|
Gideon Toury
Tel Aviv University (Israel)
How Come the Translation of a Limerick
Can Have Four Lines (Or Can It?)
... a text is to be translated like a particle in an electric
field attracted
by the opposing forces of the two cultures ... (Peter Newmark 1981: 20)
An introductory Note
A recurring claim is that translation involves texts and nothing but texts.
These texts may be approached differently, but it is always a text that is
submitted for translation, and it is always a text which comes out of it. This
problematic claim notwithstanding, the present discussion will not revolve
around texts at all. Rather, its key notion will be that of MODELS; those
hypothesised entities which underlie the way texts are created, classified and
assessed within a particular cultural set-up (cf. Sheffy, 1997). The notion of model
itself will be used first and foremost in its generative sense. It will thus be conceived
of as a set of guidelines for the production of an infinite number of texts
recognised as belonging to one and the same type, whether these texts (or the
model they supposedly embody) have also been given a collective name or not.
Naming is, of course, an important indicator of the institutionalisation of a
model within a particular culture, but it is far from necessary for its generative
capacity.
A model as a set of guidelines for text production should always be seen as
hierarchically ordered: not every principle derived from it is equally essential to
the text-type in question, and hence to individual acts of performance pertaining
to it. What we will in fact be doing is to isolate for closer inspection one feature
of the model of the English limerick. However, in order to enhance the
significance of our discussion, not just any feature will be singled out but one of
the most salient ones. The claim that what applies to a high-level generative
principle is also applicable to lower-level principles is, of course, much easier to
maintain than its opposite.
Something about the English limerick
In many respects, the following text (Lear, 1965) could be regarded as a
164
prototypical limerick:
There was an old man with a beard,
Who said, `It is just as I feared!
Two owls and a hen,
Four larks and a wren,
Have all built their nests in my beard!'
Like all English texts which embody the model underlying this text, thus
pertaining to what has come to be known as the limerick,1 this text comprises
one stanza occupying exactly five lines which also rhyme, using the aabba
scheme. It is the omnipresence of BEING A FIVE-LINER which renders this feature
the most salient single generative and classificatory principle of the English
limerick: a texter failing to subscribe to it would hardly be said to have produced
a limerick.
To be more precise, what we have here is more of the kind of five PROSODIC UNITS: two
of the lines, the third and fourth, which are shorter than the rest (dimeters versus
trimeters), can be, and have indeed sometimes been, joined together in print to
form the visually longest line of a seeming four-liner, thus partly normalising
the limerick even for the English reader, accustomed to considering rhyming
verses in pairs. However, the resulting typographic tetrameter always featured
an internal rhyme, and at its exact middle at that. It thus fell into two equal, and
rhyming parts (the bs in the scheme), which still permits us to regard the English
limerick as a five-liner throughout.
A prototypical limerick is also basically ANAPESTIC in meter, even though
not all that piously: quite often, metrical irregularities may occur. It also tends
to be COMIC, bordering on complete nonsense. To a substantial extent, this
comic-to-nonsensical impact hinges on the limerick's CLOSING LINE, especially
the way it ties in with the rest of the text, which renders it a kind of punchline.
The OPENING LINE enjoys special status too, the more so as it often imitates the
beginning of an English fairy tale. Finally, the final line is often a repetition, or
varied repetition, of the first.
The following text constitutes another prototypical realisation of the model,
the enumeration of whose features has far from been exhausted:
There was an Old Man of the Dee,
Who was sadly annoyed by a Flea;
When he said, `I will scratch it,'
They gave him a hatchet,
Which grieved that Old Man of the Dee.
Most readers must have known all along what I was driving at. On
the other hand, I suspect that, even among experienced readers of English, there
would be some who have just encountered their first (and second) limericks.
Then there would be those who may never have come across the word `limerick'
before, even though they have encountered texts manifesting the
aforementioned features, or -- if they have -- would have been hard put to
specify the generative rules of the limerick, or even differentiate between better-
and worse-formed texts embodying the model. The requirement seems obvious:
acquaintance with a particular cultural tradition rather than mere knowledge of
a language.
