Innovative Approaches in the Training of Lexicographers

The following questions are addressed in this paper: (a) What should the nature of the pre-training of the prospective lexicographer be in order to give him or her the necessary basis to be successfully trained as a practising lexicographer? (b) What qualities should the prospective lexicographer have if he or she wants to succeed as a

268 D.J. van Schalkwyk (W AT 1993), it is clear to me that there is still a big difference of opinion regarding the pre-training of the lexicographer.
Some of these Editors-in-Chief and managers claim that experience has shown them that formal tertiary education is not important for the lexicographer.Others argue that this pre-training should not be in linguistics.
The opposition of some to linguists stretches further than the training phase.One of the people I talked to said: "I do not think contact with linguists is of any real importance to the lexicographer" (W A T 1991: 4).Another made an even stronger claim: "The existence of a division between the linguist and the lexicographer is healthy" (WAT 1991: 37).
These are not isolated opinions.Landau (1984: 236) said the following regarding lexicographers: "They need not be linguists; in fact, knowledge of linguistics is of no particular help in dictionary defining".He then quotes Patrick Hanks who said: "I have not found that people with a sound theoretical understanding of current linguistic theory make good definition writers".Hanks motivates this claim by saying: "It is a literary, not a scientific activity".Landau (1984: 236) also refers to the claim by Bosley Woolf that "graduate study, while it may give the new definer an initial advantage, does not ultimately make him more fit for the job than 'last year's college graduate"'.
I assume that these claims applied primarily before metalexicography became part of linguistics courses at universities, or before the effect of this training became apparent in lexicography.On the other hand, one should note that training in metalexicography still is not commonly provided at universities.
We are lucky in South Africa, because metalexicography is taught at at least five universities that I know of, viz. at the Rand Afrikaans University, UNISA, and the Universities of Pretoria, Stellenbosch and Western Cape.
At some of these universities the training started in the seventies, but only really gained momentum in the eighties and nineties.
The pre-training of the prospective lexicographer is very important at the Bureau of the WAT.On co-editorial level the entrance requirement is a master's degree in Afrikaans or Afrikaans and Dutch, with a specialization in linguistics or literature, or a master's degree in general linguistics, communication or journalism, with Afrikaans or Afrikaans and Dutch on at least a third year level.

2.
Qualities of the prospective lexicographer In answer to the question which qualities a good definer should have, Landau (1984: 235) mentions four.He says: (i) "He must be able to write well and easily".
(ii) "He must have an analytical mind".(iii) "(He] must have a broad, but not necessarily deep, fund of information", and (iv) "(He] must have a feeling for the language, Sprachgefuhl, a sense of aptness of expression, an appreciation of nuance, style, and idiom."(1984: 236) also points out that the editors of a general monolingual dictionary are almost always mother tongue speakers of the language in question, but that they do not have to be male.The Bureau of the W A T agrees with the qualities which Landau stresses.However,' the Bureau added additional qualities when it set out the specifications for the "ideal incumbent" during its personnel planning (W AT 1990).

He
Regarding the personality profile of the ideal incumbent, the Bureau is convinced that he or she should have determination, must be able to work well in a team, must be goal-oriented, and have a good ability to complete tasks.
Lexicography as part of a larger lexicographic project, is by nature a team effort.Someone who cannot work in a team and who does not have the determination to complete a task in a goal-oriented manner, will never become a good lexicographer.
The only way to ensure that the right personnel are recruited to be trained as lexicographers, is by means of proper personnel planning and job specifications.The recruitment of personnel should then take place in accordance with such planning and job specifications.Proper attention should therefore be paid to the contents and wording of advertisements for vacancies.

