MICHELA IMPERIO I think-aloud protocol come strumento per indagare il processo mentale della traduzione

I think-aloud protocol come strumento per indagare il processo mentale della traduzione

MICHELA IMPERIO

Scuole Civiche di Milano Fondazione di partecipazione Dipartimento Lingue
Scuola Superiore per Mediatori Linguistici via Alex Visconti, 18 20151 MILANO

Relatore: Professor Bruno OSIMO

Diploma in Scienze della Mediazione Linguistica Marzo 2008

I think-aloud protocol come strumento per indagare il processo mentale della traduzione.

Think-aloud protocols as a method to investigate the mental process of translation.
Michela Imperio

ABSTRACT IN ITALIANO

I processi cognitivi dell’uomo – e in particolare il processo mentale della traduzione – sono stati studiati attraverso diversi metodi di indagine, dall’osservazione di reazioni a stimoli specifici, all’analisi degli errori e dei risultati relativi a un compito svolto. Nel vasto panorama di ricerca in questo campo, think-aloud protocol (TAPs) si distinguono in quanto strumento più adeguato per indagare il complesso processo creativo del tradurre. Essi danno la possibilità di raccogliere dati sui pensieri del traduttore nello stesso momento in cui quest’ultimo li verbalizza, riducendo al minimo il rischio di ottenere informazioni errate o incomplete, dovuto ai limiti della memoria umana. La base teorica su cui poggiano i TAPs sono gli studi sul processo cognitivo umano inteso come processo di elaborazione delle informazioni. Verso la fine degli anni ottanta del novecento, alcuni studiosi pionieri hanno iniziato ad applicare TAPs all’attività traduttiva svolta da studenti di lingue straniere, con fini principalmente pedagogici. La successiva generazione di ricercatori, sulle orme dei primi esperimenti, ha applicato questo metodo con modalità diverse, proponendosi nuovi obbiettivi e avanzando nuove ipotesi (studio delle differenze tra traduttori professionali e non; confronto tra TAPs e joint translation). I TAPs continuano ad avere una grande varietà di applicazione che, se da una parte mette in luce la complessità del processo traduttivo, dall’altra non permette di mettere a confronto gli esperimenti realizzati e di controllare i risultati ottenuti.

ENGLISH ABSTRACT

Human cognitive processes – and particularly the mental process of translation – have been investigated in different ways, e.g. observing reaction to specific stimuli, analyzing the errors and the results of a task performance etc. Think-aloud protocols are the best suited method to investigate the complex and creative process of translating. This method allows to collect data about the translator’s thoughts at the same time he verbalizes them, reducing the risk of getting wrong or incomplete information caused by memory limitations. The theoretical framework for TAP experiments is provided mainly by the studies on human cognition as information processing. In the late 1980s, some pioneer scholars began to apply TAPs to translation tasks carried out by foreign language learners, with pedagogical aims. Following the early experiments, the next generation of researchers applied TAPs in different ways, investigating more specific hypotheses (the study of professional vs. non-professional translator; think-aloud protocol vs. joint translation). The different interests and backgrounds of the researchers involved have resulted in a large variety of independent approaches. TAPs involve different methods of analysis, which, from one hand sheds light on the complexity of the translation process, and from the other makes it difficult to compare different experiments and to test the data collected.

ABSTRACT EN ESPAÑOL

Los procesos cognitivos del hombre – en particular el proceso mental de la traducción –han sido estudiados con distintos métodos de investigación, de la observación de las

reacciones a estímulos específicos, al análisis de las faltas y de los resultados relativos a una tarea desarrollada. En el marco de la investigación en este ámbito, los protocolos de pensamiento en voz alta (TAPs) se destacan como el instrumento más adecuado para estudiar el complejo y creativo proceso traslativo. Ellos ofrecen la posibilidad de recoger datos sobre los pensamientos del traductor al mismo tiempo que éste los verbaliza. De esta manera se reduce el riesgo de obtener informaciones erradas o incompletas, debido a los límites de la memoria humana. La base teórica sobre la cual descansan los TAPs son los estudios sobre el proceso cognitivo humano como proceso de elaboración de las informaciones. Hacia la fin de los años ochenta del siglo XIX, algunos estudiosos pioneros empezaron a aplicar los TAPs a la actividad traslativa desempeñada por estudiantes de lenguas extranjeras, con fines pedagógicos. Los investigadores de la generación sucesiva, siguiendo los primeros experimentos, aplicaron este método de distintas maneras, se propusieron nuevos objetivos y propusieron nuevas hipótesis (estudio de las diferencias entre traductores profesionales y no; comparación entre TAPs y actividad en grupos). Hoy, se sigue a aplicar los TAPs en muchas maneras distintas; esto, por un lado saca a luz la complejidad del proceso traslativo, y por el otro, no permite poner en comparación los experimentos desarrollados y averiguar los resultados obtenidos.

SOMMARIO

Prefazione 4

Problem solving 6
Methods of data collection 7
2. 1. Different verbalizing procedures 8 2. 1. 1. Introspection 8 2. 1. 2. Retrospection 8 2. 1. 3. Questions and prompting 10 2. 1. 4. Thinking aloud 10 2. 1. 5. Dialogue protocols 11
Think-aloud protocols – theoretical framework 12 3. 1. The study of the human mind 12 3. 2. Ericsson and Simon’s model 13
3. 2. 1. Implications of Ericsson and Simon’s model 14 3. 3. Goals 17
TAPs in translation studies 18 4. 1. First studies on foreign language learners 20 4. 2. Further studies: different aims and hypothesis 23
Thinking aloud vs. Joint translation 26 5. 1. Joint-translation’s limits 29
Traduzione con testo a fronte 31
Conclusion 66 Appendice: il testo di riferimento per gli esempi 68 69 70

Ringraziamenti Riferimenti bibliografici

PREFAZIONE

Una delle sfide più grandi con cui l’uomo, da sempre, si confronta è quella di capire come agisce la propria mente, quali sono i meccanismi che stanno alla base dei propri ragionamenti. Perché certe situazioni fanno scaturire determinate reazioni comportamentali, perché si fanno o dicono determinate cose, che cosa ci spinge ad affrontare un problema in un modo piuttosto che in un altro, da che cosa dipendono le decisioni che prendiamo? Non sempre, trovandosi di fronte a situazioni problematiche – dalla risoluzione di un’equazione matematica, alla scelta del vestito da mettersi – l’uomo reagisce ponderando tutti i pro e contro del caso in modo conscio; molte reazioni, gesti, parole, nascono da un ragionamento inconscio, che non siamo in grado di spiegare a noi stessi e agli altri, soprattutto se a distanza di tempo.

La soluzione di situazioni problematiche dipende da molti fattori: l’esperienza personale accumulata, la capacità di raccogliere e analizzare le informazioni necessarie dalla propria memoria o dall’ambiente esterno, il fattore emotivo e così via. Lo studio dei processi mediante i quali le informazioni vengono acquisite dal sistema cognitivo, trasformate, elaborate, archiviate e recuperate è affidato alla psicologia cognitiva; essa analizza principalmente processi mentali come la percezione, l’apprendimento, la risoluzione dei problemi, la memoria, l’attenzione, il linguaggio e le emozioni.

La traduzione, fra molte altre, è un’attività complessa, che comporta la risoluzione di tanti piccoli problemi; è un processo di scomposizione, comprensione e analisi del prototesto, di ricerca di strategie traduttive, di nuova sintesi per la creazione del metatesto nella lingua della cultura ricevente: attività che qualsiasi traduttore compie in modo più o meno consapevole, ma che richiedono un grande sforzo di concentrazione, in quanto parte di un’opera creativa. La traduzione, quindi, è frutto di un complicato processo mentale, soggettivo e creativo, di cui è impossibile cogliere tutti i passaggi e le sfumature, senza affidarsi a metodi di studio specifici.

È in tempi relativamente recenti (anni ottanta del novecento) che alcuni studiosi europei, insoddisfatti degli studi fino ad allora realizzati, hanno

iniziato a sondare più a fondo le dinamiche del processo mentale del traduttore, affidandosi a metodi di indagine induttivi ed empirici. I think- aloud protocol sono stati, finora, lo strumento più diffuso e adatto a questo tipo di indagine; si tratta, infatti, di una procedura di verbalizzazione che avviene simultaneamente al compito svolto dal traduttore, al quale viene chiesto di riferire ad alta voce ciò che avviene nella sua mente, mentre traduce.

Lo scopo di questo studio è quello di offrire una panoramica dei diversi metodi utilizzati nel corso degli anni per indagare il processo mentale della traduzione e, nello specifico, di presentare – attraverso alcuni studi esemplari realizzati negli ultimi trent’anni – la grande varietà di applicazione dei think- aloud protocol, varietà che, se da una parte apre orizzonti verso nuovi oggetti di studio e nuovi metodi di indagine, dall’altra non permette di mettere a confronto le ricerche condotte e di controllare i risultati ottenuti.

1. PROBLEM SOLVING

Problem solving is the process of finding solutions to complex problems for which the answer is not necessarily evident. It can be described as a goal- directed cognitive process that requires effort and concentration. This can be caused by the fact that we can’t retrieve the answer directly from memory, but we must construct it from the information available in memory or information obtained from the environment (for example, the givens of a problem or extra information that can be requested). In other cases, “finding the answer involves exploring many possible answers none of which is immediately recognized as the solution to the problem” (Someren, Barnard and Sandberg 1994: 8). Therefore, we don’t find the solution directly in a single step but via intermediate reasoning steps, some of which may later appear useless or false.

A person’s ability to solve problems relies on his innate ability to mentally organize stimuli into relevant and useful schemas that can be used to deduce a solution from a limited stock of information. People frequently have to deal with problem-solving activities, professionally as well as privately. Some of these problems are well defined, for example algebraic equations, questions in a school chemistry test, medical diagnosis problems in a standard setting; in other cases, the problem itself and its potential solution are not so well defined and it is not so easy to evaluate the solution in terms of correctness. Examples of these activities are: designing websites, buying a new house, selecting a new bass player for a band, translating a text. Such activities require the solution of many smaller problems.

In everyday life one has to solve a lot of problems. What cloths shall we put on today, what is the most efficient route to the university, how much does a kilo of apples cost given the price of a pound, how to discuss troubles with one’s friend etc. Sometimes we are well aware of the fact that we are trying to solve a problem, for instance when we are trying to calculate how much something costs in a foreign currency: we consciously try to remember the exchange rate, in order to perform the mathematical operations required. But at times problem solving goes on without noticing, i.e. we may not perceive our mental process as problem solving.

There can be different reasons to study problem-solving processes. For example, a psychologist may want to investigate what mechanisms underlie human reasoning, what causes errors, the character and origin of people’s different performances. An educational scientist may be interested in the effect of education or in children’s difficulties in solving exercise problems. A knowledge engineer may want to analyze how a subject carries out a task, in order to try to build a computer system that can do the same. The aim the researcher may have partly determines the nature of the procedure he follows when using protocol analysis for collecting data about the cognitive process (Someren, Barnard and Sandberg 1994).

2. METHODS OF DATA COLLECTION

One class of methods of data collection on the problem-solving process is based on observation of problem-solving behavior. The first one, product analysis, uses the results of problem solving: the solution to a problem may reveal aspects of the problem-solving behavior. It is possible to obtain further information by observing the problem-solving behavior concurrently while it takes place.

Besides simple observation of results and behavior during problem solving, properties that are not directly visible or audible may be examined by using special equipment: researchers may register eye movements during problem solving or even measure activity in various parts of the brain by special techniques, which may provide data on what information is being focused on and processed at a certain moment.

Behavioral observations are registered as action protocols. One of the few techniques that give access to data about the problem-solving process is storing a behavior trace if a person manipulates objects during problem solving (for example when using a computer).

Another class of techniques used both in psychology and in knowledge acquisition, is based on predefined forms in which the subject should express his knowledge. According to the task and the purpose of the research, it is possible to use an infinite variety of formats. One of them are questions with

predefined answers from which one or more is selected (Someren, Barnard and Sandberg 1994).

A final class of methods involves unstructured verbal reports of problem solving, which can be obtained in different ways.

2. 1. DIFFERENT VERBALIZING PROCEDURES

“It is assumed that those mental activities which are dealt with in working memory (i.e. which are to some degree conscious) can be verbalized” (Jääskeläinen 1999: 62).

According to Ericsson and Simon (1984) a distinction should be made between various kinds of verbal report procedures (or introspective methods), and particularly between classical introspective reports, retrospective responses to specific probes and think-aloud protocols. This distinction is crucial in determining the reliability and the validity of these methods of data collection (Jääskeläinen 1999).

2. 1. 1. INTROSPECTION

Classic introspection consists in instructing the subject to report his thoughts at intermediate points of the problem-solving task, which are chosen by him. As used by psychologists in the 1920s and 1930s, researchers also ask the subject to give an accurate, complete and coherent report on his cognitive processes. As a result, introspective reports involve the use of psychological terminology and interpretation by the subject; for this reason, they are also more subject to memory errors and misinterpretations than other methods (Someren, Barnard and Sandberg 1994).

The problem involved is that “the informant is also expected to act as the analyst/researcher. Consequently, both the data and the analysis are subjective” (Jääskeläinen 1999: 63); there is no objectivity in the sense of the object of research being independent of the researcher.

2. 1. 2. RETROSPECTION

In retrospective responses to specific probes subjects are invited to perform a task and afterwards they are asked questions about their behavior

during the performance. The problem-solving session can also be recorded on video; this way, the experimenter can then review the video-tape together with the subject, who can give his interpretation of what happened” (Someren, Barnard and Sandberg 1994).

“The informant is therefore no longer the analyst, which makes the analysis more objective and the findings open to falsification” (Jääskeläinen 1999: 66).

However, there seem to be some problems. In fact, a person could find it difficult to remember exactly what he did, especially if some time has passed after the task has been carried out. Sometimes even, one is not aware of what he is doing. Furthermore, subjects may tend to report their thought process as more coherent and intelligent than it originally was, giving the false impression of perfectly rational behavior (Someren, Barnard and Sandberg 1994).

This kind of post hoc rationalizing can be intentional or unintentional. In fact, humans tend to reconstruct events as more structured than they originally were, because their memory is guided by their knowledge of the result.

Some researchers have shown that the data obtained by retrospection are not always valid (Nisbett and Wilson, 1979; Ericsson and Simon, 1993). They examined closely the conditions under which reports are considered unreliable and they discovered that

all discrepancies were found in situations in which there was either a delay in time between the cognitive process and the report, or there was a question by the experimenter that required an interpretation rather than a direct report, (‘Why did you do X instead of Y?’), or both. (Someren, Barnard and Sandberg 1994: 22)

When subjects are asked for memories, explanations or motivations, they don’t answer from direct memory of the cognitive process but from an interpretation of it that can be influenced by expectations.

