Contrastive Functional Analysis, Andrew Chesterman

Andrew Chesterman is a Professor of Multilingual Communication at the Department of General Linguistics, University of Helsinki. He is the author of several books including ″Constractive Functional Analysis″ and "Can Theory Help Translators? A dialogue between the ivory tower and the wordface" (with Emma Wagner, 2002).

Contrastive Functional Analysis

Chapter One ″Contrastive Analysis″ Chapter One goes into some general issues of contrastive methodology in some detail. We start with the concept of similarity, how it can be defined, analysed and assessed. This leads to a comparison of the ways in which the crucial concept of equivalence has been understood and analysed in the two related disciplines of Translation Theory and Contrastive Analysis. The contrastive functional approach advocated in the book is closely related to issues of translation. It also links up with the psycholinguistic concept of interference: the general issue of psychological realism in Contrastive Analysis is discussed, and related to a recent proposal in neurology. The first chapter concludes with an outline of a falsificationist methodology built around the idea that contrastive studies should produce hypotheses than can be empirically tested.

1. Similarity Assessment

Main problem:

•

Theoretically, what does it mean to compare or contrast two things? What is the “same” or “similar”? Is similarity transitive?

We often compare things in order to give them evaluation. Translation Theories usually handle the issue of ‘equivalence’ between the two texts from this point of view. Contrastive grammar for instance analyses languages. The process of looking for similarities is present everywhere. The fact is that there are different similarities between things that can be perceived. Similarity as such depends on the context. It is not necessarily transitive (A = B ; B = C ; A ≠ C) and not necessarily symmetrical: e.g. the weather is exceptional today = the weather is abnormal today

my son is exceptional → my son is abnormal?

Conclusion: comparison and similarity as such are determined by relevance.

2. Equivalence in Translation Theory

Main problem:

•

In what way has the “similarity” as such been present in translation theories? The equative view, the taxonomic view, the relativist view.

The main thought that the equative view expresses is that the meaning should remain identical in translation; we can change the form but not the meaning. It’s the oldest approach towards translation. In the taxonomic view (Nida) the effect has a crucial role: the target text has to have the same impact on the reader as the source text, so the words may be changed to achieve this (I have arrived → я пришла); it means to say that with some types of equivalence, identity is quite impossible. The relativist view (Reiss, Vermeer) tells that aiming at the equivalence is a pure self-delusion, this argument rejects sameness and similarity: translation takes shape in translator’s head. *Already the languages are too different to speak about similarity: every language has its unique mechanisms, which cannot be transferred in any way—translator makes up his mind how to “explain” them using the mechanisms of the other language.

Conclusion: similarity and equivalence as such are being regarded differently both in different translation theories and by different linguists.

3. Equivalence in Contrastive Analysis

Main problem:

•

What is the translation exactly concerned with and what is the contrastive analysis concerned with? Do they coincide?

It is sometimes suggested that whereas translation is concerned with communication via texts (particular instances of language use in particular situations, as parole), the focus of Contrastive Analysis is on differences and similarities between language systems, grammars, language as langue, while in reality they actually coincide. In the end translational theories aim to translate such texts that “they evoke maximally similar cognitive reactions in the users of these texts” and Contrastive Functional Analysis (CFA) aims to be compatible with psycholinguistic research in order to understand the human mind.

Conclusion: they analyze the same issues aiming to understand the best way of translating and learning the language.

4. On Psychological Realism

Main problem:

•

How is CFA concerned with the process of the language learning?

The original aim of CFA was strongly motivated by the need to improve language-teaching methods and materials. Contrastive studies were regarded psychologically real for scholars were interested in the language-learner’s mind—what difficulties occur in the process of learning? It was noticed that when learning a language various difficulties tend to occur; one is connected with the obtaining new information, other with the mother tongue of the learner: Foreign language learning is based on the mother tongue Positive transfer: similarities facilitate learning Negative transfer/Interference: differences cause problems

Yet as the causes of difficulties are not always linguistic (lots of them are non-linguistic) CFA aims deeper; it aims at the level where CFA can formulate protypes of comparison, which also may include culture and psychology in the process just as the antibodies aim to adjust to deal seccessfully with bacteria in neurology.

