Tõlkesemiootika lugemised

Delabastita, Dirk 1993. There's a Double Tongue: An investigation into the translation of Shakespeare's wordplay, with special reference to Hamlet. Amsterdam; Atlanta: Rodopi. .

If mere curiosity is an indispensable incentive to any scholarly endeavour, its exact objectives and methods will largely depend on the research tradition that serves as its context, both positively in terms of available research results and theories, including the unspoken assumptions underlying them, and negitavely in terms of its real or putative inadequacies. (Delabastita 1993: xi)

Curiosity is indeed indispensable, but it almost seems that all else kind of is dispensable, at least to a degree. For example, one aspect that I'm especially interested in is "speaking about" the "unspoken assumptions". For often these are the most curious factors in any research.

The theory of translation that will be advanced in the first chapter and that underlies this study as a whole has close affinities with the so-called descriptive or polysystems approach in translation studies, and more particularly with the work of James S Holmes and Gideon Toury. This explains why I will avoid any ahistorical, normative definition of translation. Translation will be viewed as a norm-governed concept - as one type of 'text-processing' or 'rewriting' besides others - to be discussed in terms of the functional relationships between and within the source and targen (linguistic, literary, cultural), polysystems in question. (Delabastita 1993: xii)

I can see myself getting into the polysystems approach, given that I can view my "concourse" as just another subsystem among many. This affinity is partly based on cursory acquaintance with the work of Itamar Even-Zohan, whose ideas are comparable to those of Juri Lotman.

[...] there is a point in distinguishing between the 'meaning' of a sentence and its 'interpretation'; by the former I mean "the inherent semantic structure of a sentence as a type, such as it is specified in a linguistic description" and by the latter "the various ways in which one and the same sentence can be understood in each unique case of language use" (Kooij 1971: 117); (Delabastita 1993: xiii)

Hmm. This is actually quite weird, as it assumes that the meaning of a sentence is something distinct from its interpretation and not, for example, that meaning is the product of interpretation. It presupposes a "final interpretant" without interpretation. I'm sure someone somewhere has a name for this type of thinking, but I'm not aware what it could be. In any case, in nonverbal matters this would not fly, because body motion without interpretation is meaningless - if it has no meaning for anyone, thus there is no meaning. But I can see that in linguistic matters you can presume that every sentence definitely has some meaning.

Furthermore, I will make use of typically Saussurian distinctions such as signifier vs signified, paradigmatic vs syntagmatic, and diachrony vs synchrony. I will also try to apply these concepts to non-linguistic semiotic systems and processes. (Delabastita 1993: xiv)

Oh god. With the premises he is working with, I'm not sure if want. The Saussurean theory doesn't usually work well with nonlinguistic semiotic systems and processes. Maybe it's just my prejudice towards semiology.

My choice of structuralist semantic thinking as a general framework is not grounded on some naive belief in its inherent superiority, but first and foremost on the pragmatic consideration that it is naturally more compatible with my polysystemic view on translation, and, more generally, that it is embedded in a rich tradition that has proved to be very stimulating to semiotics and various forms of literary and cultural studies permitting me to achieve a degree of a conceptual coherence that otherwise would have been beyond my grasp. (Delabastita 1993: xiv)

At least it is well-meaning.

A language utterance does not possess an inherent meaning that is readily available to all who simply open their eyes to it. Utterances can become meaningful signs only to those who can handle the natural language in which they are worded. Meaning results not so much from the isolated 'messages' as from the multiple relations of the messages to the conventional 'codes' or sign systems used: what elements from the code's repertory have been selected and according to what rules have they been combined? Such phrases may be commonplace in our semiotic era, but in everyday reality language users rarely seem to appreciate their full implications. (Delabastita 1993: 2)

Didn't the preface say to the contrary that the author assumes a distinction between the "meaning" of a sentence and it's "interpretation"? Damn you, semiotics of language!

There can be no meaningful signs without sign systems, codes have to be shared for communication to take place. (Delabastita 1993: 2)

Ballsy, but not true. There can be no meaningful language sign without a language sign system. This does not apply to all signs... And this is also part of my prejudice towards French semiology which, as Sebeok put it, consigns "the sign behavior of well over two million extant species of animals beyond the semiotic pale" (1990a: 39), all because we can't talk of a language sign system in other species. Ironically it was a Frenchman, Hadamard, who (probably) first proposed the concept private signs, and Morris most likely just followed suit. // The problem is really about identifying communication and signification, as if something can be meaningful only if it is communicable. This is not the case.

