1.

Why editing and revising are necessary

Why is it necessary for someone other than the writer or translator to check a text, and perhaps make changes,
before it is sent off to readers? A very general simple answer is that we human beings do not do perfection. In every
realm of activity, we make mistakes, sometimes serious ones, regardless of how experienced we are. Indeed, highly
experienced people can be overly confident in their ability to avoid error.

Of course, the impossibility of perfection also applies to editing and revising: no matter how carefully or how
often you check a text, you can be sure that you will not find every single problem.
In this chapter, we’ll look at several more specific reasons why editing and revising are necessary. First, it is
extraordinarily easy to write sentences that are structured in such a way that readers will misunderstand them or
have difficulty understanding them. Second, it is easy, while writing, to forget about the future readers and
write something which is not suited to them or to the use they will make of the text. Third, a text may fail to
conform to linguistic norms or to the reigning ideas about the proper way to translate or to write in a particular
genre. Finally, what the author or translator has written may conflict with the publisher’s goals.
To deal with these problems, revisers and editors amend texts in two ways. First, they act as gatekeepers who
ensure that the text conforms to society’s linguistic and textual norms and achieves the publisher’s goals.
Second, they act as language therapists to ensure ease of mental processing and suitability of the text for its
future users. This latter function is certainly important in the English-speaking world, but some language
cultures do not value reader-orientation as highly; readers are expected to do more of the work of understanding
themselves, bringing their background knowledge to bear on the task. In this kind of linguaculture, one would
not start an article by giving the reader a helpful overview of its structure (first I shall do this, then that); to do
so would seem patronizing.
Editors and revisers often find themselves faced with conflicting demands and needs. There are demands
from the client—the company, ministry or publishing house which has commissioned a writing or translating
job. Then there are standards required by professional associations to which the editor/reviser belongs, and
edicts from language-standardization or terminology-standardization bodies. Authors too make certain
demands, and finally, editors and revisers must constantly keep in mind the requirements of readers.
Editing or revising is thus not a matter of a vague ‘looking over’. There are specific things the editor or
reviser is looking for. Here are just a few of the many different ways in which a text might be defective:

• There are many typing errors.

• Sometimes the main numbered headings are bolded, and sometimes they are italicized.
• There are unidiomatic word combinations.
• You often have to read a sentence twice to get the point.
• You often come across a word like ‘it’ or ‘they’ and you cannot tell what it refers to.
• The text contains a great many words which the readers won’t understand because they are not very highly
educated, or because they are not experts in the subject matter of the text.
• The text is not written in a way appropriate to the genre. For example, it is a recipe, but it does not begin
with a list of ingredients, it is rather vague about how to make the dish, and it is full of commentary on the
history of the dish and the chefs who are famous for making it.
• If the text is a narrative, it is hard to follow the sequence of events. If it is an argument, it is hard to follow
the steps.
• There are passages which contradict each other.

1.1 The difficulty of writing

In this section, we’ll look at why texts need therapy to help readers. Writing is difficult work. In this it is quite
different from speaking, which is easy (despite being highly complex in terms of the physiological processes
involved). We all learn to converse, without any formal instruction, during infancy. Writing, on the contrary,
requires long years of apprenticeship and even then, many people never learn to do it well. Indeed, even the
best writers and translators make mistakes— sometimes serious ones. There is no point in seeking out writers
and translators who are so good that their work never needs to be checked.
Why is writing so difficult? There are three main reasons. First, there is no immediate feedback from readers.
If you are conversing, a question from your interlocutor or a puzzled expression on their face will lead you to
repeat or rephrase in order to make your message clear. If you are writing, however, you may create an
ambiguous sentence, or use a word the reader doesn’t know, but there is no one there to react to the problem, so
you do not notice it. This is part of a larger difference between speech and writing: a conversation is jointly
constructed by at least two people who are together in a situation, while in writing (other than text messaging)
the main burden of successful communication falls on the writer. The writer must imagine the reactions of an
 Why editing and revising are necessary 2
often-unknown reader in an unknown future situation, anticipate the reader’s problems in receiving the intended
message and act to forestall them.
The clerk who posted the message ‘back in 30 minutes’ on the shop door failed to anticipate the reader’s
problems. If I come along and read this message, I don’t know whether the shop will re-open in 1 minute, 10
minutes or 29 minutes. The clerk should have written ‘back at 10:45’.
To write successfully, it is necessary to be constantly aware of what your future readers do not know (it’s not
part of their likely background knowledge or you have not already told them earlier in the text). Poor writers
forget this. They treat writing as self-expression rather than communication with others. They seem to operate
on the principle that if they have a certain meaning in mind as they write, that meaning will automatically come
across to readers.
A second reason writing is difficult: written documents tend to be lengthy. When speaking, you typically
need to organize what you are saying over a stretch of a couple of words to a couple of dozen words (the
delivery of lengthy monologues such as formal speeches is usually assisted by speakers’ notes or scripts). In
writing, things are quite different. Unless you are tweeting, preparing a grocery list or sending a very brief
email, you typically need to organize a stretch of a few hundred or a few thousand words in the case of a report
or article, or a few tens of thousands or hundreds of thousands of words if you are writing a book.
Third, it is easy to forget to compensate for lack of intonation and gestures. In conversation, much meaning is
conveyed through intonation, and to some extent also by gestures (facial expressions, body posture, hand
movements such as pointing). It is very easy to forget to compensate for the lack of intonation in writing, and
the result will be ambiguity, or an unclear connection between successive passages.

Consider this sentence:

As these studies tend to show the form translation has taken in Canada, both on an institutional level and on
the level of the actual practice of translation, is specific to our particular national context.

Here the reader might wrongly take ‘the form’ to be the object of ‘show’, whereas in fact, it is the subject of ‘is
specific’. In speech, the voice would drop slightly when pronouncing the word ‘show’ and there would be a
slight pause. The writer forgot to place a comma after ‘show’ to ensure a correct reading.
Writing a translation, aside from being subject to the three difficulties just described, is also difficult because
of the need to convey someone else’s meaning. The translator is often not a member of the intended readership
of either the source text or the translation. As a result, it’s easy to convey to readers a meaning not present in
the source text, or to write in a way that will confuse the intended readership. In addition, it is very difficult
when translating to avoid undesirable linguistic influences seeping in from the source language.
Good writers and translators recognize how easy it is to err. To minimize errors in their final output, they
engage in some combination of planning and self- editing. One study of writing strategies found four basic
strategies:
Writing strategy Planning before drafting Self-editing
Architect Major Minimal, after
drafting
Bricklayer Major Major, during drafting
Watercolourist Minimal Minimal, during
drafting
Oil painter Minimal Major, during & after
drafting
Some writers (‘architects’ and ‘bricklayers’) forestall error by thinking through their message carefully before
they start composing; sometimes they will even prepare a detailed outline. A few of these writers—the
‘architects’—are apparently so good at planning that they manage to produce good writing on the first draft,
writing that requires only minimal self-editing after they have got the draft down. ‘Bricklayers’, on the
contrary, do major self-editing as they draft.
Quite different are the ‘watercolourists’ and ‘oil painters’. They tend to think by writing, so there is little
planning. They simply start writing, perhaps with just a theme or a single idea in mind, or a few scribbled notes.
Watercolourists, in addition to their minimalist planning, also engage in little self-editing. As a result,
watercolourists are generally not very good writers. Oil painters compensate for their lack of planning by
engaging in major self-editing both during and after drafting.