Conclusion: The aim of CFA is to help to optimize the process of learning, but it’s merely a platform, which concerns both linguistic and cultural issues.

5. CFA Methodology

Main problem:

•

How does CFA work?

Traditional CA methods as such are described as following:

James

Krzeszowsky

1.assemble the data 2. formulate the description 3.supplement of the data as required 4. formulate the contrasts

1.description 2. juxtaposition 3. comparison

The problem occurs with the unobservable systems; speakers of the language A use certain forms, speakers of the language B use the other forms. In this case the moment of description plays role. Chesterman referres to Pinker’s “The Language Instinct” at this point comparing the process with “putting the ideas into someone’s head”: we say something out loud and something that we’ve said takes a certain form in the listener’s head. There is another methodology of CFA, it is using a perceived similarity as a platform of comparison. Yet language behaviour is predictable to a certain extend only, and CFA doesn’t answer all the questions concerned with the process of learning the language; it is only hopefully better than the theory.

Conclusion: Although CFA is helpful to some extend, it’s good to test the theory against a corpus, other speaker’s intuitions; the more information, the more corroboration.

Chapter Two ″Functional″

Chapter Two specifies what is meant by "functional" in this approach. It provides a brief preliminary outline of a semantic framework forming the basis of a functional syntax, and indicates its main affinities with other functional and contrastive theories. Some practical justifications are offered for adopting a contrastive model based on such an approach.

1. Grammar as a Tool Factory

Functionalists believe that the language is a system in which the means (how we say it?) are shaped by purposes (what we want to say?). Over time there have been several linguistic schools with different approaches; Prague linguists speak of the “needs of communication and expressions” saying that it has to be examined how these needs are satisfied, Bűhler divides functions of the languages into three major points: o Darstellung (representation) o Apell (effect on the receiver) o Ausdruck (self-expression of the speaker) And Halliday combined both Jakobson and Bűhler expressing his own functions of the language: • • •

Ideational function (to express content, to talk/write about sth) Interpersonal function (to establish social relations, to talk/write about someone) Textual function (to organise the form of the talk or text itself)

Contrastive analysis exploits only some of these (based on Halliday): *the idea of lexicogrammar *there’s no sharp border between syntax and lexis On the studies above functional grammars have been practiced: they’re pragmatic, they work within both rules and tendencies, they’re oriented towards structures and words and are context-sensitive.

2. Interpreting the Constraint of Relevant Similarity

If we take the following phrases:

Ein Fixstern hat Jeder Fixstern hat (Die) Fixsterne haben

} eigenes Licht.

Alle Fixsterne haben Fixsterne haben insgesammt

sip: We see that the initial meaning is the same. However, each different form affects the initial meaning in some way, for there are no such things as entirely synonymous expressions. This is exactly what constraint of relevant similarity means: expressions are not equivalent, but similar and similarity is operated on semantic grounds. CFA seeks to encompass the variation being a bridge of the conceptual gap between translation-theoretical notions of equivalence and contrastive-analytical ones; similarity constraint is prioritized, not equivalence. A functional grammar (in this sense) of a single language will thus aim to state the options available for expressing a given initial meaning within the constraint of relevant similarity—describing a language (grammar, semantics).

3. An Outline Model of Semantic Structure

As mentioned previously, any framework for CFA seeks ultimately to do three things: • • •

Provide a theoretical model of semantic structure in general Provide a description of the (primary syntactic) forms of expression of particular semantic structures of two or more languages Provide a description of the conditions of use determining the differential distribution of the various forms of expression of a given semantic structure in the languages concerned.