In-service training
The editor of Newsletter (1993-94: 3) of the Dictionary Society of North America (DSNA) said the following regarding lexicographic in-service training: "Obviously an apprenticeship in a functioning dictionary office would provide valuable practical training, but I know of none that offer such opportunities routinely".Not all dictionary projects possess training programmes, not even for their own staff.In the past there was also no training program for editors at the Bureau of the WAT.It was the responsibility of each new editorial staff member to acquaint himself or herself with the editorial system up to the point where manuscript can be prepared independently.In the past it was accepted that this process took about four years.
There exists a considerable difference of opinion among different dictionary projects regarding the duration of the in-service training of a lexicographer.Landau (1984: 235) says: ''It usually takes about a year to train someone with aptitude to be a definer, and even then he cannot be trusted, as a rule, to work without supervision."But Landau (1984: 235) is not totally pessimistic, because he continues: "Of course, some learn faster ... ".He then expresses an unsettling thought.He says that if you employ ten carefully selected candidates, "one is fortunate if two develop ultimately into good definers capable of independent work".
This training process takes two years at the Oxford English Dictionary (Shenker 1989: 94), and three years at the Deutsches Worterbuch (WAT 1991: 51).
It is interesting to note that Landau does not place a high premium on proper academic training.He says (1984: 236): "Beginners (with advanced degrees, whether in linguistics or another field) often imagine that their academic training qualifies them for advanced work in lexicography and are chagrined and hurt when they discover that it doesn't".Yet he still manages to train a lexicographer in one year.This of course only applies to those who are, according to him, trainable (1984: 235).The Oxford English Dictionary prefers a candidate to have a degree.There the training takes two years (W AT 1991: 5).The Deutsches Worterbuch prefers a preliminary training period of ten semesters, while the in-service training lasts three years (W AT 1991: 51).At the Bureau of the W A T the admission requirement has always been a master's degree, and here training originally took four years.This creates the impression that the longer the preliminary training was, the longer the period of inservice training needed would be.I am, however, not convinced that this is the essence of the problem.

Strategic planning
In the past, the Bureau of the W A T was placed under great pressure, and commissions of enquiry were appointed because the work on the WAT did not progress satisfactorily (Gericke 1991: 52-54 and 66-68).In 1987 the Bureau of the W AT started to use the computer for its tasks (Harteveld 1994: 1) and in 1989 the Board of Control of the WAT instructed the Bureau to carry out strategic planning in order to find ways to speed up the work on the dictionary (WAT 1989: 1).
Ten performance areas or strategic focus areas for the Bureau were identified during the strategic planning (WAT 1989: 71).These are areas in which the Bureau need to excel if it wanted to fulfil its task.Each of these performance areas were then planned to the finest detail.Personnel planning (W A T 1990) was one of these areas.This planning led to a new organizational and personnel structure for the Bureau.
In addition to the goals and job specifications, a set of production norms were developed for all the positions in the Bureau (WAT 1990: Appendix E-Appendix J).

Production norms for lexicographers
During my contact with other lexicographic projects it became apparent that several projects had no production norms.In cases where production norms were present, they were aimed primarily at input and not also at output.Furthermore, it was apparent that these norms differed from each other to a great extent.
At the Woordenboek der Nederlandsche Taal, each editor is required to process apprOximately 7 600 citations per annum.These citations are already checked (WAT 1991: 31).At the Deutsches Worterbuch every editor has to process 15 000 unchecked citations (WAT 1991: 51).For various reasons, it is very difficult to determine whether these two norms make more or less the same demands on lexicographers.On the other hand, it must also be kept in mind that the same norms cannot apply to all lexicographic projects.
For some or other inexplicable reason there exists the common perception that the output of a lexicographer can not be measured.The following was said in the subsidy formula which the erstwhile Department of National Education developed for the Woordeboek van die Afrikaanse Taal and for The Dictionary of South African English: "In order to encourage increased productivity, a subsidy formula should ideally be based upon outputs.However, the outputs of dictionary units ... are difficult if not impossible to measure" (NATED 1988: 3).
We at the Bureau of the W AT argued that it was impossible to increase productivity and consequently accelerate the rate of the lexicographic process if the output of the editors was not measurable.Therefore a method had to be found to measure output against time.Four matters are brought into play regarding a quality norm, viz.