Exercise 1
Take a few minutes to consider the following questions and then tell the group about your approach to
writing.
 Why editing and revising are necessary 3
a) When you are writing in (not translating into) your own language, which of Chandler’s four strategies
do you adopt? Are you an architect, bricklayer, watercolourist or oil painter? Or do you use more than
one of the strategies, depending on the nature of the writing project?
b) Do you identify with none of the four strategies? Say why not.
c) If you identify yourself as, say, a bricklayer, have you always been a bricklayer? Did you learn one
strategy at school and then switch later?
d) Do you use similar strategies when writing and when translating? For example, if you plan your writing
extensively, do you also do a lot of preparation before you begin to draft your translations? If you make
many changes while writing, do you also make many changes while drafting your translations?

1.2 Enforcing rules

In this section, we’ll consider why texts need gatekeepers. Writing differs from speech in that it is usually
subject to external regulation in a way that conversation is not. This is so in two senses. First, texts are usually
written in a standard language, which has more or less clear-cut rules set out in dictionaries, grammars and
recognized usage authorities. (Exceptions may be allowed for innovative work, often called ‘creative’ writing,
but the editing of such work will not be considered here.) Publishers of texts may also have special rules about
a host of matters such as when to write ‘eight’ and when to write ‘8’, whether ‘he or she’ should be replaced by
‘they’, and whether quotations are to be separated from the main text and indented. In addition, writing in
specialized fields is subject to standardized terminology. Finally, every language community or subcommunity
has rhetorical habits and genre traditions; there are widely accepted principles for constructing an argument or
for writing a recipe.
The second kind of external regulation stems from the fact that much writing is commissioned; that is, there
is a publisher who has asked the writer or translator to prepare the text. The publisher has certain goals, and
someone has to ensure that these goals are achieved. For example, corrections may be needed to deal with
departures from appropriate content, such as political or sexual content. The rules here may be current social
conventions (or laws!) or they may be imposed by a particular publisher. Publishers will also want to maintain a
certain reputation, and this will require correcting inaccuracies (factual and mathematical errors, erroneous
quotations). Editing will of course be especially important if the text is to be sold rather than given away.
Much commissioned original writing at workplaces (e.g. minutes of meetings, progress reports) does not
need to be edited because it is ephemeral and circulates within a very restricted group of readers; no great harm
is done if no one checks and corrects such writing. Also, private writing (diaries, personal letters, shopping
lists) is not edited because there is either no audience at all or none beyond the writer’s immediate circle.
Editing is important when a text bearing a message deemed important by its publisher is being prepared for a
large audience of strangers, or for audiences who will be reading it over a long period of time. Editing gives
such a text the ability to reach out into space and time, by ensuring that it carries enough contextual information
to enable people outside the immediate world of the writer to interpret it in the intended way.
Before the computer age, it was very hard to make writings available to a wide audience without going
through the editor of a newspaper, magazine or publishing house. To get around editors, you had to type up and
copy leaflets, or take them to a printer, and hand them out in the street or mail them out through the post office,
or else make posters and paste them up around town. Now however, through blogs and the like, it is very easy,
quick and cheap to self-publish to a potential readership of millions. This has democratized public writing, by
taking it out of the hands of editors and publishers with their various biases, though it has also dramatically
increased the volume of writing which has not been checked for language or facts and may not have been self-
edited at all.
Turning to translations, since they are almost always commissioned, there will be a client to satisfy. People
who decide on their own to translate a Wikipedia article must still meet Wikipedia’s publishing rules even
though actual revision by a second person is optional. Also, some degree of revision is needed even with
ephemeral texts to correct errors which are peculiar to translational writing: mistranslations, omissions, and the
strange unidiomatic language which is so hard to avoid when translating (odd word combinations or sentence
structures inappropriately calqued from the source text). Finally, someone is needed to ensure conformance
with current norms governing translation in a given genre: Must the translation reflect the source-text message
in tiny detail or only in broad outline? To what extent must the actual wording of the source text be reflected?

1.3 Quality in translation

Translations need to be revised in order to achieve quality. But just what is quality?

Stated and implied needs

The International Organization for Standardization, in its 2015 standard ISO 9000, entitled ‘Quality
management systems: fundamentals and vocabulary’, defines quality as the degree to which the inherent
 Why editing and revising are necessary 4
characteristics of an object fulfil a set of stated or implied requirements, with ‘requirements’ defined to include
needs, expectations and obligations. Implied requirements are those which are common practice or customary.
There are two important things to note here. First, quality is always relative to needs. There is no such thing
as absolute quality. Different jobs will have different quality criteria because the texts are meeting different
needs. In one job, an editor or reviser must improve the readability of the text to a very high level; in another
job, a lower degree of readability will suffice. Sometimes several degrees of quality are recognized, sometimes
two; in the latter case, a frequent distinction is between information-quality (the document will be used in-
house, usually by a small number of people for information only, and then be discarded) and publication-quality
(the document will be read by a large number of outside readers over a fairly lengthy period of time).
The second thing to note in the ISO definition is that needs are not just those stated but also those implied.
The most important implied need in translation is accuracy. People who use the services of translators don’t ask
for an accurate translation; they just assume that it will be accurate. Ensuring accuracy is a key task of revisers.
Another implied need is successful communication of the text’s message to the readers. Achieving this may
require an editor or reviser to override the publisher’s or client’s instructions (sometimes called the brief or
commission). This is particularly true in translation, because clients unfamiliar with the target-language
community may have a mistaken or incomplete understanding of the cause-and-effect involved. For example,
the client asks the translator to follow the paragraphing of the source text, but the paragraphing habits in the
target language are quite different for the genre in question. In no profession can one always bow to the client’s
wishes. Imagine some people who are renovating their house. They tell the architect that they want a certain
wall removed. When the plans come back, they see that the wall is still there. Why? Because the architect has
determined that it’s a bearing wall—the house would fall down without it. Similarly in translation, it is up to the
reviser to ensure that communication will not break down when the message is read by members of the target-
language community.
In some cases, it may be possible to change the client’s stated expectations through education. Most clients
know next to nothing about what translation involves, how much can be translated in a given time, why
translators need documentation, and so on. However, educating clients is not easy, for a variety of reasons (lack
of interest, frequent changes in the person who represents the client in dealings with the translator).
Consequently, education should probably not be seen, in most instances, as a way of overcoming problems
related to clients’ stated needs.