165
Thus, it would hardly be surprising if it were to emerge that readers who have
never encountered the word `limerick', or who would run into difficulties trying
to account for its contents, were found to have had limited exposure to English
culture -- or to any other culture which has had close contacts with it, especially
contacts leading to the appropriation of the limerick by that other culture --
whether lack of exposure to the English culture was merely accidental or typical
of the reader's culture as a whole. In the latter, more significant case, it is quite
safe to assume that no limerick-like texts, or only very few, would have figured
in that culture, whether translated or original, or -- if they had -- that the
products, and especially the model they realise, have settled down somewhere
in the periphery.2 To be sure, if limerick-like texts were to have been produced
in the reader's culture, they would not necessarily bear a generic label of any
kind, much less so the label `limerick'.3 Finally, even if they were to have been
called limericks, the features of what is referred to by that name in this other
culture need not be identical to those expected of an English limerick, in either
quantity or salience (i.e. in terms of sheer number of features or their position in
the hierarchy of the text-type model).
Which brings us directly to our Title Question:
if an English limerick is indeed characterised first and foremost by its being
a five-liner, how come the translation of a limerick into the language of
another culture can have four lines?
This question has both theoretical and descriptive aspects. On the one hand, its
formulation suggests a mere possibility; and questions of possibility, while
definitely theoretical in nature, are obviously located in a most elementary kind
of theory. On the other hand, if and when a possibility is realised, the choice
underlying it cannot remain unexplained, namely, in terms of the circumstances
which can be said to have motivated such a decision. Whatever the explanation,
it would be applied to an empirical finding, a real-life instance of behaviour,
within a descriptive kind of approach. However, its terms of reference would
necessarily be theoretical again, only this time taken from a much more elaborate
kind of theory which has already taken into account previous instances of actual
behaviour. (This spiral mode of progression in Translation Studies is described
in Toury, 1995a:14-17.)
Before dealing with these issues in connection with the English limerick, some
basic assumptions concerning translation as an object for study will be sketched.
These are the assumptions underlying my Descriptive Translation Studies and
Beyond (Toury 1995a). The interested reader is referred especially to the
methodological sections comprising Part Two of the book.
Assumed Translations and the Limerick
For descriptive-explanatory purposes, any text which is assumed to be a
translation, on no matter what grounds, will be taken into consideration. This
notion of ASSUMED TRANSLATION implies at least three interdependent postu-
lates:
166
(1) There is another text, in another culture/language, which has both chrono-
logical and logical priority over it and which can be assumed to have served
as its immediate source (The Source-Text (ST) Postulate);
(2) The process whereby the assumed translation came into being involved the
transference of certain features from the corresponding assumed ST (The Transfer Postulate);
(3) There are accountable relationships which tie the assumed translation to its
assumed original, an obvious function of the transferred features that the
two now share in the context of the differences that exist between them (The
Relationship Postulate).
How these postulates are realised in any particular situation constitutes an open
question rather than a precondition for regarding a textual entity, or the act in
which it came into being, as a translation within the culture in question.4
In fact, there is a whole list of open questions here which will have to be
answered during the execution of the descriptive-explanatory study; for instance:
Was there indeed an ST?
Was there just one such text, or more than one?
Was the assumed ST the ultimate original (the first in the line, so to speak)
or another text, notably a mediating translation?
And what are the relationships actually tying the two textual entities (or
parts thereof) to each other?
To what extent and in what way do these observed relationships realise the
norm of appropriateness as it is held by the culture in question?
And what was the reconstructable process of translation like? What
strategies did the translator have recourse to? (And if there was more than
one pair of hands involved in the passage from the assumed ST to the
assumed target text (TT) -- for instance, a translator and a reviser, or editor,
-- what exactly did each one of them do?)
The starting point for any attempt to account for an assumed translation as the
translation it assumedly is is the assumption that it was designed to fulfil a need
of the culture that hosts it, or rather to fulfil what that need was taken to be by
the translator. However, individual translators are always productsof a given
societal group, and therefore they are bound to make their decisions with
reference to a culturally-determined normative framework, whether they are
fully aware of it or not and whether they wish to enhance or undermine that
framework.