The semiotic concept of code derives from information and telecommunication studies:

a code is an agreed transformation, usually one to one and reversible, by which messages may be converted from one set of signs to another. Morse code, semaphore, and the deaf-and-dumb code represent typical examples. In our terminology then, we distinguish sharply between language, which is developed organically over long periods of time, and codes, which are invented for some specific purpose and follow explicit rules that have been invented. (Cherry 1978: 8)

In terms of the properties listed by Colin Cherry, human languages certainly fall short of the technical definition of code - as opponents of the extended use of the concept have not failed to point out (e.g. Schmitz 1980). For instance, communication through language is not at all based on prior agreement between speaker and hearer. Rather, languages are rooted in a lonng history of social behaviour; they are t obe learnt in practice, and are, moreover, liable to change. (Delabastita 1993: 3)

At least the author acknowledges that the concept of code is problematic (as he put it, "has definite conceptual implications", ibid).

The notion of language as a code and the text as messages (and, ultimately, of translation as 'recoding') can in fact be perceived as a modern technological version of one of those basic metaphors we live by (Lakoff & Johnson 1980), more precisely the so-called conduit metaphor, according to which speakers put pre-existing ideas into words and send these words-cum-ideas to the receivers, who merely have to unpack the ideas from their verbal wrapping (Reddy 1979). Words count as mere containers that are sent along a conduit. Similar notions about language, meaning, and the nature of communication demonstrably have a very long history. (Delabastita 1993: 4)

Jakobson traced this history to at least ancient India.

(1) The decomposition or analysis of the S.T. into its constituent S.ling.code elements and structures. (ii) The replacement of the S.ling.code items (elements and grammatical structures) resulting from the S.T. analysis by corresponding T.ling.code items (elements and grammatical structures). The question on what grounds these corresponding T.ling.code items can be or have to be selected from the complete repertory of the T.ling.code, i.e. the question of the exact nature of this 'correspondence' between S.T. units and T.T. units, is evidently one of the cruxes of translation theory. (iii) The recomposition or synthesis of the T.T. on the basis of the output of the transfer phrase, whereby the T.ling.code items that have been selected are fitted together to form the T.T. (Delabastita 1993: 6)

This is a common crux in everyday translation practices as well - the issue of finding a corresponding or equivalent word or phrase to substitute a word or phrase. Sometimes, especially in case of very technical texts, there are no corresponding units and the translation has to take up the responsibility of coining a new word. (Personally, I find this responsibility troublesome.)

Decoding-recoding models of translation tend to assume that the translation process can (and shoud, for that is usually the normative implication) guarantee a relation of identity or near-identity between the meaning of the S.T. and T.T. This is a logical consequence of the way such models assess the cognitive role of language. If language is only the dress of meaning, the linguistic expression can be replaced without really affecting meaning itself. Structuralist theory calls into question the validity of this argument:

We never will have, and in fact never have had, a 'transport' of pure signifiers from one language to another, or within one and the same language, that the signifying intrument - or 'vehicle' - would leave virgin and untouched. (Jacques Derrida, quoted in Van den Broeck 1988a: 272)

The structuralist alternative will therefore present a totally different view on the relation between the meannig of the S.T. and T.T. Rejecting any claims of semantic identity between both, it will insist on the inevitability of meaning shifts in translation. (Delabastita 1993: 6)

This is quite reasonable and I can only assume (at this stage), that culture and other aspects of the text constitute some factors that influence the shifts in meaning.

In spite of their privileged status, our natural languages are apparently not the only means of signification, communication, or organization that human cultures have at their disposal and are built on. Different examples of other sign systems spring to mind: codes of ethics, legal codes, military signals, computer languages, ciphers, culinary codes, vestimentary codes, architectural codes, pictorial codes, musical codes, kinesic and proxemic codes, ideological value systems, daily life routines and rituals, and so forth. Such lists leave one wondering:

How can one bring this inventory to a close? Everything is a sign: presents, our houses, our furniture, our domestic animals. (Guiraud 1975: 90)

In the following pages I will use the term cultural code (and the abbreviations S.cult.code and T.cult.code) to indicate a particular aggregate of such cultural sign systems as have just been listed. I will proceed on the double assumption that a culture is a kind of complex structure that produces meaning (and not just expresses it, so that any strict interpretation of the code metaphor is out of the question) and that linguistic codes and cultural codes do not coincide (even though they are organized according to the same structural principles). (Delabastita 1993: 13)

But kinesic and proxemic codes overlap! These two names were proposed for the same set of phenomena. You could add Wescott's coenetic codes for sake of inclusiveness, but no one uses this term as well (even "kinesic" and "proxemic" are vestiges of a by-gone era). So I'd suggest going nuts with it and have myself found it useful to talk about concursive codes, especially when it comes to verbal representations of nonverbal behaviour.