Based on Mustajoki’s model some description:

The semantic structure of a simple clause centres round a nucleus called a predication. Around this central nucleus there may be also complicators, commentators, and conjucators. Predicates fall into eight main semantic types: 1. 2. 3. 4. 5. 6. 7. 8.

action (Sam washed up, Sam did the washing-up) relation (Sam can’t stand sport, Sam has a hatred of sport) possession (That horse is Sue’s, that horse belongs to Sue) location (Sam lay in bed all day, Sam spent all day in bed) existence (There were papers in the bag, the bag had papers in it) state (Sam has the flu, Sam is suffering from flu) characterisation (Sue is completely trilingual, Sue speaks three languages perfectly) identification (Today is Thursday, it’s Thursday today)

The same subtypes have also the complicators, commentators, and conjucators.

As a conclusion it may be said that a semantic structure is not a “meaning” in itself, it is its analysis of a meaning. Semantic structures can be totally different (the style etc is concerned). Semantic structures represent here hypotheses, based ultimately on argument and evidence that the meaning of such-andsuch an expression can indeed be thought of as having such-and-such structure.

4. Other Functionalist Models

There are other functional models of course; here are some of them. Leech and Svartvik organise their 1st part of communicative grammar of English mainly in terms of general semantic concepts. Foley and van Valin aim to reveal the “contextual independencies, both linguistic and social”. Mustajoki’s model of grammar seeks to be compatible as far as possible with what is known about human cognition. To sum up there are two basic lines: one school of construction grammar “discovers” syntactic constructions (being currently developed by Fillmore) and the other school means an inventory of semantic structures plus their possible forms of extension (Mustajoki).

5. Other Contrastive Models

Contrastive Analysis has to be meaning-based. What is to be compared are the ways of expressing the same meaning in different languages. The contrastivists look both at the selection of wording-signs and their possible combination. There are obvious parallels here with the present model.

Chapter Three ″Analysis″

Chapter Three offers four sample contrastive mini-studies using this approach, at the clause level or below. The samples have been selected to illustrate different aspects of the model; the languages concerned are English, Finnish, German, Swedish and French.

1. States of Disease

Chapter demonstrates a sample of Contrastive Analysis starting with one single language. Ways of saying different diseases can be divided into following sections: o o o o

Any disease Infectious disease Recurrent disease Recurrent, very serious disease

For any disease For any infectious For any recurrent Disease to be felt structure “subject + disease “subject + disease, non-fatal; serious or recurrent HAVE + D” can be CATCH + D” temporary “subject + “subject + SUFFER used BE + off/down with + + from + disease” D” He has measles, a cold, She has caught chicken He is off with flu Alzheimer’s pox He’s got cancer, malaria, He has picked up an Half of the school was down AIDS infection with flu He has a cancerous He is in bed with malaria growth, a heart condition She is laid up with a cold, a touch of bronchial fever He had an attack of panic*

He suffers from asthma, hay fever He is suffering from depression at the moment

It’s also possible to distinguish between the state itself (My tooth aches) and the element of change (he blushed, he died of cancer)

This brief illustration concerns contrasts within a single language, but it also suggests avenues for contrastive studies across languages. So, contrastive functional analysis could be also called crosslinguistic variation analysis.

2. Inclusion

Even though English and Finnish do not belong into the same family-tree, there are still some similarities:

English

Finnish

She belongs to the group of the most Iiri kuuluu kelttiläisiin kieliin influential people at the ministry. (Irish belongs to the Celtic language group)

This chapter is a part of the compulsory Tiotokonepőytä on osa toimiston kalustoa literature (The computer table is a part of the office furniture)

As seen, according to the examples from both languages, both make use of structures with “part” being coded as syntactic subject. The only difference is perhaps the frequency of using either one or other variant of expressing in what way the “part” belongs to the larger group. In English etymological image seems to be one of closure, of enclosing within. In Finnish the viewpoint is that of belonging to the larger group.

3. Invitation to Eat

English Supper! Lunchtime! Come and eat!

German Essen! Bitte Essen! Zum Essen!