Whatever the need, translating satisfies it in a way of its own which is different
from any other way of satisfying that need, namely, by introducing into the
culture in question a version of something which has already been in existence
elsewhere and which is deemed worthy of introduction into it. This worthiness
too is established within the receiving culture and in its own terms, even if the
status which that something enjoyed in the source culture is being heeded. After
all, the decision to take such a status into account can only be made in the culture
in and for which the translator is operating.
The introduced entity itself, the way it enters the target culture, is something
167
which has never been there before: even in the case of retranslation, what is
actually incorporated in that culture will definitely not have been there before,
no matter how many translations preceded it. Needless to say, the decision to
retranslate itself (rather than reprint, submit to editing, or simply ignore) is
another matter determined by the recipient culture.
At the same time, what is introduced through translation is never completely
new either, never alien to the receiving culture on all possible accounts, otherwise
there would be little sense in saying that it has been introduced into that culture,
let alone incorporated in it. To quote an analogy from medicine: to prevent
immediate rejection, transplantation always requires some measure of matching
between the transplanted organ and the recipient's systems. Moreover, even if
initial matching exists, it will have to be backed up and enhanced through
constant medication to safeguard against alienation and delayed rejection.
Thus, much as translation always entails the retention of aspects of an ST (which
forms a basis for many of the novelties it may introduce into the receiving
culture), it also involves an element of adjustment to its requirements, and not
necessarily in terms of language alone: alien models, and especially their most
deviant features (from the point of view of the prospective target system), if
retained, may be grounds enough for rejection.
Obviously, the greater the proximity of the model embodied by an ST to models
which are already included in the repertoire of the target culture, or the `tool kit'
at the disposal of its texters (Swidler, 1986), the greater the initial possibility of
constituting such a match, whether this proximity (and match) are then
brought to bear on particular acts of translation or not. By contrast, the greater
the distance between the systemic constellations of the two cultures involved in
the contact, the greater the need to make adjustments and allowances -- to the
extent that the act of translation is governed by an initial norm of acceptability.
Once the focus is on the behaviour of flesh-and-blood translators within
definable sociocultural conditions rather than on any ideal(ised) notion of
translation, there is no way of taking for granted any kind of balance between
retention and adjustment -- not even when the opening conditions of the game
have been established. Rather, the realisation of this crucial balance constitutes
a question, namely, of the kind we have just been posing.
In the simplest of cases, the only thing which is added to the recipient culture
by means of translation seems to be an individual text. In more complex cases,
however, fully-fledged models (or elements thereof) may also be imported,
normally with groups of texts which either embody recurring patterns or else
are translated in a way which unifies the end-products. Nor is the novelty an
immediate function of what the ST itself manifests in view of the options offered
by the recipient culture, however striking the incongruity between the two.
Rather, any novelty is a matter of what the target culture is willing (or allowed)
to accept versus what it feels obliged to submit to modification, or occasionally
even reject.
It is precisely for this reason that an act of translation performed on an English
limerick can yield a four-line entity. Thus, not only is there nothing in principle
to exclude this possibility (which would have been reason enough to retain it in
any theoretical account), but there may be circumstances in the recipient
168
culture which would reduce the acceptability of a translated limerick, if
transferred in an unmodified fashion to the verge of total rejection.
Under such unfavourable circumstances, a target-language (TL) five-liner
may well constitute an irregular textual entity, unrecognisable as a poem, or even
as a literary text. If such an entity exists at all, it may be reserved for other uses
and hence regarded as unfit for the recasting of an English limerick. If a translated
limerick is recast in five lines nevertheless, the result is likely to look strange, at
best.
Of course, no translation of a limerick can be expected to reflect all the features
of the English model, let alone all the characteristics the immediate ST
as an act of verbal performance. This is precisely why I have chosen to single
out the most salient feature of the underlying model. Needless to say, a culture
rejecting certain features of an alien model, or even the model as a whole, may
still have its reasons for wishing to possess a version of a text pertaining to this model,
and a version which would not be rejected too readily either. Under such
circumstances, it may well favour, even encourage, the adoption of any number
of solutions, among them the four-line option.
