension is expanded to include three
categories: the text, the discourse and the genre of a given narrative; they are
interrelated and can be seen as inner, middle and outer frame. The categories, however,
appear to be just a duplication of the dimensions and could as easily as not be
assimilated so that what we get is: an outer frame, i.e. the semiotic/generic dimension,
strictly linked to the social occasion giving rise to the written act of communication; a
middle, pragmatic and interpersonal, frame where the translator considers how the
discourse of the narrative is structured; and an inner frame where peculiar textual and
stylistic aspects are investigated.
This ternary make-up, devised to account for the three main dimensions of every
narrative context, can be conveniently applied to IPS . As much as every utterance is a
multi-level speech act, Hatim and Mason argue that narratives can be seen as made up
of a sequence of meaningful text-acts (cf. divisions such as paragraphs, chapters, etc.)
constituting one big semiotic narrative-act where social occasion (genre), illocutionary
intention (discourse) and perlocutionary effect (text) are to be considered by the
translator. In the ‘conventional’ semiotic dimension of genre, Manzoni’s IPS will be
seen in relation to TL historical novels like Scott’s and Edgeworth’s. This will serve as
preparatory work before in-depth analysis of translation strategies for culture-bound
lexicon is carried out in the ‘attitudinal’
9
pragmatic dimension of discourse, along with
other relevant stylistic matters. Finally, analysis of ST and TT specimens is in order to
verify how discoursal and stylistic considerations find concrete expression in IPS and
TB . Clearly, Hatim and Mason’s hierarchical ternary model of context 10
is here slightly
8
Hatim and Mason, p. 60.
9
Ibid., p. 71.
10
Ibid., cf. figures on pp. 58, 71, 74, 75 and 237
7
adapted to fit a complex narrative such as IPS and single out apt translation strategies,
as in the figure below, where IPS is seen as an historical novel with exemplary purpose
but with comic and theatrical elements:
1. GENERIC FRAME: – linked to social occasions and conventions of genre (historical
novel)
– experiential, locutionary and author-centred
– linked to semiotics and translating culture-bound items
– SL biased, semantic translation 2. DISCOURSAL FRAME: – linked to pragmatic and stylistic matters (didactic purpose)
– interpersonal, illocutionary and narrator-centred – linked to pragmatics and translating irony, puns and proverbs – TL biased, equivalence of intention in translation
3. TEXTUAL FRAME: – linked to register analysis and textual typology (theatricality)
– communicative, perlocutionary and character-centred – linked to communicativeness and translating idioms and dialogues
– TL biased, equivalence of effect in translation The framework established by Hatim and Mason, from which the above operational
outline is derived, is still too programmatic and needs further specifications. However,
its relevance to the translation problems posed by IPS /TB lies in the preference granted
to the semiotic dimension and its three lower-degree categories, in other words on its
focus on cultural signs which link the text to the particular extra-linguistic
context/situation of the text producer by means of generic and discoursal considerations.
In particular, the translator is supposed to recognise a semiotic entity, however obsolete,
trace it back to the language-user’s where and when and transform it into another in the
TL, ‘under certain equivalence conditions to do with semiotic codes, pragmatic action
and general communicative requirements’
11
. Particular culture-bound rhetorical devices
such as irony, puns, proverbs, idioms are to be seen as units of translation requiring a
number of ad-hoc procedures; Hatim and Mason, for example, outline four of them:
identification , information , explication and transformation . Such procedures will be
11
Ibid., p. 105.
8
resumed and discussed in detail later on, for now it suffices to say that, in order to
understand and translate cultural signs, a discussion of the novelist’s environment and
motivations is called for.
In order to get to a more practical and systematic method of work, Juliane House’s
model for translation quality assessment (1977) deserves attention. First of all, it should
be specified that House’s focus is on careful textual analysis and evaluation of a given
translation. Hatim and Mason, instead, outline general translation strategies with a view
to transferability into the target language in the semiotic and discoursal dimensions
above all (there is as yet no translation, as the translation is their goal). The model
developed by House is useful because it merges ST translator and TT critic, and
prompts a comparative analysis of typical texts from IPS and TB . It is also worth noting
that her starting point is an attempt to avoid bias against either SL or TL. Nida’s concept
of dynamic equivalence in translation is discussed and the conclusion reached that ‘It is,
of course, undeniably true that a translation should produce equivalent responses’
12
; but
she also admits, as pointed out by the more subjectively oriented Newmark, that the
equivalent response principle is too mentalistic and superordinate to semantic adherence
to original form and content, which is less arbitrary and, therefore, safer to gauge.
