Sysrem Vol. 8. pp. 59-70 Pergamon Press Ltd. 1980. Printed in Great Britain

LINGUISTIC AWARENESS OF THE TRANSLATION PROCESS

PATRICIA A. SMELCER, NIKET PATWARDHAN, ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

Vanderbilt University, Nashville, Tenessee

This paper examines the linguistic awareness of the translation process by analyzing the self-corrections made by native subjects. Thirty (30) subjects, twenty-seven (27) of whom were native English speakers who had acquired French as a foreign language, and three (3) of whom were native French speakers who had acquired English as a foreign language, were requested to make a simultaneous interpretation of a variant of Little Red Riding Hood from English into French. The story, which was 1: 14 minutes long, had previously been recorded on audio tape at a rate of 164 wpm. Subjects listened to the story through head-phones while speaking into a microphone. Immediately following the experiment, subjects were asked questions about the content of the story and about their translation strategies. Subjects were videotaped during the experiment and during the interview. The analysis of the transcripts focused on the spontaneous self-corrections made by the subjects while interpreting. The self-corrections were divided into three categories: A. Corrections of Content; B. Corrections of Grammar; C. Corrections of Style. The largest distribution of errors (55.56%:15) fell into the grammatical category and the majority of these (93.33%:14) dealt with adjustments of the masculine and feminine gender of French nouns expressed in definite articles, indefinite articles, and possessive pronouns. The self-corrections confirm previous findings by Gerver (1974) that subjects do monitor themselves while interpreting. The fact that valuable time was spent correcting grammatical points that cause no misunderstanding suggest that the foreign language teaching programs in the United States train students to strive for grammatical correctness at the cost of meaningful communication.

Introduction Little, if anything, is known about the psychological process involved in translating from one language into another. Neither psychological nor linguistic theories are adequate to cover the psycho-linguistic process involved in the transfer of form and meaning from one code to the form and meaning of another code. In practice, however, little children and not especially intelligent adults can perform this task without great effort (Harris and Sherwood, 1977). Small children who are bilingual will perform spontaneous self-

59

60 PATRICIA A. SMELCER. NIKET PATWARDHAN, ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

corrections when they translate (von Raffler-Engel 1970). More puzzling even is the human capacity to perform simultaneous interpreting. Given the large extent of bilingual populations all over the world, given the ease with which bilingual children can and do translate, and given the common practice of engaging translators and interpreters in our society, it comes as a surprise that most of thedata from empirical research on the subject are incompatible, and that theories that have been formulated so far are conflicting. Some of the problems have been identified but no large scale experiment has been conducted on any single issue. Given the diversity of approaches the cumulative effect is nil. The literature on the subject is scattered in a limited number of books and in a variety of journals in psychology, linguistics, translation theory, and foreign language education. These works fall into two groups. The first are general treatises on psycholinguistics dealing with aspects that are relevant to the study of the translation process such as memory span (Norman 1968) and Perceptual Chunking (summarized in Miller 1967). Besides these background works there are a number of empirical studies dealing specifically with translation (Barik 197 1; Goldman-Eisler 1972; Gerver 1974). As our experiment concerns only simultaneous translation, works dealing exclusively with consecutive translation were not included in the survey.

Research Goal Several investigators have noted that interpreters state that they often do not know what they are saying while they are saying it. Spilka (1979) recently pointed this out by means of an anecdote about one of her students in translation who after interpreting a text about bovine and aviary tuberculosis asked whether the two scientists mentioned, Bovine and Aviaire, were Russians. Earlier, Welford (1968), who had noted the same phenomenon, suggested that interpreters were able to acquire the skill of speaking and listening at the same time, because with practice they learned to ignore the feedback of their own voices. The same phenomenon led Gerver (1974) to the opposite conclusion that self-corrections demonstrate that interpreters do monitor their own output, for if they were not monitoring they would not know an error had been made. Gerver believes that the monitoring procedure used by interpreters is similar to the TOTE or “Test-Operate-Test-Exit” model proposed by Miller et al. (1960: 27-38). According to Gerver (1974: 166) “The interpreter generates a target language response, which passes a first test and is uttered, the utterance is then tested again. If the second test is passed, the interpreter proceeds to the next item, if not he ‘operates’ again by generating a further response, and so on.” The goal of our research was to discover to what extent and to what degree people were conscious of what they are doing when they perform a simple translation. We tried to approach this subject from the standpoint of linguistic consciousness. This concept was originally formulated by Leopold (1952: 111,67) in the context of child language, and looks at people’s overt consciousness of their communicative performance. While Bloomfield’s (1944) secondary and tertiary responses express social attitudes towards language, linguisticconsciousness manifests the individual’s awareness of his own language. Research into linguistic consciousness is basically different from research into more covert aspects of psycho-linguistic awareness such as can be discovered through the study of the Tip of the Tongue Phenomenon, Error Analysis, Pauses, and Perceptual Chunking. To discover overt consciousness in translating we researched self-corrections. Self-corrections have been analyzed by Gerver (1974) in the course of his work on the effect of noise on simultaneous interpretation. This, to our knowledge, is the only analysis available on self- corrections in translations. The earliest mention of the presence of self-correction was made in 1957 by Paneth in her Master’s thesis. This author, however, did not perform any actual research on self-correction.

