Does Machine Translation have a role in language learning?

1
Does Machine Translation have a role in language
learning?
Harold Somers Centre for Computational
Linguistics UMIST, Manchester
2
Outline

• MT state of the art
• Language learners trainee translators?
• Previous suggestions for MT as CALL
• Some suggestions/open questions
• Using MT as a bad model
• The place of translation in language teaching
  can MT play a part?
• Final thoughts a bit disorganised what do you
  think?

3
1. MT the state of the art

• 50 years old the original application of
  language engineering
• FAHQT not achieved
• and no longer sought
• Stability and maturity with users who (mostly)
  understand its pros and cons
• Available for 12 major languages, often free
  (WWW)

4
1. MT the state of the art

• Experts stress the distinction between
  translation for assimilation and translation for
  dissemination
• MT adequate for former
• Text must be highly constrained for latter
• Why?
• Linguistic knowledge can be quite sophisticated
• But Bar Hillels semantic barrier (1959) still
  there real world knowledge, common sense
  reasoning, understanding

5
1. MT the state of the art

• Most MT systems translate rather literally
• structure preserving if not word-for-word
• Therefore most CALL uses of MT exploit MTs
  weaknesses rather than its strengths

6
2. Language learners as trainee translators

• To what extent is translation a legitimate
  activity for language learners?
• As an exercise?
• As a vocational activity?
• We will return to this question
• Inasmuch as language learners may become
  translators, they should be made aware of
  translation technology in all its forms
• But this is not what I understand as CALL

7
3. Previous studies

• Loffler-Laurian (1983, 1985, 1987)
• rather general comments
• need for trainee translators to be aware of
  technological advances
• role of post-editing and revision
• question of style and ladaptation du style aux
  besoins spécifiques de la communication
• MT output can be useful in reconsidering the
  traditional notions of mistake and error.

8
Corness (1985, 1988)

• uses ALPSs interactive MT system with advanced
  learners of German
• interactive translation mode user chooses
  among alternative interpretations of an ambiguous
  phrase
• e.g. a big computer user
• heightening awareness of varieties of possible
  translation due to differences in interpretation,
  or stylistic difference

9
MT as a bad model

• Typical view in 1980s
• As language learning aids they are woefully
  inadequate, but might provide a teacher with an
  interesting peg on which to hang a discussion of
  grammar, asking the students to spot the
  machines howlers and account for them. (Higgins
  and Johns, 1984)
• withholding the source text and inviting the
  student to reconstruct it from the raw
  translation. This can be quite useful for
  drawing attention to half-forgotten points of
  grammar and usage. (Ball, 1989)

10
Anderson (1995)

• bidirectional EnglishHebrew MT system
• Students manually enter sentences one by one from
  a corpus provided to them, note results, and then
  use native-speaker intuition and/or L2 reference
  works (depending on the translation direction) to
  identify and correct the errors
• If into L1, can reinforce students awareness of
  differences between the languages by showing them
  a bad translation into their own language

11
Doing it backwards (Richmond 1994)

• But if into L2, danger of showing learners
  examples of bad L2
• Overcome by providing a model translation
• Students asked to type in L1 sentence, note that
  system gets it wrong.
• Modify the L1 sentence until appropriate target
  text is obtained.
• In order to get desired output, L1 text has to be
  modified to make it more like the L2 target text!
  
• This is, of course, the reverse of normal
  student behaviour, which so often consists of
  producing incorrect French that sounds like
  English.
• No danger of reinforcing L2 errors, nor of
  introducing L1 errors
• by increasing the students awareness of the
  differences between their first language and
  the target-language, the backwards translation
  method places the emphasis on linguistic
  processes and linguistic input rather than on
  linguistic forms and output.

12
Pre-editing (Shei 2002)

• both L1 and L2 text, either the students own, or
  a given (native speakers) text
• Chinese-English
• L1 pre-editing encourages students to reflect
  on their knowledge in the target language
• Editing native quality L2 to coerce a better
  translation is controversial approach. Mixed
  reactions some say it only reveals how
  translating is not a good way to learn L2, or how
  poor students L2 grammatical competence was

13
Evaluation (Belam 2002)

• compare alternative translations both human and
  MT systems
• focus on question of wider context in which
  translations are done
• question of exposing students to L2 texts of
  varying quality
• assumes they are competent to give a relative
  judgment about L2 translation quality

14
Post-editing (Niño 2003)

• Work in progress
• Post-editing (revising) L2 MT output to develop
  L2 writing skills
• Advanced students

15
Previous studies summary

• Focus on translation
• MT as a bad model
• Danger of exposure to bad L2
• (Mainly) advanced students
• Heightens awareness of contrastive aspects

16
4. Some suggestions

• Using MT as a bad model
• Agree that showing ill-formed L2 may be
  counterproductive
• Useful to link bad model activity with
  understanding of how MT works
• Which in turn can focus attention on how
  languages differ

17
Example (1)

• On a donné le livre à Paul.
• On a dormi dans ce lit.
• One gave the book to Paul.
• One slept in this bed.