(i)
the number of citations processed, the number of definitions formulated, (iii) the nature of the semantic structure of the lemmas treated, and (iv) the number of computer printout pages of manuscript produced.
Senior staff members evaluate the quality of manuscript produced.The time the editor takes to process all the commentary of the senior staff members, form an important aspect of performance evaluation.It is important to note that when we speak of the number of computer printout pages produced in a certain time, we mean the number of approved pages of manuscript.This ensures its quality.

The necessity of in-service training
The Pero,mnes evaluation technique was used for the job evaluation during the personnel planning of the Bureau.From this it was clear that the measure of difficulty of produ(;:ing manuscript was very high.Everyone who has attempted the defining task of the lexicographer knows the high demands it sets.
It should be kept in mind that training in metalexicography is not a substitute for lexicographic in-service training, because lexicographic practice cannot be simulated in the lecture hall.Nor should it be.
In an organization where it is expected from employees to meet production'norms, in-service training is a necessity.Employees should not be placed in a position by production norms where they will necessarily fail.This would constitute an unfair labour practice.Furthermore, it would also discredit the production norms.
The introduction of production norms and the establishment of an in-service training program should go hand in hand so that employees will be placed in a position to meet production norms.In this way all employees can contribute to the realization of the goals of the institution.

Contents of the in-service training of the Bureau
The in-service training program of the Bureau consists of three courses, viz. a course in general lexicography, which is presented on an introductory and advanced level; a course in computerized lexicography and a course in the planning and management of a lexicographic project.All these courses are presented in Afrikaans and English.

General lexicography
The following themes are handled in the course on general lexicography: ( This course was developed after various articles in the WAT were ana-lyzed with a view to the development of a computerized system for the WAT.This analysis showed what kind of information needed to be included in the macro-and microstructure of the WA T. These themes are treated in both the introductory and the advanced course in general lexicography.Only the level of treatment differs.

Computerized lexicography
In the course on computerized lexicography, the candidates are trained to use the computer hardware and software of the Bureau.Attention is paid to the theoretical development, technical structuring, compilation and USe of an electronic database, as well as practical processes such as data input, data retrieval, the preparation of manuscript and even typesetting by means of the computer.
As regards the preparation of manuscript, the course consists primarily of a course in general lexicography, together with computer use.The advanced course in computerized lexicography carries the practical application further than the beginner's course, using actual language material.

Course in planning and management
The Bureau also offers a course in the planning and management of a lexicographic project for managers of other lexicographic projects.This course is designed in such a way that it meets the needs of the project concerned.

Duration of the training
An introductory course of six weeks is conducted for editors of the Bureau.The aim of this course is to give candidates a general overview of the work of the lexicographer so that they can decide whether they are suited to the job, and to give the Bureau an opportunity to evaluate candidates.Then follows an intensive training program of six months.
Staff alternate in the training -each within his or her field of speciality in order to ensure specialist training and to prevent too much time being spent by one of the senior staff members of the Bureau.
Courses of three or six weeks are offered for outside persons.

Monitoring of the candidates
With the help of a professional guidance program, candidates are integrated in the Bureau from the moment they first report for duty.This guidance program was developed in conjunction with the post specifications of each post in the Bureau.
The objectives of the Bureau, the values that it pursues, its goals and planning in general are introduced to new colleagues.This forms part of the training, and the lexicographic training of Co-editors flows from this.
The guidance program was developed in such a manner that the progress of each candidate can be monitored continuously.In this way the trainer constantly gains an indication of the development of the technical or functional skills, the conceptual skills and the interpersonal skills of the candidate.
The evaluation of candidates takes place in a transparent and democratic manner, because the candidate takes part in it.Each evaluation is discussed in detail with the candidate, who is then given an opportunity to comment on the evaluation.The candidate evaluates his or her own progress and expresses himself or herself regarding shortcomings in his or her capabilities.These shortcomings receive attention until the trainer and the candidate are satisfied that they have been completely eliminated.
The output of the trainee lexicographer is monitored by means of inservice training courses and the guidance program until the trainers are satisfied that the pupil can produce independent work.