French

Swedish

Finnish

À table! Mat! Ruokka! Ruoalle! C’est prêt. Maten är färtig. Ruoka-aika! Le diner / le Ja, var så goda då. Syömään! repas est prêt. Nu är maten färtig.

Would you like to come Komm un iß! and eat now? Lunch is ready. It’s Komm essen! ready! Dinner is served Das Essen ist fertig! Lunch is on the table Gehen wir essen? / Gehen wir essen!

Venez manger!

Kom och ät!

Tulkaa pöytään!

À la bouffe!

Det är mat!

Täällä olisi nyt syötävää keittiössä. Ruoka on pöydässä.

Bien! Passons à Maten är serverad. table. Nous pouvons Nu är det färdigt. Pöytä on nyt katettu, passer à table. olkaa hyvät. Bon appétit! Grub’s up! Das Essen ist Voulez-vous Maten står på Ruoka on valmis/ serviert. passer à table? bordet. valmiina / valmista Please come and sit Darf ich Sie zum Si vous voulez Det är serverad. Saanko/ saisinko down. Tisch bitten? bien passer à pyytää pöytään? table. Isn’t anybody hungry? So bitte, setzen wir Le diner est Var so vänliga, Jospa tulisitte nyt uns an den Tisch. servi. middagen är syömään. serverad. It’s getting cold! Olkaa hyvät ja käykää pöytään. Mr. and Mrs. Gump Tervetuloa syömään request the pleasure of / pöytään. your company at the dinner…

Already within one language there are certain differences when we start with contrastive analysis: “please come and sit down” and “it’s getting cold” may have nothing to do with eating, it’s just a default interpretation connected with eating nowadays (British English).

If we move on to the other languages, we see that in the other languages the same rule applies—there are certain forms of invitation to eat, which, by the way, cannot be literally translated to English, and so the contrastive analysis between the languages begin: • • •

• •

English avoid bear infinitive, À table! Does not exist in English English, Swedish, and Finnish regard the food itself very important in such phrases, while in German and French the “table” is also frequently used In English unlike in German phrases like Darf ich Sie zum Tisch bitten? and So bitte, setzen wir uns an den Tisch is not associated with food—in English it is more common to ask someone to sit at the table is when you want to discuss something, rather formal usage Bon appétit! and Mahlzeit! As such don’t exist in English, there is no natural equivalent There are differences in using “please”, the English “please” is to benefit the speaker, it can be added quite freely and is actually more use in requests such as “Could you please tell me…” while in other languages, especially in Swedish and French “please” is to benefit the hearer— Swedish—literally “be good” and French “if it please you”—their “please” is a part of an expression when inviting or offering.

*such topics may be of course freely raised and debated

4. Genericity (English and French)

There have been many contrastive studies of specifiers. Research on the expression of time, aspect, modality, negation, and definiteness across different languages has been a popular subject over the past few decades. This particular division pays attention to the usage of article in English and French. Seemingly the use of articles is practically the same both in English and French:

English French

plural zero les

mass zero le/la

sing. def. article the le/la

sing. indef. art. a un/une

Yet if we employ contrastive analysis we’ll see: •

Even though it’s possible both in English and in French not to use article at all, the rules of article usage have different extensions in either language: in English it’s possible not to use article more often than in French; firstly because French words have gender and the English don’t, secondly, because in French generalisation as a grammatical and stylistic device is not used as widely. e.g. Man first set foot on the moon in 1969 (“man” without any article = human) L’homme a mis le pied sur la lune en 1969

•

On the other hand there are also rules that apply both in English and French. E.g.

Le lapin se reproduit en moyenne tous les six mois Un lapin se reprodiut en moyenne tous les six mois In the first variant (the rabbit reproduces in average every six months) is that of an externeral point of view; rabbit is being opposed to the other animals, and the second variant is that of an internal point of view—rabbit is a part of a group, one of the common species.

As a conclusion the rules of the target language must be always followed unless the context helps us to come up with total grammatical equivalents (in this case wording will be changed a bit).

Research in the field of articles between English and French has been conducted by Guillaume and Kleiber.