Paul was given the book. The bed was slept in.
18
Example (2)
Mon cousin est beau. Ma cousine est belle. Ma
cousine est riche. My cousin is beautiful. My
cousin is beautiful. My cousin is rich.

• My cousin is handsome. My cousin is beautiful. My
  cousin is a rich woman.

19
Translation in language teaching

• Classical GrammarTranslation method much
  derided
• But mainly for
• Dullness of grammar bit
• Choice of texts
• Note also, original model involves only
  translation L2-L1
• aim is to ensure comprehension and, perhaps, to
  improve L1 writing skills

20
More reasons not to teach translation

• It is independent of the four skills which define
  language competence (reading, writing, speaking,
  listening)
• It is radically different from these
• It takes up valuable time which could be better
  used teaching these
• It is unnatural

21
More reasons not to teach translation (cont.)

• It misleads students into believing in 11
  correspondences between languages
• It prevents students from thinking in the L2
• It produces interference
• It is a bad test of L2 skills
• It is only appropriate for trainee translators

22

• Many of these can be refuted
• Indeed translation persists as a classroom
  activity
• Both formally as an exercise
• And informally, as a quick means to explain things

23
Contemporary use of translation in language
classroom

• French thème (into L2) vs. version (into L1)
• Different uses of translation at different stages
  of learning
• In the elementary stages, translation from L1 to
  L2 may be useful as a form of control and
  consolidation of basic grammar and vocabulary.
  
• In the middle stages, translation from L2 to L1
  of words and clauses may be useful in dealing
  with errors therefore interference,
  interlanguage or unconscious translationese can
  be illuminated by back-translation.
• In the advanced stage of language teaching,
  translation form L1 to L2 and L2 to L1 is
  recognised as a fifth skill (Newmark 1991).

24
Translation into L2

• Generally not seen as something translators do
• Useful as a measure of L2 acquisition
• Its what people think linguists do
• Exercises can give some insight into
  interlanguage development
• Little or no attention paid to concept of
  interlanguage in TS community, though translators
  often produce a style which is neither L1 nor L2
  Campbell 1992
• Easier to control vocabulary and structures to be
  tested
• Easier to assess, thanks to model answers (albeit
  multiple)
• Are these pedagogically sound reasons?

25
How can MT help?

• Keep coming back to MT as bad model to
  reinforce awareness of differences
• Alternative to full MT some CAT tools
• Translation memory can provide some nice
  exercises (alternative to gap-filling)
• Interactive translation pinpoints areas of
  divergence

26
Translation memory
27
(No Transcript)
28
Translation into L1

• Seems a natural thing for learners to do
• But it develops L1 skills, not L2 skills
• Brings us back to recurrent themes
• Grammar-Translation model
• language learner as trainee translator

29
Regardless of direction

• Translation in either direction an important
  element in lexical acquisition
• Highlighting general differences, e.g. motion
  verbs in E and F
• Focussing on shared cognates and false friends
• (Anderman 1998)

30
How to conclude?

• It would seem that very many lovers of languages
  love to translate, it is a very motivating
  activity, more so perhaps than some other
  language learning activities conducted
  exclusively in the target language. This feature
  is perhaps something teachers can capitalize on.
  (Sewell 1996)

31
How to conclude?

• Anecdotal evidence is that using MT is an
  enjoyable exercise which makes a change for
  some students. The strange and often humorous
  L1 constructions produced by the students help to
  fix the correct L2 constructions in their minds.
  (Richmond 1994)

32
How to conclude?

• It would, of course, be foolish to claim that a
  study of MT should be part of the standard
  repertoire of language-learning activities.
  However, many students expressed the view that
  they have increased their cognitive knowledge of
  L2 grammar through having to enter information
  in the systems dictionaries for those students
  whose command of formal grammar is weak, the MT
  dictionaries appear to provide a stimulus for
  researching areas of basic grammatical structure
  .... (Lewis 1997)

33
Tentative (surprise?) conclusion

• MT may be a nice toy a novelty but its not
  designed as a language-teaching tool, so you
  shouldnt use it as one
• If you want to haul hay, get a tractor