Kernerman Dictionary News • Number 10 • July 2002


Translation, the Key or the Equivalent?
a study of the dictionary use strategies of Finnish senior secondary school students

Seppo Raudaskoski

1. Introduction
It has long been the desire of lexicographers and publishers alike to find out as much as they can about the needs, wishes and skills of dictionary users, so that they can customize their products accordingly. To this end, numerous questionnaire and test studies have been conducted over the past few decades. The study under discussion was based on a test aimed to compare how well Finnish senior secondary school students make use of the information available to them in the representatives of two dictionary archetypes, the bilingual dictionary (represented in the test by English-Finnish General Dictionary and Finnish-English General Dictionary) and the bilingualized dictionary (represented by Englannin opiskelijan sanakirja, a bilingualized version of the Collins Cobuild New Student's Dictionary). The author was part of the editorial team responsible for the bilingualization of the Cobuild dictionary. A similar type of bilingualized dictionary was first published in Finland as part of Kernerman Semi-Bilingual Dictionaries in 1993 (Password English Dictionary for Speakers of Finnish).

2. Different dictionary types

It is well documented that the EFL dictionary market is still characterized by a rigid dichotomy, that is, the battle between the monolingual and the bilingual dictionary. The monolingual dictionary is favored by language teachers, who feel that monolinguals contain more information about the foreign language (L2) than bilinguals (see, for example, Atkins 1985). More importantly, monolinguals present their L2 information in L2. With their definitions and examples, they make every dictionary search a useful experience in more ways than the one perhaps originally intended; besides pinpointing the meaning of a headword, the user finds out about its collocations, learns how to paraphrase it, and receives several good examples of how to use it in a sentence. In addition, the user learns to think in L2 instead of relating every new word he or she comes across to his or her own mother tongue (L1). The drawback of monolinguals is that they are often difficult to use for a beginner. With their L2 definitions, grammar codes and lengthy entries, they may leave the user confused and unsatisfied, but their main problem is that they are inherently circular; the L2 definitions may send the user searching all over the dictionary for the meanings of the words contained by the definition. What is more, if the user wants to express something in L2 but does not know the necessary words, he/she is unable to start searching from among the L2 headwords of the monolingual dictionary.

Bilingual dictionaries, on the other hand, are reviled by EFL teachers because they help students maintain a "translation barrier": by concentrating on isolated headwords and their equivalents, they keep up the students' habit of relating every new word they learn to their L1. The listing of equivalents is particularly harmful because of the anisomorphic nature of languages (Zgusta 1971). All languages have a unique way of naming and organising reality, which means that full equivalence is, in fact, quite rare outside of terminology. Neat juxtaposition of headwords and equivalents may keep the student under the illusion that there is always full equivalence between the lexemes of two different languages and, what is worse, that the equivalent can be inserted to all contexts the student might come across. The illusion is made all the more dangerous by the fact that bilinguals rarely provide enough information on how to use the headwords or the equivalents in an actual textual environment. Bilinguals are nevertheless easier to use than monolinguals and they provide instant answers. For these reasons, bilingual dictionaries are the popular choice among students, especially in the beginner and intermediate levels.

The bilingualized dictionary is, of course, the supposedly happy marriage of the two above-mentioned paradigms. It contains the L2 definitions and examples of the monolingual dictionary and the easy-to-use L1 equivalents of the bilingual dictionary. (This type of dictionary is often based on an existing monolingual learner's dictionary.) The emphasis in the entries is on the L2 material, and for this reason the equivalents are often called 'keys', as they are rather aids for understanding than stand-alone translations of the headword. The user is supposed to turn to the definitions and examples first, and if the meaning of the headword still remains somewhat unclear, the key is there to provide clarification and reassurance (cf. Reif 1987). If the bilingualized dictionary is equipped with an index of all the keys used, the user also has handles by which to access the L2 headwords when in need of an L1-L2 translation. In short, the bilingualized dictionary can be seen as an all-in-one solution to the needs of a learner's dictionary user.

The bilingualized paradigm, however, does not escape all criticism. The concept of the key is slightly problematic, as the key should be a competent L1 translation, but simultaneously draw as little attention to itself as possible. There is a danger that the user may skip definitions and examples altogether and only pick up the instant translation proffered by the key (Nakamoto 1995). Furthermore, the index is a double-edged sword in the hands of an inexperienced user. Since it contains only the keys used in the entries, it is by no means a representative sample of the L1. It merely puts on display the reactions of the dictionary editors to a series of L2 situations, that is, the entries of the original monolingual dictionary. At worst, the index could be used as a misleading and incomplete L1-L2 dictionary.

3. The test
The bilingualized dictionary used in the test, Englannin opiskelijan sanakirja, is aimed specifically at senior secondary school students. As its Cobuild background implies, all its definitions are in simple English consisting of complete sentences ("if you X something, you Y it"; "an X is a Y"), and its examples are culled from the Bank of English, a corpus of newspaper, literary and spoken texts. There are few symbols or abbreviations in the entries, and each headword is complemented with at least one key in Finnish. There is only one key per headword whenever possible, as it is crucial that the user not get bogged down in the Finnish part of the entry, but concentrate on the information in English instead. With 35,000 headwords, it displays only the essential vocabulary of the English language.