To be sure, there was good reason to select the number four, in this connection,
even though any number other than five would have done: in most Western
literatures, four-liners have been regarded as highly regular poetic entities. Texts
couched in four lines thus come closer to the target culture's conventions, making
this number a paradigmatic representation of the `adjustment' option. (It might
be recalled that even in the English culture itself, limericks have sometimes been
partly `normalised' by having them printed as four-liners.)
In Search of Explanations
Claiming that the translational replacement of an English limerick may have
four lines is, of course, a far cry from claiming that it will in fact have them, under
all or any circumstances. Having agreed on the theoretical claim, the next step
should be to proceed descriptively and try to find out:
(1) whether a limerick has indeed been replaced by a four-liner which has then
been presented and/or regarded as its translation; and if so --
(2) under what circumstances this kind of replacement and presentation
occurred; and finally,
(3) how the correlation between the circumstances (2) and a specific type of
behaviour (1) could be accounted for.
The first question is easy enough to answer. All that is needed for a positive reply
is a single instance of behaviour of the required type, and since such an instance
can always be produced at will, potentiality is proof enough of existence. By
contrast, a truly negative answer could be given only after all assumed
translations, past, present and future, have been scanned -- to no avail. For all
practical purposes, this is tantamount to never, which reinforces the positive
answer yet again.
A descriptivist pursuing regularities of behaviour would, however, welcome
more than a single instance of any type; many more, in fact, and contextualisable
instances, at that. Only a significant volume of material can attest to the real
169
weight of a particular mode of behaviour, and it is only recurrent patterns of
behaviour under definable circumstances that make it possible to extract the
relevant variables and determine their role as conditioning factors of that
behaviour.
Nor would anything be solved by calling a practice such as the substitution
of a TL four-liner for a source language (SL) five-liner `adaptation'. It is not that
adaptation is inferior in any way. It is only that the distinction implied by the
term is not necessarily in keeping with the culturally pertinent one. Any use of
such a label would therefore blur the issue. Above all, it would make it all the
more difficult to establish the circumstances under which a particular strategy,
such as the substitution of a four-liner for a five-liner, tends to be operative,
which can be achieved with no recourse to any misleading, non-cultural and
a-historical distinctions and labels.
As already indicated, a similar set of questions could be directed to any feature
of the limerick and its behaviour under translation. For instance:
Under what circumstances would a TT be produced which rhymes
according to a scheme other than aabba? What rhyme scheme would tend
to replace it, and where would it be taken from? And under what
circumstances would completely non-rhymed entities (for example, jokes
in prose) come into being, in preference, perhaps, to rhymed ones?
Under what circumstances would the first, second and fifth lines of an
assumed translation which is a five-liner be anything but trimeters and/or
its third and fourth lines anything but dimeters?
Under what circumstances would that text be non-anapestic? Non-metri-
cal? Non-comical?
Under what circumstances would its closing verse not act as a kind of
punchline?
and so on and so forth.
However, once a positively-oriented answer has been given to the most extreme
question, there should be no problem in extending that same answer to less
extreme ones, applying it to less central features of the original model and its
realisation in individual texts, on the way to establishing the correlations
among all of them.
To be sure, in spite of the claim that translational considerations and decisions
are basically governed by the needs of a recipient culture, the ST -- and sometimes
the tradition underlying it as well -- still has an impact on the way the act of
translation is performed. It is, however, the case that the way these are taken to
bear on translation will (again) only be determined at the target end of the
process.
In this vein, it would be interesting to find out to what extent the rejection of
features in translational practice runs parallel to their relative position in the
hierarchical organisation of the original model, to the effect that -- as many a
wishful thinker would have it -- it is the less important ones which are the first
to be obliterated. My experience with various case studies makes me doubt the
generalisability of such a parallelism.
Thus, for instance, `having seventeen syllables' was one of the first features to
170
be obliterated when Japanese haiku poems were first translated into various
European languages, even though it is no doubt one of the most important single
features of the original model (cf. Toury, 1995a: 176-80). A similar conclusion
arises from my attempt to establish the circumstances under which Shakespear-
ean sonnets translated into Hebrew lost their most distinctive features and
became Italian-like (Toury, 1995a: 114-22), and especially the detailed analysis
of the reasons why the model underlying a German text of the Schlaraffenland
tradition was not only rejected in its translation into Hebrew, but also replaced
by a (mediating) Russian model (Toury, 1995a: 147-65).