House’s overview of previous translation quality assessment methods is meant to
display their main fault which she identifies as the lack of reference to the original.
After remarking on the inconclusiveness of TT biased experimental models, House
‘points to the necessity of developing a comprehensive, linguistic model of translation
quality assessment’ based on ST and TT alike. She outlines three main programmatic
stages towards the formulation of an operational model which to date did not exist:
1. source text criticism with a view to transferability into the target language 2. translation comparison in which the particular methods of translation used in the
production of the given translated text are described 3. evaluation of the translation, not according to vague, general criteria such as “good” or
“highly intelligible”, etc., but according to “adequate” or not “adequate” given the text-
specific features derived in 1
13
12
Juliane House, A Model for Translation Quality Assessment (Tubinger. TBL-Verlag Narr, 1977) p. 9.
13
House, p. 22. These preliminary stages are drawn from W. Koller, ‘Anmerkungen zu Definitionen des
Ubersetzungs “vorgangs” und der Ubersetzungskritik’; Aspekte der Theoretischen,
Sprachenpaarbezogenen und Angewandten Sprachwissenschaft , ed. by W. Wilss and G. Thome
(Heidelberg: Groos. 1974) pp. 35-45.
9
The first point is concerned with discussion of source text features (register, see
below), but it is here and now assimilated to analysis of Hatim and Mason’s generic
dimension of a literary work: Manzoni’s Betrothed . The second point suggests an
overview of the characteristics peculiar to given translations. In the case of IPS (1840),
the translations nearest in time by Daniel J. Connor (1924), Archibald Colquhoun
(1951), and Bruce Penman (1972) have been acknowledged as the best ones, so their
orientations to translation problems deserve foremost discussion. The faults and the
merits of the early, often abridged, 19
th century renditions will also be briefly
considered, as well as the history of the reception of Manzoni’s novel in Britain and the
United States. Finally, the third key-point requires an evaluation of objective textual
criteria in source and target versions to assess degree of efficiency in translation. What
these criteria ought to be, House goes on to explain.
First of all, she proceeds to formulate a tentative definition of translation: ‘translation
is the replacement of a text in the source language by a semantically and pragmatically
equivalent text in the target language’
14
. Clearly expressed, this is not, of course, really
innovative. Therefore, House narrows her focus to broadly equivalent key-concepts
such as text function or text type, which, she argues, is what needs be kept equivalent in
source and translation text. Given sincerity conditions, it could be identified with ‘the
text producer’s intention in producing that text’
15
. It can also be more empirically
obtained by the text itself by ascertaining whether the text serves referential or non-
referential purposes. Adopting Halliday’s and Buhler’s terminology with regard to
language in general, texts are said to display above all ideational/representational
(semantic/referential), and interpersonal/conative (pragmatic/non-referential)
16
functions. Manzoni’s novel is no exception: it is made up of narratorial exposition
(mainly ideational), of narratee-centred comments (interpersonal) and dialogic passages
(interpersonal) above all. Being still too simplistic, the focus is further restricted to
careful analysis of those linguistic feaures in short texts which determine a particular
register profile, which is the key to generalise and identify a text type.
14
Ibid., pp. 29-30.
15
House, p. 30.
16
Cf. K. Buhler, Sprachtheorie (Stuttgart: Fischer, 1965) pp. 28-33; M. A. K. Halliday, ‘Language
Structure and Language Function’, in New Horizons in Linguistics , ed. by John Lyons (Harmondsworth:
Penguin 1970) pp. 140-65. Buhler adds an ‘expressive’ function, it will be accounted for in the second
chapter, devoted to discourse and style in IPS /TB .
10
Ricevi informazioni sui nostri servizi, sulle offerte e non perdere news e consigli su università e lavoro.
