in Linguistica Antverpiensa, 1999, pp. 127-137)
The discourse of localization, along with variants of multilingual
information management, seamless documentation workflows
and the like, was originally developed in the realms of
software marketing but is now more generally associated
with highly adaptive and technologized translation in a
teamwork environment. Since the discourse threatens to outflank
the traditional categories of translation studies, there
are theoretical as well as labour-market reasons behind
increasing calls for localization techniques, along with
training in various electronic tools, to be incorporated
into our curricula. Yet the very ideology of localization
should require that the discourse itself be modified to
suit the teaching environment. The hype of the project managers
and the promises of ever more efficient paradises may yet
be questioned by our more humanistic university disciplines.
The discourse of
localization began in the specific field of software marketing,
probably in the United States and probably in the 1980s
(I have been unable to find reliable information for either
dimension). For most laypersons, to “localize” a programme
means just about the same thing as to “translate” it. Yet
the newer verb would seem to be spreading out as a sexy
substitute for our tired old jargon: hurried intimes insist
on calling it “L10N”; international news items are increasingly
“localized” rather than “translated”; marketing material
is likely to undergo the same revamp; indeed any text requiring
what traditional theorists once called “translation and
adaptation” is now probable prey for a localization branding.
This is happening out there, in the markets that surround
our training institutions. With a few timid exceptions,
it is not happening in here, within our training curricula
or academic debates. Or more exactly, when the term does
jump the gap--as has happened in discussion groups in Spain
in recent months--, it is either as a fancy replacement
for “adaptation” or as an open question: What’s in “localization”
that isn’t already implied by “translation”, understood
and taught in the best communicative sense?
Or, to wrap it into
one simple doubt: Is there any reason why we should take
the discourse of localization at all seriously?
Here is one possible
reason: Since at least 1984, if not for several thousand
years, translation theorists have been arguing over the
meaning of the term “translation” and its cognates. If we
can now do away with the term, we might also pull the plug
on a lot of quite stupid nominalistic debates. For example,
I cite 1984 in honour of Justa Holz-Mänttäri’s
renaming programme published as Translatorisches Handeln--translative
action--rather than Übersetzen, Translation, or any
other term consecrated in the world of training. This “translative
action” was, and remains, a vague blanket term for everything
that translators might be called upon to do, including many
wonderfully adaptive things (such as writing a text for
which there is no source text) that lay beyond any restricted
definition of “translation”. In German-language theorizing,
that tradition can be followed right down to explicit calls
for various new professions able to embrace the various
extras. When Vermeer envisages translators entering a new
profession called “the ‘intercultural management assistant’
or ‘consultant’” (1998: 62) and Austermühl et al. talk
about the translator becoming a Sprachdienstleister--language-services
provider-- or a cultural consultant--they leave the term
in English--(1998: 336) or occasionally even an “information
broker” (1999), they are using new terms basically to describe
what Gregory Shreve at Kent State more modestly names as
the “multitasking translator” (cit. Austermühl 1999).
That is, since more or less 1984, some parts of translation
theory have been recognizing that, beyond translation theory,
the market has been calling on translators to do more than
translate. Hence the various names for this “more than”.
What is at
stake here? Generous minds might suppose that translator
training has been adjusting efficiently to labour-market
demands. Ungenerous minds could venture that little academic
empires have been built as students labour to learn new
theories instead of practising the communication strategies
required. But more honestly, in terms of theory, what was
and remains at issue is really no more than a traditionalist
technicality. For the newish terms, translators can now
be called upon to add and delete without reference to source-text
constraints. For good old translation, even in its more
adaptive moments in Nida, it was legitimate neither to add
nor to take away (cf. Duet. 4:2; 12:32; Rev. 22:18-19).
Everything that was said or meant in the source had to be
somewhere in the target. Whether or not the term “translation”
can really be stretched out to include a lot of adding and
subtracting has largely been a matter of questioning this
time-honoured biblical legitimacy, historically behind the
general trust accorded to the translator as conduit.
Of the many other
things that have happened in translation theory over the
past two decades, perhaps the most important is the similarly
idealized shift from prescriptive to descriptive approaches.