The overall question as to what features of an ST (and model) are the first to be
obliterated will certainly have to be tackled on its own terms, and answered on
the basis of a series of well-targeted studies. There may be laws here, but one
should resist the temptation to formulate them until ample data has been
collected and analysed.
Translating Literary Texts versus Literary Translation
It is numerous observations of this kind, and attempts to account for them
within Translation Studies rather than having the cases they involve dismissed
as simply non-pertinent (in terms of an a priori, inevitably idealised concept of
translation of whatever kind), which led me to climb higher up the ladder
of generalisation and question the notion of LITERARY TRANSLATION as a whole.
I soon came to the conclusion that the term `literary translation' -- as it had
come to be used -- was afflicted with ambiguity, referring as it does to two
different things:
(a) the translation of STs which function5 as literary texts in their home culture:
in an extreme formulation, which has become rather obsolete, any transla-
tion of such texts; in a modified version, a translation where the focus is on
the reconstruction of the ST's internal `web of relationships', the one which
makes that text a unique instance of performance (for example, Snell-
Hornby, 1988: 69ff.);
(b) the translation of a text (in principle, at least, any text, of any type
whatsoever) in such a way that the product be acceptable as literary to the
recipient culture.
The manifestations of the two senses of `literary translation' may, of course,
overlap: there may be circumstances under which a close reconstruction of a
text's web of relationships would indeed override any problems of acceptability.
This may occur, for instance, when the two cultures involved have had very
similar literary traditions anyway, often as a result of continuous contact
between them, or else when the recipient system is considerably weaker vis-a-vis
the source system in question, and is hence willing to use it as a source of
enrichment in ways which transcend the individual piece of text, or, finally,
when the translator him/herself has gained a position in his/her culture which
allows him/her to deviate from sanctioned behavioural patterns and get away
with it, sometimes to the point of being permitted to introduce changes into that
culture. What differentiates the two senses of `literary translation' is not that they
can never coincide, then, but the fact that there is no inherent need for them to
171
do so: even though reality may bring them together, they still remain different
in essence, namely, source-oriented and target-oriented, respectively.
Thus, neither the literariness of the ST, nor even careful embodiment of
its web of relationships within a textual entity in the TL, would secure a position
for the end product in the recipient literature, much less a position parallel to
that which the ST has enjoyed in its own setting. As we have seen, a translation
may well be eligible for rejection, precisely on the grounds that it reflects its
source too closely, thus constituting an irregular entity within the receiving
literature, according to its own norms.
Nor is the distinction between the mere `translation of texts of a particular
kind' and `a kind of translation designed to fulfil a particular function' a
peculiarity of literature alone. One could easily draw an analogical distinction
between, say, the translation of legal documents (the products of which would, for instance, be incorporated in a TL history book and serve as a source of
detailed information about the original document) and strictly legal translation
(where the products are designed to serve downright legal purposes, whether
similar to, or different from, those served by the STs). By the same token, there
are translated versions of many national anthems yet few of these versions also
function as anthems. Moreover, those that do (for example, in bilingual countries
such as Canada or Belgium), do not necessarily reflect the most salient features
of the texts in the other language, from which they were derived. (Cf., for example, Harris 1983.)
Similar things can be said of the translation of a text such as the Hebrew Bible,
which, at one point, was canonised as a Jewish religious text: of the many
possible options only one would be for the translated version to represent the
original Bible in this particular function (let alone to replace it), and even
then, there would still be a difference between functioning as a Jewish and a
non-Jewish religious text. In this sense, not just any translation of the Bible, nor
even necessarily one particular mode of translation, would amount to biblical
translation. Needless to say, the Bible, or a national anthem, or sometimes even
a legal document, can also undergo literary translation in sense (b), to the
extent that attempts are made to have the end products incorporated in the
literary system of the target culture.
Literature is just an example of a general phenomenon, then. It is, however, a
very convenient one for a critical discussion of the concept of translation,
precisely because, on the one hand, it is not normally regarded as so obvious a
case as, say, the translation of legal texts, and on the other hand, it evades many
of the ideology-laden reservations which any discussion of the translation of
national anthems, let alone the Bible, is bound to entail. By the same token, the
limerick is just a convenient example for discussing literature under translation
without sliding into too many issues involving value judgement.