Three concepts of quality

There are several broad concepts of quality current in the world of translation, and these lead to differing
‘philosophies’ of revision. Note that ‘quality’ here means for the most part linguistic quality. We will not be
much concerned with visual quality, quality of service or the technical quality of electronic files. Also,
translation companies and the organizations representing them have in recent years advanced a procedural
concept of quality that is focused not on the relationship between the source text and the translation, or on the
quality of writing in the translation, but on the process used to prepare the translation, the idea being to forestall
errors before they are made.
One common concept of quality is that achieving acceptable quality means satisfying clients. This may lead
you to pay most attention, when revising, to finding errors that will be easily noticed by the client, such as
typing errors and client-related terminology. This approach has a contractual version in which an actual
agreement is prepared between translation provider and client. The U.S. standardization organization ASTM, in
its 2006 document F2575 (second edition 2014), entitled ‘Standard Guide for Quality Assurance in Translation’
defines translation quality as ‘the degree to which the characteristics of a translation fulfil the requirements of
the agreed-upon specifications’. This concept is satisfactory only if it is understood that the translation provider
cannot agree to specifications that conflict with professional standards.
An extreme version of such agreements would see quality as being achieved if the benefits to client and
provider in terms of price, revenues and repeat business outweigh the cost of production. This is ethically
dubious since the actual characteristics of the translation are not considered. For example, the practice of
requiring translators to accept 100% matches found in Translation Memories without revision will certainly
reduce production costs but it will also pose an unacceptable risk to accuracy and language quality.
An extension of the client-satisfaction concept of quality in the Web 2.0 age is reader satisfaction, since it is
now possible for a ‘crowd’ of translation readers on Facebook or Twitter to ‘vote’ on proposed translations.
A second concept of quality sees it as doing what is necessary to protect and promote the target language.
This view will typically be found in language communities where translators want to counter the effects of a
formerly or currently dominating foreign or majority language—these days very often English. Revision then
becomes a quasi-literary writing exercise in language and style improvement. Revisers working under this
concept of quality will keep making changes in texts until they fit a certain ideal of authentic and excellent
writing in the target language, regardless of the time that takes, and thus regardless of the added cost. Here
quality refers more to how well the translation is written than to its correspondence with the source text.
 Why editing and revising are necessary 5
A third view of quality is that a translation is of acceptable quality if it is ‘fit for purpose’. This means that in
the reviser’s view, the translation is suited to the people who will be reading it and the reason they will be
reading it, as specified by the client. A reviser working under this concept of quality will read the draft
translation with the client’s purpose in mind, and then make only such changes as are needed to make the
translation suitable for that purpose. The translation needs to be ‘good enough’ to serve its purpose, and no
better. The notion of quality as fitness for purpose is endorsed by the International Organization for
Standardization in its 2015 standard ISO 17100 ‘Translation Services – Requirements for Translation Services’,
which says that ‘The reviser shall examine the [translation]… for its suitability for purpose’.
Some people speak rather disparagingly about this approach to quality. They say that we should be aiming at
‘good’ translations, implying that ‘good enough’ is second best. In my view, the concept of a ‘good translation’
is useless for revision purposes because it is almost entirely subjective: what one person considers to be a good
translation someone else may find mediocre. The concept ‘good enough’, on the contrary, implies the question
‘good enough for what?’ and the answer to this question can pick out a set of fairly objective criteria for
deciding whether changes in wording are needed; for example: Are the wordings suited to the education level or
subject-matter knowledge of the potential readers? A translation that is intelligible and fairly accurate but has
awkward and unidiomatic wordings may sometimes be ‘good enough’; in other cases, extreme accuracy and
very high writing quality may be required if the translation is to be ‘good enough’.
The purpose of a translation can to some extent be read off the target-language genre: the purpose of an
obituary is obviously different from the purpose of a patent or a tourist guide, and wordings will be acceptable
or unacceptable accordingly. To sum up on fitness for purpose: there are not just two kinds of quality: excellent
and dreadful. Rather, the criteria for acceptable quality vary with the type of text and its future readership.
The various quality concepts provide a focus for the reviser’s work: will it be on the specifications for the
job, on language protection or on appropriateness for users and use? The chosen focus will dictate what
happens during revision. If your focus is language protection, you may make a change which you would not
make if your focus were appropriateness or specifications. The concept of quality also has an influence on how
you handle ‘accuracy’ and ‘readability’ as goals. If you are operating under the concept of language protection,
readability may be sacrificed to some degree in order to follow the prescriptions of a conservative language-
regulating body. On the other hand, if what you mean by language protection is keeping target-language genre-
related rhetoric free of English influence, then readability may be your highest value. If you are operating under
the concept of fitness for purpose, accuracy may either be extremely important (with legal texts, where extreme
accuracy may be required even at the expense of readability) or it may be not so important (for the in-house
employee newsletter of a multilingual government ministry, the reviser may well accept wordings where the
translator has added to or subtracted from the source text message in order to make it more lively or funnier, if
that seems appropriate for target-language readers).
The concept of quality under which revisers work may vary from country to country, or from language pair
to language pair. It may even vary by direction. In Canada, for example, English-to-French revisers tend to be
quite concerned with the ‘language protection’ aspect of revision, whereas this is not the case for French-to-
English revisers, except perhaps for those who work in Quebec, where English is a minority language
influenced by French. Revisers and editors who work in countries where their language of work is not the local
language (say, English speakers working in the Netherlands) may also be more concerned with the ‘protection’
aspect of translation than those who work in majority English-speaking countries.
Note that the ‘fit-for-purpose’ concept cannot be expanded to include the other two concepts. Protecting the
target language and pleasing the client are not ‘purposes’ in the intended sense because they are not specific to
the text at hand. Suppose a translation is to be read by subject-matter experts who are starting a research project
and have come across the source text, written in a language they
 6
don’t know, during their literature search. The question for the reviser then is whether the draft
translation is suited to these expert readers and to their need to find out what their colleagues, writing in
the source language, have discovered on the subject of the research project. The wording suited to this
may not please guardians of the target language or conform to what the client specified.
Note too that with the contractual, please-the-client view, you may end up after revision with a
translation that is less than fit for purpose (depending on how detailed the contractual specifications
were). With the target-language protection view, on the other hand, you may well end up with a
translation that is more than fit for purpose, since many of the changes you have made will not be
necessary for fitness.
A difficulty with the fit-for-purpose conception of quality is that it may be hard to determine the
purpose. In addition, some readers who are not members of the audience which the reviser had in mind
may nevertheless retrieve the translation from an electronic archive, possibly months or years later, and
use it in a way which the reviser was not contemplating. However, this is simply a risk one takes when
using the fit-for-purpose approach. The purpose in question has to be the currently known purpose. One
cannot revise a text to be suitable for any possible use or readership now or in the distant future.
A final point: it may be psychologically handy, when editing or revising, to think of quality in
negative terms: don’t ask yourself whether a passage is fit for purpose, for example, but whether it is not
fit; does it diverge unacceptably from expectations, however defined?