The memory model can explain why. Retrospection means that people must retrieve information from long-term memory and then verbalize it. The inconvenience is that the retrieval process may not reproduce all the information that was actually present in working memory during the problem- solving activity.

Furthermore, it is also possible that people retrieve information that was not actually in working memory as if it was. “After solving the problem, the solution will help to remember the steps that actually led to it” and to reconstruct them easily. “However, odd and fruitless steps that occurred on the way are less likely to be retrieved” (Someren, Barnard and Sandberg 1994: 22).

2. 1. 3. QUESTIONS AND PROMPTING

Another verbalizing procedure implies actually interrupting the problem- solving process: subjects are asked questions during the activity or are prompted at given intervals to tell what they are thinking or doing. Therefore, they don’t have the chance to smooth over the answer as in retrospection. The drawback of this method is that it interrupts the problem-solving process and subjects may have difficulty in taking up the thread. Moreover, prompts that require interpretation may affect the problem-solving process (Someren, Barnard and Sandberg 1994).

2. 1. 4. THINKING ALOUD

The thinking aloud method differs from classical introspection and retrospection in that it is undirected and concurrent. The verbalizations are produced simultaneously with the task performance, but the subject is not as a rule required to verbalize specific information. “Due to memory limitations concurrent and undirected reporting is likely to capture more of the process (less is forgotten) more reliably (less is distorted)” (Jääskeläinen 1999: 66).

According to Ericsson & Simon (1993), thinking aloud does not interfere with the task performance and the thought process. The subject solves a problem while the talking is executed almost automatically; in fact, almost all

of his conscious effort is aimed at solving the problem, and there is no room left for reflecting on what he is doing. For this reason, there is no delay and the data gathered are direct; the subject does not interpret his thoughts nor is he required to bring them into a predefined form, but he renders them just as they come to mind. However, think aloud protocols are not necessarily complete because a subject may verbalize only part of his thoughts (Someren, Barnard and Sandberg 1994).

With tasks in which thinking aloud is not possible (e.g. simultaneous interpreting), data can be collected through retrospective verbal reports.

“They ought to be elicited immediately after the task performance (immediate retrospection) and with as little interference from the experimenter as possible” (Ericsson and Simon 1984: 19).

2. 1. 5. DIALOGUE PROTOCOLS

Although monologue protocols are still predominantly the main tool for collecting data, the artificiality that still remains has led some researchers (House 1988; Hönig 1990 and 1991; Kussmaul 1989a, 1989b, 1993 and 1994; Schmid 1994) to get subjects to talk to each other. In a small-scale experiment, House compared monologue and dialogue protocols applied to translation tasks. The findings show that in monologue protocols processes as selecting target language items, weighing alternatives and choosing a particular translation equivalent remained unverbalized (House 1988). In contrast, when people collaborate they will sometimes have differing opinions. Thus they are forced to give arguments, to clarify steps of their thinking processes. In fact, when talking in pairs, subjects negotiated solutions to translation problems and each individual’s thoughts appeared to have been consistently shaped due to the necessity of having to verbalize them. House concluded that the dialogue situation provided richer data than monologue protocols (House 1988). “Later TAP experiments have shown, however, that the richness of data depends on the type of subjects and the translation brief, and, above all, on the priorities of the researcher” (Kussmaul and Tirkkonen-Condit: 180).

3. THINK-ALOUD PROTOCOLS – THEORETICAL FRAMEWORK

Over the last three decades, think-aloud protocols have become a widely- used method in the study of cognitive processes as problem solving, reading and writing and human-computer interactions.

As I mentioned before, the method of thinking aloud consists in organizing an experiment in which subjects are asked to carry out a task and to verbalize their thoughts while performing. The task performance is recorded on audio- or, preferably, on video-tape. The resulting recordings are then transcribed (think-aloud protocols, or TAPs) and subjected to analysis.

It is important to note, however, that “thinking aloud” as a method of eliciting data is not the same as “thinking aloud” in the everyday sense; it entails more than sitting people down next to tape-recorder and asking them to talk (Jääskeläinen 1999: 9).

3.1. THE STUDY OF THE HUMAN MIND

The think-aloud method has its roots in psychological research.

One of the hardest problems in research dealing with mental process is that the workings of the human mind cannot be observed directly the way some other objects of scientific endeavors can be. Instead, indirect means are necessary, which creates obvious problems for research (Jääskeläinen 1999: 53).

One of the first approaches to the study of the mind (in the late 19th century) was to train people to introspect upon their own thought process. Classical introspection is based on the idea that events that take place in consciousness can be observed, more or less the same way events in the outside world can be. It is a problematic research method, mainly because the events that take place in consciousness, which are to be analyzed and explained, are accessible only to a single observer, who also performs the thought process. The source of data also provides the analysis of the data; therefore the analysis is totally subjective and it is impossible to replicate empirical studies and thereby to settle scientific discussions about thought

processes. Due to the built-in limitation of the introspective method, psychologists turn away from it and from all associated theories. But introspection was a central method in studying cognitive processes and consequently psychological research turned away from cognitive processes too (Someren, Barnard and Sandberg 1994).

Understandable counter-reactions followed. One of them was the behaviorist paradigm (1930s) which promoted psychology as a hard positivist science. Its purpose was to limit psychological research to objectively observable behavior. This entailed abandoning subjective research methods like introspection, because “the objective, i.e. scientific, study of the human behavior could only be based on the analysis of the relationship between external stimuli and behavioral responses” (Jääskeläinen 1999: 55).

Behaviorism dominated American psychology, while European researchers, particularly representatives of the Gestalt Psychology school of thought, had a slightly different view of how to do psychology. Although they also rejected classical introspection as a research method, they wanted to study thought and not just behavioral responses to stimuli. They also developed more sophisticated methods of data collection on the thought process: phenomenological observation, phenomenological introspection, and the method of thinking aloud (Börsch 1986). The beginnings of the cognitive paradigm are usually dated in the mid 1950s and picked up where Gestalt Psychology left off (due to the beginning of the Second World War) (Jääskeläinen 1999).

3. 2. ERICSSON AND SIMON’S MODEL

Thinking aloud as a method for scientific research rests on a solid scientific foundation in cognitive psychology, a science that studies human cognition, i.e. how humans receive, store, manipulate, and use knowledge.

The theoretical framework for TAP experiments is provided mainly by the work of Ericsson and Simon (1984). These scholars base their theory of verbalization on the information-processing approach in cognitive psychology,

i.e. they assume that human cognition is information processing. According to Ericsson and Simon’s model, humans keep information in different memory stores, characterized by different access and storage capabilities: short-term memory (STM) present easy access but severely limited storage space, whereas long-term memory (LTM) is characterized by more difficult access and larger storage space (Bernardini 1999).

Information input will first be heeded by the STM and when its capacity and storage time is exhausted, the information is transferred to the LTM. A certain loss both prior and during this transfer is assumed, but it does not seem to be a substantial loss (M. A. Schmidt 2005: 27).

Only information present in STM, that is information which is currently being processed, can be directly accessed and reported; LTM contains information which has left consciousness, but which can later be retrieved back to STM for further processing.

The STM is also called working memory (WM); it is the primary site of the procedural memory. LTM, by contrast, serves as the vessel for the declarative memory (M. A. Schmidt 2005). As far as the translation process is concerned, it is important to consider the function and capacity of the STM because translating relies as much on procedural knowledge as on declarative knowledge. “This distinction is crucial because the cognitive processes, as well as information that is not currently being processed, cannot be reported but must be inferred by the analyst on the basis of the verbalizations” (Bernardini 1999: 2).

The implications of Ericsson and Simon’s model are manifold.

3. 2. 1. IMPLICATIONS OF ERICSSON AND SIMON’S MODEL

First of all, according to Ericsson and Simon’s model, only concurrent verbalization of thoughts exhaustively reflect the mental states of a subject carrying out a relatively long task, which takes longer than ten seconds to complete, according to Ericsson and Simon (Bernardini 1999). It is important to notice that a cognitive process takes longer when the subject thinks aloud.

“This means that people are able to slow down the normal process to synchronize it with verbalization” (Someren, Barnard and Sandberg 1994: 33). When the subject has completed the task, part of the information moves

on to LTM, leaving behind retrieval cues in STM: in such cases, it has been found that post hoc verbalization is difficult and often incomplete (Ericsson and Simon, 1993). Moreover, under these circumstances, it can be extremely problematic to exclude the possibility that a subject is interpreting his own thought processes or even generating them anew, instead of retrieving them from LTM.

Secondly, to make sure that the subject actually reports his mental states without distorting them, it is important that he does not feel he is taking part in a social interaction: although conversation is obviously a natural situation, it involves reworking thoughts to conform them to socially established norms; this process might sensibly alter the information attended to (Bernardini 1999). Emotional and motivational factors can produce a cognitive process different from the process that would take place without thinking aloud. “There is not much evidence that thinking aloud adds much to the effect of being studied and evaluated that is inevitable in knowledge acquisition and experimental settings” (Someren, Barnard and Sandberg 1994: 33). The interaction between subject and experimenter (or between subjects) should therefore be avoided or at least reduced to a minimum. There is one other cause for concern: if the subject keeps silent for a long time, the verbalization will become useless, because significant parts of the cognitive process in STM may not be tracked down. To avoid this, the experimenter is allowed to repeat to the subject to think aloud with a short and non-intrusive reminder; Ericsson and Simon propose to use the phrase “keep talking” (Krahmer and Ummelen 2004).

Thirdly, “practice and experience may affect the amount of processing carried out in STM, so that fewer mental states will be available for verbalization to subjects experienced in a task” (Bernardini 1999: 2). This process, known as ‘automation’ refers to the fact that “as particular processes

become highly practiced, they become more and more fully automated” (Ericsson and Simon 1984:15) and do not require active processing in working memory, i.e. they are executed at an unconscious level and become less accessible for verbalization. “To give a simple example, a novice driver has to focus all of his attention to driving; after most of the process involved in driving has become automatized, it is possible to engage in a conversation while driving” (Jääskeläinen 1999: 59). These kinds of processes are faster and more efficient than those under conscious control, but they are also less flexible and more difficult to modify at need. However, it is possible to bring at least some automatized processes back to conscious attention; otherwise teaching would be virtually impossible.

There exist several other obstacles on access to process, for example, a heavy cognitive load during a task performance. Due to STM’s storage limitations, subjects tend to stop verbalizing if they have to pay attention to too many things at the same time. “In some cases, processing uses all the available capacity and none is left for producing verbalizations” (Jääskeläinen 1999: 59).

For example, if reasoning takes place in verbal form, verbalizing the working memory’s contents is easy and doesn’t use memory capacity. However, if the information is non-verbal and complicated, verbalization will take time and space in working memory because it becomes a cognitive process itself. Consequently, the report of the original process will be incomplete and the process itself can even be disrupted (Someren, Barnard and Sandberg 1994).

Finally, Ericsson and Simon take into account the effects of personality and personal history over the data collected through TAPs. Individual differences in knowledge and ability to verbalize thoughts can heavily bias the data collected through TAPs. The problem here is the object of study and not the methodology used: individual differences exist, and research should not conceal them. However, it seems advisable to try to limit the effects of

individual differences and to take them into account when analyzing the data, in order to obtain more reliable and generalizable data (Bernardini 1999).

All these limitations imply that verbalizations represent but a minute fraction of the total amount of mental activities occurring at any moment in time. However, this does not mean that this fraction would be somehow unimportant or uninteresting for research. Moreover, the fragmentary verbal reports can be completed with other kinds of data, such as questionnaires, process-product comparisons, eye-movements, pauses, etc. (Jääskeläinen 1999).

Thinking aloud is unnatural. Therefore, Ericsson and Simon recommend an initial warm-up session in which subjects are taught to verbalize their thoughts. During this practicing phase, the experimenter should feel free to disrupt the task and talk to the subject, whereas during the experiment he should be very concerned not to interfere. During warming-up

subjects should learn the difference between describing what they are doing (“I now move a disk from here to there”) and thinking aloud (“since this disk is smaller than that one, I put it on another pin first” (Krahmer and Ummelen 2004: 3).

They are also instructed to avoid making analytic comments about their tasks. Verbalizations are non-limited: the participants are instructed to say

aloud what comes into their minds without any restrictions.

3. 3. GOALS

Think-aloud protocols have been used for three types of goals:
1. To find evidence for models and theories of cognitive processes: Newell & Simon for instance, used TAPs for collecting data to develop and support a theory of human problem solving. Many other researchers have been working at the development of models of the cognitive processes involved in writing. One of the most known is Flower and Hayes’ model, presented in 1981. It was

the starting point for a discussion about the use of the think-aloud method in writing research.
2. “To discover and understand general patterns of behavior in the interaction with documents or applications, in order to create a scientific basis for designing them” (Krahmer and Ummelen 2004: 1). Carroll, for example, used TAPs to investigate how learners interacted with new software. It has been found that they were annoyed by the huge quantity of irrelevant information contained in tutorial manuals. In fact, manuals appeared not to satisfy the users’ goals and questions. TAPs analyses also showed how software users learn to work with a new system; consequently, researchers were able to develop a new design for software manuals: the minimal manual.

3. To test and revise functional documents and applications such as manuals and websites. Researchers like Schriver and Nielsen used verbal protocols (usability testing, pre-testing, formative testing) to gather users’ information to support the design of a specific product (Krahmer and Ummelen 2004).

4. TAPs IN TRANSLATION STUDIES

The analysis of think-aloud protocols (TAPs) in translation studies began in Europe in the late 1980s. Scholars felt the necessity to develop empirical and inductive methods in order to complement the predominantly deductive and often also normative models of the translation process presented until then, which usually described what ideally happened or rather – with a pedagogical aim – what should happen, in translating. It was researchers like Krings, Königs and Lörscher in Germany, Dechert and Sandrock in Britain, Jääskeläinen and Tirkkonen-Condit in Finland, who began to ask what actually happens when people translate (Kussmaul and Tirkkonen-Condit).

This new trend can be partly explained by developments in the adjacent disciplines: psychology had renewed its interest in the study of the mental process (as opposed to patterns of external behavior) with the consequent choice of appropriate or legitimate methods of research. This change had an

impact on psycholinguistic research, including research on second language learning, and, via L2 research, on translation studies (Jääskeläinen 1999).

“There has always been a kind of empirical research, like translation criticism and error analysis, but this was product- and not process-oriented” (Kussmaul and Tirkkonen-Condit: 177). In fact, when comparing the target text with the source text or when looking at errors, one could at best speculate in retrospect about what had occurred in the translator’s mind during translation. What was needed was a way to discover what actually happens, “to get a glimpse into the ‘black box’, as it were” (Kussmaul and Tirkkonen- Condit: 178).