The bilingual dictionaries that were used, English-Finnish General Dictionary and Finnish-English General Dictionary, are much more comprehensive (90,000 and 160,000 headwords, respectively) than Englannin opiskelijan sanakirja. The information contained in them is packed very densely with the help of abbreviations, symbols, parentheses, tildes and other space-saving methods. In addition, there can be dozens of headwords in a single entry, which sometimes makes finding the necessary information a time-consuming task. The dictionaries contain some made-up examples of how to use the equivalents, but these are often short phrases lacking vital collocations.

The test described in this study was devised to determine which type of dictionary, bilingual or bilingualized, would be more helpful to a completely untutored user working in an actual textual environment. The test consisted of sixteen translation assignments, eight from English into Finnish and eight from Finnish into English. The study was decided to be conducted in the form of a test, because observation studies would have required too much time and manpower, and surveys can be a rather unreliable source of information: the subject might give answers that he/she thinks are appropriate, or he/she might misunderstand the questions. A translation test was chosen over a reading comprehension test on the grounds that in a reading comprehension test, the subjects could use guessing techniques to deduce the correct answer from the textual context. Finally, open-form assignments were chosen over a multiple-choice study so that the subjects could not reach the correct answer by way of eliminating the least plausible options. As the subjects were confronted with English source text words they did not know, or with Finnish source text words they did not know how to translate into English, they resorted to dictionaries in a natural, unforced manner. In other words, dictionary use was dictated by the situation, not the test form. More than one word was usually required in the translation, which made it possible to reach an acceptable answer in more ways than one. To avoid the pitfalls that have proved to be the undoing of many dictionary tests and surveys in the past, the work of Nesi (2000) proved to be a useful guide.

The test group comprised of twenty Finnish senior secondary school students, all of whom had at least 9 (out of 10) as their previous English module grade number. The point in choosing apt students was to prevent the test from deteriorating into a cavalcade of simple grammar mistakes, which would have undermined the original intent of testing language learners for their dictionary use skills rather than their elementary language skills. The students were divided into two groups of ten, one using the bilingualized dictionary and the other using the bilingual dictionaries. The students had had no prior guidance in the use of dictionaries apart from the exhortations of their teachers to use monolinguals and distrust bilinguals. They had 105 minutes to fill sixteen blank spots with the help of the Finnish and English source texts. The texts were fairly long, so there was little time to contemplate proper search strategies. Hopefully, this made it possible to record the students' instinctive reaction to the information on offer in the dictionaries.

The test was completed twice, once without any dictionary and once with a chance to make use of the dictionaries. In the first round, the students were asked under every blank spot whether they were satisfied with the translation themselves. In the dictionary round, two new questions were asked in addition to the satisfaction question: on what page(s) the student had found information useful for the translation, and how many searches he/she had made in the dictionary.

To illustrate, one English-Finnish blank spot in the dictionary round looked like this (the English passage requiring translation is is likely to be diluted or shelved):

...a ban on snowmobiles in the park, due to come into effect in two years' time, is likely to be diluted or shelved.

...puiston moottorikelkkakielto, jonka pitäisi astua voimaan kahden vuoden päästä, ________________________________.

page(s): ___________
no. of searches: _____
satisfied (y/n)? ______

4. The results
When the answers to the dictionary use questions were analyzed and compared to the actual translations, it was possible to "triangulate" quite reliably, whether the students had used their dictionary, where they had gone to search for information, what kind of information they had found there and what they thought of its usefulness.

The test translations were marked according to two criteria: they had to fit in to the sentence around them and they had to represent the meaning of the source text accurately, leaving nothing out. The open-endedness of the translations may have left the test scores somewhat open for debate, as there were quite a few translation proposals that were not clear-cut correct or incorrect cases; in future tests, it would be advisable to have more than one marker available in order to reach some sort of consensus in such cases. The blank spots were chosen so that the students could not simply copy a key or an equivalent from the dictionary. Instead, they were often forced to adapt the translations provided in the entry to make the translation adequate. It was important to gauge the adaptability of the students; should a user fail to do any thinking on her/his own and simply accept the key or the equivalent at face value, any extra information present in the entries, such as the definitions and examples of the bilingualized dictionary, is rendered useless.

Content analysis of the students' translations revealed that the students often used their dictionaries uncritically. A prime example would be the translation of the word seething in the context seething sulphur spring. The bilingualized dictionary offered the key kuhiseva, an adjective used to describe a place full of something that is animate. Most students went for this key, with the resulting translation, kuhiseva rikkilähde, being something of an absurdity, since a seething sulphur spring can hardly sustain much life. In another case, the text natural assets was translated word-for-ford with help of the dictionary, a strategy which resulted in a nonsense concept luonnolliset varat, "nature-like assets". The most serious problem was, however, that the students used the Finnish-English index as a dictionary of its own and seldom bothered to consult the actual dictionary after they had located an English word in the index. This resulted in errors when there was something unusual about the inflection of the word or there were several headwords to choose from, but nothing to give clues about the suitability of each word for the context at hand.