The difference between the two senses of `literary translation' stems from the
fact that literature does not boil down to a body of texts, or even a repertoire of
features which have something inherently literary about them. Rather, literature
is first and foremost a CULTURAL INSTITUTION. Thus, in every culture, certain
features, models, techniques (including modes of translation), and -- by
extension -- texts utilising them, are perceived as, rather than being literary, in
172
any essentialistic sense. What lends a phenomenon or a text its position, as what
Jurij Tynjanov called a `literary fact', is a systemic constellation -- a network of
relationships into which it enters and by virtue of which it functions. Its
literariness is thus established in terms of a given cultural system, and never in
isolation. And, indeed, features, models and techniques, as well as texts utilising
them, may both become literary or lose their literariness in the course of time
without undergoing any change of textual organisation or linguistic formula-
tion. It is thus clear that it is the systemic position which makes the difference,
not any of the surface realisations.
Obviously, only rarely would two different cultural or literary systems fully
concur; and since the (functional) identity of a phenomenon is governed first and
foremost by the internal organisation of the system which hosts it, the literariness
of an act of translation and/or its result can be said to be determined by the way
the requirements of the target literature are brought to bear on it. Of course, these
requirements can bear on an act of translation to various extents too, which
makes `literary translation' in sense (b) a graded notion rather than a matter of
either/or. It is in fact no less graded than the possibility of reconstructing an ST's
features, and, as we have already noted, very often the two stand in inverse
proportion.
`Literary' as a qualifier of `translation' can of course be added to `linguistic'
and `textual', to form a series which is hierarchically ordered in terms of the
specificity of the conditions imposed on the act, while also presenting basic
homology. Thus, with respect to an ST which, institutionally speaking, is literary
itself (as is the case we have been dealing with here), the following would apply:
LINGUISTIC TRANSLATION would be any act yielding a product which is
linguistically well-formed, even if it does not conform to any model of text
formation within the repertoire of the target culture. In this case, at least
partial interference of the model underlying the ST is to be expected.
TEXTUAL TRANSLATION would, in turn, yield products which are well-formed
in terms of general conventions of text formation pertinent to the target culture,
even if they do not conform to any recognised literary model within it.
Interference of the model underlying the ST may still be expected, namely,
in terms of its literary-specific features.
Finally, LITERARY TRANSLATION would involve the imposition of conformity
conditions beyond the linguistic and/or general textual ones, namely, to
models and norms which are deemed literary at the target end. It thus yields more
or less well-formed texts from the point of view of the literary requirements
of the target culture, at various possible costs in the realm of the
reconstruction of the ST's own features.6
Subjugation to target literary models and norms may thus involve the
suppression of some of the ST's features, sometimes even those which have
marked it as literary, or as truly representative of a specific literary model, in the
first place (such as the five-line structure of an English limerick, or the features
which are unique to a Shakespearean sonnet, or the paratactic nature of a
German Schlaraffenland text). It may also entail the reshuffling of features, not to
mention the introduction of new ones, in an attempt to enhance the acceptability
173
of the translation as a target literary text, or even one of a particular literary type.
The added features may go so far as to occupy central positions within the
translation (when considered as a text in its own right), or even serve as markers
of its own literariness within the target culture, despite their having no basis in
the original.
Completing the Circle
For all these reasons, it is easy to see how, conversely, a non-limerick-like text
could be filtered through the limerick model when translated into English; and
it is easy to see how the resulting entity could have five lines even if the original
had, say, four. The translation of a tetrametric entity can be a five-liner, then; the
question is, again, under what circumstances it would tend to be one, which is
where the circle I have tried to outline closes, namely, with the beginning of
another loop. I trust we can now trace this new loop from a more enlightened
vantage point.
Notes
1 The history of the English limerick itself is of course different from the history of the
use of `limerick' as a label. Not surprisingly, it is much longer too. Even the classification
of limerick-like texts as pertaining to one and the same type precedes the introduction
of the label, which seems to have occurred towards the end of the 19th century. For
both histories cf. e.g. Legman, 1964 v. Belknap, 1981.