In this sense, viewing translation mainly as a problem-solving activity, some scholars proposed that it should be possible to study it by means of TAPs, and set up experiments to test this hypothesis. The different interests and backgrounds of the researchers involved have resulted in a large variety of independent approaches (Bernardini 1999).

This kind of analyses increases our potential for describing and explaining the translation processes, and thus our theoretical understanding; moreover, they have at least two pedagogical purposes. (1) The different strategies observed in the TAPs may serve as models for successful translating (Lörscher 1992a; Jääskeläinen 1993; Krings 1988; Kussmaul 1993). (2) If translation students are used as subjects, TAPs may be used to find out where they have problems. The data collected can then form a basis for translation pedagogy (Krings, 1988; Kussmaul, 1989a+b, 1994). It might be argued that teachers of translation, from years of experience, already know which strategies to recommend to learners. But sometimes they draw the wrong conclusions from their students’ translations. Teachers may, for instance, have the impression that students have problems with text-comprehension while, when talking to them, they find that students actually have problems expressing what they had understood. TAPs can help teachers to see matters more clearly (Kussmaul and Tirkkonen-Condit).

In fact, one of the first areas to apply verbal reports procedures to the study of language use was research on foreign language (FL) learning; other areas were research on writing processes and on FL reading processes. Some of the researches on L2 learning/acquisition have used translation tasks to elicit data on students’ text processing strategies (Gerloff 1986) or on the organization of cognitive planning in a translation task. Consequently, these studies may offer interesting insights and hypothesis for translation-oriented research.

The empirical investigation of the translation process, (data are collected asking subjects to think aloud during a translation task) vary in terms of subject population (language learners, translation students, professional translators), translation task (oral or written translation), text-types (news articles, advertisements, editorials, etc.); source and target languages, access to reference material, translation briefs, limited or unlimited time, etc. In addition, and more importantly, TAP studies offer a lot of definitions of translating, research interest and objectives, and methods of analysis. “Failure to recognize the variety of approaches taken by TAP researchers can lead to misleading over-generalizations (or even discarding all such studies as ‘uninteresting’ for translation studies)” (Jääskeläinen 1999: 39).

It is important to know what are the general aims and frames of reference and the experimental details of different TAP studies, in order to assess their findings and relate them to other kinds of research. The first essential distinction is to see whether the emphasis of a research is on translation studies or psycholinguistics. This means, whether the aim is to understand the nature of translating or whether translation is used as an experimental task to collect data on the nature of language processing (Jääskeläinen 1999).

4. 1. FIRST STUDIES ON FOREIGN LANGUAGE LEARNERS

The very first studies by Sandrock (1982) and Krings (1986) already show the advantages and the limitations of this method of elicitation and set the

standards for the design of similar studies. Kings’ extensive study of the translation process of eight advanced students of French demonstrates the immense wealth and richness of data that can be obtained by TAP as well as the necessity to choose among all the possible variables for both the aim and the analysis.

From the point of view of Translation Studies though, the research has a drawback: the participants were not involved in translation as a professional or even potentially professional activity; they were foreign language students and teachers-in-training and translated their tasks the same way they would have translated an ordinary assignment in a language class. In fact, the translation brief specified they should translate in their usual manner.

Thus, the objective of Kings’ study was translating in a pedagogical context or didactic translation, which is a rather different task than translation as a professional activity. Nevertheless, the study provided a number of research questions and categories to apply to analysis as well as a highly fruitful way to use TAP in the study of the translation process (Schmidt 2005: 22).

Gerloff published her study in 1988. She investigated and compared the translation process of three different subject populations: four college students of French, four bilingual speakers English/French without any experience of translation, and four professional translators, normally translating from French into English. The study focused on translation in one direction i.e. from L2 (French) to L1 (English). Gerloff used essentially the same coding and classifying categories as Krings. The most important finding is that more experienced translators (experience here is defined in the context of translating being an innate ability in bilinguals), such as both the professionals and the bilinguals in her sample, do not necessarily translate more easily or faster than the less experienced translators (here defined as the foreign language students). From that, along with other indicators, she concluded that experienced translators are more aware of the difficulty of the problems they find and of their possible solutions; furthermore, they set higher

standards for their performance than novices. From this comes the quality of their translations and of their text production.

According to Krings and Gerloff, the results of their experiments are determined by a difference in strategy. Inexperienced translators, in fact, employ more local strategies, i.e. they are concerned only with the fragment they are working on, without considering the text as a whole. Moreover, they don’t relate to their own world-knowledge. More experienced translators, in contrast, use more global strategies, that relate the problem to their world- knowledge, to the text as a whole and to its overall theme.

The two studies also shared other important results: the subjects mostly rendered small syntactic units, working their way through the task in a linear way, i.e. “translation is done proceeding from item A to item B in a text without looking forward or backward further than to the next sentence boundary” (Schmidt 2005: 23).

Lörscher (1991) investigated the translation process in foreign language learners using as subjects of his experiment first- or second-year students of English at the university he was working at (they were not even advanced learners). He assumed that oral translation would provide richer material then written translation. Therefore, he instructed his participants to translate a written text orally and recorded their spoken translations, including all concurrent verbalization. With this research, Lörscher claims to investigate the translation process itself, even if he recognizes that his model does not resemble a “real mediating situation”, as he calls it because:

[…] it is still unknown whether translation processes in real mediating situations are different – in detail or in principle – from translation processes in artificial mediating situations (Lörscher 1991: 4).

Despite the discrepancies between the design and the aim of his study, Lörscher developed a refined model for analyzing TAP, providing a useful tool for further research.

In all of these experiments, the subjects were foreign language learners, rather than students of translation, which has received a fair amount of criticism, because their findings can hardly account for professional translators’ performance. However, they have laid the methodological basis for subsequent TAP studies and have provided important information about translating by foreign language learners, which can be used for comparisons between non-professional, semi-professional (translation students) and professional translation, a design feature that can be seen in the studies by Königs 1987, Kussmaul 1998 and Jonasson 1998 (Jääskeläinen 1999).

4. 2. FURTHER STUDIES: DIFFERENT AIMS AND HYPOTHESIS

Following these pioneers, a number of other translation researchers have since used TAP to elicit data for their studies. These studies have different settings and involve different subject populations (translation students, professional translators, teacher of translation, laypersons, bilingual, a combination of these categories); different language pairs (depending partly on where the research has been carried out); different types of task and experimental conditions (translating a written text orally, producing a written translation, translating alone, in pairs or in small groups; translating with o without access to reference material, limited or unlimited available time); different text-types (political satire, newspaper editorials, tourist brochures, government documents) and translation briefs (faithful translation, cultural adaptation, shortening or popularizing the ST, rewriting procedures); different categories of analysis (identification of translation problems and problem- solving strategies, focus on conscious attention, role of affective factors on translation).

“TAP studies on translating could also be conceptualized in terms of their general purposes and the specificity of their hypothesis” (Jääskeläinen 1999: 44). On this basis, they could be divided into first- and second-generation studies; the first group would include those with a relatively general aim of discovering what happens in translation (Krings and Lörscher) or what

distinguishes between professional vs. non-professional translation (Jääskeläinen, Tirkkonen-Condit). The second group, in turn, focus on investigating more specific hypotheses, often derived from the findings of the first-generation studies. TAP research carried out at the Savolinna School of Translation Studies makes part of this second kind of studies.

The purpose of the first TAP experiments at Savolinna was to identify differences between professional vs. non-professional translation (Jääskeläinen and Tirkkonen-Condit). The designation of ‘professional’ and ‘non-professional’ referred, misleadingly, to fifth-year and first-year students of translation respectively. However, it has been frequently pointed out that the differences between first-year and fifth-year students’ translation processes may not depend on different levels of translation competence alone, but on other factors, such as differences in their world knowledge (in its quantity, because the quality of world knowledge is naturally different in every single individual). As a consequence, Pöntinen and Romanov (1989) organized a TAP experiment with two subjects: the first one was a teacher of translation and free-lance translator; the second one was a subject specialist. They were about the same age with a high level of education. The data collected showed some interesting differences between the two subjects’ decision criteria: the translator relied more on textual knowledge than the subject specialist (Jääskeläinen 1999).

A further generation of researchers on the translation process turned their interest to even more specific aspects, e.g. the semantic change and the reading and comprehension process involved in translation (Dancette 1994).

Several of the most recent TAP studies on the translation process, are aimed to explore the difference between categories of translators such as professionals, advanced students in translation training programs and language students with respect to their translational behavior (Englund Dimitrova 2005, Norberg 2003). Some of them wanted to discover what kind of linguistic and extra linguistic factors influence the production of “good”

translations. Jensen (2000) and Jääskeläinen looked at the influence of routine vs. non-routine tasks on task performance and investigate the differences between professionals and laymen. Künzli (2003) explored the impact of emotional and affective states on subjects’ performance.

Some researchers in Denmark proposed including methods of logging the writing process during translation to develop and corroborate data collected by means of TAP. Their studies point towards a possible design combining different analyzing methods that could be able to elicit and evaluate data telling us more about the complex structures that govern the translation process (Schmidt 2005).

The list of TAP studies presented above shows that it exists a heterogeneous group of investigations. In fact, the first process-oriented research projects started in isolation, independently of each other; therefore they reflect very different translation theoretical frameworks ad research aims.

Furthermore, in the absence of previous research, methods of analysis have been developed to describe a particular body of data. As a result, applying the methods of analysis to other kinds of data has, as a rule, resulted in modifications or in the introduction of new methods of analysis (Jääskeläinen 1999: 46).

On the whole, the presence of different researches has the advantage to shed light on different aspects of different kinds of translation processes. This increases our understanding of the complex mechanisms underlying translation. “Indeed the great variety of TAP approaches has highlighted the fact that ‘the’ translation process does not exist; instead, there are many different translation processes which are the outcome of many kinds of factors. However, the differences in the kinds of data collected, analyses carried out, and the overall goals of research have made it more difficult to test the methods employed in previous experiments.

Moreover, due to the difficult and time-consuming methods of data collection and analysis involved in TAP research, the numbers of subjects have remained relatively small (ranging from one to 48) and investigators have been extremely careful in generalizing on the basis of the data collected, even if it would be very important to be able to test research findings and conclusions with those of other studies, particularly at the early stages of process-oriented research (Jääskeläinen 1999).

5. THINKING ALOUD VS. JOINT TRANSLATION

Due to the limitations involved in the use of TAPs, it has been suggested that a better and more natural way to investigate the translation process would be to ask subjects to translate in small groups (“joint translation”, Matrat 1992).

House’s, Matrat’s and Séguinot’s experiments provide data to discuss on the validity of this method.

House asked to German university students of English (not translation students) to translate a text from English into German. One group translated in pairs and another alone, while thinking aloud. The students in the think- aloud session were not trained to spontaneous think aloud with the help of a warm-up task, and this makes it difficult to compare the two bodies of data.

House’s findings show that the students translating in pairs were using more sophisticated strategies. For example, while students in the think-aloud session focused exclusively on lexicon-semantic problems in a text which was chosen for its syntactic difficulty, the student-pairs frequently dealt with grammatical problems.

House concludes that “the introspective data produced by pairs is less artificial, richer in translation strategies and simply much more interesting” (House 1988:95).

But, until a systematic methodological survey is carried out, it can also be speculated that translating in pairs may help externalize knowledge which is poorly employed or access when students are translating alone.

Matrat (1992) carried out a systematic comparison of three categories of subjects (novice, advanced and expert translators) performing translation tasks in a think-aloud vs. joint activity translating experiment.

Matrat’s approach is embedded in Vygotsky’s psychological theory that proposes consciousness as “the highest level of organization of mental functions comprising both intellect and affect”, and thus as “the fundamental object of psychological research” (Jääskeläinen 1999: 75).

Vygotsky rejected introspective methods and reducing psychology to the studying of isolated components of mind; he argued that interdisciplinary research could account for the interrelationship between cultural, linguistic and psychological phenomena (Matrat 1992). According to him, consciousness is socially constructed and consequently he proposed that observing joint activity would be the appropriate method (genetic method) to investigate human cognitive processes. He also proposed to introduce obstacles and difficulties into the experimental task in order to disrupt routine methods of problem-solving and, thus to discover new skills.

In Matrat’s experiment, the three groups of students produced a written translation of a written source text from English into Italian (their native language). The same subjects took part in two experiments: first a think-aloud experiment and then a joint translating activity. The source texts were different for the two activities, but to retain the same level of difficulty they were different paragraphs of the same text. The use of dictionary was not allowed and the time was limited. There appears to have been no articulated translation brief. The experimental sessions were video-taped.

The setting of Matrat’s experiments shows that more variables may have contributed to her findings (choice to use text excerpts from the same text, limited time, no access to reference books). Moreover, a problem arises: when the subjects started the joint activity, they were already familiar with the source text.

Matrat compares the collected data in terms of (1) problem definition and structure and (2) strategic processing. The findings indicate that in joint

translating, students identified problems more clearly and recognized they have a complex structure. Furthermore, evidence of strategic processing was more easily identifiable than in think-aloud protocols.

As far as TAPs are concerned, advanced students were the best subjects, as their training had provided them with the metalanguage to discuss translation problems. This observation shows Matrat expects the subjects to provide sophisticated analyses, i.e. to introspect rather than to think aloud, which, in turn, reflects her interest in the emergence of metacognition.

“One of the most puzzling findings is that none of the protocols showed evidence of decision-making strategies or decision criteria, whereas other TAP studies contain plenty of verbalizations on decision-making” (Jääskeläinen 1999: 78). In fact, the subjects discussed problems, but were not able to decide on a solution, then moved on and never came back to the problem (Matrat 1992). This may have been the result of time constraint or of the fact that the text was incomplete.

In addition, the subjects’ interpretation of the purpose of the experiment may also have played a role: in one example of joint translation, one of the subjects said that the experimenters were interested in what subjects said and not in how they translated. This comment is important in relation to the methodological comparison: if the subjects felt that their ability to talk about translating was being investigated, they might have been intimidated by the demands of the task when translating alone, while it seems reasonable to assume that tackling the task together (and with some previous experience with the text) would be less face-threatening to the subjects.

In sum, it seems that Matrat’s investigation in not only trying to compare the appropriateness of the two methods of data collection, but also to argue for the appropriateness of Vygotsky’s theory on human consciousness in relation with translation. This complicates the assessment of the validity of her methodological comparison.