How important is accuracy?

Aside from different general concepts of quality, there may be differences about which aspect of
translation is most important: Is accuracy the prime quality of a good translation, that is, the translation
conveys more or less all and only the meaning which the reviser believes to be present, explicitly or
implicitly, in the source text? Or is the most important quality of a good translation that it satisfies the
agenda of the commissioner, regardless of correspondence to the source text? A rather inaccurate
translation of a tourist guide (one with several additions and subtractions) may nevertheless be very well
written and be found to be very useful by the tourists who buy it, quite possibly more useful than an
‘accurate’ translation, because the translator will have a good idea of the culture-based interests of
target-language readers who are visiting the country or city in question. Some people refuse to call such
texts translations, and insist they be called adaptations, but since many translators produce them, the
distinction seems pointless in the world of translation practice (though it may be useful in theoretical
writing). An ‘accurate’ translation can be of high quality, and so can an ‘agenda-based’ one.

1.4. Limits to editing and revision

Theoretically one could edit or revise any text until it fits a given quality ideal. However, from a
business point of view, some texts are not worth editing or revising. They are so badly written or
translated that it would take a very long time to edit or revise them and, consequently, it would cost too
much. Consider this badly written version of a passage near the beginning of the section entitled ‘The
difficulty of writing’ in this chapter:

Its hard to write but speaking is very easy even though its complicated, we all learned to speak as
children without any insrtuction. But it takes a very long time to learn to write and many people’s
writing is still awful.
If an entire article of this quality were submitted for publication in a magazine, it would probably be
rejected. The problem is not so much the mechanical errors (its, insrtuction, the comma after
‘complicated’); even large numbers of mechanical errors can be corrected easily and fairly quickly. The
problem lies in the lack of flow and poor focus—an ordering of words that does not bring out the logic
of the argument. A few such problems scattered through a long text are fixable, but if almost every
sentence needs recasting, then it simply would not be economical to proceed. The above passage does
not call for editing; it calls for rewriting— preferably by a different writer.
Some clients may wish to proceed with the editing of bad writing despite economic considerations.
Perhaps some political or ideological concern is more important than cost. Or perhaps the author is an
Important Person. You may therefore find yourself editing the work of people who need help writing:
learners of English as a second language; people who had insufficient or ineffective schooling and have
still not mastered the differences between speech and writing; or people who think that they will impress
readers if they write very long sentences chock full of subordinate clauses, clauses within clauses, and
parenthetical expressions.
 7
Just as editing is not rewriting, so revising is not retranslating. If a translation is full of unidiomatic
word combinations, if the sentence structures are so influenced by the source text that the result is
unreadable, and of course, if the translator has clearly misunderstood numerous passages of the original
text, the solution is to retranslate, not revise.

Exercise 2
Machine translation (MT) output is often unrevisable. Create your own example by pasting a
paragraph from an online news article (written in your second or third language) into the source-text
box at www.babelfish. com, translate.google.com or any other free site you can reach by entering
‘machine translation sites’ in your search engine. Ask for a translation into your first language. Does
the output seem revisable? Try to actually revise it, removing only the worst mistakes. Does this
confirm your initial impression about its revisability?

1.5 The proper role of revision

Revising is necessary because translators make mistakes, but it is important not to place too great a
burden on it. It should not be the main way of ensuring quality. Quality is best ensured by preventive
action: hiring translators whose training and talent has given them the skills required for a given
workplace, using the right translator for a given job, making sure the specifications for the job are
known to the translator, passing on any client feedback from previous translations, making sure the
translator has access to appropriate technological tools and to the necessary documentation, terminology
resources, previous translations on the subject and subject-matter experts.

When working with a new translator, the reviser should forestall predictable errors. For example, if
the text is a set of instructions, point out that the source language may convey instructions in the passive
(‘the green button should now be pressed’), but in the target language, the imperative is probably best
(‘now press the green button’).
Revision should be seen as a necessary final resort to clean up the inevitable errors that will occur
despite such precautions. Unfortunately, revision is often— perhaps increasingly—used as a way of
dealing with the problems that arise when translation is outsourced to cheap but unqualified contractors.
The main burden of ensuring quality should fall on translators (once they have sufficient experience),
not revisers.
Now, while there can be too much emphasis placed on revision, there can also be too little. Many
translation services and agencies see the reduction of revision as a way to increase productivity and thus
income; perhaps many freelances do so as well. Revision by a second translator is applied to fewer and
fewer texts, or entirely replaced by self-revision, except for novices. Perhaps even self-revision is
reduced. From a financial point of view, the revision process is very different from the drafting process.
If you have a 20-page source text and so far, you have only translated 5 pages, obviously the translation
is incomplete and cannot be sent off to the client. However, once you have translated all 20 pages, you
have a complete translation, which could be sent off to the client with little or no revision. The
temptation to do this is great; one wonders how often it is happening.
2 The work of an editor

A full description of the organization of work within publishing organizations is beyond the scope of
this book, we’ll just look briefly at the jobs of people who work as editors. We’ll then distinguish
editing from adapting and rewriting and look at the editing work done mentally by translators when
their source texts are poorly written, as well as the editing of non-native English. Texts are of course
also edited by people who are not professional editors or translators, but this topic will not be
considered here. The chapter concludes with a discussion of degrees of editing.
In this book, editing means reading a text which is not a translation in order to spot problematic
passages, and making or suggesting any corrections or improvements needed to achieve some concept
of quality. In some cases, the text may happen to be a translation but the editor either does not know
this or does know but treats the text as if it were not a translation. For example, someone familiar with
the subject matter of a machine-produced translation may edit it without consulting the source text or
even knowing the source language.

2.1 Tasks of editors

Dictionary definitions of the verb ‘edit’ present a considerable variety of meanings. Here is a sample,
culled from various dictionaries:

• assemble, prepare or adapt (an article, a book) so that it is suitable for publication;
• prepare an edition of (a literary author’s work), especially by researching manuscripts;
• be in overall charge of the content and arrangement (of a newspaper, journal, etc.);
• reword, revise or alter (a text) to correct, alter the emphasis, etc.

As for the occupation ‘editor’, here is what we find in the 2016 National Occupational Classification
published by Canada’s employment ministry:

Editors review, evaluate and edit manuscripts, articles, news reports and other material for
publication, broadcast or interactive media and co-ordinate the activities of writers, journalists and
other staff. They are employed by publishing firms, magazines, journals, newspapers, radio and
television networks and stations, and by companies and government departments that produce
publications such as newsletters, handbooks, manuals and Web sites. Editors may also work on a
freelance basis.