The bilingual dictionary, with its dense entries full of symbols and abbreviations, caused difficulties for many students, especially when the necessary headword or equivalent was concealed inside a long entry. Lack of entry navigation skills was a problem especially during the Finnish-English test, as the students could not simply deduce the correct English equivalent from a long string like they could when faced with Finnish equivalents.

One major source of translation errors in both groups was the students' inability to make some translations fit in with the surrounding text. A missing definite or indefinite article or a wrong case ending could make all the difference between a successful and an unsuccessful translation. In such cases, there was little any dictionary could do to help, and in some instances the test measured the language skills and translator's instincts of the students as much as their dictionary use skills.

Examination of the test scores revealed that the bilingualized dictionary users improved their performance from the first (non-dictionary) round more than the bilingual dictionary users. Due to the small size of the sample, it is impossible to make any universal statements. It can be said, however, that despite all the translation errors caused by poor use of the bilingualized dictionary and its index, the test group made better use of it than the bilingual dictionary. Regardless of the dictionary used, the English-Finnish test scores improved markedly in the dictionary round, whereas the influence of dictionaries was only marginal in the Finnish-English test. This could have been partly because the students only made roughly half the number of searches while translating from Finnish into English when compared to the English-Finnish test.
In the non-dictionary round, the students scored significantly better in the Finnish-English test than the English-Finnish test. In a way, it may have been easier to translate into L2, since students knew the exact semantic content of the source text and could attempt to paraphrase it in any number of ways. When translating from L2 into L1, however, students were stuck with the source text words they did not know; one cannot paraphrase something one does not quite grasp in the first place.

Both dictionaries eroded the students' ability to evaluate the adequateness of their translations, the effect being more pronounced in Finnish-English translations. This could be detected by comparing the adequateness of translations with the answers to questions concerning satisfaction. The comparison showed that the number of correct evaluations (correct-satisfied, incorrect-unsatisfied) decreased in the dictionary round. This is a slightly alarming trend, as finding a vaguely appropriate-feeling word during a dictionary search should not be a universal seal of approval that makes the user oblivious to all errors in his or her text.

5. Conclusion
The main lesson learned from the test results is that effective dictionary use requires some rudimentary skills and a healthy attitude towards dictionaries. The two contradictory traits of the users, that is, not bothering to find out about the proper uses of a dictionary while simultaneously accepting dictionary information as final truth, often defeat the best efforts of the lexicographer.

Before any further studies are conducted utilizing the method described here, its biggest flaw needs to be addressed: a test group of twenty is far too small to make any sweeping statements about any dictionary type. Considerably larger groups, with more than one person available to mark the translations, would be the way to go. Ideally, future studies would dovetail with courses on dictionary use; the test could be used to compare the scores attained by an untutored group with the scores of a group that has been given dictionary training. The results of such a study might well be helpful to anyone devising a short course module on dictionary use.

Atkins, S.
1985. 'Monolingual and bilingual learners' dictionaries: a comparison.' In ELT Documents 120: Dictionaries, Lexicography and Language Learning, Robert Ilson (ed.). Oxford: Pergamon Press, 15-24.
Englannin opiskelijan sanakirja. 2001. Helsinki: Otava.
English-Finnish General Dictionary. 1990. Porvoo: WSOY.
Finnish-English General Dictionary. 1984. Porvoo: WSOY.
Nakamoto, K. 1995. 'Monolingual or Bilingual, that is not the Question: the 'Bilingualised' Dictionary.' In Kernerman Dictionary News, 2.
Nesi, H. 2000. The Use and Abuse of EFL Dictionaries. How learners of English as a foreign language read and interpret dictionary entries. Tübingen: Max Niemeyer Verlag.
Password English Dictionary for Speakers of Finnish. 1995. Helsinki: WSOY.
Reif, J. A. 1987. 'The development of a dictionary concept: an English learner's dictionary and an exotic alphabet.' In The Dictionary and the Language Learner: Papers from the EURALEX Seminar at the University of Leeds, 1-3 April 1985, Anthony Cowie (ed.). Tübingen: Max Niemeyer Verlag, 146-158.
Zgusta, L. 1971. Manual of Lexicography. Praha: Academia.

About the author
Seppo Raudaskoski holds a masters degree in English Translation and Interpreting, and is a project researcher in the School of Modern Languages and Translation Studies at the University of Tampere, Finland. This article is a summary of his MA thesis.

K Dictionaries Ltd
10 Nahum Street, Tel Aviv 63503 Israel
tel: 972-3-5468102 • fax: 972-3-5468103