Continuing Professional Development: How Does a Translator Stay Up to Date?

Professional development for translators is one of the least discussed subjects of the period after graduation. Learning is assumed to end when training ends; yet translation is a profession in which knowledge ages at different speeds in different layers.

Contents

The problem is not time but priority. If you try to keep up with every new tool and every new method, you learn none of them. The first step in a development plan is therefore to work out what is genuinely going out of date.

Below we look at how quickly the different kinds of knowledge age, at the criteria for choosing sources, at ways of setting time aside, at the feedback loop, and at the effect that tools assisted by artificial intelligence have on this picture.

Why is continuing development necessary?

The first reason is that the profession is unregulated. Translation is not a regulated profession in the European Union (2005/36/EC; Pym et al. 2012). There is no compulsory renewal mechanism; staying up to date is entirely a matter of personal initiative.

The second reason is the pace of technology. The tool ecosystem changes faster than curricula or personal habits. A set of settings learned a year ago may give way to another after a few releases.

The third reason is the structure of the market. In-house translation departments are shrinking and work is moving outside; the number of in-house translators at the European Commission has been reported to have fallen by around 50 per cent over twenty years (Strandvik 2017). A translator working outside is also responsible for their own training.

The fourth reason is the nature of specialisation. The law changes, medical terminology is updated, product names are renewed. Subject knowledge is not a stock that needs no maintenance.

The fifth reason is the blinding effect of routine. As experience grows, solutions become automatic; and an automatic solution that goes unquestioned turns into a source of error. Regular development makes the routine visible again and reduces the trap the experienced translator falls into most often, that of mistaking habit for correctness.

Telling what has gone out of date

Not all knowledge ages at the same rate. The table below separates the translator’s layers of knowledge according to how much renewal they need.

Layer of knowledge	Rate of ageing	How it is renewed
Tool version and interface	Fast	Release notes, short hands-on trials
Workflow and file formats	Medium to fast	Deliberately taking on new types of project
Subject terminology	Medium	Reading original texts in the field, termbase maintenance
Text-type norms	Slow	Comparisons with parallel texts
Linguistic and stylistic competence	Very slow	Regular reading and writing practice
Ethics and legislation	Variable	Following changes in legislation

The practical conclusion from the table is this: the greater part of the development budget should not go to the layer that ages fastest. Knowledge of the interface is updated anyway in the course of the work; the real investment should go to the layers that age at a medium rate.

The second conclusion is this: the layer that ages most slowly is the most valuable one. Linguistic and stylistic competence is the component no tool can compensate for, and the component whose loss is noticed last when it is neglected.

What to look for when choosing sources

Development resources are more than plentiful today. That abundance makes the criteria for choosing them all the more important. The first criterion is applicability: can you use what you have learned on a real job within a week?

The second criterion is the impartiality of the source. Training provided by tool vendors is useful in functional terms but does not explain the limits of the product. Independent publications and academic work fill that gap.

The third criterion is whether it involves feedback. A course that is watched is less lasting than an exercise that is done. Workshops that include practice and correction give a higher return.

The fourth criterion is the peer environment. Communities of colleagues show not how a problem is solved but which problems are widespread. That knowledge cannot be acquired by research on your own.

The fifth criterion is cost. Paid training is necessary from time to time, but starting with paid content when a foundation could be built from free resources is inefficient. First identify the gap, then choose the source.

The sixth criterion is timing. Learning a subject immediately before the job on which you will use it is the most efficient order. Knowledge learned months earlier and never applied has to be learned again when the time comes to use it.

Setting time aside: the development budget

The greatest enemy of development is workload. When work comes in, learning is the first item to be postponed, and the postponement builds up over years.

The method that works in practice is to break development into small, regular blocks. A short block set aside once a week gives a more lasting result than an intensive week set aside once a year.

The second method is to embed development in the work itself. Trying one small new thing on every project reduces the need to set aside separate time. Writing a new quality control rule, or handling a file format for the first time, are examples.

The third method is an annual self-assessment. Asking yourself once a year, layer by layer through the table above, which layer you have slipped back in ties the plan to reality.

The fourth method is setting limits. Rather than trying to learn every new tool, aiming at one or two new competences a year is more realistic. A small number of targets raises the likelihood of completing them.

The fifth method is to make development visible. When every new skill learned is recorded in a short note, a concrete list emerges at the end of the year. That list both sustains motivation and provides evidence that can be shown in conversations with clients.

Building a feedback loop

The greatest disadvantage of the freelance translator is the lack of feedback. Someone working in-house gets regular responses from a reviser; someone working alone often only hears anything when a complaint arrives.