AWARENESS OF THE TRANSLATION PROCESS 61

Research design

Subjects We selected a total of thirty (30) subjects all of whom were untrained in simultaneous interpretation and none of whom was a professional translator. Fifteen (15) of the subjects were men and fiteeen (15) were women, ranging in age between twenty (20) and fifty-four (54) years. Twenty-seven (27) of the subjects were undergraduate and graduate students at Vanderbilt University and three (3) were professors of French at this same university. Among the students twenty-five (25) were native speakers of American English and two (2) were foreign students and native speakers of French. Among the professors two (2) were native speakers of American English and one (1) was a native speaker of French from Belgium. None of the subjects was a native bilingual of French and English, all had acquired the other language as a learned language. Each subject filled out a questionnaire and was then classified according to his self-rated knowledge of French into beginners, intermediate, and fluent speakers. Each subject was tested alone. After examining the results of their translation from English into French it became apparent that subjects tended to rate their knowledge of French more positively than would appear from their actual performance. Whether this discrepancy is due to an actual illusion in favour of their self-image or whether for many subjects the unfamiliarity of simultaneous translation and the strain of being recorded created an extreme hardship cannot be ascertained at this time. On the basis of their performance subjects were divided into the following four categories: Category A: The eight (8) members of this group gave a relatively fluent interpretation

and included most of the details of the source text into their version. Category B: The two (2) members of this group had false starts and omissions but their

interpretations were complete enough in detail and sufficiently connected so that they sounded like stories.

Category C: The seven (7) members of this category gave a very hesitant interpretation. Their versions did not qualify as stories in that they consisted of disconnected, individual sentences, either complete or incomplete.

Category D: The thirteen (13) members of this category fall into one of three sub- groups: Sub-group 1 consists of participants who did not give interpretations, i.e. said nothing or very few words. Sub-group 2 consists of subjects whose interpretations were incomprehensible. Sub-group 3 consists of subjects who, regardless of the quality of their interpretations, displayed belligerent behaviour in that they listened to the tape before attempting to interpret. This extra hearing of the story introduced an uncontrolled variable into the experiment.

From the original pool of thirty (30) subjects we were able to select a total of seventeen (17) subjects, nine (9) females and eight (8) males, between the ages of twenty (20) and fifty (50), including the three (3) native speakers of French. The selected subjects were then grouped into the three categories of:

A. fluent translators (n. 8) B. intermediate (n. 2) C. beginners (n. 7)

Subjects from Category D were all discarded. Table 1 (see Appendix 1) shows the relationship between how subjects rated themselves as to fluency in French and how they actually performed. For “Category A” 50% (4) of the subjects rated their spoken French as “excellent”; 37.5% (3) as “near excellent”; and

62 PATRICIA A. SMELCER, NIKET PATWARDHAN, ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

12.5% (1) as average. (It should be noted that “Category A” had two native French speakers, one of whom rated his English as excellent and the other as near excellent.) 50% (1) of the members of “Category B” rated their French as excellent and 50% (1) as near excellent. For “Category C” 28.57% (2) of the subjects rated their French as excellent; 28.57% (2) as near excellent; and 42.85% (3) as average. (One member of “Category C” was a native French speaker and she rated her English as below average.) For the purposes of our analysis the self-rated categories had to be abandoned and objective categories were established based on actual performance.