5. Speaker Perspective (English, German Finnish)

This last study is dedicated to the potential range of syntactic expressions for any semantic structure in English, German, and Finnish. We might find, for instance, that certain forms of expression of a given structure-type are typically dispreferred in a particular language.

English Fiona gave Fred a new CD Fred got a new CD from Fiona This new CD came from Fiona

German

Finnish

Fiona gab Fred diese neue CD

Fiona antoi Fredille tämän uuden CD Fred bekam diese neue CD von Fred sai tämän uuden CD Fiona Fionalta Diese neue CD kam von Fiona Tämä uusi CD tuli Fionalta

When looking at these particular examples, there are no major semantic differences, but when looking further it is revealed that English sentences correspond to quite a variety of semantic structures while the same range of variation is not so characteristic of German or Finnish

Fred has lost his earphones Sam burped The cat has the flu Sue loves Sam This key will open that door

Fred hat dir Ohrhörer verloren Fred on kadottanut kuulokkeet Sam hat sich übergegeben Sam röyhtäisi Die Katze hat Fieber Kissa on flunssassa Sue liebt Sam Sue rakastaa Samia Mit diesem Schlüssel kann man Tämä avain avaa tuon oven jene Tür öffnen His 12 novels won him the Nobel Seine 12 Romane brachten ihm Hänen 12 romaanian toivat prise den Nobelpreis ein hänelle Nobel-palkinnon Cancer kills many people Krebs ist die Uhrsache für den Syöpä tappaa paljon ihmisiä Tod vieler Menschen This new computer can do Dieser neue Komputer kann Tämä uusi tietokone tekee melkein practically anything praktisch alles mitä tahansa 5.000 marks could not buy you a Für 5,000 Mark kannst du heute 5,000 markalla ei saisi ostetuksi decent car nowadays kein anständiges Auto kaufen kunnon autoa nykyään The fall of the Berlin wall began a Mit dem Fall der Berliner Mauer Berliinin muurin purkamisest alkoi new era begann eine neue Ära uusi aika A fist banged angrily on the door Eine Faust schlug zornig an die Oveen hakattiin vihaseisti nyrkillä Tür The notice said No Smoking Auf dem Zettel stand Nicht Ilmotuiksessa luki tupakointi Rauchen keilletty

The papers criticized the Pope’s Die Zeitungen kritisierten den Lehdet kritisoivat paavin statement Papst für seine Äußerungen lausuntoa The radio announced that Im Radio wurde angekündigt, daß Radiossa ilmoitettiin, että hurricane was imminent ein Hurrikan im Kommen war hurrikaani oli tulossa The centre of town saw a Das Zentrum der Stadt hat im Kaupungin keskusta koki valtavan remarkable development in 1989 Jahre 1989 eine gewaltige kehityksen vuonna 1989 Entwicklung erlebt The tent sleeps four Das Zelt ist für vier Telttaan mahtu neljä The Norwegians were the subject Die moisten Objekte seiner Witze Norjalaiset ovat hänen useimpien of most his jokes waren die Norweger vitsiensä kohteena The recession did not figure Das Thema der Rezession spielt Lama ei tullut juurikaan esille largely in the President’s speech keine große Rolle in der Rede des presidentin puheessa Präsidenten

German, as seen, is less tolerant of metaphoric expressions (the tend sleeps four → Das Zelt ist für vier), the same with Finnish. It may be hypothesized though that such truths change only according to the genre or text-type, or with general diachronic developments in the two languages.

Chapter Four ″Rhetoric″

Chapter Four suggests ways in which Contrastive Functional Analysis can be extended beyond clauselevel phenomena. The contrastive analysis of textual meaning needs a model of contrastive functional rhetoric. This in turn can be further extended to account for interactional phenomena of contrastive discourse.

To begin with, contrastive functional rhetoric starts with the idea of a message, and then explores the different ways in which it can be expressed. As before, the aim is to establish the nature and range of possible variants of message expression, and determine the appropriateness conditions for their use in different cultures.