A structuralist theory of culture in which language is assigned the central position was to be developed many decades later by the Tartu-school semioticians of literature and culture. These theorists regard language as a primary modelling system and culture as a secondary modelling system. The former models our perception and understanding of the world around us in a Sapir-Whorf-like manner. So does the latter, but in the second degree: it acts on, or is engrafted upon the categories of language, and aims to modify and intensify its primary modes of perception. The central role of language is, in fact, one of the most salient features of this theory of culture:

The structure of language, the model of the world it represents, in turn influences the way in which the individual and the culture perceive and understand reality. The extent of this interaction is disputed but the process itself is not. Thus natural language is in fact a primary modeling system in the most literal sense. Virtually all aspects of man's perception and understanding are in some measure shaped by this primary interpretive system. [...] Language as the central mode of human communication underlies a large number of other social systems - tradition, social convention, ritual, religion, [...] the representational arts. (Lotman 1976: xiv)

Even non-verbal art forms such as painting, etc. are reckoned to be "im Banne der Sprache" (Framend 1985). (Delabastita 1993: 15)

But we must keep in mind that the modeling systems are relational and accord to a typology. That is, the influence between the systems can be viewed as reciprocal. That is, it doesn't have to be logocentric if we don't want it to be.

[...] it is sometimes pointed out that cultural analogues are likely to be short-lived themselves owing to diachronic changes within the T.cult.code. The more references to contemporary target culture, the sooner the T.T. will be obsolete. (Delabastita 1993: 18)

This is something that the Nostalgia Critic (a film reviewer on youtube) points out constantly. Disney and other studios often use modern slangs ond inside jokes that are already worn out and ridiculous a decade later.

Bear in mind that Lotman (e.g. 1976: 127-131) describes the relation between the author's code and the reader's code in terms of a struggle. As Culler (1976: 105) puts it,

literature continually undermines, parodies, and escapes anything which threatens to become a rigid code or explicit rules for interpretation. [...] Literary works never lie wholly within the codes that define them, and this is what makes the semiological investigation of literature such a tantalizing enterprise.

(Delabastita 1993: 23)

This is the making-strange of dominant-deformation view taken to the extreme. In actuality it rather seems that many texts are not bent on undermining or deforming literary canons but embrace them. Not everyone is as rebellious as the futurists, there are also "potpoilers" (works written to pay for the creator's daily expenses) and "pulp fiction".

Using Tynjanov's own terms autofunction and synfunction we arrive at the following formulation: the autofunction of an item belonging to a literary work (i.e. that item's relational value within its code) does not necessarily coincide with, and is in each specific textual occurrence subordinate to its synfunction (i.e. its relational value within the literary text in question). The difference between the two is of great importance for the practice of literary translation and by the same token also for translation theory: a new question has entered the field. (Delabastita 1993: 30)

And despite its great importance I have yet to see anyone actually put this distinction in practical use. It rather seems to be another repeat-o-matic bit of theory that gets mentioned again and again but still remains inactive.

The assumptions that implicitly support this approcah can be paraphrased as follows: the surplus value that the synfunction of a S.T. item represents with respect to its autofunction is the result of the structural relationships that the S.T. item in question entertains with the other signs that constitute the S.T.; to the extent, then, that the translator succeeds in finding corresponding target items which have equivalent autofunctions within the target codes, the same sets of structural relationships will automatically emerge in the T.T.; hence, the synfunctions of the S.T. will be preserved. In short, the meaning surplus that results from the interplay of S.T. items will be reproduced in the T.T., if the individual elements themselves can all be rendered equivalently: if the translator looks after the autofunctions, the synfunctions will take care of themselves. (Delabastita 1993: 31)

I didn't catch any of this. Am I too dumb? I'll have to return to this paragraph when I'm more versed in both Marty and Tynjanov.

However, the inadequacies inherent in narrowly linguistic conceptions of translation are due to their unjustified neglect of the 'non-linguistic' dimension of translation and not to their supposedly fallacious focus on valeurs (instead of significations). (Delabastita 1993: 43)

Would the inclusion of intersemiotic and extratextual translation improve the inadequacies?