Testing procedure Upon completing the questionnaire, subjects were videotaped while attempting a simultaneous translation of a variant of James Thurber’s version of the familiar story of Little Red Riding Hood (see Appendix 2) which had been previously recorded on audio tape and which subjects heard through headphones while speaking into a microphone. The test was administered by the first author, a senior in the Linguistics Program at Vanderbilt University, female and in her early twenties. Each subject was tested alone at the Vanderbilt University Learning Resource Center and was instructed to translate the English story into French as he listened to it, without waiting for the end of the whole story or for any pauses within the story. As the experimenter had previously prepared the story recorded on tape, the voice of the experimenter and the voice heard over the earphones were the same. The total duration of the story was 1: 14 minutes and it was read at the speed of 164 words per minute. The speed of reading on the upper limit of efficiency (between 150-200 words per minute, according to Seleskovitch, 1965) made it difficult for subject to silently monitor self-corrections before speaking out. We thus hoped to hear most, if not all, self-corrections. As soon as subjects had finished their translation, the experimenter asked them various questions concerning the content of the story; and unobtrusively in the midst of these questions the subjects were also asked about how they actually performed the act of translating (see Appendix 3). The interview was also video-taped.

Analysis The translations and the interviews were transcribed. The transcripts of the translations were marked for selected paralinguistic features (pauses, rise in intonation, increase in tempo) according to an ad hoc notational system. On the translation transcript we isolated all lexical items that were not uttered in French (some English words) and all self- corrections. Only one instance of a self-correction from English to French was encountered (“say” le petit fiile - “dit” le petit fille). By matching the pertinent information about subjects’ self-perception of what they had done with what they actually did we were able to establish a set of parameters of linguistic awareness. Of consistent interest to researchers in translation strategies on the covert level is the segmenteation of the continuous flow of the source language and the possible utilization of pauses in the source language (Goldman-Eisler 1972; Barik 1973). A survey of the earliest studies on EVS or ear-voice span, between 1957 and 1974, is available in Gerver (1976). As the statistics from preliminary research on the use of pauses in our sample show random distribution, we realize that our research is open to the question of individual variation and we see the desirability, if not the necessity, for a large sample population. Given the limited number of subjects in our present experiment we cannot rule out individual variation which

AWARENESS OF THE TRANSLATION PROCESS 63

may have biased our results. The latter is inherent in most pilot projects and will be overcome once we proceed with the main study. In this study we concentrated on self-perception of the translation process as described by subjects during the interview, and on self-corrections as they were performed by subjects during the actual translation. As said above, subjects tended to overrate their competence in the target language. We did not research this aspect because we would have needed extensive work to isolate the compounded variables. More importantly, our interest lay in examining linguistic awareness during the actual performance of the translation process rather than in subjects’ self- perception of their general capacities in the target language. When subjects were asked whether they began translating at the first word or at the first pause (see Appendix 3 question number 5) their answers were frequently at variance with the actual data (see Table 2, Appendix 4). From a linguistic standpoint the most interesting discrepancy was that subjects thought they had started at the first pause of the source language when they actually started at the last word before the first pause (“wolf”). What they unconsciously reacted to was the intonation contour which signalled to them that a pause was coming. Not being trained in linguistics all they knew on the conscious level was the presence of pauses in discourse. Fundamentally, there are two means for testing linguistic consciousness. One as distinct from other features of linguistic awareness is the avoidance phenomenon when speakers may try to avoid an embarassing item, substitute synonyms for words containing a difficult sound (von Raffler-Engel 1965). This type of manifestation of linguistic consciousness was not feasible to test in our project. The second means for testing overt linguistic consciousness is the analysis of spontaneous self-corrections (Gerver 1974). From the literature that we were able to consult, all researchers in translation utilized, either entirely or in part, subjects that were professional translators or students in training, with the exception of Lawson (1967) and Treisman (1965). Lawson’s and Treisman’s objectives were, however, different from ours. They were concerned with selective attention span and their experimental procedures were also different from ours. Treisman looked at ear-voice span and Lawson experimented with dichotomous listening. For lack of time and availability we were unable to consult works written in languages other than French and English. We were also unable to read a number of Ph.D and M.A. dissertations and unpublished research reports. This literature will be covered before we initiate the main project. In our sample, subjects’ self-corrections were divided into three categories (see Table 3, Appendix 5).