Here, attention will be restricted to print publishing as opposed to e-publishing or broadcasting. In

some sectors, notably news, the advent of electronic media has created financial problems for print
publishing because advertisers have migrated away from print, and one result is that there is less
money for editorial work on text and, consequently, more uncorrected errors.
Editors have many duties, and different editors have different duties. An editor’s daily routine will
be rather different at a newspaper from what it will be at the office of a firm publishing a medical
journal. The description of a particular editor’s job might include one or more of the following:

• deciding what kinds of materials will be published (e.g. at a newspaper, selecting stories for
coverage);
• finding or assigning writers and handling relations with them;
• evaluating the suitability of manuscripts and recommending changes in content, style or
organization;
• dealing with reviewers (subject-matter experts who comment on the content of specialized
writing);
• scheduling the publication process;
• designing page layouts, with incorporation of graphics;
• marking up manuscripts with instructions for printers;
• ensuring permission has been obtained to use copyrighted material, and dealing with other legal
concerns such as libel;
• managing the financial and material resources, and the employees, of a publishing enterprise or
department;
• amending the text submitted by a writer.
 The work of an editor 9
An editor’s work will also vary greatly with the type of writer being edited. Editing the work of
professional writers is quite different from editing the writing efforts of, for example, scientists
writing articles for a journal or employees who are required to prepare reports as part of their job but
do not actually like writing or are not very good at it. For many editors, relations with professional
writers may be the most difficult feature of their work.
In translation, the situation is rather different. While literary translators must often negotiate rather
carefully with living source-text authors, non-literary translators—the great majority—are usually
dealing with non-professional writers. The translator is therefore the writing expert in the relationship,
and in addition, often enjoys the advantage of being a native speaker of the target language. In non-
literary translation, difficulties tend to arise in the relationship between translator and reviser, or
translator and client, and not so much in the relationship between translator and author.

Types of editing work

Notable in the job description for editors near the beginning of this chapter is that editing in the sense
of checking and amending a text is mentioned as just one among many tasks. In this book, we will not
be looking at the full range of editors’ duties or the various situations in which they work. Instead, we
will be concerned almost entirely with the checking/amending task—a task that some people with the
job title Editor do not perform at all. There are four broad types of checking/ amending work:

• Copyediting. This is the work of correcting a manuscript to bring it into conformance with pre-set
rules—the generally recognized grammar and spelling rules of a language community, rules of
‘good usage’, the publisher’s ‘house style’, and page layout and typographical matters. The
copyeditor must also ensure a degree of consistency in such matters as terminology and the
positioning, numbering and appearance of section headings and subheadings. As a result of the
above-mentioned financial difficulties being encountered in some sectors of print publishing, the
effort devoted to copyediting may be reduced.
• Stylistic editing. This is work done to improve rather than correct the text. It involves tailoring
vocabulary and sentence structure to the readership and creating a readable text by fixing
awkwardly constructed sentences, ensuring the connections between sentences are clear, and so
on.
• Structural editing. This is the work of reorganizing the text to achieve a better order of
presentation of the material, or to help the readers by signaling the relationships among the parts
of the message.
• Content editing. This is the work of suggesting additions to or subtractions from the coverage of
the topic. The editor may (perhaps with the assistance of a researcher) personally have to write the
additions if the author for some reason cannot or will not do so. Aside from such ‘macro-level’
work, content editing also includes the ‘micro-level’ tasks of correcting factual, mathematical and
logical errors. Correcting logical errors is part of the broader task of ensuring that every passage in
a text is intelligible.

If you have simply been asked to ‘edit’, you should inquire about which kind(s) of editing are wanted,
so as to avoid wasting large amounts of time on unwanted editing; stylistic editing, in particular, can
be very time-consuming. It’s also a good idea to know where you fit into the overall scheme of
preparation of a document for publication. Perhaps structural editing has already been done, in which
case decisions about combining or splitting paragraphs, or may already have been made. Perhaps a
copyeditor will be working on the text after you, in which case you do not need to worry about things
like consistency or compliance with house style.
Note that in some editing situations, changes are made without consulting the author; in others, the
text is sent back to the author with suggested amendments and perhaps comments next to certain
passages (either handwritten or made with the Comment function of a word processor), or else
separate sheets are sent containing specific or general suggestions and questions.
One final type of amending work is making changes of all kinds to produce a new edition of a
previously published work. Somewhat confusingly for our purposes in this book, the term ‘revise’ is
sometimes used to refer to the process of reviewing the original edition with an eye to such
amendments, or to refer to both the reviewing and the consequent amending, the final result being a
‘revised edition’. Occasionally, ‘revise’ is used in this sense for translations, usually when someone
makes amendments to a previously published translation of a literary work.
 The work of an editor 10
In large organizations, a similar process may go on before a document is published: a draft is
prepared and sent out for comment; amendments are made on the basis of the comments, the outcome
being labelled ‘version 2’. The process repeats until a satisfactory result is achieved. ‘Version’, like
‘revision’, is a term which has a different meaning in the worlds of editors and translators; for the
latter, it is a synonym of ‘translation’, mostly used in combination with a language name (‘the German
version’).

Division of labour
In large publishing companies, and in corporations or government ministries that have a publications
department, there may be a considerable division of labour— a hierarchy of employees working under
a variety of titles such as senior editor, assignment editor, editorial assistant, copyeditor, production
editor, fact-checker or proofreader. The title Editor is often used just to designate the person in charge
of some area of work, such as the photo editor or the sports editor at a newspaper. Also, titles do not
necessarily reflect tasks: a copyeditor may also do stylistic editing and micro-level content editing.
This may happen unofficially: an editor’s official task may be copyediting, but in practice, they may
see that other types of editing have not been done and will intervene.

Senior editors
A senior editor will oversee a publishing project, deal with authors and reviewers, and suggest macro-
level content changes in texts. All the more detailed textual work, as well as the layout and printing
work, will be left to others. Senior editors at publishing companies and newspapers often find
themselves at the interface between the creative and the commercial aspects of publishing. They may
want to promote a certain writing style or certain innovative ideas, but the marketing department
(responsible for selling the publication) or in some publishing sectors the advertising department
(responsible for perhaps much of the publisher’s income) may not be supportive, or the budget for
hiring a sufficient number of good editors and writers may not be available.
Senior editors in the publication departments of government ministries, churches or other
institutions may often not be bothered by such commercial considerations, but like newspaper editors
they will be responsible for ensuring that publications are consistent with, or actively promote, the
political or ideological goals of the publishing organization. As a result, the editor may from the outset
exclude all or most authors whose ideas are not acceptable to the publisher.
Thus, far from simply dealing with words on a page, the editor becomes the focal point of
negotiation among the sometimes-conflicting interests of publisher, writers, marketers and buyers
(readers).