A norm is both a sort of performance instruction and a criterion for evaluating the performance afterwards. It acts as a constraint on the members of a community whenever they want to carry out the kind of behavioural activities that the norm bears on. A norm is based on the principle of mimetic or imitative behaviour, insofar as it requires that these activities should be performed in the manner of a certain model. A norm can, but need not, be explicitly formulated. In fact, norms occupy a very large middle-ground between two extremes:

objective, relatively absolute rules (in certain behavioral domans, even stable, formulated laws) on the one hand, and fully subjective idiosyncrasies on the other. [...] The norms themselves do not occupy merely one point of the scale, but a graduated section of the entire continuum. (Toury 1980: 51)

Thus, norms may be more or less stringent. (Delabastita 1993: 47)

I haven't been into the norms discourse for years now (ever since reading Holy and Stuchlik 1983), but this is by far the weirdest slash useful definitions of norms.

Adherence to the first type of initial norm results in the production of adequate (or source-oriented) translations. While there is a good chance that such translations will fail to meet the standard text expectations of the target readership, the translations of the second type are called acceptable (or target-oriented) precisely on account of their striving to come up to target readership in this first chapter, translations of the adequate type would attempt to give a full rendering of the entire semantic organization of the S.T., including its synfunctional hierarchies; homological methods would be adopted for cultural and textual S.T. signs; linguistically, analogues would be preferred at the cost of the necessary rule-governedd shifts. On the other hand, translations of the acceptable type would show an overall predilection for analogues on all levels; low-equivalence analogues, additions, and deletions would not be eschewed. (Delabastita 1993: 48)

This is a weird choice of terms, as adequate and acceptable are pretty much synonymous. In the 1970s there were countless iterations of the opposition between adequate and inadequate (even in terms of signs), but for us Tartu semioticians the most familiar opposition should probably between perfect and adequate communication (MacLead 1974: 60; Lotman 1990). I'm counting on understanding the syn- and autofunction distinction applied here later, when I'm on better terms with the distinction.

Semiotically speaking, it will be clear that it is the target or recipient culture, or a certain section of it, which serves as the initiator of the decision to translate and of the translating process [...]. Translating as a teleological activity par excellence is to a large extent conditioned by the goals it is designed to serve, and these goals are set in, and by, the prospective receptor system(s). Consequently, transators operate first and foremost in the interest of the culture into which they are translating, and not in the interest of the source text, let alone the source culture. [...] [T]ranslations are facts of one system only: the target system. (Toury 1985: 18-19; in Delabastita 1993: 52)

In case of Powys, it is a small cluster of students who initiate the process of translation in the interest of a certain portion of the culture into which it will be translated. Thus I would add to Toury's scheme the fact that the target is also first and foremost a "certain section" of the target culture, not target culture as such. There is an unnecessary assumption of cultural homogeneity. In case of my example, it is clear that not every Estonian would be interested in reading Powys.

Holmes, James S. 1988. Forms of Verse Translation and the Translation of Verse Form. In: Translated!: Papers on Literary Translation and Translation Studies. Amsterdam: Rodopi, 23-33.

Literature, Roland Barthes has suggested, is of two classes. In the first place is the class of poetry, fiction, and drama, in which a writer uses language to, as he says, "speak about objects and phenomena which, whether imaginary or not, are external and anterior to language." Besides this there is a class of writing which "deals not with 'the world', but with the linguistic formulations made by others; it is a comment on a comment." (Holmes 1988: 23)

My writings are obviously of the second class, but the picture gets much more complex when you consider that a lot of what I comment on has to do with linguistic formulations about phenomena that are external and anterior to language; and my commentaries do sometimes refer to these phenomena themselves. So the picture is not just black and white; there is a grey space in between.

In defining this "secondary language or meta-language" (as he calls it), Barthes is concerned primarily with criticism, but the range of what I should prefer to designate meta-literature is actually much broader. Round a poem, for instance, a wide variety of meta-literature can accumulate; (Holmes 1988: 23)

And now we have to navigate between such notions as meta-language, meta-literature, meta-fiction, and meta-text. Meta-language is certainly not the same as meta-literature, just as meta-textuality is not meta-communication (in the strict, synchronic, sense). But meta-literature as a term does make sense - it is literature that is written about literature. Literary criticism would be an obvious example but even countless literary works themselves include writings or thoughts on other literature. E.g. a novel by John O'Lughlin casually quoting John Cowper Powys on how beauty can only be attributed to living things or to a soul, but not to nonliving things like a car.