The first solution is to record requests for correction. If every correction that comes in is classified by type, a personal error profile emerges within a few months. That profile is the most reliable input to a development plan.

The second solution is delayed reading. Reading a delivered text again a week later creates a distance from your own writing and offers a view freed from the shadow of the source text.

The third solution is peer exchange. Two translators reading each other’s texts at set intervals sets up a free and reciprocal checking mechanism. As long as confidentiality obligations are observed, this method is workable.

The fourth solution is to review the working environment. Research on ergonomics shows that an interrupted workflow and badly designed interfaces exhaust the capacity for attention (Ehrensberger-Dow and O’Brien 2015). A drop in performance does not always come from a lack of knowledge.

Staying up to date in the age of artificial intelligence

In recent years the most contested item in development plans has been tools assisted by artificial intelligence. Because these tools change so quickly, investing in learning them can look risky; not learning them creates a risk of another kind.

The distinction can be drawn here: the interface of a particular product is knowledge that ages quickly. The ability to judge where the output is reliable and where it is not, by contrast, is a lasting competence. The investment should go to the second group.

At the core of that ability to judge lies an old question: how do you spot a text that is fluent but wrong? Machine translation output creates risk far less through obvious breakdowns than through quiet shifts of meaning, and seeing that risk requires subject knowledge.

The second lasting heading is data security. Knowing which text may enter which environment is an area of decision that is independent of any product. It has been discussed before that the terms of use of free services can include broad licensing rights (Novović 2017, p. 202).

The third heading is the definition of the job. Post-editing is a different job from translation and is priced differently. Defining the level of editing at the outset protects the expectations of translator and client alike. Being able to make that definition is itself a development topic.

The Türkiye context

In Türkiye, continuing professional development is supported by workshops run through translators’ associations and the continuing education units of universities. Participation is voluntary; there is no compulsory renewal system. That leaves the planning entirely to the individual.

The practice of sworn translation adds a distinctive layer to this picture. Authority obtained by swearing an oath before a notary is not tied to any time-limited professional development requirement. The permanence of that authority turns the responsibility for staying up to date into a personal choice.

The position of Turkish also shapes development priorities. Turkish has less parallel data than the major language pairs. The ability to evaluate machine translation output is therefore a priority development topic for a translator working with Turkish.

On the legislative side, KVKK is a topic that has to be followed regularly. How long client data is retained and the environment in which it is processed have become standard clauses in contracts. This article is for general information and is not legal advice.

What does this mean for you?

Your supplier’s habits of development are reflected directly onto you. A team that does not stay up to date turns work away when faced with new file formats, cannot evaluate machine translation output and cannot give clear answers to questions about data security.

You can use the following steps to test their maturity:

1. Ask which new type of work or file format they have taken on in the past year.
2. Find out whether your requests for correction are recorded and carried over to later projects.
3. If machine translation is used, ask for the level of post-editing to be defined in writing.
4. Ask where your data is processed and how long it is kept.
5. Confirm that proofreading within the team is done by a separate person.

These five points let you predict not only your supplier’s performance today but their performance in the future. We have summarised how our process is organised on the services page; if you would like to talk about what kind of structure would suit your own project, you can use the quotation form.

Frequently asked questions

How much time should a translator set aside for professional development?

There is no fixed proportion. The method that works in practice is to break development into small, regular blocks set aside in the calendar; in busy periods development is the first item to be sacrificed.

Which knowledge ages fastest?

Tool versions and knowledge of the interface age fastest. Subject terminology changes at a medium rate. Knowledge of language and of text types changes far more slowly.

Is it necessary to obtain a certificate?

It is not compulsory. Translation is not a regulated profession in the European Union (2005/36/EC). A certificate can open doors, but a portfolio and experience in the field are usually more decisive.

What should I do if I cannot get feedback?

You can set up checks of your own: reading a text you have delivered again a week later, recording and classifying requests for correction, and peer exchange all help.

Sources

• Pym, A. et al. (2012). The Status of the Translation Profession in the European Union. European Commission, pp. 30-45.
• Strandvik, I. (2017). Evaluation of Outsourced Translations. Berlin: Language Science Press.
• Ehrensberger-Dow, M. and O’Brien, S. (2015). Ergonomics of the Translation Workplace. Translation Spaces, 4(1), pp. 98-118.
• Pietrzak, P. and Kornacki, M. (2021). Using CAT Tools in Freelance Translation. London: Routledge Focus, Chapter 6.