Subject-matter reviewers
When it comes to highly specialized documents, some publishers will draw on the services of experts;
for example, a manuscript in the field of atmospheric physics will be looked at by a meteorologist.
Such an expert may review the manuscript, prior to acceptance for publication, in order to determine
whether it is original work and represents a contribution to the field, point out gaps in the argument,
and so on. Experts may also be employed to do content editing for factual and conceptual accuracy
and any other matters calling for specialist knowledge. Alternatively, such a text may be edited by
someone who specializes in editing scientific texts. ‘Scientific and technical editor’ and ‘medical
editor’ are now occupations engaged in by people who are not themselves technicians, engineers,
scientists or medical researchers. This is less often the case with other specialized areas: people who
edit specialized works in law or music will usually be subject matter experts themselves.

Proofreaders
After a manuscript (usually in the form of a Word document) has been edited, it goes to the
publisher’s production department for page design and typesetting (the old term is still used, though
nowadays literal setting of metal type is an artisanal pursuit; typographical decisions are made on a
computer screen). In some publishing sectors, authors are given electronic templates into which they
must insert their text, so that no further page layout and typographical work is required.
The outcome of the design and typesetting process, the ‘proof’, may then be compared with the
original Word document in order to catch any remaining errors in the Word document, or errors
introduced during design and typesetting. This task, known as proofreading, may be assigned to the
 The work of an editor 11
author, the editor, or a specialized proofreader employed by the publisher. Nowadays the proof most
commonly takes the form of a PDF file. Adobe Reader and Adobe Acrobat contain tools that make it
easy to mark up this file.
Proofreaders working on paper use special paired marks to indicate errors: one mark appears within
the text itself (the copy mark) and the other in the margin, to draw the printer’s attention to the
change. Note that not all proofreaders in the English-speaking world use the same set of marks. For
exercises, use the copy marks your instructor recommends. Proofreading per se lies outside the scope
of this book.

Terminology note. The term proofreading is sometimes used by translators to mean copyediting. It
is also used by some translators to refer to the procedure called unilingual re-reading. It is even
used as a synonym of revision.

2.2 Editing, rewriting and adapting

Editing needs to be distinguished from rewriting and adapting. When editing, you start from an
existing text and make changes in its wording. Sometimes, however, the existing text is so badly
written that it is easier to abandon the existing wording and re-express the text’s content with newly
composed sentences and possibly a new text structure. This is rewriting. In Complete Plain Words,
Ernest Gowers provides a treasure trove of real examples of bad bureaucratic writing and discusses
the principles that should be used in rewriting them. Here’s an example that illustrates the single
greatest problem with such writing—overuse of nouns and sequences of nouns:

This compulsion is much regretted, but a large vehicle fleet operator restriction in mileage has
now been made imperative in meeting the demand for petrol economy.

This sentence may not pose a problem for specialists in road transportation, but non-specialists will
find it easier to read:

We much regret having to do this, but we have been obliged to greatly reduce the use of our fleet
of large vehicles in order to meet the demand that we economize on petrol.

Clearly this second sentence was not created by adding, subtracting and moving words in the first
sentence. Sometimes such rewriting is needed only in the occasional sentence; sometimes most of a
text has to be overhauled in this way.
Now, both editing and rewriting aim to create a text that is maximally suitable for the original
intended audience. Sometimes, however, people don’t want to replace the old poorly written
document with one that is better written; instead, they want to prepare an additional document for a
new audience. In this case we’ll call the activity adaptation. This may involve either complete
recomposition (as in the above example of rewriting) or relatively minor rewording of the existing
sentences (as in editing).
First, let’s look at a case where adaptation will typically require complete recomposition. English
legal documents traditionally address an audience of lawyers and judges; a legal editor would check
that such documents were suited to that readership. However, in recent decades the English-speaking
world has seen a movement demanding ‘plain writing’ of legal documents so that they can be read by
non-lawyers. In some jurisdictions, there has even been ‘plain writing’ legislation, requiring for
example that consumer financial documents such as mortgages be in readily understandable language.
This will generally call for complete recomposition of sentences in order to achieve a high level of
readability, perhaps at the expense of precision. Legal language is often hard to read because the
writer was trying to be extremely precise, eliminating as much vagueness and ambiguity as possible;
often this cannot be accomplished without sacrificing ease of reading.

Now, let’s look at two cases where minor rewording would probably suffice to adapt a text:

• Start from a document originally written for a British audience and adjust it for an American
audience (e.g. make adjustments in vocabulary and spelling).
• Start from a document originally written for an audience of native readers and adjust it for an
international audience that includes mainly non-native readers.
 The work of an editor 12
These cases exemplify two common procedures for document adaptation. In the first, features are
added to a document that are specific to a local readership, while features specific to other localities
are subtracted. In the second, all local features are subtracted, in order to address the broadest
possible, international audience. Preparing a document for such a broad audience is especially
difficult for adapters who are native speakers of the language in which the text is written and also
members of the culture from which it originates. That is because they must have a knowledge of what
others do not know, whether it be difficult aspects of the language or local history. Thus a reference to
the ‘44th president’ (of the United States) is likely to be obscure to readers in other countries.
A final case of preparing a supplementary document through adjustment of an existing one is
repurposing. Here material is adjusted for use with a new medium. For example, text might be
adjusted for use in a printed brochure, on a Web page, or in a slideshow presentation. The adjustments
might include changes to the wording but also to visual appearance (e.g. some fonts work better on
paper, others on screen).
Finally, it should be noted that the distinction between authoring (original writing) and editing is
becoming blurred. It has become common to ‘write’ by editing bits and pieces of existing text from
various sources and pasting them together, with or without original additions. Some of the existing
text may be rewritten or adapted.

Terminology Note. The terms adapt and rewrite have been used here to denote activities within a
single language. The terms are also used, with a variety of meanings, by translation theorists.

2.3 Mental editing during translation

Another activity which is similar to editing is a regular feature of professional translation. It is often
noted that translations are easier to read than their sources. That is because experienced translators of
non-literary texts tend to produce translations whose writing quality is much superior to that of the
source text. To accomplish this, they do not actually prepare an edited version of the source text;
instead, they engage in what might be called mental stylistic editing and mental structure/content
editing while they translate. Three examples (with an English gloss of the source text):

• If the source text has ‘necessary pre-requisites’, the translator will just write ‘pre-requisites’,
eliminating the redundancy.
• If the source text has ‘fish and animals’, the translator will write ‘fish and other animals’, since
fish are themselves animals.
• If the source text has ‘with a view to the need for a clear definition of the concept of violence at
the very outset of the preventive work, an inclusive definition is to be preferred’, the translator
will write something much simpler, such as ‘the first step in prevention is to define violence
clearly, and the definition should be an inclusive one’.