A. Corrections of content B. Corrections of grammar C. Corrections of style

A correction in style is when the translator adjusts to an equally correct stylistic variant, such original, like when he corrects “le petit loup”, the little wolf, to “le gros loup”, the big wolf. A correction in style is when the translator adjusts to an equally correct stylistic variant, such as the use of the noun instead of the pronoun, “Lorsqu’elle” becomes “lorsque la petite fille”. For the individual groups and for the groups as a whole the largest distribution fell into the grammatical category. Among a total of twenty-seven (27) self-corrections, fifteen (15) (55.56%) were grammatical. Fourteen of these fifteen (93.33%) dealt with adjustments of the masculine and feminine gender of French nouns (either incorrect to correct or vice versa) expressed in definite articles, indefinite articles, and possessive pronouns.

64 PATRICIA A. SMELCER. NIKET PATWARDHAN, ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

Conclusion Theoretical Linguistics All those who have done research in self-corrections have concluded that interpreters must be aware of what they say, given that they monitor their own voices and make the necessary adjustments. Considering the observation reported by Spilka (1979) that interpreters perform like typists, who at the end of a perfect job do not recall the content of what they have covered, a conflict arises between the conscious performance of the moment and the impossibility of subsequent recall. One of our subjects who did a creditable translation could not remember whether the voice on the tape was a male or a female voice. From our pilot study we are inclined to conclude that translating is governed by short-term memory. This would adequately explain monitoring and self-correcting. Neither the original text in the source language nor the translation in the target language may reach long- term memory. More research into long- and short-term memory storage in translation and interpretation is warranted. That one other aspect of linguistic consciousness needs further investigation became apparent from the fact that subjects perceived that they were listening to the pauses rather than the intonation contour preceding the pauses, which is what they actually reacted to. This problem ties in with research on Perceptual Chunking but our findings point to a somewhat dissimilar interpretation. Students and teachers of foreign languages, and other persons not trained in linguistics, have rarely, if ever, heard of intonation contours. Linguistic consciousness, like all other manifestations of conscious awareness, is largely a function of learned categorization. Unconsciously, subjects reacted to intonational cues which were interpreted as pauses on the conscious level. Unconscious strategies are perceived in the light of a conscious category.

Applied Linguistics The present research, by showing that valuable time was spent on minor grammatical points that cause no misunderstanding of the content, in our opinion, can be attributed to the methods of teaching foreign languages in the college classroom in the United States. Considering the weight of errors of grammar in the grading of foreign language tests and considering the fact that teachers will consistently interrupt a student whenever he makes such a mistake, without distinction of whether the mistake is causing a breakdown in the flow of information or not, appears to condition the student to attribute greater importance to grammatical correctness than to the adequate coverage of the content within the allotted time span. Our finding, we hope, may be of some practical value in foreign language teaching.

AWARENESS OF THE TRANSLATION PROCESS 65

PARTICIPANT

MZ

JR

DC

AS

PB

PS

AJ

JLB

RS

JE

JAM

AB

SM

JJ

JL

GM

ED

Appendix 1

Table 1

1 SELF-RATED FLUENCY CATEGORY

BY 3 =

PERFORMANCE 8 65

A VE

B

66 PATRICIA A. SMELCER, NIKET PATWARDHAN. ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

Appendix 2

1. The little girl and the wolf

2. One afternoon a big wolf waited in a dark forest for a little girl to come along carrying a basket of food to her grandmother.

3. Finally a little girl did come along and she was carrying a basket of food.

4. “Are you carrying that basket of food to your grandmother?” asked the wolf.

5. The little girl said “Yes, I am”.

6. So the wolf asked her where her grandmother lived and the little girl told him and then he disappeared into the wood.

7. When the little girl opened the door to her grandmother’s house she saw that there was somebody in bed with a nightcap and a nightgown on.

8. She had approached no nearer than twenty-five feet from the bed when she saw that it was not her grandmother but the wolf, for even in a nightcap a wolf looks no more like your grandmother than Georgie Jesse1 looks like the Metro-Goldwyn lion.

9. So the little girl took an automatic out of her basket and shot the wolf dead.

10. The moral of the story:

11. It’s not as easy to fool little girls nowadays as it used to be.

Appendix 3

STANDARD QUESTIONS

1. What type of work have you done in translating or interpreting?

2. Briefly, can you tell us the story of the text?

3. Is there anything funny or unusual about this story?

4. Tell us how you went about interpreting the story, i.e., did you try to do it word for word or did you try to go by units that make sense.

5. When did you begin to interpret, at the first word or the first pause?

6. Was the voice of the storyteller that of a male or a female?

7. Did the storyteller do or say anything that distracted you, that is did you notice any obliterated words, strange intonational or inflectional patterns?