5. 1. JOINT-TRANSLATION’S LIMITS

Someone could argue that “making people translate together is as artificial as asking them to think aloud while translating, since most translators usually work alone” (Jääskeläinen 1999: 80). For this reason, Séguinot studied the translation processes of two professional translators who were used to work together. Findings show that dialogue protocols illuminate the ‘non-rational’ element in translation: during the translation process, the translators’ discussion shifted to areas which had nothing to do with the task at hand (Séguinot 1996). The findings also “indicate that translation is non- linear and iterative, i.e. the mind keeps looking for alternatives even after a translation problem has been solved” (Jääskeläinen 1999: 80). Furthermore, there is evidence of parallel processing during translating.

Data from joint activity may be richer, more natural and even more interesting than think-aloud reports, but they do not provide access to the solitary translation process. The object of the research is different in the two experimental conditions. Moreover, it seems that joint activity elicits more sophisticated strategies form the subjects. This can be due to the re-activation of automatized processes in the case of professional translators or externalizing unused strategies in the case of translation students or, as House or Matrat argue, that joint activity is better able to capture the underlying mental processes than thinking-aloud. However, on the bases of these studies, the latter conclusion seems premature.

Another problem with joint activities is that they may distort the results. One of the subjects may assume a leading role, because of his personality. Thus, other subjects may accept solutions not because they are better but because they are proposed by the more dynamic person. In other cases, subjects may hold back their ideas for reasons of politeness, or even chivalry. When analyzing the dialogue protocols the researcher should therefore take care to observe only those processes where subjects take an equal part in solution-finding. One way of minimizing this kind of problems would be to choose “matching” subjects, with no psychological or social superiority of one over the other and where personalities are quite similar (Kussmaul, 1995).

However, one should be aware of the fact that variables cannot be completely controlled (Kussmaul and Tirkkonen-Condit).

6. TRADUZIONE CON TESTO A FRONTE

The following section talks about the importance of taking into account subjects’ social-psychological factors, when carrying out a TAP experiment and analyzing the data collected. Variables such as subject’s personal history and emotional factors can result in unexpected behavior by the subjects and alter the results of the experiment.

The whole section is a text taken from Tapping the process: an explorative study of the cognitive and affective factors involved in translating (Jääskeläinen 1999: 137-151), which I have also translated into Italian.

SUBJECTS’ BACKGROUND AND SOCIAL-PSYCHOLOGICAL FACTORS

As has been mentioned in the previous sections the unexpected features of the subjects’ behaviour in the experimental task may be explained by their personal histories. In addition, social-psychological factors, such as Goffman’s (1961) notion of role distancing, may also provide explanations. I will begin with a discussion of the subjects’ personal histories as an explaining factor and then move on to describing the types of behaviour which I see as potential displays of role distancing in the present data.

The first example of a subject’s personal history as an explanation deals with one of the non-professional translators, Laura. As was mentioned in section 4. 1., Laura had written her doctoral dissertation in English. In her professional career she had also reported on her research both in English and in Finnish, i.e. she had considerable experience in reporting her own thoughts in two languages. Nida argues (1964: 242) that ‘if a person is to serve as a translator, – he must have had a good deal of experience in language shifting’. Although Laura had had no experience in translating, which she explicitly pointed out in the background questionnaire, she had accumulated a great deal of experience in language shifting. As a consequence, Laura also seemed to have moved away from the ‘school translation’ approach to translating and was able to produce a relatively fluent and idiomatic Finnish text. Furthermore, Laura can also be regarded as an experienced writer, which makes her a less ‘naive’ language user than the other non-professional translators in the present data.

BACKGROUND DEI SOGGETTI E FATTORI SOCIO-PSICOLOGICI

Come accennato nelle sezioni precedenti, le caratteristiche inattese del comportamento dei soggetti durante il compito sperimentale possono essere chiarite attraverso la storia personale dei soggetti stessi. Possono fornire spiegazioni anche fattori socio-psicologici, come il concetto di «distanza dal ruolo» delineato da Goffman (2003). Inizierò con un approfondimento sulla storia personale dei soggetti intesa come fattore esplicativo, per poi descrivere i modelli di comportamento che ritengo siano potenziali manifestazioni di distanza dal ruolo, in questo studio.

Il primo esempio di storia personale dei soggetti come fattore esplicativo riguarda uno dei traduttori non professionali, Laura. Come accennato nella sezione 4.1., Laura aveva scritto la sua tesi dottorale in inglese. Nel corso della sua carriera professionale aveva anche scritto relazioni sulla sua ricerca sia in inglese sia in finlandese e aveva quindi notevole esperienza nel riferire i propri pensieri in due lingue. Nida sostiene (1964: 242) che «se una persona svolge la funzione di traduttore, deve avere avuto molta esperienza nel passaggio da una lingua all’altra». Nonostante Laura non avesse esperienza nel campo della traduzione, come ha affermato esplicitamente nel questionario sul background dei soggetti, aveva accumulato molta esperienza nel passaggio da una lingua all’altra. Di conseguenza, sembra anche che Laura si fosse distaccata dall’approccio “scolastico” alla traduzione e fosse in grado di produrre un testo in un finlandese relativamente scorrevole e idiomatico. Inoltre, Laura può essere considerata una professionista della scrittura, il che fa di lei un utente del linguaggio meno “ingenuo” rispetto agli altri traduttori non professionali, in questo studio.

However, Laura’s liberal views of translating are not entirely unproblematic. Laura treats the ST as if it were her own creation; as a result, her translation is factually vague, even incorrect, at places. This can be illustrated by two examples. The first of these relates to the translation of the ST phrase ‘mop all the excess fats’. Laura translated this as poistaa kaikki ylimääräiset rasvakudokset (‘remove all the excess fatty tissues’). Laura’s verbalisations indicate that she did not stop to ponder what kind of fats are meant here, but she simply weighed alternative Finnish expressions, rasvakerrokset and rasvakudokset (‘layers of fat’ and ‘fatty tissues’), and eventually chose the one which to her seemed to be more idiomatic Finnish. As a result, her translation of the phrase is factually incorrect, since the ST is concerned with the types of fats contained in blood, for instance, and not with the visible fatty tissues in a human body. Laura may also have been misled by the headline, which talks about staying slim, or she may have failed to utilise textual information which would have made clear the nature of the ‘fats’ referred to.

The second example deals with Laura’s translation of the expression ‘to feed a diet to rats’. Her solution runs as follows: rotat noudattivat epätavallisen rasvaista ruokavaliota (‘rats followed an unusually fatty diet’). Here, as elsewhere, Laura’s main concern was to find idiomatic TL expressions by using the ST only as a relatively flexible framework for the search. As a consequence, Laura’s translation in the above example involves a shift in meaning; the Finnish expression noudattaa roukavaliota implies that the rats chose be on a fatty diet, instead of being forced to eat whatever the experimenters decided to feed them.

Tuttavia, la concezione libera che Laura ha della traduzione non è del tutto acritica. Laura tratta il prototesto come fosse una creazione propria; ne consegue una traduzione di fatto vaga, in alcuni punti anzi scorretta. Questo può essere dimostrato attraverso due esempi. Il primo riguarda la traduzione della frase «mop all the excess fats» [eliminare tutti i grassi in eccesso]. Laura ha tradotto «poistaa kaikki ylimääräiset rasvakudokset» («rimuovere tutti i tessuti adiposi in eccesso»). Come indicano le verbalizzazioni, Laura non si è fermata a pensare a che tipo di grassi alludesse il testo, ma ha semplicemente valutato le alternative espressioni finlandesi, «rasvakerrokset» e «rasvakudokset» («strati di grasso» e «tessuti adiposi»), e infine ha scelto quella che secondo lei era la più idiomatica in finlandese. Ne risulta una traduzione denotativamente scorretta, visto che il prototesto prende in considerazione i tipi di grassi contenuti nel sangue, per esempio, ma non i tessuti adiposi visibili nel corpo umano. Probabilmente Laura è stata fuorviata dal titolo, che parla di mantenere la linea, o forse non ha utilizzato le informazioni testuali che avrebbero chiarito la natura dei grassi cui si fa riferimento.

Il secondo esempio riguarda il modo in cui Laura ha tradotto l’espressione «to feed a diet to rats» [sottoporre i ratti a una dieta]. Ha risolto in questo modo: «rotat noudattivat epätavallisen rasvaista ruokavaliota» («i ratti hanno seguito una dieta insolitamente grassa»). Qui, come in altri punti, la preoccupazione principale di Laura è stata quella di trovare espressioni idiomatiche da inserire nel metatesto, servendosi del prototesto solo come un modello di riferimento relativamente flessibile per la ricerca. Di conseguenza, la traduzione dell’espressione nell’esempio comporta un cambiamento di significato; il finlandese «noudattaa roukavaliota» significa che i ratti hanno scelto di seguire una dieta ricca di grassi, e non che sono stati costretti a mangiare tutto quello che gli sperimentatori gli davano.

In sum, Laura seems to work exclusively with her own ‘text’ as it were and ignore what is the ST saying1. While translators always work with their own interpretation of the ST, it seems that Laura has taken this one step further: she is working with her own response to the ST and not checking it against the ST. Tirkkonen-Condit (1992) reports on a similar finding in an experiment with a professional translator and a subject specialist as subjects (Pöntinen and Romanov 1989). One of the significant differences between the two was that while the subject specialist relied on her own world knowledge, the professional translator used a great deal of textual knowledge to be able to determine what the ST was saying. Laffling (1993: 124f.), in turn, draws attention to a similar incident reported in Krings (1988a) of a professional translator overlooking textual information and relying (misguidedly) on her own encyclopedic knowledge instead. These findings illustrate the complicated nature of translating; while understanding a text is necessarily subjective, in translation the subjective interpretation is usually checked against the ST to maintain a balance between the ST author’s ideas and the translator’s interpretation of them. In sum, Laura’s experience in language shifting seems to have been a double-edged sword; on the one hand, it has freed her from the confines of the ST and helped her produce a relatively fluent Finnish text. On the other hand, Laura ignores the ST to such an extent that her translation contains wrong information.

1 The two examples discussed here differ in their gravity as translation ‘errors’. The former I would regard as an error because it changes the meaning; i.e. it promises a slimming effect which is not in the text. (The headline may have influenced this interpretation as well). The latter, however, I would consider as a borderline case; although it gives the wrong idea, the effect is humorous rather than disastrous, which could be acceptable in a text to be published in the target column.

In breve, pare che Laura lavori, per così dire, esclusivamente con il proprio “testo” ignorando il contenuto del prototesto.2 Mentre i traduttori lavorano sempre con la propria interpretazione del prototesto, pare che Laura abbia fatto un passo avanti: lavora con la sua personale risposta al prototesto senza confrontarla con il prototesto. Tirkkonen-Condit (1992) parla di un risultato simile in un esperimento che aveva come soggetti un traduttore professionale e una specialista in materia (Pöntinen e Romanov 1989). Una delle differenze più significative tra i due era che mentre la specialista in materia si basava sulla propria conoscenza del mondo, il traduttore professionale utilizzava moltissime informazioni testuali per essere in grado di determinare il contenuto del prototesto. Laffling (1993: 124f.), a sua volta, pone l’attenzione su un caso simile descritto in Krings (1998a) riguardo a una traduttrice professionale che ignorava le informazioni testuali basandosi invece (in modo fuorviante) sulla propria cultura enciclopedica. Questi risultati illustrano la complicata natura del processo traduttivo; mentre la comprensione di un testo è necessariamente soggettiva, nella traduzione, l’interpretazione soggettiva viene solitamente messa a confronto con il prototesto, per mantenere un equilibrio tra le idee dell’autore del prototesto e l’interpretazione delle stesse da parte del traduttore. Per concludere, l’esperienza di Laura nel passare da una lingua all’altra si è rivelata un’arma a doppio taglio: da una parte, l’ha liberata dai vincoli del prototesto e l’ha aiutata a produrre un testo in un finlandese relativamente scorrevole; dall’altra, Laura ignora il prototesto a tal punto che la sua traduzione contiene informazioni errate.

2 I due esempi qui analizzati differiscono per la loro gravità di “errori” di traduzione. Considererei il primo un errore, perchè comporta un cambiamento del significato; infatti, promette un effetto dimagrante, non menzionato nel testo (questa interpretazione potrebbe anche essere stata influenzata dal titolo). Tuttavia, considererei il secondo un caso limite; nonostante trasmetta l’informazione sbagliata, l’effetto è comico e non disastroso, e potrebbe essere accettato in un testo da pubblicare in quel periodico.

The second example of personal history as an explanation concerns Penny, a professional translator whose translation was rated as ‘mediocre’. Penny’s verbalisations as well as her comments in the follow-up letter (see examples below) imply that her experience with translating medical texts resulted in her translating for the wrong audience in the experiment. This observation is linked to the effects of the experimental situation. In cognitive psychology (e.g. Saariluoma 1988b: 56), it has been observed that in problem- solving situations subjects often make so-called cognitive errors, i.e. the subjects do not perceive the situation correctly and, as a result, are not able to collect and utilise all the information relevant to successful task performance. Cognitive errors take place in ‘natural’ problem-solving situations, too, but it stands to reason that the additional strain created by the experimental situation might increase their probability.

Evidence of Penny’s misinterpretation of the task is shown in example (4). Here Penny decides to retain the original English explanation of ‘NADPH’ in the translation, because she thinks that the readers of the translation will be familiar with the acronym. This implies that the potential readers Penny has in mind were experts in medicine rather than ordinary newspaper readers.
(4) ja (.) sitten laitan sulkuihin ihan tos englanninkielisessä

kirjotusasuasussaan tää nicotineamide adenine dinucleotide phosphate hybri-hybridi
mä en tietäs muita mitää muuta ku että se on fosfaattihybridi mutta (2.0) mut touta (1.0) mä luulen että (.) et et ne ihmiset jotka (3.0) tai olenkin varma et et (.) et joille toi NADPH jotaki tarkottaa ni (1.0) ni ne he (.) he tietää sen ihan tolla lyhenteellä (.) (Penny: P)

Il secondo esempio di storia personale come fattore esplicativo riguarda Penny, una traduttrice professionale la cui traduzione è stata giudicata «mediocre». Le verbalizzazioni di Penny e i commenti contenuti nella lettera di follow-up (vedi esempio sottostante) implicano che la sua esperienza nel tradurre testi medici l’ha indotta a rivolgere la traduzione dell’esperimento a un pubblico non adeguato. Quest’osservazione è legata all’influenza esercitata dalla situazione sperimentale. Nella psicologia cognitiva (per esempio, Saariluoma 1988b: 56), è stato osservato che in situazioni di problem-solving i soggetti commettono spesso i cosiddetti «errori cognitivi»; non interpretando la situazione nel modo corretto, non sono in grado di raccogliere e utilizzare tutte le informazioni rilevanti per svolgere con successo il compito. Si fanno errori cognitivi anche in situazioni di problem-solving “naturali”, ma va da sé che la maggiore tensione creata dalla situazione sperimentale può aumentare la loro occorrenza.