In each of Chapters 4 to 6, there is a short section devoted to this quasi-editing work. Just how much
such cleaning up is permissible? It’s not possible to formulate any precise answer. There is a
permissible range: some translators do more cleaning up than others, just as some translate more
freely than others. You learn the permissible range through advice from experienced translators in a
particular translation service or agency. The most common type of improvement is paring down the
convoluted, verbose sentences and eliminating the high-flown vocabulary or jargon commonly used in
bureaucratic writing to express rather simple ideas. The obvious limitation here is that clients might
wonder about a translation that is only half the length of the source text!
In some cases, large-scale structural and content editing is required while translating. This activity,
known as trans-editing.
One view often voiced is that the burden of editing should not be placed on the translator. That is,
the source text should be edited before submission for translation. In some cases, this is just a matter
of timing. The source text is going to be published and does have to be edited; the only question is
whether this will occur before or after translation. In other cases, the situation is quite different.
Within a multilingual bureaucracy, someone who is either a poor writer or not a native speaker writes
a document which will be circulated as a draft rather than a final version. Spending time and money to
edit it is not thought worthwhile by the powers that be. In these cases, the translator’s desire for a
well-written source text is likely to remain a dream.
 The work of an editor 13
2.4 Editing non-native English
In many organizations and countries, texts are very frequently written in English by people who are
not native speakers. For example, as the Web site of the South African Translators’ Institute mentions
in its definition of editing: ‘In a country like South Africa, where many people are forced to write in a
language that is not their mother tongue, the work of editors is extremely important.’ (This is a
reference to speakers of Afrikaans and of the indigenous languages such as Xhosa and
Zulu, who find themselves having to write in English.)
More often than not, texts written in English within the European Union bureaucracy are written by
non-native speakers; the Directorate General for Translation at the European Commission has a unit to
edit these writings before they are sent for translation into other EU languages.
Then there is the case of science writing. These days, scientists very often write directly in English
rather than in their own language. Many scientific and other scholarly publications insist that such
writers have their work edited by a native speaker of English prior to submission. Here is a sentence
from an article written in English by a French-speaking scientist:

Activity levels were not correlated to brains or bodies mass.

A native speaker would never use the plurals ‘brains’ and ‘bodies’ here. One has to write ‘brain or
body mass’, even though the meaning is ‘mass of the brains or bodies’.
People who attempt to write in English as a second language are often quite good or even excellent
speakers of English, but poor writers of the language. Their justified confidence in their speaking
ability may lead them to overestimate their writing ability. They make all sorts of elementary errors
(they fail to capitalize the days of the week if their native language does not do so) as well as errors in
such matters as language level (they use overly informal language that is acceptable in speech but not
in writing, or odd mixtures of formal and informal language). Also, if their native language is
historically related to English in some way (Dutch or French for example), they may frequently use
‘false friends’: a French speaker might use ‘library’ to mean ‘bookshop’ because the French word
‘librairie’ means ‘bookshop’.
The biggest problems seen in non-native English are not micro-errors such as failure to capitalize or
a wrong lexical choice. The biggest problems are failures in English composition: since the writers
were not educated in English, they may never have learned how to organize sentences in an English
manner, using English methods of ensuring inter-sentence cohesion and positioning of focused
information. They may also not have learned how to organize paragraphs, or entire arguments or
narratives, in the English manner. Instead, they will inappropriately use the sentence-organizing, text-
composing and rhetorical devices of their own language which they learned as children at school. As a
result, you may have to do a good deal of structural editing, such as reorganizing paragraph divisions.
If you edit non-native English, you may be employed directly by the author of the text, not by the
publisher. You are acting as an ‘author’s editor’ rather than a ‘publisher’s editor’, but you will still
want to know about the intended publisher’s requirements, since your task is to increase the likelihood
that the manuscript will be accepted for publication.
Ideally, editors of non-native English should be native speakers who were educated in English.
However in many countries, it is not always practical to find such a person, and the editor may be
someone with near-native ability. It is also a good idea for the editor to know the native language of
the writer, since it will then be easier to reconstruct what the writer had in mind in passages which are
obscure (the writer may have been engaging in literal mental translating from his or her own
language). Translators who work from that language are obviously well positioned to accept such
editing work. Thus, a translator who works from German to English, or at least has some knowledge
of German, will have an easier time with a passage such as the following, taken from a text written by
a German speaker about how to design roads in a way that will reduce accidents:

Some new opened roads unfortunately show accident concentrations (black spots) in a short time.
In these road sections have to be done a Road Safety Inspection to detect the deficiencies causing
accidents.

In the first sentence, a native English reader who knows no German might think that the writer is
talking about new roads, but that is not the intended meaning. They can also be old roads that have
been closed for modifications or repairs and have now been newly opened. ‘New’ needs to be
understood as an adverb modifying ‘opened’, not as an adjective modifying ‘roads’. The German
word ‘neu’ (new) can function either as an adverb or an adjective, whereas in English, the adverb
 The work of an editor 14
form ‘newly’ is needed to make the meaning immediately clear. The second sentence manifests the
so-called ‘verb second’ construction that is compulsory in main clauses in all the Germanic languages
except English. In the other Germanic languages, one says ‘yesterday saw she an elephant’: since
‘yesterday’ is occupying the first structural position in the sentence, the next position must be
occupied by the verb (‘saw’). An editor who does not know another Germanic language may well
become confused upon reading this sentence, especially since the verb ‘have’ does not agree in
number with ‘inspection’ (this is not an influence of German but just a plain old mistake on the
writer’s part, perhaps arising from the plural noun ‘sections’ that immediately precedes ‘have’). The
sentence needs to be edited to read ‘…a road safety inspection has to be done in these…’. In this
particular case, an editor will probably be able to deduce the meaning from world knowledge, that is,
by relating the words ‘sections’, ‘inspection’, ‘deficiencies’ and ‘accidents’ to what he or she already
knows about road safety. But that will not always be the case, either because the editor does not have
the requisite world knowledge, or because he/she makes an incorrect deduction from world
knowledge. At any rate, the failure of syntax to signal the meaning in this sentence will slow down the
editing process for editors unfamiliar with German word order.
Here is a case where the editor will probably not be able to work properly unless he or she knows
the writer’s native language:

To be effective, the committee should be subjected to the support of local management.

The ideas expressed by ‘subjected to’ and ‘support’ do not fit together. However, if you know the
writer’s native language, French, you will recognize a word that may have inspired the English,
namely ‘assujetti’. This word is indeed often translated by ‘subject(ed) to’. However, it also has the
sense ‘secured’, as when a boat is secured to a dock. By extension then, the committee here can only
be effective if it has a secure tie to management, which supports its work. Unfortunately, this meaning
cannot be borne by English ‘subjected to’. Even if ‘subjected to’ is changed to ‘subject to’, it still
sounds like local management is constraining rather than assisting the work of the committee.
If as the editor you are not familiar with the writer’s native language (and sometimes even if you
are), there is an important technique you can use to identify the meaning of obscure passages.
Consider this baffling passage from a text about weather observations in developing countries:

…the difficulty of maintaining observation sites with recordings homogeneous with development

What does ‘recordings homogenous with development’ mean? In such cases, the best approach is to
hope that the writer repeats the point using a different wording later in the text. You should therefore
keep an eye out for such wordings. In the case under consideration, a later passage reads:

With economic and social development, it is difficult to maintain observation sites in operation,
protect them from deterioration and maintain homogeneous data series.