As this brief morphrology re-emphasizes, all translation is an act of critical interpretation, [...] (Holmes 1988: 24)

Translation is not merely interpretation, but necessarily a critical interpretation. From this standpoint one could argue that any serious interpretation or true interpretation is necessarily critical. There are other more solid (hermeneutical?) bases for arguing this, but for my purposes it is clear that if you wish to fully understand a piece of writing, you must not only take it's claims as truth as they are, but examine them and argue with them, to the point that you can agree or disagree at will. Total interpretation in this sense implies total control over both the form and content of the object of interpretation.

[...] no verse form in any one language can be entirely identical with a verse form in any other, however similar their nomenclatures and however cognate the languages. What in reality happens is that, much as one dancer may perform a pattern of steps closely resembling another's, yet always somehow different, because the two dancers are different, in the same way the translator taking this first approach will imitate the form of the original as best he can, [...] (Holmes 1988: 26)

I'm not sure how good this analogy is, because some dance groups are based on not only practicing until the routine has an "identical form" in all members of the group, but go even so far as to include only members that are physically as similar as possible.

The organic form of the metapoem, on the other hand, is a corollary of an organic and monistic approach to poetry as a whole: since form and content are inseparable (are, in fact, one and the same thing within the reality of the poem), it is impossible to find any predetermined extrinsic form into which a poem can be poured in translation, and the only solution is to allow a new intrinsic form to develop from the inward workings of the text itself. (Holmes 1988: 28)

This sounds like a radical version of Jakobson's poetic function.

Jakobson, Roman 2010 [1959]. Tõlkimise keelelistest aspektidest. Tõlkinud Elin Sütiste. Acta Semiotica Estica VII: 299-306.

Bertrand Russelli meelest "ei ole võimalik mõista sõna 'juust', kui ei olda juustuga tuttav keeleväliselt". (Jakobson 2010 [1959]: 299)

Kas meil on võimalik mõista kes või mis on Bertrand Russell kui me ei ole selle või temaga tuttav keeleväliselt?

Sõnade "juust", "õun", "nektar", "tutvus", "aga", "pelk" ja ükstapuha millise muu sõna või väljendi tähendus on kahtlemata keeleline või - kui väljenduda ühtaegu täpsemalt ja laiemalt - semiootiline fakt. (Jakobson 2010 [1959]: 299)

Kui väljenduda veel täpsemalt, aga kitsamalt, siis on selliste sõnade või väljendite tähendus autosemantiline fakt.

Tundmatu sõna tutvustamiseks on tarvis tervet hulka keelelisi märke. (Jakobson 2010 [1959]: 299)

Jällegi: vaja on sünsemantilist konteksti.

Meile, keeleteadlastele ja harilikele keelekasutajatele, on mis tahes keelemärgi tähendus selle tõlge mõneks järgmiseks, teistsuguseks märgiks, eriti selliseks, "milles ta on täielikumalt avatud", nagu on toonitanud Peirce, too märkide olemuse põhjalikem tundmaõppija. (Jakobson 2010 [1959]: 299-300)

Inglisekeelne originaalväljend on, kui ma ei eksi, more developed. Kuidas sellest sai "täielikumalt avatud" on müstiline. Või vähemalt tundub sisaldavat moonutust. Rohkem arenum märk võib rohkem areneda lõputult. Täielikumalt avatud märk saab mingil hetkel täielikult avatud. Meenub mingi metafoor - kahjuks ei mäleta kelle, kas Ricoeuri, Brentano, või jumal teab kelle - et märgi (või teksti?) avamine ei ole nagu karbi avamine, mille põhi muutub avades nähtavaks, vaid rohkem nagu lõputult avanev sügavus.

Me eristame sõnalise märgi kolme tõlgendamisvõimalust: sõnalist märki saab tõlkida sellesama keele teisteks märkideks, teise keelde, või teistsugusesse, mittesõnaliste sümbolite süsteemi. Nimetame neid kolme tõlketüüpi erinevat moodi:

1. Keelesisene tõlkimine [intralingual translation] ehk ümbersõnastamine on sõnaliste märkide tõlgendamine sellesama keele teiste märkide abil.
2. Keeltevaheline tõlkimine [interlingual translation] ehk päris tõlkimine on sõnaliste märkide tõlgendamine mõne teise keele abil.
3. Märgisüsteemide vaheline tolkimine [intersemiotic translation] ehk transmutatsioon on sõnaliste märkide tõlgendamine mittesõnaliste märgisüsteemide märkide abil.