L’esempio (4) fornisce alcune prove del travisamento da parte di Penny. Qui Penny decide di mantenere la spiegazione inglese originale di «NADPH», perchè pensa che i lettori della traduzione conoscano l’acronimo. Questo implica che i lettori modello che Penny aveva in mente fossero esperti di medicina e non normali lettori di un giornale.
(4) ja (.) sitten laitan sulkuihin ihan tos englanninkielisessä

kirjotusasuasussaan tää nicotineamide adenine dinucleotide phosphate hybri-hybridi

mä en tietäs muita mitää muuta ku että se on fosfaattihybridi mutta (2.0) mut touta (1.0) mä luulen että (.) et et ne ihmiset jotka (3.0) tai olenkin varma et et (.) et joille toi NADPH jotaki tarkottaa ni (1.0) ni ne he (.) he tietää sen ihan tolla lyhenteellä (.) (Penny: P)

and (.) then I’ll put that in brackets using that English spel- spelling this NICOTINEAMIDE ADENINE DINUCLEOTIDE PHOSPHATE HYBRI-hybride
I wouldn’t know any other parts except that it is phosphate hybride but (2.0) but well (1.0) I think that (.) that that those people who (3.0) or actually I’m sure that (1.0) those for whom that NADPH means something (1.0) that that they (.) will be familiar with the abbreviation

As Penny did not refer explicitly to the translation brief during the experiment, it was underlined in the post-experimental questionnaire (cf. section 4. 4.). In the post-experimental stage, Penny crossed out the English explanation of ‘NADPH’ as well as the source reference to ‘Experientia’. In the follow-up letter Penny gave the following comment (my translation from English) on both omissions:

(5) I leave these out now that I know where the translation will be published. Somehow I thought in the experiment that the intention was to produce a translation which would be as faithful to the original as possible. Therefore I did not ask, at least not explicitly, where the translation will be published. (Penny: P)

Penny’s comment corroborates the assumption that she ignored the task description in the experiment. In fact, she seems to suspect that she has been deliberately misled in this respect. It is highly probable that when Penny saw, in the somewhat unnerving experimental situation, what the source text dealt with, she immediately slipped into her old familiar role as a medical translator.

e (.) allora quello lo metto tra parentesi usando quello spel- spelling inglese questo NICOTINEAMIDE ADENINE DINUCLEOTIDE PHOSPHATE HYBRI-hybride
non saperi cosa siano tutte le altre parti a parte che è un fosfato ma (2.0) ma be’ (1.0) penso che (.) che che quelle persone che (3.0) o a dire il vero sono sicura che (1.0) coloro per i quali quel NADPH significa qualcosa (1.0) che loro loro (.) conosceranno l’abbreviazione

Visto che, durante l’esperimento, Penny non ha fatto esplicitamente riferimento al translation brief, ciò è stato sottolineato nel questionario post- sperimentale (cfr. sezione 4.4.). Nella fase post-sperimentale, Penny ha cancellato la spiegazione inglese di «NADPH» così come la fonte di riferimento «Experientia». Nella lettera di follow-up, Penny ha commentato come segue (mia traduzione dal finlandese) su entrambe le omissioni:

(5) Ora che so dove verrà pubblicata la traduzione, lascio via queste. Per qualche ragione, durante l’esperimento, ho pensato che l’intenzione fosse quella di produrre una traduzione che fosse fedele il più possibile all’originale. Perciò non ho chiesto, o per lo meno non esplicitamente, dove sarebbe stata pubblicata la traduzione (Penny:P)

Il commento di Penny conferma l’ipotesi secondo la quale avrebbe ignorato la descrizione del compito dell’esperimento. Infatti, in questo senso, sembra che Penny sospetti di essere stata fuorviata deliberatamente. È molto probabile che, all’interno di una situazione sperimentale alquanto snervante, quando ha visto di che cosa trattava il prototesto, Penny sia immediatamente ricaduta nel suo vecchio e familiare ruolo di traduttrice di testi medici.

As a result, she overlooked the translation brief and translated the text for a wrong target group, which may explain the lower-than-expected quality of her translation.3 Penny’s behaviour seems a particularly conspicuous example of a cognitive error, i.e. misinterpreting the situation and therefore not being able to use all the information relevant to successful task performance. However, and somewhat more alarmingly, Penny’s implicit assumption that a faithful translation was required could also reflect her understanding (and her experiences) of the variety of translating prevailing in translator training.

The third example of the role of personal history relates to Lucy’s (professional translator) poor success in the experimental task. Lucy differed from the other professional translators in terms of her occupation; she worked as a ‘business correspondent’, while Fran, John and Penny worked as free lance translators at the time. Lucy also reported that she worked with four non-native languages, whereas the free lance translators reported only two (in fact, Fran reported that she worked almost solely with English). As a consequence, the working conditions of Lucy vs. the other professionals can be assumed to be quite different. According to Toury (1984: 191), ‘the norms which govern [the] ‘well-formedness’ of translated utterances involve, like any other norm, sanctions’. The nature of these sanctions, which partly determine how the translator approaches each translation task, depends on the nature of each translating situation.

3 In spite of the ‘cosmetic’ revisions (i.e. the above-mentioned omissions), Penny’s translation remained too difficult for the target column. It is possible that Penny used heavy structures with low readability, because she was translating for an expert audience.

Di conseguenza, ha ignorato il translation brief e ha tradotto il testo per un lettore modello errato, fatto che spiegherebbe perchè la qualità della sua traduzione deluda le aspettative.4 Il comportamento di Penny sembra un esempio particolarmente evidente di errore cognitivo, ossia travisare la situazione e quindi non essere in grado di usare tutte le informazioni rilevanti per una buona riuscita del compito. Tuttavia, e in modo più allarmante, il fatto che Penny abbia implicitamente supposto che venisse richiesta una traduzione fedele, riflette la sua conoscenza (e la sua esperienza) del tipo di traduzione prevalente nella formazione professionale dei traduttori.

Il terzo esempio del ruolo della storia personale riguarda gli scarsi risultati ottenuti da Lucy (traduttrice professionale) durante il compito di traduzione sperimentale. Lucy si distingueva dagli altri traduttori professionali per la sua occupazione: era corrispondente commerciale, mentre Fran, John e Penny al momento erano traduttori free lance. Lucy ha dichiarato di lavorare con quattro lingue B, mentre i traduttori free lance solo con due (in realtà, Fran ha dichiarato di lavorare quasi esclusivamente con l’inglese). Di conseguenza, potremmo ipotizzare che le condizioni di lavoro di Lucy fossero abbastanza diverse rispetto a quelle degli altri professionisti. Secondo Toury (1984: 91), «le norme che governano la “bella forma” delle frasi tradotte implicano, come qualsiasi altra norma, delle sanzioni». La natura di queste sanzioni, che in parte determinano il tipo di approccio del traduttore ad ogni lavoro di traduzione, dipende dalla natura di ogni situazione traduttiva.

4 Nonostante la revisione “superficiale” (le omissioni sopraccitate), la traduzione di Penny è ancora troppo difficile per il periodico a cui è destinata. È possibile che Penny abbia usato delle strutture complicate con un basso grado di leggibilità, perché stava traducendo per un pubblico esperto.

Even though Toury clearly refers to translation quality (‘well-formedness’) here, it could be argued that the demands and sanctions imposed upon translators often relate to both quantity and quality of translating. Usually the translator must compromise between the opposed demands of quality and quantity, but, in some translating situations the emphasis is clearly on quantity, that is, the translator is expected to produce enormous quantities of text and to do it fast.

Lucy’s behaviour in the experiment seems to indicate that the sanctions imposed upon Lucy as a translator have emphasised speed and efficiency to such an extent that the quality of the product is affected. Lucy’s attitude towards the use of time can be illustrated by her comment in our telephone conversation about scheduling the experiment. I told Lucy that the experiment would take about one hour. When she found out how short the source text was, she burst into laughter and said: ‘If I spent one hour for translating such short texts I would’ve been fired ages ago!’ Lucy’s comment points to a ‘quantitative’ attitude to translation, i.e. a short source text will take only a short time to translate. In contrast, a ‘qualitative’ attitude to translation would imply, among other things, that before any estimate can be made as to how long the translation process is likely to take, one should at least see the source text. Moreover, Lucy’s above comment reveals the nature of the sanctions imposed upon her: ‘If you do not work fast, you will be fired’.5

Lucy’s attitudes can also be observed in the excerpt in example (6) in which she is working on the medical term ‘NADPH (nicotineamide adenine dinucleotide phosphate hybride)’.

5 Here we could also talk about translational ‘sub-cultures’ to refer to the translational norms prevailing in different working environments (cf. chapter 6)

Anche se Toury si riferisce chiaramente alla qualità della traduzione (“bella forma”), qui, si può affermare che le esigenze e le sanzioni imposte ai traduttori spesso riguardano sia la quantità sia la qualità della traduzione. Solitamente il traduttore deve trovare un compromesso tra le esigenze opposte di qualità e quantità, ma in alcune situazioni traduttive l’enfasi è chiaramente posta sulla quantità, quindi ci si aspetta che il traduttore produca enormi quantità di testo, e che lo faccia velocemente.

Il comportamento di Lucy durante l’esperimento sembra indicare che le sanzioni imposte al suo ruolo di traduttrice hanno enfatizzato la velocità e l’efficienza a tal punto da compromettere la qualità del prodotto. L’atteggiamento di Lucy riguardo all’uso del tempo può essere chiarito dal suo commento durante la nostra telefonata per la programmazione dell’esperimento. Ho detto a Lucy che l’esperimento sarebbe durato circa un’ora. Quando ha visto quanto era breve il testo, è scoppiata a ridere e ha detto: «Se ci mettessi un’ora a tradurre un testo così breve, mi avrebbero licenziata da un bel pezzo!». Il commento di Lucy rivela un approccio “quantitativo” alla traduzione: la traduzione di un prototesto breve dovrà richiedere un lasso di tempo breve. Al contrario, un approccio “qualitativo” alla traduzione implicherebbe, tra le altre cose, che prima di poter avanzare qualsiasi tipo di ipotesi sul tempo che il processo traduttivo potrebbe richiedere, bisognerebbe almeno leggere il prototesto. Inoltre, il commento di Lucy sopraccitato, rivela la natura delle sanzioni a lei imposte: «Se non lavori in fretta, verrai licenziata».6

Possiamo osservare l’atteggiamento di Lucy nell’estratto portato come esempio (6) nel quale sta lavorando al termine medico «NADPH (nicotineamide adenine dinucleotide phosphate hybride)».

6 Qui si potrebbe anche parlare di “sottoculture” traduttive, per far riferimento alle norme traduttive prevalenti nei diversi ambienti di lavoro (cfr. Capitolo 6)

After writing down the acronym, Lucy reads aloud the first two components of the explanation, after which she points out that the easiest way of solving the problem would be to call up a doctor. Lucy makes this decision rather quickly (in two or three seconds) and effortlessly (i.e. it is clearly a non-problematic decision).

(6) jonka lyhenne (5.0) on N A P (1.0) eiku D (.) P H (2.0) nikotiiniamidi adeniini ni ni di di di (1.0)

ja tätä tällästä minä en edes etsi sanakirjasta koska (1.0) hyvin paljon helpommalla pääsee ku soittaa jollekii lääkärille (2.0) (Lucy:P)

whose abbreviation (5.0) is N A P (1.0) no D (.) P H (2.0) nicotineamide adenine ne ne de de de (1.0)

and this sort of a term I won’t even look for in a dictionary because (1.0) you get off much more easily when you call up some doctor

Example (6) illustrates Lucy’s inclination to high efficiency; she does not want to waste time on trying to find the Finnish term in reference books and simply points to the easiest and quickest way to solve the problem. Admittedly, asking for expert help with terminological problems is in principle sound professional practice. However, Lucy did not call up a doctor to check the Finnish term; thus her translation remained incomplete. This also seems to imply that she was not motivated enough to produce a refined translation product (cf. section 5. 2. 2. 3.).

On the basis of these observations it can be argues that Lucy’s view of the qualities of a good translator seems to be more or less quantitative; that is, she seems to place most weight on speed and efficient problem-solving in translation.

Dopo aver scritto l’acronimo, Lucy legge ad alta voce le prime due parole della spiegazione, dopo di che fa notare che il modo più semplice per risolvere il problema sarebbe telefonare a un dottore. Lucy prende questa decisione abbastanza in fretta (in due o tre secondi) e spontaneamente (è chiaramente una decisione non problematica).

(6) jonka lyhenne (5.0) on N A P (1.0) eiku D (.) P H (2.0) nikotiiniamidi adeniini ni ni di di di (1.0)

ja tätä tällästä minä en edes etsi sanakirjasta koska (1.0) hyvin paljon helpommalla pääsee ku soittaa jollekii lääkärille (2.0) (Lucy: P)

la cui abbreviazione (5.0) è N A P (1.0) no D (.) P H (2.0) nicotinamide adenin ne ne di di di (1.0)
e questo genere di termini non lo cercherò nemmeno sul dizionario perchè (1.0) perchè te la cavi molto più semplicemente se chiami un dottore

L’esempio (6) dimostra l’inclinazione di Lucy verso l’alta efficienza; non vuole sprecare tempo cercando di trovare il termine finlandese in testi di consultazione e semplicemente indica il modo più facile e veloce di risolvere il problema. In verità, chiedere l’aiuto di un esperto per i problemi terminologici è una pratica professionale valida. Tuttavia, Lucy non ha chiamato un dottore per controllare il termine finlandese e, di conseguenza, la sua traduzione è rimasta incompleta. Questo sembra anche implicare che non era abbastanza motivata per produrre una traduzione raffinata (cfr. sezione 5.2.2.3.).

Sulla base di queste osservazioni, è possibile sostenere che la concezione che Lucy ha delle qualità di un buon traduttore è per lo più quantitativa; infatti, in ambito traduttivo, sembra dare più peso alla velocità e all’efficienza nel problem-solving.

These are naturally desirable qualities of all translators; however as far as the experimental translation task is concerned, the emphasis on quantity has clearly resulted in low quality. However, it should be stressed that Lucy’s poor success in translating may be limited to the experimental situation; her conduct may be successful in translating routine texts on the job.

Lucy’s nonchalant translational behaviour can also be explained through social-psychological factors, namely by looking at the subjects’ role behaviour in the experiment. It seems that role theory, particularly the notion of role distance (Goffman 1961)7 can offer useful insights for understanding the subjects’ behaviour (see also Jääskeläinen 1996b).