2 And cf. for example, what Alfred Liede (1963:266) has to say about the attempts to
domesticate the limerick in the German-speaking world, which came to naught in spite
of the existence of thousands of texts, including a substantial number of books,
competitions for the composition of limericks (or for supplying them with the missing
punchline), etc.
3 For instance, in the fifties, when the first Hebrew texts were produced which were based
on the English limerick model, often directly on actual texts pertaining to it, they were
usually referred to simply as shirim ( = poems). When they were addressed to children,
an attempt was made to introduce a new word as a label; namely, lahadamim (a plural
form of the acronym lahadam -- = completely false; or even: fully nonsensical). Later
on the word xamshir was invented, a portmanteau word made out of xamesh ( = five)
and shir ( = poem). This word caught on and gradually became the umbrella term for
all sorts of texts, including English limericks, all sharing the use of a five-line stanza.
4 For a fuller discussion of the rationale underlying these claims cf. Toury, 1995b.
5 It should be clear that the term `function' is used here in its semiotic sense, as the `value'
assigned to an item belonging in a certain system by virtue of the network of relations
it enters into. (Cf. e.g. Even-Zohar, 1990:10.) As such, it is not tantamount to the mere
`use' of the end product, as seems to be the case with other uses of the term, most notably
in the so-called Skopostheorie.
6 In this connection, see also Roda Roberts's discussion of the differences between
functions of language, functions of (source) text and functions of translation (Roberts,
1992), even though she uses the term `function' within a different frame of reference
(see note 5).
References
BELKNAP, G.N. (1981) History of the limerick. The Papers of the Bibliographical Society of
America 75(1), 1-32.
EVEN-ZOHAR, I. (1990) Polysystem Studies. Tel Aviv: The Porter Institute for Poetics
and Semiotics, and Durham: Duke University Press. [= Poetics Today 11(1).]
HARRIS, B. (1983) Co-writing: A Canadian technique for communicative equivalence. In
G. JÄGER and A. NEUBERT (eds) Semantik und Übersetzungswissenschaft: Materialien der III.
Internationalen Konferenz "Grundfragen der Übersetzungswissenschaft" (pp. 121-32).
Leipzig: VEB Verlag Enzyklopaedie.
LEAR, E. (1965) The Complete Nonsense of Edward Lear, ed. H. JACKSON. London: Faber and
Faber.
LEGMAN, G. (1964) The limerick: A history in brief. In G. LEGMAN. The Horn Book: Studies
in Erotic Folklore and Bibliography (pp. 427-53). New Hyde Park, New York: University
Books.
LIEDE, A. (1963) Dichtung als Spiel: Studien zur Unsinnpoesie und der Grenzen der Sprache 2.
Berlin: de Gruyter.
NEWMARK, P. (1981) Approaches to Translation. Oxford etc.: Pergamon Press.
ROBERTS, R.P. (1992) The concept of function of translation and its application to literary
texts. Target 4 (1), 1-16.
SHEFFY, R. (1997) Models and habituses: Problems in the idea of cultural repertoires.
Canadian Review of Comparative Literature 24 (1), 35-47.
SNELL-HORNBY, M. (1988) Translation Studies: An Integrated Approach. Amsterdam-Philadel-
phia: John Benjamins.
SWIDLER, A. (1986) Culture in action: Symbols and Strategies. American Sociological Review
51 (2), 273-86.
TOURY, G. (1995a) Descriptive Translation Studies and Beyond. Amsterdam-Philadelphia:
John Benjamins.
TOURY, G. (1995b) The notion of `assumed translation': An invitation to a new discussion.
In H. BLOEMEN, E. HERTOG and W. SEGERS (eds) Letterlijkheid/Woordelijkheid: Literal-
ity/Verbality (pp. 135-47). Antwerpen-Hermelen: Fantom.
TYNJANOV, J. (1967) Das literarische Faktum. In J. TYNJANOV. Die literarischen Kunstmittel und
die Evolution in der Literatur (pp. 7-36), from Russian original by A. Kaempfe. Frankfurt am Main: Suhrkamp.
|