When Toury (1980) started to see equivalence as something
automatically produced by all ostensible translations no
matter how bad, and when he later proposed that for the
purposes of selecting a descriptive corpus, “a ‘translation’
will be taken to be any target-language utterance which
is presented or regarded as such [i.e. as a ‘translation’],
on whatever grounds” (1985:20), he effectively challenged
pedagogically restricted definitions of “translation”, and
thereby created much trouble. Discussion and critique of
the descriptivist move is not hard to find (cf. Gutt 1991:
7; Hermans 1995; Toury 1995: 33; even Pym 1998: 58-61),
but it’s all rather ho-hum scholastic stuff. Worse, as we
can see now, the move was fatally flawed in its immediate
historical context. After all, if one of the major growth
areas in the labour market starts dropping the term “translation”--and
that’s why we have the discourse of “localization”--, then
any theory that relies on that context for its definitions
will eventually be left dancing without a partner.
problems here can become as knotty as you like. For translator
training, though, the underlying aim of the words on or
instead of “translation” remains a fairly straightforward
matter of preparing people for a changing market. From this
perspective, theorists’ acrobatics over the meanings or
non-meanings of the term “translation” have been to remarkably
little avail; they would seem not to have solved many problems.
In such circumstances, as Althusser was wont to say (in
1965, if we must), it pays to look at the way theoretical
problems are solved in practice. More exactly, it pays to
look at the theory (little-t) that is a part of quotidian
praxis, rather than the Theory (big-T) that is our own reflexive
praxis (sorry, this epistemology is so old that not even
Venuti cares to cite it anymore). And that is precisely
what I now propose to do with the discourse of localization.
I will ask, seriously, if it has solved some of the problems
that our more up-front translation theorists have reached
such little consensus on.
What is Localization?
A basic research
method is to do an Altavista internet search for the term
‘localization’ (as I am doing right now). You’ll get about...
let’s see, 160,490 webpages, since this is a fairly popular
term. The most useful of those pages, and among the first
in the list, are written by language-service companies and
consultants. I’ve picked just one sample from the pack,
largely because of its overt pedagogical virtues: it is
101. The document has been produced by a California
entity called International Consulting, which would appear
to be the brand-name of someone called Willem Stoeller.
Like many of the documents listed, it is designed to explain
how localization can help software providers market their
products internationally; it is addressed to prospective
First remark: Some
theorists, including Gouadec and various proponents of Skopostheorie,
have long been talking of the need to teach clients about
translation. Here we have a document that is actually doing
it, albeit under a different name. There might be something
strategic to learn.
Second remark: As
the name “Localization 101” suggests, the document is structured
like notes from a first-year university course, even though
it is entirely within the commercial domain. I suppose it’s
nice to know universities are still prestigious enough to
imitate, but perhaps there is something else going on here.
Perhaps, just as our theorists could learn something from
the discourse of localization, perhaps the discourse itself
requires something from the more traditional fields of education.
But let’s get to
the meat of the matter.
The secrets of localization
are recounted in two easy steps, worth citing:
Locale: Those features
of the customer’s environment that are dependent upon language,
country/region and cultural conventions.
is a way of designing and producing products that can be
easily adapted to different locales. This requires extracting
all language, country/regional and culturally dependent
elements from a product. In other words, the process of
developing an application whose feature design and code
design do not make assumptions based on a single locale,
and whose source code simplifies the creation of different
local editions of a program, is called internationalization.
then to be understood as more or less the opposite of internationalization.
But let’s consider just these two terms for a moment.
I propose, has serious virtues as a short word for all the
“language and/or culture” collocations that occasionally
litter our more pretentious or exacting theories. Further,
the very newness of “locale” usefully shrugs off attempts
to associate our target loci with whole national societies
or cultures; locales may even be quite small professional
on the other hand, would seem to offer rather fewer ideological
virtues in that, like the dream of pure meaning without
language, it requires a utopia without locale. As far as
I am aware, there is no such place (and no such meaning)
in the life that we know. Yet there is probably little philosophy
at stake here. What localizers have in mind has more to
do with the various international codes for language scripts,
where a good deal of the world’s different letters and diacritics
are indeed encoded. There is also no doubt the naturalizing
assumption that the world of programme developers is so
refined as to be not worth calling a locale (the zero-locale
natural language is elsewhere glossed as American English).
But let us leave such naïvetés lie. Our interests
are more with the term thus ushered in:
is the process of creating or adapting a product to a specific
locale. With a properly localized product a user can interact
with this product using their own language and cultural
conventions. It also means that all user-visible strings
and all user documentation (printed and electronic) use
the language and cultural conventions of the user. Finally,
the properly localized product meets all regulatory and
other requirements if the user’s country/region.
So that’s it; now
we know (and terribly written it is too). The point of most
interest here is, of course, the notion of adapting to “the
language and cultural conventions of the user”. What is
translation if not this? The document seems to pre-empt
the question, however, by immediately offering a list of
things to be localized:
in the world use very different notational conventions for
writing time, date, numbers, currency, etc.
1. Time is represented
by both a 12-hour clock and a 24-hour clock. The separators
can be period or colon.
2. Date has many
representations: sequence of month, day and year; different
separators; different calendars (lunar, Gregorian); different
abbreviated names for the months, etc.