In role theory human behaviour is analysed in terms of the roles people perform in society. Goffman (1961: 93) defines roles as ‘the typical response of individuals in a particular position’. However, people do not always produce ‘the typical response’ expected of them, i.e. people do not always live up to their roles. Goffman therefore maintains that typical roles must ‘be distinguished from the actual role performance of a concrete individual in a given position’ and, it could be added, in a given situation. Roles carry with them a large array of expectations which serve to identify individuals; in a sense, we are defined (correctly or incorrectly) by our roles. Furthermore, roles involve norms which govern the behaviour of individuals in particular positions. Consequently, roles are closely related to issues of self-image and the image conveyed to other people, which may or may not coincide. According to Goffman (1961: 87):

7 I am grateful to Stephen Condit for bringing this reference to my attention.

Queste sono qualità naturalmente auspicabili in tutti i traduttori; tuttavia, per quanto riguarda il compito di traduzione sperimentale, l’enfasi sulla quantità ha chiaramente portato a una bassa qualità. Ad ogni modo, bisogna sottolineare che l’esito mediocre ottenuto da Lucy nella traduzione potrebbe essere limitato alla situazione sperimentale; in ambito lavorativo, il suo comportamento potrebbe portare risultati positivi nella traduzione di testi di routine.

Il comportamento disinvolto di Lucy nel tradurre può essere chiarito anche attraverso fattori socio-spicologici, ossia osservando il comportamento di ruolo dei soggetti durante l’esperimento. Pare che la teoria dei ruoli, in particolare il concetto di distanza dal ruolo (Goffman 2003)8, possa offrire intuizioni utili per comprendere il comportamento dei soggetti (vedi anche Jääskeläinen 1996b).

Nella teoria dei ruoli, il comportamento umano è analizzato in base al ruolo che le persone svolgono nella società. Goffman (2003: 108) definisce il ruolo come «la risposta tipica degli individui che si trovano in una posizione particolare». Tuttavia, le persone non sempre producono «la risposta tipica» che da loro ci si aspetta, cioè non sempre tengono fede al loro ruolo. Perciò Goffman sostiene che «si deve distinguere il ruolo tipico dall’esecuzione di ruolo effettiva da parte di un individuo concreto in una data posizione» e, potremmo aggiungere, in una data situazione. I ruoli portano con sé una lunga serie di aspettative che servono a identificare gli individui; in un certo senso, siamo definiti (in modo corretto o no) in base ai nostri ruoli. Inoltre, i ruoli implicano norme che governano il comportamento degli individui in particolari posizioni. Di conseguenza, sono strettamente legati a questioni riguardanti l’immagine di sé e l’immagine trasmessa agli altri, che possono coincidere o no. Secondo Goffman (2003: 103):

8 Sono riconoscente a Stephen Condit per aver sottoposto questo riferimento alla mia attenzione.

— in performing a role the individual must see to it that the impressions of him that are conveyed in the situation are compatible with [the] role-appropriate personal qualities effectively imputed to him: a judge is supposed to be deliberate and sober; a pilot, in a cockpit, to be cool; a book-keeper to be accurate and neat doing his work. These personal qualities, — provide a basis of self-image for the incumbent and a basis for the image that his role others9.

In terms of role theory the subjects in the present study were faced with different situations, the professional translators were acting in their familiar role of a ‘translator’ which, presumably, forms a part of their self-image. The non-professional translators, in turn, were asked to act in an unfamiliar role. As a result, there were two kinds of constraints operating within the experimental group. For the professional translators the experimental situation is potentially face-threatening, because it entails exposing a part of their ‘self’ to outside observation. The non-professionals, in turn, may have found the situation threatening because they were asked to perform a role which was outside their competence. Instead, as adults they possessed other roles by which they wished to be defined.

To describe actual role performances of individuals (instead of typical roles on the basis of observing groups of people), Goffman introduces a number of new role concepts (1961: 95ff.), of which the notions of ‘role distance’ and ‘attachment to a role’ seem to have direct relevance to explaining the subjects’ behaviour in the present study.

9 “Role others” refer to the other people invvolved in the situation where roles are performed, i.e. “relevant audiences” (Goffman 1961: 85).

[…] nell’eseguire un ruolo, l’individuo deve far sì che le impressioni di se stesso che vengono comunicate nella situazione siano compatibili con le qualità personali appropriate al ruolo che gli sono attribuite nei fatti: si presume che un giudice sia ponderato e non ubriaco; un pilota in una cabina di pilotaggio non deve apparire agitato; un contabile dev’essere preciso e ordinato nel fare il suo lavoro.

Queste qualità personali, […] forniscono una base per l’immagine del sé, e una base per l’immagine che avranno su di lui i suoi altri di ruolo10.

Per quanto riguarda la teoria dei ruoli, i soggetti di questo studio sono stati sottoposti a diverse situazioni. I traduttori professionali eseguivano il loro abituale ruolo di traduttori che, presumibilmente, fa parte della loro immagine di sé. Ai traduttori non professionali, a loro volta, è stato chiesto di eseguire un ruolo non abituale. Di conseguenza, sul gruppo sperimentale agivano due forzature. Per i traduttori professionali la situazione sperimentale è potenzialmente face-threatening, in quanto richiede l’esposizione di una parte del loro Sé all’osservazione esterna. I traduttori non professionali, a loro volta, potrebbero aver trovato la situazione minacciosa, perchè è stato chiesto loro di svolgere un ruolo al di fuori della loro competenza. In quanto adulti, invece, possedevano altri ruoli in base ai quali desideravano essere definiti.

Per descrivere le effettive performance di ruolo degli individui (anziché il ruolo tipico, sulla base dell’osservazione di gruppi di persone), Goffman introduce alcuni nuovi concetti di ruolo (2003: 105), tra cui i concetti di «distanza dal ruolo» e «attaccamento a un ruolo» sembrano particolarmente rilevanti per spiegare il comportamento dei soggetti in questo studio.

10 «Altri di ruolo» si riferisce alle altre persone coinvolte nella situazione all’interno della quale si svolgono i ruoli, cioè «pubblici rilevanti» (Goffman 1961: 85).

Goffman uses the notion of role distance ‘to refer to [the] actions which effectively convey some disdainful detachment of the performer from a role he is performing’ (1961: 110). In contrast, attachment to a role is described as follows (1961: 89)

The self-image available for anyone entering a particular position is one of which he may become affectively and cognitively enamoured, desiring and expecting to see himself in terms of the enactment of the role and the self-identification emerging form this enactment.

As a rule, people are attached to the roles they perform regularly. In fact, Goffman points out that it is considered to be ‘sound mental hygiene for an individual to be attached to the role he performs’ (1961: 89f.). However, on some occasions it may be equally necessary, partly in terms of ‘mental hygiene’, to express detachment from the role which is being performed (see examples below).

Role distancing can be manifested by explanations, apologies or joking which are, according to Goffman, ‘ways in which the individual makes a plea for disqualifying some of the expressive features of the situation as sources of definitions of himself’ (1961: 105). The reasons for expressing role distance, i.e. the functions of role distancing, may vary greatly, depending on the role performer and on the situation. Expressions of role distance may show a general detachment of the role (‘This is not the real me’); alternatively, role distance can also be a momentary escape from a role, to which the role performer is in fact attached. The function of the latter kind of role distance can be, for instance, to secure the functionability of a whole system of roles, which Goffman calls ‘situated activity systems’, such as a surgical operation (Goffman 1961: 126ff.).

Goffman usa il concetto di «distanza dal ruolo» «per descrivere atti che comunicano efficacemente un certo sprezzante distacco dell’esecutore da un ruolo che sta eseguendo» (2003: 127). Al contrario, l’attaccamento a un ruolo è descritto come segue (2003: 105):

L’immagine di sé che viene offerta da una certa posizione a chi la occupa può essere qualcosa di cui ci si innamora col cuore e con l’intelletto; non si aspetta altro che di calarsi nel ruolo e di sfruttare i vantaggi in termini di identità che ciò può dare.

Di norma, le persone sono attaccate al ruolo che svolgono regolarmente. In realtà, Goffman fa notare che si considera «buona igiene mentale per un individuo sentirsi attaccato al ruolo che svolge» (2003: 105.). Tuttavia, in alcune occasioni potrebbe essere ugualmente necessario, in parte in termini di «igiene mentale», esprimere la distanza dal ruolo che si sta svolgendo (vedi esempio sottostante).

La distanza dal ruolo può manifestarsi attraverso spiegazioni, scuse o battute che sono, secondo Goffman, «modi in cui l’individuo pretende di screditare alcuni degli aspetti espressivi della situazione in quanto fonti di definizione del suo Sé» (2003: 121). I motivi della manifestazione della distanza dal ruolo, cioè le funzioni della distanza dal ruolo, possono variare molto, in base all’esecutore del ruolo e alla situazione. Le manifestazioni della distanza dal ruolo possono mostrare un generale distacco dal ruolo («questo non è il vero me»); alternativamente, la distanza dal ruolo può anche essere una fuga momentanea dal proprio ruolo, al quale l’esecutore è di fatto attaccato. La funzione dell’ultimo tipo di distanza dal ruolo può essere, per esempio, quella di assicurare la funzionalità di un intero sistema dei ruoli, che Goffman chiama «sistema situato di attività», come le operazioni chirurgiche (Goffman 2003: 111.).

As think-aloud experiments cannot be defined as ‘situated activity systems’ in the same sense as surgical operations, the manifestations of role distancing in the present data seem to be connected to the function of role distancing as a form o psychological self-defence. Conspicuous displays of role distancing were observable only in the new data, and in the non-professional subjects’ behaviour in particular. There are two potential reasons for the apparent absence of displays of role distancing in the students’ translation processes. First, Goffman maintains (1961: 139f.) that learners are granted certain liberties in the period of learning, i.e. in the period of ‘role-taking’. The role-taker is allowed to make mistakes which would otherwise be considered discreditable, because ‘he has a learner’s period of grace in which to make them – a period in which he is not yet quite the person he will shortly be, and, therefore, cannot badly damage himself by the damaging expression of his maladroit actions’ (1961: 139). It seems possible that the students of translation see themselves as learners and, therefore, do not feel threatened by exposing themselves to observation n the experimental situation. Second, Goffman argues (1961: 109) that ‘immediate audiences figure very directly in the display of role distance’. In the two parts of my experiment, the immediate audience, i.e. the experimenter, may have occupied a slightly different role. In the first sessions, which were organised to collect data for my pro gradu thesis, the experimenter was ‘just a fellow student’ and, as a result, did not pose any kind of thereat to the students acting as subjects. In the new sessions, in addition to all the subjects having more firmly established roles in society, the experimenter as ‘a researcher’ might also have occupied a slightly different position, which, in turn, might have created a more face-threatening situation favourable to expressions of role distance.

Dato che gli esperimenti di think-aloud non possono essere definiti «sistemi situati di attività» al pari delle operazioni chirurgiche, le manifestazioni di distanza dal ruolo in questo studio sembrano legate alla funzione della distanza dal ruolo come forma di autodifesa psicologica. Dimostrazioni evidenti di distanza dal ruolo sono state osservate solo nei nuovi dati, e in particolare nel comportamento dei soggetti non professionali. Ci sono due potenziali cause dell’apparente assenza di manifestazioni della distanza dal ruolo nei processi traduttivi degli studenti. In primo luogo, Goffman afferma (2003: 156) che, durante il periodo di apprendimento, cioè il periodo in cui un individuo comincia a entrare nel proprio ruolo, ai principianti vengono concesse certe libertà. Al principiante viene concesso di commettere errori che altrimenti sarebbero considerati screditanti, perchè «dispone del periodo di grazia dell’apprendista per cui questi sbagli gli sono concessi: un periodo in cui egli non è ancora del tutto la persona che sarà tra poco e non può quindi danneggiare gravemente se stesso con la dannosa manifestazione delle sue azioni maldestre» (2003: 156). È possibile che gli studenti di traduzione si vedano come apprendisti e, perciò, non si sentano minacciati nell’esporsi all’osservazione in una situazione sperimentale. In secondo luogo, Goffman afferma (2003: 126) che «il pubblico presente ha una parte diretta nell’esibizione della distanza dal ruolo». Nelle due parti del mio esperimento, il pubblico presente, cioè lo sperimentatore, avrebbe occupato un ruolo leggermente diverso. Nelle prime sessioni, organizzate per raccogliere dati per la mia tesi di laurea, lo sperimentatore era “solo un compagno di corso” e, di conseguenza, non rappresentava alcun tipo di minaccia per gli studenti soggetti dello studio. Nella nuova sessione, oltre al fatto che tutti i soggetti avevano ruoli ben definiti nella società, lo sperimentatore in qualità di ricercatore avrebbe occupato una posizione leggermente diversa, che, a sua volta, avrebbe potuto creare una situazione più face-threatening che avrebbe favorito la manifestazione della distanza dal ruolo.

As was mentioned earlier, the most conspicuous examples of role distancing can be found in the non-professional translators’ (particularly Ann’s, Laura’s and Paul’s) behaviour. As they found themselves performing a role in which they did not feel comfortable, or with which they did not wish to identify themselves, they seemed to want to convince the observers that ‘this is not the real me, you must not take my actions seriously’. This was expressed by constant laughter (especially Ann and Laura) and joking. Consider, for instance, the following examples (verbalisations of interest are printed in bold).