(Jakobson 2010 [1959]: 300)

Märkasin alles nüüd, et nonverbal system of symbols ehk "mittesõnaliste sümbolite süsteem" eeldab süsteemsust. St mitte lihtsalt mitteverbaalne märgisüsteem, aga selline, mis sisaldab põhiliselt sümboleid. Varem oleksin ma oma kriitikat osutanud süsteemsusele kui sellisele, aga ka sümboolsust tuleb arvestada. St märgisüsteemide vaheline tõlkimine (intersemiotic translation) lõdvemas tähenduses hülgab nii süsteemsuse kui sümboolsuse. See on üksjagu oluline punkt, sest "tõlkimise" aspekt muutub sellistel juhtudel eriti kahtlaseks, või vähemalt "tõlgendamisega" asendatavaks.

Sõna keelesisesel tõlkimisel kasutatakse kas mõnda teist, enam või vähem sünonüümset sõna või võetakse appi ümberütlemine. (Jakobson 2010 [1959]: 300)

Jumal tänatud. Ma olen eelmisest semestrist saadik aeg-ajalt kasutanud igapäevase väljendina mugandust tsirkumlokutsioon. Ümberütlemine on ilmselgelt parem.

Luulekunstis valitseb sõnamängi, või, kasutades erudeeritumat ja ehk täpsemat sõna, paronomaasia, ning olgu tema võim absoluutne või piiratud, on luule definitsiooni poolest tõlkimatu. Võimalik on ainult loov transponeerimine: kas keelesisene transponeerimine - ühest luulevormist teise -, või keeltevaheline transponeerimine - ühest keelest teise -, või viimaks märgisüsteemide vaheline transponeerimine - ühest märgisüsteemist teise, nt sõnakunstist muusikaks, tantsuks, filmiks või maaliks. (Jakobson 2010 [1959]: 305)

Oh wow. Ma ei märganudki seda varem, aga vastandus translation/transposition ongi tegelikult vastandus lõdva ja range määratluse vahel. St tõlkimisel peab olema mittesõnaliste sümbolite süsteem, aga transpositsiooni puhul on "märgisüsteem" vabamas tähenduses. Ehk: journal of intersemiotic translation peaks tegelikult olema journal of intersemiotic transpposition!

Torop, Peeter 2000. Tõlge ja/kui retseptsioon. In: Kultuurimärgid. Tartu: Ilmamaa, 16-26.

Vana vastuolu uuenemine lingvistilise ja kirjaliku, täpse ja vaba tõlkekäsitluse vahel 1950-ndatel asendus järk-järgult kommunikatsiooniprotsessi arvestava käsitlusega, nii et 1970-ndatel sai E. Nida eristada tõlketegevuse teoreetilises mõtestamises juba kolme suundumust: 1) filoloogilised teooriad tegelevad ilukirjandusega, žonri jt. kirjanduslike tunnuste tolgitavusega; 2) lingvistilised teooriad tegelevad sisu ja väljenduse vahekordadega keeles ning keelte struktuurse kõrvutamisega; 3) sotsiolingvistilised teooriad vaatlevad tõlget konkreetse kommunikatsiooniprotsessi osana. (Torop 2000: 16)

Keele tasand - lingvistiline; teksti tasand - kirjanduslik; kultuuri tasand - sotsiolingvistiline.

Tema [W. Wilss'i] arvates võikski rääkida eraldi prospektiivsest (tõlkimise üldised raskused, tõlkijate koolitus) ja retrospektiivsest (tõlkevigade analüüs, tõlkekriitika) tõlketeadusest. (Torop 2000: 17)

Kuna tõlkesemiootika aines ei ole me ise tõlgid, tegeleme me just retrospektiivse tõlketeadusega - sooritame tõlkekriitikat.

Üldiselt võib tõlketeaduse arengut kõige lihtsamalt kujutada rõhuasetuse nihkumisena keelelt tekstile, tekstilt kultuurile, kultuurilt tervele ühiskonnale. Seejuures on ühest küljest oluline kultuuri nägemine keeles ja tekstis ning teisalt keele ja teksti nägemine sotsiokultuurilises ümbruses. (Torop 2000: 18)

Käesolevas artiklis siirdume Jakobsoni lingvistiliselt käsitluselt Toropi tekstualistlikule käsitlusele, üritades metateoreetilisel tasandil tõlkide Jakobsoni (keelelisi) tõlketüüpe tekstiteooria sfääri.