3. Number representations
differ in separators.
4. Currency units
and presentation differ greatly from country to country.
(script) and language:
There are many
linguistic differences, some of the major ones are mentioned
1. Word order
varies between languages;
2. Word delimiters
are non-existent in certain oriental languages (Thai, Japanese);
oriental languages often do not have capitalization, some
languages allow accented capitals:
6 [sic.] Spelling
8. Character sets;
9. Cultural context:
sports and humor
I have copied out
these banalities precisely because they are such crashing
banalities. We might, for example, fit the whole of Chomsky
into the trite observation that “word order varies between
languages”, and the rest is surely just as insultingly obvious.
What (good) translator would not pay due attention to all
these things? Is this collection of commonplaces the only
material justification for the discourse of localization?
But we are not the
potential clients for whom the text was written. This list
should be seen as a reasonable piece of marketing; a sequence
of easily demonstrated items that, even if we always remember
all of them (though I must admit my students don’t), the
client may never have thought about. Horror stories about
date orders are easier to get across than the intricacies
of comparative syntax; the potential results of non-localization
can be made to sound more immediately disastrous.
A second virtue
of the list is that all these items are explicitly mentioned
as things that a client should pay for. The “all inclusive”
translation service is still often paid by the number of
words; in localization, on the other hand, the client knows
from the outset that there are more than words involved.
While we’re on the point, it is perhaps worth noting that
while the term “translation” is carefully excluded from
the above definitions and lists, it does appear elsewhere
in the text, when the first of six “major localization activities”
is described as “Translation of all natural language
strings that can be displayed or printed” (italics mine).
The other five activities are then what we would want the
client to pay extra for: adding spelling, hyphenation and
grammar checks; selecting notational and other conventions
(I have no idea what this means); adjusting menus, dialogue
boxes etc. for changes in the length of natural-language
strings (some of my Spanish-language applications have not
been localized in this respect); managing changes in the
original application in order to reflect changes in localized
versions; and testing the localized software on the target
platform. These undertakings are far from banal; they explicitly
break with the “neither add not take away” strictures of
straight translation; they should certainly not be considered
thus have one major justification: it can get clients to
pay higher fees than what they are prepared to cough up
for translation. This is a serious virtue. Yet the extras
are not necessarily for translators. There are several accounts
of how the various cakes can be cut, but a general rule
of thumb might allow up to 30% of the localization bill
for language transformation (“translation” in the narrow
sense), 30% for various engineering and publication activities,
and the rest for “management” of one kind or another (10-15%
for project management, plus communication, travel, training
and on-site work). Given the sums involved, there are very
good financial reasons why the localization discourse should
be proving enormously popular with virtually everyone except
we put localization?
I take it for granted
that some kind of introduction to software localization
should be part of any full-degree translator-training programme,
and should probably figure in many specialized short-term
postgraduate programmes. There are a few material problems,
of course (on the politics of which, see Wright 1999, among
others). First, you need computers for students to work
on (but this is necessary for virtually all translator training
anyway). Second, you need staff who know something about
localization and are prepared to work for a humble teacher’s
pay (but this is also a problem for the training of conference
interpreters, for example). A few basic localization principles
might ideally fit into courses where information technology
is already important, perhaps in terminology management,
electronic tools for translators, or alongside something
on HTML translating. However, although the word “localization”
does appear here and there in the programmes of the 280
or so translator-training institutions of which I am aware
(Caminade and Pym 1999), the only actual training courses
I have found are in Ireland, in departments of computer
science. This could mean that our translation schools simply
do not have enough of the computers and experts needed to
teach localization. Alternatively, it might mean that the
best solution is to send our translation students to the
computer-science departments for their enlightenment.
Nonetheless I am
prepared to argue that the discourse of localization, perhaps
even more than its actual techniques, be brought across
into our more general training programmes. There are several
reasons for this.
First, as I have
tried to show, parts of the discourse address and may help
to solve a few problems involving the interface between
language and culture (the newness of the term breaks clearly
with certain ideological constraints that have accrued to
the word “translation”; the discourse engages successfully
in client education; and the notion of “locale” could help
attenuate the petty mud-slinging between linguistics and
Second, there is
a growing market demand for graduates able to undertake
localization activities, especially at levels involving
more than language transformation (“translation” in the
narrow sense). If our training programmes are ethically
justified to the extent that they help students find good
jobs, this may be one major area in which we have to improve.