1. Background

Various linguists have pointed out that both in CFA and in translation studies the process of searching the links between the texts (similarities, differences) is a relevant issue. There are also parallels with cognitive models when doing this: those models are present and needed in the process of comparison and understanding (linguistic cognitive psychology)

Nowadays there’s a tendency to examine communication means of speakers of different languages above all (all kinds of analysis are used) Contrastive rhetoric has clearly pedagogical roots. It started in the 1960s with the realization that foreign students in the USA had difficulties writing the kind of English that they needed in their studies. It was not that they made grammatical mistakes, but their whole way of organizing a text seemed to be non-English. The locus classicus is Kaplan’s paper, 1966, which suggested that different languages (or rather group of languages) tended to prefer different rhetorical styles, and that students tended to transfer the rhetorical patterns of their native language into their English. Kaplan illustrated this with the drawings that have since become notorious: a linear arrow representing the AngloAmerican norm; a more irregular pattern, with great tolerance for digressions, for the Romance languages and Russian; a series of parallel structures for the Semitic norm; a spiral for Orientals. In this early paper, Kaplan speculated that such rhetorical differences might as well reflect cognitive differences. CFA combined with cognitive psychology became a flourishing field in linguistics. Methodology is pretty much the same as in the rest of the CFA.

2. Text Types

As it is possible to distinguish between text semantics it is possible to distinguish between text types. Here are some classifications:

Bühler

Werlich

Jakobson

Darstellung Appell Ausdruck

Narrative Descriptive Expository Argumentative Instructive

Refential Emotive Poetic Conative Phatic Metalinguistic

Yet even within the same text type there are different semantic means practiced according to the author’s ability to feel the language (even though text type should determine the expressions to be used). A sample study on the composition writing was reported in Connor, 1987. The aim was to compare argumentative compositions from writers in England, Finland, Germany, and the United States when they were to use a particular model of argument structure. Study revealed that in some cases the arguments were used with more variety than in others.

Another illustrative example is Ostler, 1987. She investigated differences between English and Saudi Arabian writers of English, and found that the Saudi writers tended to make relatively more use of balanced parallel and coordinate structures and also proverbial sayings (commonplaces). The inference is that the Saudi writers were transferring rhetorical conventions from their native Arabic. Ostler relates these rhetorical differences to the different histories of rhetoric in the two cultures, again showing how such cultural expectations influence writers being a linguistic transfer in the target language.

3. Text Actants: Episodes

The textual realisation of episodes may vary enormously, from a single sentence to a paragraph or longer chunk, the thing here is that episode type depends on a text type. So, different text types have different episodes. An academic paper in humanities might thus consist of the following sequence: introduction, problem, discussion, solution, and conclusion. As settings tend to be static, they can be regarded as sort of a standard. And taking it logically from here, it may be hypothesised which languages are more fit to these “standards”, which language representatives make the most use of them etc. Studies of the kind have been made (Söter, 1988, concerning the English-Arabic narratives; Indrasuta, 1988, concerning essays of the US and Thai students).

4. Text Specifiers

Any text naturally involves stylistics and while hypothesising on the differences based on the different languages it also important to notice discipline or genre for some text profiles may be rather discipline or genre-determined than culture determined. It is just important to pay attention to the text specifiers and distinguish between purely linguistic-cultural issues. Naturally, speculations always remain, for instance, the aspect of chronology is expressed differently in different languages, so a contrastivist needs to pay attention to the tenses, choices the author has made, was it justified from generic point of view or were they chosen for linguistic reasons etc. There may be several different aspects in text to pay attention to: angle, profile (linear, digressive), point of story etc.