(7) mikä tuo dicky on [D2:POCKETa] (25.0) ohhoh (5.0) dicky (3.0) nyt mie en ymmärrä ei tässä oo kyllä semmosta (laugh)
onks täällä oikein (laugh) (2.0)
dicky heart ku se on istuin (.) auton takaosassa (1.0) paidan etumus (1.0) dicky bird on tipu (3.0) (laugh)

voiks se olla lintusydän (laugh) (8.0)
Ei mut sillä täytyy olla jotain sell merkityksiä joita ei oo tähän laitettu (1.0) pitäskö se sitten olla joku isompi sanakirja (3.0) (Ann:N-P)

what that DICKY is [D2:POCKETa] (25.0) ohhoh (5.0) DICKY (3.0) now I don’t understand here’s no such thing (laugh)
it is correct here (laugh)
DICKY HEART as it’s a seat (.) at the back of a car (1.0) shirt- front (1.0) DICKY BIRD is a birdie (3.0) (laugh)

can it be a bird heart (laugh) (8.0)
no it’s got to have some meanings which haven’t been put here (1.0) should it be a bigger dictionary then (3.0)

Come accennato in precedenza, gli esempi più evidenti di distanza dal ruolo possono essere osservati nel comportamento dei traduttori non professionali (in particolare Ann, Laura e Paul). Trovandosi a svolgere un ruolo nel quale non si sentivano a loro agio, o con il quale non volevano identificarsi, sembravano voler convincere gli osservatori che «questo non è il vero me, non devi prendere le mie azioni sul serio ». Questo veniva espresso da riso continuo (in particolare Ann e Laura) e battute. Consideriamo, per esempio, quanto segue (le verbalizzazioni di maggior interesse sono in neretto).
(7) mikä tuo dicky on [D2:POCKETa] (25.0) ohhoh (5.0) dicky (3.0)

nyt mie en ymmärrä ei tässä oo kyllä semmosta (laugh) onks täällä oikein (laugh) (2.0)

dicky heart ku se on istuin (.) auton takaosassa (1.0) paidan etumus (1.0) dicky bird on tipu (3.0) (laugh)

voiks se olla lintusydän (laugh) (8.0)
Ei mut sillä täytyy olla jotain sell merkityksiä joita ei oo tähän laitettu (1.0) pitäskö se sitten olla joku isompi sanakirja (3.0) (Ann:N-P)

cos è quel DICKY [D2:POCKETa] (25.0) ohhoh (5.0) DICKY (3.0) ora non capisco qui non c’è niente del genere (ride)
è giusto qui (ride)
DICKY HEART come se fosse un sedile (.) posteriore dell’auto (1.0) pettino

(1.0) DICKY BIRD è un uccellino (3.0) (ride)
può essere il cuore di un uccello (ride) (8.0)
no deve avere qualche significato che qui non c’è (1.0) dovrebbe essere un dizionario più grande allora (3.0)

(8) nää on aina kauheen vaikeita kääntää nää nimet (2.0)
(sigh) siis (2.0) Olof Sodimu (.) Peter Joosef ja (1.0) Günter Augusti (1.0) koolla (laugh) (1.0) Kuustaa Augusti (3.0) (Paul:N-P)

these are always very difficult to translate these names (2.0)
(sigh) so (2.0) Olof Sodimu (.) Peter Joseph and (1.0) Günter Augusti (1.0) with a K (laugh) (1.0) Kustaa Augusti (3.0)
(9) mutta (3.0) voisikohan voisiko (1.0) valkosipulista (1.0) kosipulista
(1.0) olla (1.0)
käännetään vapaasti
apua (1.0)
toi on turha kääntää voisko se nyt pelastaa (1.0) pulasta (1.0) mutta voisiko valkosipulista olla apua
pannaan näin (1.0)
pannaan (laugh) olla pelastava enkeli (2.0) (Laura:N-P)

but (3.0) could could (1.0) garlic (1.0) garlic (1.0) be (1.0)
let’s translate freely
of help (1.0)
that’s unnecessary to translate that could it rescue (1.0) from trouble

(1.0) but could garlic be of help
Let’s put it like that (1.0)
let’s put (laugh) be a rescuing angel (2.0)

This type of joking, i.e. producing ludicrous or non-sensical translation

variants, which were, furthermore, clearly identifies as such by laughter, was typical of the non-professional translators, and practically non-existent

(8) nää on aina kauheen vaikeita kääntää nää nimet (2.0)
(sigh) siis (2.0) Olof Sodimu (.) Peter Joosef ja (1.0) Günter Augusti
(1.0) koolla (laugh) (1.0) Kuustaa Augusti (3.0) (Paul:N-P)

questi sono sempre difficilissimi da tradurre questi nomi (2.0) (sospira) dunque (2.0) Olof Sodimu (.) Peter Joseph e (1.0) Günter Augusti
(1.0) con la K (ride) (1.0) Kustaa Augusti (3.0)
(9) mutta (3.0) voisikohan voisiko (1.0) valkosipulista (1.0) kosipulista
(1.0) olla (1.0)
käännetään vapaasti
apua (1.0)
toi on turha kääntää voisko se nyt pelastaa (1.0) pulasta (1.0) mutta voisiko valkosipulista olla apua
pannaan näin (1.0)
pannaan (laugh) olla pelastava enkeli (2.0) (Laura:N-P)

ma (3.0) può può (1.0) l’aglio (1.0) aglio (1.0) essere (1.0) provo a tradurre liberamente
d’aiuto (1.0)
non è necessario tradurre che può salvare (1.0) dai problemi (1.0) ma l’aglio può essere d’aiuto

mettiamolo così (1.0)

mettiamo (ride) essere un angelo salvatore (2.0)

Questo tipo di battute, cioè delle varianti traduttive grottesche e senza

senso, oltretutto chiaramente identificate come tali dal riso, erano tipiche dei traduttori non professionali, e praticamente inesistenti nel

in the other subjects’ behaviour. These examples seem to be expressions of role distancing, which the subjects employed to detach themselves from the translator’s role they had been compelled to perform. In fact, Goffman maintains (1961: 112) that in situation where novices or non-experts are performing an unfamiliar role, manifesting role distance gives the performers some elbow room in which to manoeuvre. By expressing role distancing, the role performer is telling the observers that ‘I am not to be judged by this incompetence’. Goffman also points put that such ‘out-of-character situations can easily be created experimentally by asking subjects to perform tasks that are inappropriate to persons of their kind’ (1961: 112). Even though the non- professional subjects might not have felt that the translation task was inappropriate, they probably did feel that the task was not within their sphere of competence.

As was mentioned earlier, manifestations of role distance are harder to find in the professional translators’ behaviour, with one notable exception, namely Lucy. To my mind, Lucy’s behaviour in the experiment resembled closely Goffman’s description of displays of role distance in what he calls an ‘unserious setting’, i.e. the merry-go-round. Goffman describes the behaviour of the operator of the merry-go-round as follows (1961: 109):

Not only does he show that the ride itself is not – as a ride – an event to him, but he also gets off and on and around the moving platform with grace and ease that can only be displayed by safely taking what for children and even adults would be chances.

The ‘grace and ease’ of Lucy’s decisions to solve problems by contacting experts could be considered as manifestations of role distance.

comportamento degli altri soggetti. Questi esempi sembrano manifestazioni di distanza dal ruolo, usata dai soggetti per allontanarsi dal ruolo di traduttore che sono stati obbligati a svolgere. Infatti, Goffman sostiene (2003: 129) che in situazioni in cui gli apprendisti o i non esperti svolgono un ruolo non familiare, la manifestazione della distanza dal ruolo dà loro un certo spazio di manovra. Manifestando la distanza dal ruolo, l’esecutore sta dicendo agli osservatori: «Non devo essere giudicato in base a questa incompetenza». Goffman afferma anche che «analoghe situazioni “fuori personaggio” possono essere create sperimentalmente senza difficoltà; occorre solo chiedere a qualcuno di eseguire compiti che non sono appropriati a persone del suo tipo» (2003: 129). Anche se i soggetti non professionali potrebbero non aver percepito che il compito di traduzione era inappropriato, hanno probabilmente capito che il compito non apparteneva alla loro sfera di competenza.

Come detto in precedenza, è più difficile riscontrare manifestazioni della distanza dal ruolo nel comportamento dei traduttori professionali, con un’importante eccezione, e precisamente Lucy. A mio parere, il comportamento di Lucy durante l’esperimento riproduce nei minimi dettagli l’esibizione della distanza dal ruolo in quello che Goffman chiama «un setting poco serio», vale a dire la giostra. Goffman descrive il comportamento dell’addetto alla giostra come segue (2003: 126):

Non solo egli mostra che la corsa in se stessa non è (in quanto corsa) un evento per lui, ma per di più sale e scende dalla piattaforma in moto con un’eleganza e un’agilità che possono essere ottenuti solo a prezzo di correre tranquillamente dei rischi impensabili per i bambini o anche per gli adulti.

L’«eleganza e l’agilità» delle decisioni di Lucy di risolvere i problemi contattando degli esperti potrebbe considerarsi come una manifestazione di distanza dal ruolo.

That is, Lucy seemed to be trying to convey an image of a highly efficient and competent translator, who is in absolute control of the situation. Certainly, had the quality of the translations not been assessed at all, Lucy’s behaviour would, on the surface, have supported the early hypotheses about professional behaviour in translation (cf. section 2. 4.), namely that professional translators rely on automatised processing and are able to sail through translation tasks quickly and effortlessly. However, none of the other professional translators performed the experimental translation task with similar ‘grace and ease’; on the contrary, they spent considerably more time and effort on it. Moreover, there were few instances in their protocols which could be categorised offhand as manifestations of role distancing.

Goffman argues (1961: 130) that a person ‘who manifests much role distance may, in fact, be alienated from the role’, but the opposite may equally well be true, as ‘in some cases only those who feel secure in their attachment may be able to chance the expression of distance’. Without more information about the subjects, it is impossible to say conclusively what might have been the reason for the obvious manifestations of role distancing in Lucy’s behaviour, as opposed to their apparent absence from the other professional translators’ behaviour. However, Goffman contends that (1961: 102):

Whatever the individual does and however he appears, he knowingly and unknowingly makes information available concerning the attributes that might be imputed to him and hence the categories in which he might be placed.

On this basis, we can hypothesise that the differences in the professional translators’ behaviour may imply that they wished (unconsciously) to convey different images of their roles as translators,

Sembra cioè che Lucy stia tentando di trasmettere l’immagine di una traduttrice molto efficiente e competente, che ha pieno controllo della situazione. Certamente, se la qualità delle traduzioni non fosse stata valutata, il comportamento di Lucy, in superficie, avrebbe assecondato le prime ipotesi sul comportamento professionale nella traduzione, cioè che i traduttori professionali si basano su esecuzioni automatizzate e sono in grado di superare il compito di traduzione velocemente e senza sforzo. Tuttavia, nessuno degli altri traduttori professionali ha eseguito il compito di traduzione sperimentale con simile «eleganza e agilità»; al contrario, gli hanno dedicato molto più tempo e fatica. Inoltre, nei loro protocolli c’erano pochi esempi classificabili, su due piedi, come manifestazioni di distanza dal ruolo.

Goffman afferma (2003: 145) che una persona «che manifesta molta distanza dal ruolo può essere effettivamente alienata dal ruolo», ma anche l’opposto potrebbe essere vero, in quanto «in certi casi solo quelli che sono certi del loro attaccamento possono essere capaci di rischiare un’espressione di distanza». Senza ulteriori informazioni riguardo ai soggetti, è impossibile dire definitivamente quale potrebbe essere stato il motivo delle evidenti manifestazioni di distanza dal ruolo nel comportamento di Lucy, in contrasto con la loro apparente assenza nel comportamento degli altri traduttori professionali. Tuttavia, Goffman sostiene che (2003: 118):

Qualunque cosa un individuo faccia e quali che siano le sue apparenze, egli, consapevolmente o inconsapevolmente, rende disponibili delle informazioni relative alle qualifiche che possono essergli attribuite e quindi alle categorie in cui può essere collocato.

In base questo, possiamo ipotizzare che le differenze nel comportamento dei traduttori professionali implicano il loro desiderio (inconscio) di trasmettere immagini diverse dei propri ruoli di traduttore,

which may, in turn, reflect different self-images. Lucy may have wished to give the impression of a highly efficient translator who knows hoe to solve problems quickly whereas Fran may have aimed at an impression of a conscientious and meticulous translator by behaving in a potentially over- conscientious fashion in the experiment. Obviously there is also the possibility that in an experimental situation subjects may to some extent manipulate (probably unconsciously) their behaviour to please the researcher. Thus Lucy may have behaved with exaggerated ease in the experiment, at the expense of translation quality, while Fran may have overdone her ‘meticulous translator act’. If this is true, it is particularly interesting that Lucy and Fran should have manipulated their behaviour into opposite directions with regard to the demands of quantity vs. quality in translation. This seems to reveal a difference in their (implicit) definitions of a ‘good translator’, as it seems unlikely that they would have wanted to give an example of a ‘bad translator’ in the experiment.

The observations discussed in this section highlight the complexity of investigating translation process, or of any type of human behaviour for that matter. I would like to stress that it is not the purpose of the present study to discredit any of the professional translators who took part in the experiment; the success or failure in the experimental task may have little to do with how they succeed in their own work. However these observations seem to support the hypothesis that professional translators may behave differently when performing routine vs. non-routine tasks. In fact, one of the most plausible explanations for the unexpectedly poor success of the two professional translators, Penny and Lucy, is that they applied a routine approach to a non- routine task. The consequences were less dramatic for Penny, for whom the topic of the ST represented a familiar special field.

che possono, a loro volta, riflettere immagini di Sé diverse. Può darsi che Lucy desiderasse dare l’impressione di un traduttore molto efficiente, che sa come risolvere i problemi velocemente, mentre che Fran volesse dare l’impressione di un traduttore scrupoloso e meticoloso comportandosi in modo potenzialmente troppo scrupoloso nell’esperimento. Ovviamente, esiste anche la possibilità che in una situazione sperimentale i soggetti manipolino (probabilmente in modo inconscio) il loro comportamento in una certa misura per assecondare il ricercatore. Perciò Lucy potrebbe essersi comportata con esagerata “agilità” nell’esperimento, ai danni della qualità della traduzione, mentre Fran potrebbe aver esagerato il suo «atto traduttivo meticoloso». Se questo è vero, è molto interessante che Lucy e Fran avrebbero dovuto manipolare i loro comportamenti in direzioni diverse rispetto alle richieste di qualità vs. quantità in traduzione. Questo sembra rivelare una differenza nelle loro definizioni (implicite) di “buon traduttore”, visto che sembra improbabile che nell’esperimento volessero dare l’esempio di “cattivo traduttore”.

L’osservazione esaminata in questa sezione sottolinea la complessità dello studio dei processi traduttivi, o se è per quello, di qualsiasi tipo di comportamento umano. Vorrei sottolineare che non è obiettivo di questo studio screditare i traduttori professionali che hanno preso parte all’esperimento; la buona o la cattiva riuscita nel compito sperimentale può avere poco a che fare con i risultati che ottengono nel loro lavoro. Tuttavia, queste osservazioni sembrano supportare l’ipotesi secondo la quale i traduttori professionali possono comportarsi diversamente a seconda che eseguano compiti di routine o no. Infatti, una delle spiegazioni più plausibili per l’inaspettata poca riuscita delle due traduttrici professionali, Penny e Lucy, è che abbiano usato un approccio abituale per un compito non abituale. Le conseguenze sono state meno accentuate per Penny, per la quale l’argomento del prototesto apparteneva a un campo specifico familiare.

7. CONCLUSION

The purpose of the present research was to portray think-aloud protocols as the best suited method for collecting data and investigating the human mind processes, particularly the translation process. At the same time, it presents the problems this verbalizing procedure entails and the limits researchers meet, when trying to compare different TAP experiments.

According to cognitive psychological literature, verbal reports yield valid and reliable data on human thought processes. However, cognitive psychology has always dealt mainly with well-defined problem-solving tasks, for which it is possible to determine a priori correct solutions and problem-solving strategies. Translating, in turn, as a creative and subjective process, represents an fuzzy form of problem-solving. Furthermore, translating and thinking aloud are both verbal tasks, which means that they may draw on the same memory resources, and thus interfere with each other.