Kui näiteks viimaste asatate tõlketeaduses on ülekaalus huvi tõlgete kontaktide vastu kultuuriga, millesse nad satuvad, siis samal ajal tungib see problemaatika ka neisse akadeemilistesse traditsioonidesse, mis on olnud enam huvitatud originaalist, selle keelest ja kultuurist. Nii sündis nn skopos-teooria, mille väitel on iga tõlke dominandiks tema otstarve või eesmärk. (Torop 2000: 19)

Seda "otstarbe või eesmärgi" rõhutamist kohtasin ma mingis muus kontekstis, aga hetkel ei tule meelde kus. Selles tundub olevat terake ideoloogilisust, justkui telooloogia pealesurumine - kõik peab olema eesmärgipärane. Minu jaoks on siin oluline küsimus, et kas "niisama" on ka valiidne põhjus. Seda probleemi illustreerib jällegi Ancient Aliens saade, mis pidevalt eeldab, et muistsed rahvad ei teinud midagi niisama, said kogu oma tarkuse tulnukatelt ja otsisid viise, et nendega suhelda... Mitte, et äkki neile lihtsalt meeldis maaväliste eluvormide idee sama palju kui meile või, et nad joonistasid ja meisterdasid kummalisi asju lihtsalt niisama, ilma mingi ulmelise tagamõtteta.

Järgmiseks ja väga viljakaks sammuks on tõlkele orienteeritud algteksti analüüsi metodoloogia, millele vastavalt on tõlkimine funktsionaalse tulemteksti loomine, kusjuures seos algtekstiga säilitatakse sõltuvalt tõlke sihist. On oluline, et ka tulevase retseptsiooni arvestamise juures lähtutakse siiski originaali säilitamisest, mitte ainlut vastuvõtva kultuuri omahuvist. (Torop 2000: 19)

Minu arvates eristabki see tõlkimist ja tõlgendamist - esimese puhul on oluline uue teksti loomine, teise puhul uue arusaamani jõudmine. Tõlkimine ja tõlgendamine lähevad siin (artiklis) muidugi segamini. Omahuvi ja originaali säilitamise vastandus sealjuures rakendub ka tõlgendamisele. Näiteks Ricoeuri järgi iseloomustab tõlgendamist appropriation - nö enda jaoks või endast lähtuvalt tõlgendamine, aga samas jätab ta välja võimaluse, et tõlgendatakse mingeid väliseid eesmärke silmas pidades (nt teadustööde tõlgendamine mistahes spetsiifilisel otstarbel). Ise adun seda tekstide lugemisel/tõlgendamisel (ja, harva, ka tõlkimisel), et kindlat tekstide seeriat lugedes on mul väga kindel eesmärk: nt siinseid tõlkesemiootika tekste lugedes ei pea ma silmas graffitit, mentaalsust, loomade käitumist vms, vaid esmajoones tõlkimist ennast (kahjuks aga ei saa ma välja lülitada nonverbalismi moodulit, mis kaasneb minu puhul igasuguse lugemise-tõlgendamisega).

Võõra tekstina uues kultuuris omandab tõlketekst erilise omaväartuse. Tal on võime esindada väga erinevaid asju. Kõigepealt on iga tõlge originaali implitsiitne kriitika, s.t. ta peegeldab oma suhtumist algteksti samastumise-eristumise skaalal. (Torop 2000: 20)

Selline implitsiitne kriitika on tõepoolest huvitav nähtus. Ka minu kommentaaride-kirjutamises on see aspekt olemas. Kuigi ma ei seagi endale eesmärgiks iga teksti igat lõiku tõlgendada, ilmneb minu valikutes ka kindel kriitiline moment: see, mis on liiga selge, jääb välja samamoodi nagu see, mis on liiga keeruline. Nende äärmuste vahel on nö kuldne tee mis on ühest käest arusaadav, ja teisest käest annab midagi juurde.