8. Are you familiar with the Red Riding Hood fairy tale?

9. How much of the story that you told was from the text you heard and how much of it was from the story you heard as a child?

10. How many animals were mentioned in the story?

11. How far was the girl from the bed when she saw that the wolf was posing as her grandmother?

12. Did you remember a gun?

13. Was it more difficult to do than you thought?

14. Any additional comments?

CA

TE

GO

RY

A

C

AT

EG

OR

Y

B

CA

TE

GO

RY

C

Part

icip

ant

Self

- A

ctua

l Pa

rtic

ipan

t Se

lf-

Act

ual

Part

icip

ant

Self

- A

ctua

l Pe

rcei

ved

Star

t Pe

rcei

ved

Star

t Pe

rcei

ved

star

t St

art

Star

t St

art

MZ

1s

t wor

d 1s

t wor

d R

S m

id-t

itle

1st p

ause

JA

M

1st p

ause

m

id-

sent

ence

JR

1st w

ord

1st p

ause

JE

1s

t pau

se

1st p

ause

A

B

1st w

ord

wol

f

DC

1s

t pau

se

wol

f*

SM

1s

t pau

se

wol

f

AS

1 st

paus

e 1s

t pau

se

JJ

1st w

ord

1st p

ause

PB

1 st w

ord

wol

f JL

1s

t pau

se

1st p

ause

PS

neith

er

1st p

ause

G

M

1st p

ause

w

olf

AJ

1st w

ord

wol

f E

D

1st w

ord

wol

f

JLB

-

1st p

ause

Perc

enta

ge

25%

Pe

rcen

tage

50

%

Perc

enta

ge

Cor

rect

C

orre

ct

Cor

rect

14

.28’

70

*wol

f is

the

last

wor

d be

fore

th

e fi

rst

paus

e

,” ^_

.-

I

“.

68 PATRlClA A. SMELCER, NIKET PATWARDHAN, ALFRED WILSON AND WALBURGA VON RAFFLER-ENGEL

Appendix 5

Table 3

CORRECTIONS AND REVISIONS MADE WHILE INTERPRETING

PARTICIPANT CONTENT GRAMMATICAL STYLISTIC

Category A

MZ

JR

DC

a ton . . . a ta grand-mere le . . . la petite fille

de sa . . . de la maison de sa grand-mere

de la . . . du lit avec une . . dans une

AS chemise

son . . . sa grand-mere

PB

PS

AJ

JLB

Category B

la Porte de la sa Porte . . de la maison de sa grand-mere

le petit loup . . . le gros loup ce jour . . ces jours-ci

le . . . un loup

la . . . les petites filles un . . une petite fille - - -

5 (41.67%) 6 (50.00%) 1 (8.33%)

RS

JE

de sa . . . la maison de sa grand-mere

le petit . . . la petite fille la . . . le loup un . . une corbeille

son grand-mere . . . sa grand-mere

lorsqu’elle a . lorsque la petite fille

Category C

1 (16.67%) 4 (66.67%) I (16.67%)

JAM le jeune . . . la jeune fille la le . . . la leqon dans le . . dans la for&

AB le loup se . . _ te demandait quand elle voit . . . quand elle a vu

SM le . . . la petite Bile une . . . un corbeille

JJ - - -

JL les jeu (nes) . . . les petites jeunes filles

GM - - -

ED si dif (facile) . . faciie

3 (33.33%) 5 (55.56%) 1(11.11%)

Cumulative Total

9 (33.33%) I5 (55.56%) 3 (11.11%)

AWARENESS OF THE TRANSLATION PROCESS 69

REFERENCES

BARIK, Henri 1971 “A Description of Various Types of Omissions, Additions and Errors of Translation Encountered in Simultaneous Interpretation” in Meta vol. 16, no. 4, p. 199-210. (Abstracted in LLBA 1973, vol. 7, No. 3,7303494).

BARIK, Henri 1973 “Simultaneous Interpretation: Temporal and Quantitative Data” in Language and Speech vol. 16, part 3, p. 237-270.

BLOOMFIELD, Leonard 1944 “Secondary and Tertiary Responses to Language” in Language vol. 20, p. 4445.

BROADBENT, D. E. 1952 “Speaking and Listening Simultaneously” in Journalof Experi- mental Psychology vol. 43, p. 267-273.

BROS-BRANN, Eliane 1975 “Critical Comments on H. C. Barik’s Article ‘Interpreters Talk A Lot, Among Other Things”‘, in Babel vol. 21, no. 2, p. 93-94.