To avoid these kinds of problems, researchers tried to use joint translating as a method of data elicitation. Although the studies comparing think-aloud protocols and joint translation have offered interesting results, particularly with regard to didactic applications, these experiments contain other variables, which makes it impossible to state that joint translating would be a better method for studying translating than thinking aloud.

In both cases, it is important to remember that subjects are asked to verbalize their thoughts; this is not a simple task, because it requires them to reorganize their mental discourse into an oral one. And even if experimenters recommend them to say things just as they come to mind, they’ll probably try to communicate their mental ideas in a way other people would understand. Moreover, many other elements, such as the subjects’ personal history, emotional factors, the effect of the experimental situation, fear of failure etc. (cf. section 7), may alter the results of TAP experiments.

More attention should be paid to identifying and isolating these variables by, for example, using pre-experimental testing; then, they should be carefully taken into account when examining the results of verbal reports. The

experimental situation should be more carefully analyzed in terms of limitations on time, access to reference material and ST difficulty. Furthermore, the validity and reliability of various methods of data collection in relation to translating should be determined by a study specifically designed for that purpose.

There exists a wide spectrum of research interests in TAP studies on the translation process; for this reason, the methods of analysis have been equally varied, which makes it difficult to use previous methods of analysis in new experiments. There is a lack of experimental tradition in this field and consequently most TAP studies on translating suffer from methodological weakness.

Nevertheless, the great variety of aims and strategies offered by TAP experiments shows the complexity of the translation process, and all the mechanisms and factors involved. Moreover, the peculiarity of every single experiment could represent the starting point for new methods of investigation.

APPENDICE: IL TESTO DI RIFERIMENTO PER GLI ESEMPI

Stay slim – eat garlic

Everyone knows that eating fatty foods is no good for you, especially if you have a dicky heart. However, the search for a miracle drug that could safely mop all the excess fats has been somewhat difficult… but could garlic come to the rescue?

O. Sodimu, P. Joseph and K. Augusti at the University of Maidugari in Nigeria, fed an exceptionally fatty diet to rats. Not surprisingly, the creatures accumulated cholesterol in their blood, liver and kidneys. But adding garlic oil to the same high-fat diet prevented the rise in the fatty constituents: cholesterol, triglycerides, and total lipids (Experientia, vol 40, p 5).

How does the garlic work? The authors speculate that it knocks out some of the key enzymes involved in making fatty acid or cholesterol. Alternatively, garlic may nobble the energy-carrying compound NADPH (nicotineamide adenine dinucleotide phosphate hybride), which is necessary for making lipids.

RINGRAZIAMENTI

Ringrazio tutte le persone che in un modo o nell’altro, mi hanno supportato e sopportato nel corso di questi anni di studio. Non deve essere stato facile starmi vicino nei miei frequenti momenti di stress e tensione, ma è grazie alla loro presenza, alla loro comprensione, alle loro parole a volte anche dure e ai loro consigli che sono riuscita ad arrivare fin qui.

Ringrazio tutti i professori con cui ho potuto lavorare in questi anni, per i loro insegnamenti non solo scolastici, ma “di vita”, in particolare il professor Bruno Osimo, che oltre ad avermi seguito nella stesura di questa tesi, è stato disponibile e gentilissimo nell’offrirmi ascolto e consigli importanti.

Un ringraziamento speciale ai miei genitori, che hanno sempre creduto in me, nella mia determinazione, e mi hanno sempre dato la libertà di scegliere la strada che volevo seguire.

Grazie ai miei fratelli, che nonostante tutto mi vogliono bene.
Grazie alla nonna che si è sempre ricordata le date dei miei esami. Grazie ad Andrea e alla sua famiglia, da cui ho sempre ricevuto

sostegno, comprensione, conforto, simpatia e una casa dove rifugiarmi a studiare.

Grazie a Silvia, con cui ho condiviso bellissimi anni di faticosi ma soddisfacenti studi.

RIFERIMENTI BIBLIOGRAFICI

ARNTZ R. AND THOME G. (a cura di) 1990 Übersetzungswissenschaft. Ergebnisse und Perspektiven. Festschrift für Wolfram Wilsszum 65. Geburtstag, Tübingen, Narr.

BERNARDINI S. 1999 Using think-aloud protocols to investigate the translation process: methodological aspects, Bologna, University of Bologna.

BEYLARD-OZEROFF, KRÁLOVA A. J. AND MOSER-MERCER (a cura di) 1998 Translators’ strategies and creativity. Selected papers form the 9th international conference on translation and interpreting, Prague, September 1995, Amsterdam/Philadelphia, John Benjamins.

BÖRSCH S. 1986 Introspective methods in research on interlingual and intercultural communication in House J. and Blum-Kulka S. (a cura di): 195-209.

DANCETTE J. 1994 Comprehension in the translation process: an analysis of think-aloud protocols in Dollerup C. and Lindegaard A. (a cura di): 113- 120.

DIMITROVA E. 2005 Expertise and explicitation in the translation process, Amsterdam /Philadelphia, John Benjamins, ISBN 90-272-1670-3.

DOLLERUP C. AND LINDEGAARD A. (a cura di) 1994 Teaching translation and interpreting 2. Insights, aims, visions, Amsterdam/Philadelphia, John Benjamins, ISBN 90-272-1617-7.

ERICSSON K. A. and SIMON H. A. 1984 Protocol Analysis. Verbal reports as data, Cambridge, MIT Press, ISBN 0-262-55012-1.

ERICSSON K. A. and SIMON H. A. 1993 Protocol Analysis: Verbal reports as data (Revised edition), Cambridge, MIT Press, ISBN 0-262-55023-7.

GERLOFF P. 1986 Second language learners’ reports on the interpretive process: talk-aloud protocols of translation, in House J. and Blum-Kulka S. (a cura di): 243-262.

GOFFMAN E. 1961 Encounters: two studies in the sociology of interaction, Indianapolis, Bobbs-Merill, [trad. it. di Paolo Maranini 2003 Espressione e identità. Gioco, ruoli, teatralità, Bologna, Il Mulino, ISBN 88-15-09079- 7]: 101-174.

HAUTAMÄKI A. (a cura di) 1988b Kognitiotiede, Helsinki, Gaudeamus. 70

HOLZ-MÄNTTÄRI J. (a cura di) 1998 Translationstheorie – Grundlagen und Standorte, Studia translatologica, Ser. A, Vol. 1., Tampere, University of Tampere, ISBN 951-44-2387-9.

HOLZ-MÄNTTÄRI J. AND NORD C. (a cura di) 1993 Traducere navem. Festschrift für Katharina Reiß zum 70. Geburtstag, Tampere, University library, ISBN 951-44-3262-2.

HÖNIG H. G. 1990 Sagen, was man nicht weiß – wissen was man nicht sagt. Überlegungen zur übersetzerischen Intuition in Arntz R. and Thome G. (a cura di): 91-101.

HÖNIG H. G. 1991 Holmes’ “Mapping Theory” and the landscape of mental translation processes in Van Leuven-Zwart K. M. and Naaijkens T. (a cura di): 77-89.

HOUSE J. 1988 Talking to oneself or thinking with others? On using different thinking aloud methods in translation, Fremdsprachen Lehren und Lernen Vol. 17.

HOUSE J. AND BLUM-KULKA S. (a cura di) 1986 Interlingual and intercultural communication. Discourse and cognition in translation and second language acquisition studies, Tübingen, Gunter Narr.

JÄÄSKELÄINEN R. 1993 Investigating Translation Strategies in Tirkkonen- Condit S. and Laffling J. (a cura di): 99-120.

JÄÄSKELÄINEN R. 1999 Tapping the process: an explorative study of the cognitive and affective factors involved in translating, Joensuu, University of Joensuu, ISBN 951-708-734-8.

JENSEN A. 2000 The effects of time on cognitive process and strategies in translation, København, Unpublished PhD thesis, Copenhagen Business School.

JONASSON K. 1998 Degree of a text awareness in professional vs. non- professional translators in Beylard-Ozeroff, Králova A. J. and Moser- Mercer (a cura di): 189-200.

KÖNIGS, F. G. 1987 Was beim Übersetzen passiert: Theoretische Aspekte, empirische Befunde und praktische Konsequenzen, Die Neueren Sprachen.

KRAHMER E. and UMMELEN N. 2004 Thinking About Thinking Aloud. A comparison of two verbal protocols for usability testing, Netherlands, Tilburg University, ISSN 0361-1434.

KRINGS H. P. 1986. Translation problems and translation strategies of advanced German learners of French (L2) in House J and Blum-Kulka S. (a cura di): 263-276.

KRINGS H. P. 1988 Thesen zu einer empirischen Übersetzungswissenschaft in Holz-Mänttäri J. (a cura di): 58-71.

KÜNZLI A. 2003 Quelques stratégies et principes dans la traduction technique français-allemand et français-suédois, Stockholm, Akademitryck, ISBN 91- 974284-6-9.

KUSSMAUL P. 1989a Toward an Empirical Investigation of the Translation Process: Translating a Passage from S. I. Hayakawa: Symbol, Status and Personality in Von Bardeleben R. (a cura di): 369-380.

KUSSMAUL P. 1989b Interferenzen im Übersetzungsprozess – Diagnose und Therapie in Schmidt H. (a cura di): 19-28.

KUSSMAUL P. 1993 Empirische Grundlagen einer Übersetzungsdidaktik: Kreativität im Übersetzungsprozeß in Holz-Mänttäri J. and Nord C. (a cura di): 275-286.

KUSSMAUL P. 1994 Möglichkeiten einer empirisch begründeten Übersetzungsdidaktik in Snell-Hornby M., Pöchhacker F. and Kaindl K (a cura di): 377-386.

KUSSMAUL P. 1995 Training the Translator, Amsterdam, John Benjamins, ISBN 9027216231.

KUSSMAUL P. 1998 Die Erforschung von Übersetzungsprozessen: Resultate und Desiserate, München, Lebende Sprachen.

KUSSMAUL P. and TIRKKONEN-CONDIT S. 1995 Think-Aloud Protocol analysis in translation studies, TTR – Traduction Terminologie Rédaction Vol 8 n 1.

LAFFLING J. 1993a Corpus-based analysis dictionary for machine (and human) translation in Tirkkonen-Condit S. and Laffling J. (a cura di): 121-136.

LEWANDOWSKA-TOMASZCZYK B. AND THELEN M. (a cura di) 1992 Translation and meaning, Part 2, Maastricht, Rijkshogeschool Maastricht, Faculty of translation and interpreting.

LÖRSCHER W. 1992 Process-Oriented Research into Translation and Implications for Translation Teaching, Neubrandenburg, University of Greifswald.

MATRAT C. M. 1992 Investigating the translation process: thinking-aloud versus joint activity, Unpublished doctoral dissertation, Ann Arbor: University Microfilms International.

NIDA E. A. 1964 Towards a science of translating, Leiden, Brill, ISBN 9004026053

NIELSEN J. 1993 Usability Engineering, Cambridge, Academic Press. NISBETT R. E. and WILSON T. D. 1979 Telling more than we can know: verbal

reports on mental processes, Psychological review.
NORBERG 2003 Übersetzen mit doppeltem skopos. Eine empirische Prozess-

und produktstudie, Uppsala, Uppsala University Library.
PÖNTINEN T. and ROMANOV T. 1989 Professional vs. non-professional translator: a think-aloud protocol study, Joensuu, University of Joensuu,

Savonlinna School of Translation Studies.
SAARILUOMA P. 1988b Ajattelu kognitiivisena prosessina in Hautamäki A. (a

cura di): 43-70.
SANDROCK U. 1982 Thinking-aloud protocols (TAPs). Ein Instrument zur

Dekomposition des Komplexen Prozesses “Übersetzen”, Stockholm, Kassel

Svenska Akademiens ordlista.
SCHMID A. 1994 Gruppenprotokolle – ein Einblick in die black box des

Übersetzens, TexTconTexT 9.
SCHMIDT H. 1993 Interferenz in der Translation, Leipzig, Enzyklopädie, ISBN

3-324-00476-4.
SCHMIDT M. A. 2005 How do you do it anyway? A longitudinal study of three

translator students translating from Russian into Swedish, Stockholm,

Acta Universitatis Stockholmiensis, ISBN 91-85445-19-3.
SCHRIVER K .A. 1997 Dynamics in Document Design: Creating Text for

Readers, New York, Wiley, ISBN: 978-0-471-30636-8. 73

SÉGUINOT C. 1996 Some thoughts about think-aloud protocols, Target vol. 8, no1: 79-95, Amsterdam, Benjamins, ISSN 0924-1884.

SNELL-HORNBY M., PÖCHHACKER F. AND KAINDL K. (a cura di) 1994 Translation Studies. An Interdiscipline, Amsterdam, John Benjamins, ISBN 1556194781.

SOMEREN M. W., BARNARD Y. F. and SANDBERG J. A. C. (a cura di) 1994 The think aloud method. A practical guide to modelling cognitive processes, London, Academic Press, ISBN 0-12-714270-3.

TIRKKONEN-CONDIT S. 1992 The interaction of world knowledge and linguistic knowledge in the process of translation. A think-aloud protocol study in Lewandowska-Tomaszczyk B. and Thelen M. (a cura di): 433- 440.

TIRKKONEN-CONDIT S. AND LAFFLING J. (a cura di) 1993 Recent Trends in Empirical Translation Research, [Studies in Languages 28.] Joensuu, University of Joensuu, ISBN 951-708-150-2.

TOURY G. 1984 The notino of “native translator” and translation teaching in Wilss W. and Thome G. (a cura di): 186-195.

VAN LEUVEN-ZWART K. M. AND NAAIJKENS T. (a cura di) 1991

Translation Studies, The state of the art. Proceedings of the first James S. Holmes symposium on Translation Studies, Amsterdam, Rodopi, ISBN 90- 5183-257-5.

VON BARDELEBEN R. 1989 Wege amerikanischer Kultur. Ways and Byways of American Culture. Aufsätze zu Ehren von Gustav H. Blanke, Frankfurt am Main, Lang, ISBN 363142454.

WILSS W. AND THOME G. (a cura di) 1984 Die Theorie des Übersetzens und ihr Aufschlusswert für die Übersetzer- und Dolmetscherausbildung Tübingen, Gunter Narr.

trad.it, il sito di Bruno Osimo

la mia attività: scritture, letture, interviste, traduzioni, poesia, scienza della traduzione, semiotica orcid.org/0000-0002-9204-024X Scopus Author ID: 41661864200 ISNI: 0000000040670907

MICHELA IMPERIO I think-aloud protocol come strumento per indagare il processo mentale della traduzione

Leave a Reply Cancel reply