5. Appeals

When examining the text in detail we can distinguish the following features: o Phatos (expressiveness of the text, which depends on cultural-linguistic issues and author’s choice of expression, phatos usually is to have an effect on the reader; phatos may be neutral, emphatic positive, emphatic negative, schematic)

o Ethos ( author’s or speaker’s personal position; ethos is expected to mark either authorpresence in the text or ideology of the text) CFA is concerned with comparison of these logical features; for instance, another study was conducted among the US, Australian, and the Finnish students. Students were to write essays, which later were analysed; it came out that Finnish students used far more metaphoric expressions than their US and Australian peers, contrastivist think it’s due to alliteration-tradition in Finnish folklore which is transferred to the modern newspapers’ language.

6. Coherence, Interaction

Coherence or the persuasive whole may be again different. So far CFA has been concerned with text linguistics. Yet there are several other categories for CFA to operate: • •

Exchange—corresponding to the units of sentence and text, a minimal, two-part exchange is an adjacency pair Exchange types—monologic (public lecture, recipe) Dialogic (dialogue, conversation, exchange of letters) Polylogic (meeting, Internet discussion involving more than two people)

• •

Moves—interaction level on which exchange is built, and individual moves—ask, greet etc. Exchange specifiers—text-semantic choices, direct, indirect.

As a conclusion it may be said that CFR (contrastive functional rhetoric) is interested both in the culture-specific conditions governing such text-semantic choices and in the language-specific formal realizations of these choices. •

Controls—consists of turn control—requesting, starting and finishing, interrupting time control—length topic control whether new topics in the discourse are introduced or not.

The last section, controls, suggests that different language carriers estimate discourse differently—for instance, linguists Bargiela and Harris (1995) have found that Italians in business meetings tend to interrupt a lot, as compared with their British counterparts. This is said only to emphasize one more time what CFR deals with—compares texts, speeches, languages contrastively, it is not to begin a discussion on various nationalities with their cultural space—it would require a separate study, or even many of them.

Chapter Five ″Closing Arguments″

Chapter Five is a brief conclusion, reviewing the main points of the book by way of a passage from Alice in Wonderland.

1. Applications

There’s no doubt that cross-cultural behaviour needs to be studied. An instance of cross-cultural behaviour par excellence is of course translation, and I suggest that CFA is of great relevance here. It is interesting to compare the general approach of CFA with what has been called the variational approach to translation teaching. This approach is based on the idea that rewording lies at the centre of all translation activity. What a translator needs to know are the options available, plus the conditions matching each option. Hewson and Martin outline a method which stats with establishing a set of paraphrases in the source language for a given source-language sentence. CFA seeks at this point to establish the paradigm of options with the same semantic structure. In H&M a similar set of paraphrases is established in the target-language. Constraints of similarity are being searched also in CFA. Both approaches emphasize the initial search for a range of similarities rather than equivalences. CFA seems therefore to be of considerable relevance in translation training.

2. Conclusions

Since the book began with a riddle-like question “Why is a raven like a writing-desk?” It would be a good idea to come back to the initial question “Theoretically, what does it mean to compare or contrast two things? What is the “same” or “similar”?” ; in the story Alice comes up with a following answer: “Come, we shall have some fun now!” thought Alice. “I’m glad they’ve begun asking riddles—I believe I can guess that,” she added aloud. “Do you mean that you think that you can find out the answer to it?” said the March Hare. “Exactly so,” said Alice. “Then you should say what you mean,” the March Hare went on. “I do,” Alice hastily replied; at least—at least I mean what I say—that’s the same thing, you know.” “Not the same thing a bit!” said the Hatter. “Why, you might just as well say that ‛I see what I eat’ is the same thing as ‛I eat what I see’!” “You might just as well say,” added the March Hare, “that ‛I like what I get’ is the same thing as ‛I get what I like’!” This pretty much sums up the whole ida of the CFA. We need to know what to comare and how to compare, we need to understand the essence of similarity that we are looking for. A complete theory of Contrastive Functional Analysis will have three components: 1. A set of possible semantic structures

2. A set of forms whereby these are expressed in different languages 3. A set of conditions governing the distribution of various forms. However, it has to be remembered that the analysis is always open for a debate—we are just searching the best options available, but they’re not unique.