CAREY, Peter 1971 “Verbal Retention After Shadowing and After Listening” in Perception and Psychophysics vol. 9, no. lB, p. 79-83.

DHANRYPONT, L. 1962 “Lettre a un interprtte” in Babel vol. 8, no. 1, p. 23-28.

GERVER, David 1974 “The Effects of Noise on the Performance of Simultaneous Inter- preters: Accuracy of Performance” in Acta Psychofogica vol. 38, no. 3, p. 159-167.

GERVER, David 1974a “Simultaneous Listening and Speaking and Retention of Prose” in The Quarterly Journal of Experimental Psychology vol. 26, part 3, p. 337-341. (Abstracted in LLBA 1974, vol. 8, no. 4,7404869).

GERVER David 1976 “Empirical Studies of Simultaneous Interpretation: A Review and a Model” in Translations Applications and Research ed. by Richard Brislin, New York: Gardner Press, Inc., p. 165-207.

GOLD, David 1976 “On Quality in Interpretation: II” in Babelvol. 22, no. 3, p. 117-l 19.

GOLDMAN-EISLER, Frieda 1972 “Segmentation of Input in Simultaneous Translation” in Journal of Psycholinguistic Research vol. 1, no. 2, p. 127-140.

HARRIS, Brian and Sherwood, Bianca 1977 “Translation as an Innate Skill”, paper pre- sented at the NATO Symposium on Language, Interpretation and Communication, Venice: Giorgio Cini Foundation. To appear in the proceedings, New York: Plenum Press.

KADE, Otto and Cartellieri, Claus 1971 “Some Methodological Aspects of Simul- taneous Interpreting” in Babel vol. 17, no. 2, p. 12-16.

LAWSON, Everdina 1967 “Attention and Simultaneous Translation” in Language and Speech vol. 10, p. 29-35. (Abstracted in LLBA 1967, vol. 1, no. 3, p. 494).

LEOPOLD, Werner F. 1952 Bibliography of Child Language, Evanston, Illinois: Northwestern Press.

MARQUIS, M. 1971 “Le Courrier des Lecteurs” in Meta vol. 17, no. 3, p. 177.

MILLER, George, Galanter, Eugene and Pribram, Karl H. 1960 Pfans and the Structure of Behavior, New York: Henry Holt & Co.

MILLER, G. A. 1967 “Psychological Approaches to the Study of Communication” in Journeys in Science ed. by D. L. Arm, Albuquerque: The University of New Mexico Press, p. 22-73.

70 PATRICIA A. SMELCER, NIKET PATWARDHAN, ALFRED WlLSON AND WALBURGA VON RAFFLER-ENGEL

NORMAN, Donald 1968 “Toward a Theory of Memory and Attention” in Psychological Review vol. 75, no. 6, p. 522-536.

PANETH, E. 1957 An Investigation into Conference Interpreting. Unpublished Master’s thesis. London University. (Mentioned in Gerver 1976, p. 187).

SELESKOVITCH, Danica 1965 Colloque sur I’enseignement de l’interpretation, Paris: Association internationale des interpretes de conference (discussed in Gerver 1976, p. 172).

SPILKA, Irene 1979 On the Translation Process. Lecture delivered at the Nashville Linguistic Circle. Unpublished video tape. Nashville, Tennessee: Vanderbilt University.

TREISMAN, A. 1965 “The effects of redundancy and familiarity on translating and repeating back a foreign and a native language” in British Journal of Phsychology vol. 56, p. 369-379. (Mentioned in Gerver 1976, p. 169).

UMANSKY, J. 1962 “Les interpretes de conference et leur marche” in Babelvol. 8, no. 1, p. 31-32.

VON RAFFLER-ENGEL, Walburga 1965 “Un esempio di ‘linguistic consciousness’ nel bambino Piccolo” in Orientamenti pedagogici vol. 13, no. 3, p. 63 l-633. English translation in Studies of Child Language Development ed. by C. A. Ferguson and D. I. Slobin, New York: Holt, Rinehart and Winston, 1973, p. 155-158.

VON RAFFLER-ENGEL, Walburga 1970 “The Concept of Sets in a Bilingual Child” in Proceedings of the Tenth International Congress of Linguists, Bucharest, 1967. Bucharest: Rumanian National Academy, p. 18 l-l 84.

WELFORD, A. I. 1968 The Fundamentals of Skill, London: Methuen. (Mentioned in Gerver 1976, p. 187).