Lexicographic teamwork
As soon as the aspirant lexicographer is able to prepare manuscript independently, he or she is included in a team consisting of a Senior Co-editor and two other Co-editors.
The editorial team can work together in different ways.One possibility is that the whole team work together on a project.All examine the allocated material and provisionally formulate the semantic aspects of the simplex, if a simplex is involved.In such a case the Senior Co-editor continues with the lexicographic treatment of the simplex, while the Co-editors process the compounds, derivatives and specialized and idiomatic expressions.
This procedure ensures that the team members do not work in isolation.Team members support each other because all are to a greater or lesser extent acquainted with the problems of the particular section of the manuscript.The Senior Co-editor is also always available to offer guidance.
If the Senior Co-editor has no solution, the assistance of the Final Editor may be called in, or even that of the Editor-in-Chief.This is one way in which the team can work together.There are also other methods of co-operation with which the Bureau has experimented.

An integrated perfonnance management system
The team provides security to each editorial member so that the fear of failure Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2011) is reduced.The production norms give the editorial member a target to aim at and the Bureau uses an integrated performance management system to evaluate the performance of each colleague every month.This management system enables the Editor-in-Chief to determine whether his colleagues are measuring up to the set norms.It also gives the Editor-in-Chief an indication of whether the whole organization is reaching its goals in an integrated manner.
An achievement bonus is negotiated for those who perform above the norm.The Bureau is also examining a salary increase system based on achievement, because by these means it will demonstrate that it is willing to reward achievement.
If colleagues perform below the norm, the causes are determined.In some cases performance below the norm indicates that the editor is busy with a section of the manuscript which is particularly complicated, but if performance below the norm persists, retraining is usually the answer.

Results achieved with in-service training
The Bureau uses two nOrInS to evaluate the success of its internal in-service training program: (a) The output of each editorial member. (b) The time it takes to produce a volume of the WAT.
It is expected of an editorial member to produce twenty pages of approved manuscript per month.All the editors were involved in the determination of this norm, because the Bureau believes in team management.Yet there remained some doubt whether this could be achieved.The fact of the matter is that after nine months all the editorial members could satisfy this norm, and after a year, half of the editorial members were regularly producing thirty pages per month.The average production for the financial year 1994-95 was twenty-five approved pages of manuscript per person per month, with an individual best performance of thirty-six pages per month, taken over the whole year.The best achievement reached in one month was fifty-seven pages.
This increased production tempo meant that volume IX of the WA Twas completed within two years and four months.The previous volumes took between seven and eight years on average.At the moment the editorial staff are working on volume X, and all indications are there that the production time of volume IX will be improved upon.
This increased production cannot only be ascribed to the in-service training program.There are also other factors which played a role, such as the strategic planning and computerization of the Bureau.It can, however, not be disputed that the in-service training program of the Bureau contributed substantially to the acceleration of the work on the WAT.

Interest in the in-service training program
In addition to the staff of the Bureau who have been trained with the aid of this program, lecturers who teach metalexicography at universities have also been trained with it.Prof. Mini of the Xhosa Dictionary Project attended the course in planning and management as well as the beginner's course in computerized lexicography in January and February this year, and she is reporting for advanced training in computerized lexicography shortly.On the same day language practitioner students of the Pretoria Technikon are reporting for training in general lexicography and the beginner's course in computerized lexicography.
Requests for training have already been received from thirteen African countries, including Egypt, Tanzania, Kenya, Angola and Gabon.

Conclusion
The involvement of the Bureau in in-service training programs has resulted from a conviction that there is world-wide a great need for this kind of training.South Africa, with its eleven official languages, is no exception.
After repeated testing of its program, the Bureau has come to the conclusion that it yields good results.These results are obtained because the training program is well designed and need-oriented, but also because it is supported by a planned management framework.
The Bureau considers this training program to be its contribution to making a success of multilingualism in South Africa and invite those who are interested to take part in it.
In addition to the lexicographic society for Africa which was founded on 14 July 1995, the Bureau of the W A T also envisages an institute for Southern African lexicography under the jurisdiction of the Pan-South African Language Board.
It will be the task of such an institute to assist all lexicographic projects which enjoy government assistance with planning, management, computerization and the development of databases.However, this institute should also manage the in-service training of lexicographers.