Just consider, if you will, the three diverging classes
of language-service providers that I see developing around
me. At the bottom of the heap might be the part-time language
teachers and unspecialized freelance translators who get
paid just enough to reproduce themselves, or better, are
exploited until, many of them female, they seek the greater
joys of marriage and children. Second, then, would come
tenured teachers, non-salaried conference interpreters,
in-house translators and stable-client freelancers, all
of whom might drive reasonable cars and have fun paying
off mortgages. The third group, where I find a few of my
former colleagues and increasingly my better, more street-wise
former students, comprises language professionals who have
either specialized in business or legal domains, or who
have managed to combine their language competence with training
in information technology, marketing or commerce. They are
generally earning two to four times what I earn as tenured
academic staff; their work is usually their life; though
they know what time they start in the morning, they never
know what time they’ll get home at night; they buy luxury
goods that they don’t have time to use. That third group
includes many of the software localizers, the people who
do rather more than just translate, and who are paid rather
more than are translators. As trainers, we might feel we
have some moral obligation to help our students get into
group three; localization and its association hype might
actually bring our graduates closer to innovative and enriching
There is one final
reason why the discourse of localization might be brought
across into translator training. It involves the following
catch. If localization is based on the supposed need to
adapt a discourse to the standards and conventions
of the receiving locale, then the same adaptation should
of course apply to any displacement of the discourse of
localization itself. That is, in its transfer from the commercial
to the educational fields, localization must be localized.
And this might be good not just for our current training
practices but perhaps also for the discourse of localization
itself, which might one day do rather more than imitate
first-year undergraduate classes.
What might this
localization of localization involve? First, one would hope,
the trivialities could be dispensed with as such, since
what works with clients tends to ruffle the thinning feathers
of know-all academics. Second, perhaps more ambitiously,
the actual checklists and techniques could be repackaged
so as not to restrict “translation” to the simple mule work
of moving between languages; translators could and should
be invited to take on the more ambitious roles and move
into the more fertile fields of expertise. And third, now
as a merely pious hope, the discourse of localization might
actually gain a few humanized considerations, if not quite
a humanistic conscience. I mean, if we look at the fragments
cited above, or even at the ten or so pages they have been
snipped from, we find no mention of people. There is no
consideration of why users might need localized products,
nor of what effects such things might have on the world
configuration of locales, nor of any other kind of value,
beyond money, that might motivate localizers or make them
humanize their lifestyles. The absence of people is more
generally a feature of the many current discourses on globalized
communications, where the odd recycled McLuhanism rubs shoulders
with numbers about the rapid growth of data (sometimes confused
with knowledge) and the disastrous consequences of missing
whatever boat is thought to be leaving. Scarcely a word
about the regional economies now beyond democratic control,
about hugely asymmetric relations of production, about our
widely diverging classes of intellectual labour, about,
in a word, ethics.
I have argued that
the discourse of localization might help us solve a few
problems in translation theory and translator training.
But my more profound hope, beyond the games and polemics,
is that an academic localization of localization discourse
might eventually transfer some of our secular humanism to
the power-holding professionals of tomorrow.
1965. Pour Marx. Paris: François Maspero.
Frank, Evelone Einhauser, Joachim Kornelius. 1998. “Die
elektronischen Hilfsmittel des Übersetzers”. AREAS
Annual Report on English and American Studies 15. Trier:
WVT Wissenschaftlicher Verlag Trier. 335-381.
Frank. 1999. "Between Babel and Bytes. The Discipline of
Translation in the Information Age". AREAS Annual Report
on English and American Studies 16. Trier: WVT Wissenschaftlicher
Verlag Trier. 439-450.
and Anthony Pym. 1999. List of Translator-Training Institutions.
1991. Translation and Relevance. Cognition and Context.
Oxford: Basil Blackwell.
Hermans, Theo. 1995.
“Toury’s Empiricism Version One”. The Translator 1/2. 215-223.
Justa.1984. Translatorisches Handeln. Theorie und Methode.
Helsinki: Academia Scientiarum Fennica.
Pym, Anthony. 1998.
Method in Translation History. Manchester: St Jerome.
Visited May 1999. “Localization 101”. http://www.intlconsultg.com/localiza.html.
Toury, Gideon. 1980.
“Equivalence and Non-Equivalence as a Function of Norms”.
In Search of a Theory of Translation. Tel Aviv: Porter Institute.
Toury, Gideon. 1985.
“A Rationale for Descriptive Translation Studies”. Theo
Hermans, ed. The Manipulation of Literature: Studies in
Literary Translation. London and Sydney: Croom Helm, 16-41.
Toury, Gideon. 1995.
Descriptive Translation Studies and beyond, Amsterdam and
Vermeer, Hans J.
1998. “Didactics of Translation”. Mona Baker, ed. Routledge
Encyclopedia of Translation Studies. London and New York:
1999. “Strange Bedfellows No Longer. Reconciling University
Language Studies with the Interests of Industry”. Language
International 11/2. 14.
- Last update
11 March 1999