Tõlketekst ei toimi üheski kultuuris iial üksinda. Autonoomse tekstina võib ta sisu kaudu esindada mingit teemat, võib vahendada originaali autori seisukohti ja stiili, võib vastuvõtvat ühiskonda rikastada uute kultuurielementidega, tuua vastuvõtvasse kirjandusse uusi kirjandusvorme ja vastuvõtvasse keelde uusi keelevorme, võib aidata lugejal tõlke kaudu mõista originaali ja võib luua tõlketeose, millest saab vastuvõtva kirjanduse osa. (Torop 2000: 21)

Minu jaoks isiklikult on oluline, et Powyse Üksinduse Filosoofia saaks millalgi lähiaastakümnel eesti keelde tõlgitud, sest see võib meie kultuuri rikastada. Siia võiks lisada nalja sellest kuidas Eestimaa jookseb inimestest tühjaks ja üksinduse filosoofia tekib siin iseenesest, aga ennekõike näen just vajadust sellise mõttesuuna järele - et eestlane õpiks oma üksindust paremini hindama.

Kirjanduslik kommunikatsioon piirdub harva ahelaga Autor-Teos-Lugeja. Kultuuris kui eellugemise, lugemise ja ülelugemise süsteemis on loomulik, et kommunikatsioon jätkub metakommunikatsiooniga. Lugejast saab kultuuritegija ja autor, olgu ta tõlkija, kriitik, õpetaja, tele- või raadiokommentaator, instseneerija või ekraniseerija. Igaüks neist loob oma teksti kellegi teise teksti põhjal, seega metateksti. Semiootilises mõttes on nad kõik tõlkijad. Et aga käesoleva artikli teemaks ei ole tõlkesemiootika, siin piirdun otseselt tõlkimist puudutava osaga metakommunikatsioonis, mida tähistam mõistega metatekstiline tõlge. (Torop 2000: 23)

Esimesel lugemisel (2013 suvi) juba heietasin, et kui nõuda, et metatekst on metakommunikatsioon, siis saab see olla vaid kvalifikatsiooniga - diakoroonne metakommunikatsioon, sest tõenäoliselt ei loeta ühte teksti ja teist, esimese kohta kirjutatud, teksti samaaegselt (kuigi ka selline võimalus on täiesti olemas, on see pigem erandjuhtum). Kõige huvitavam siinses arutelus on minu arvates "eellugemine", millest Bayard kirjutas palju - st me teame päris palju raamatutest mida me ise ei ole lugenud.

Probleemiks on aga eritüübiliste metatekstide vahekord. Tõlke kõrval võivad seista presupositsioonilise või interpreteeriva dominandiga metatekstid. Esimesed annavad vajaliku tausta, teised loova suhtumise, pakuvad lugemisversiooni. Omad ohud on mõlemal. Esimene võib muutuda faktide ja daatumite ballastiks, teine võib võtta lugemisvabaduse. Kuid sama võib juhtuda ka kultuuris. (Torop 2000: 24)

Iseenesest kasulik eristus, aga enamus eessõnasi, nt, mida mina olen lugenud, sooritavad mõlemat. St annavad taustateadmisi, aga samaaegselt ka navad lugemist kindla suhtumise poole. Humoorikas näide mis hetkel meenub on Youtube'i filmiarvustus "The Onion Reviews 'Lee Daniels' The Butler'", mis korduvalt rõhutab: "[it is] a very important film that features very important people and covers very important themes. Overall this is a big important movie that you will watch and think: oh my god, this is all very important."

[...] tekstist välja tõlkimise viisi kaudu on võimalik tuletada žanri eripära ja poeetika. Kõigil viidatud juhtudel on oluline teadvustada, et vastuvõtja ees on sekundaartekst ehk metatekst ehk kaksik- või mitmiktekst, s.t. sõnumi väärtus ja sisukus tõuseb, kui ta algteksti(de)st midagi teatakse. (Torop 2000: 26)

Huvitav kas Mikita mitmiktaju saab sellega võrrelda ja leida mingisugune seni märkamatuks jäänud metatasand? Tekstist välja tõlkimist illustreerib minu jaoks endiselt hästi Michael Radfordi linastus Orwelly teosest 1984, mis on samavõrd hall ja gloomy kui romaan ise. Sellele vastanduvad mõlemad (nii USA ja UK) filmiadaptsioonid Huxley Brave New World-ist, sest kumbki ei näi žanri eripära ja poeetikat hästi tõlkivat. (Samas see viimane näide hellitab lootust, et source material'i ei ole veel ära ammendatud ja erinevalt 1984-st võib ootama jääda ekraaniadaptsiooni mis on kirjandusteosele lähemal kui eelnevad katsed.)