Here are some general things to watch for in non-native writing. The writers may not know that a
word or turn of phrase is very formal, very informal, old- fashioned or infrequent. They may not know
that a certain phrasing may be viewed as impolite or, conversely, overly polite. Languages differ in
how direct one can be: in the writer’s language, one might need to write ‘we wonder if you might not
possibly send us a letter’ whereas in English that seems overdone; ‘send us a letter’ would be normal,
or at most ‘please send us a letter’. In a text that praises an individual (an announcement of a
promotion for example), it may be customary in the writer’s language to keep repeating praise
expressions like ‘absolutely outstanding’ and ‘incomparable’. In English, such effusiveness will seem
insincere, defeating the purpose of the text. Non-native writers may also not know about genre
conventions: a French speaker writing up the minutes of a meeting may use the present tense, not
knowing that minutes are written in the past tense in English (not ‘Mary suggests postponing the
decision’ but ‘Mary suggested…’).
Aside from such ‘rhetorical’ editing, you may also have to do some content editing if the writer has
made cultural assumptions that will not be shared by readers or has wrongly assumed certain kinds of
factual (for example, geographical) knowledge.
One possibly important question that arises when editing non-native (and sometimes native)
English is: which regional variety of the language should the editor adopt? The answer is fairly
simple. If the readers will be speakers of your own variety, use that. If the readers will be
international, then avoid usages which are specific to American, British or some other local variety.
 The work of an editor 15
Like any other text, one written by a non-native can be stylistically edited or simply copyedited. In
some cases, non-native English which is very badly written may be treated like the output of machine
translation (even though the types of error will be quite different); the text will be edited just enough
to make it intelligible.
A final word of warning: when editing non-native English, the risk of accidentally changing
meaning is far greater than with native writing. The problem is not the truly opaque passages where
you are simply baffled about the intended meaning. In these cases, you can consult the author (if
available) or a subject-matter expert, or place question marks around the passage. The real problem is
that with non-native English, much more of your attention will be devoted to language and style
matters, and as a result, you will more frequently find yourself making guesses about what the author
intends without realizing that you are guessing, and that other meanings are possible. Ideally, your
edited version will go back to the author, whose reading knowledge of English will be much better
than their writing ability, and they may be able to spot your incorrect guesses about what they meant.
Another possibility is to have both the unedited and edited versions of the text read by a subject-
matter expert, who may be able to spot places where you have written something which is unlikely to
be what the author intended. In the end, however, you are bound to make more editing mistakes with
non-native than with native writing.

2.5 Crowd-sourced editing of User Generated Content

With the advent of social media and Wikipedia in the first decade of the 21st century, it became easy
for anyone to make their own writings public without going through a publishing company’s editorial
process. Such User Generated Content, as it is called, may nevertheless be edited. Wikipedia articles
can be edited by any registered Wikipedia user, with a few restrictions. Editing is thus crowd-sourced
rather than being performed by an official ‘editor’. Any article not conforming to the posted writing
guidelines can be marked as needing correction: a box appears at the top of the screen stating that the
article is in need of certain kinds of correction.

2.6 Degrees of editing

Professional editors do not apply equal editing effort to every text. In the first place, an editor may
only have time to read through a particular text once, doing all four kinds of editing simultaneously,
or perhaps just copyediting to eliminate the most obvious errors. With other texts, it may be possible
to perform two or more separate edits. Additional factors are the nature of the publication and the
reputation of the publisher: the editors of a scientific journal whose publisher wishes to be known
internationally for readability and freedom from factual error will need to edit very carefully.
Since editors may well have several jobs going at once, they need to consider whether all merit
equal attention; there is not much point in spending vast amounts of time on the stylistic editing of a
text which is relatively ephemeral, like a report which only a limited number of people within an
organization will look through, fairly quickly, and then discard. Readers will likely have a higher
tolerance for uncaught errors in this type of text.
Another important consideration is the reaction of writers. If a freelance editor hopes to work for a
writer at some time in the future, it will be a good idea to keep changes to a minimum. The editor may
have thought of a brilliant new wording for a sentence, but if the writer’s sentence is satisfactory, it
may be best to leave it. Writers will be more inclined to work with an editor who appears to be
helping them get their message across, and less inclined to work with one who appears to be
competing with them as a substitute writer.
Some books on editing refer to a specific number of degrees of editing. For example, a distinction
may be made between light, medium and heavy editing, each being appropriate to a particular type of
job. As many as nine degrees may be distinguished, with particular tasks specified for each degree.
This approach is a kind of summary of the experience of people who have been editing for a long
time. Terms such as ‘light’ and ‘heavy’ are a bit misleading since they may seem to refer to the
number of changes the editor makes—something which of course depends on the quality of the text
submitted for editing. The actual intent of the terms is to refer to the number of aspects of a text which
are checked: in a light edit, you might just check for grammar, spelling and punctuation errors.
Another interesting way to look at the issue of degrees is to consider the writing effort required of
the editor, as opposed to the reading effort.

Here are five degrees of increasing writing effort:

 The work of an editor 16
1. make no changes
2. make changes for correctness
3. make changes for intelligibility
4. make changes for writing quality (‘style’)
5. rewrite

Left aside here are structural editing issues such as reparagraphing as well as content editing issues
(unless one wants to consider intelligibility as a content issue).
Level 1 involves no writing by the editor, who simply reads a sentence, thinks about it, and then
decides to do nothing. Level 2 requires some writing effort but less than at level 3: it takes more effort
to make an unintelligible sentence intelligible than it does to make it correct, since correcting (i.e.
making changes during copyediting) is simply a matter of applying rules. As for level 4, it will usually
take a very great effort to achieve a high level of writing quality if the author has failed to achieve it.
That is because the effort may well be required over much of the text rather than in just a few
sentences here and there, as is typical with levels 2 and 3 (if almost every sentence needs to be
corrected or made intelligible, then the text is probably not worth editing!). Finally, the degree of
writing effort required at level 5 will vary with the editor: some editors may find it easier to cross out
a sentence and write an entirely new one (whether for reasons of intelligibility or for reasons of style)
than to achieve the desired quality by making small changes here and there in a sentence.

2.7 Editing procedure

When you are just learning, the first step is to master each of the types of editing. This will be done
here through exercises in individual aspects of copyediting, stylistic editing, structural editing, content
editing and consistency checking. Thus, you will be practicing punctuation editing in one exercise and
spelling correction in another; then you will practice doing all aspects of copyediting at once.
The next step is to do a full edit of a text several pages long. Do this in separate run-throughs of the
entire text. Begin with the structural editing of the whole text, then proceed to check the content for
factual, mathematical and logical errors, then do the stylistic editing, then the copyediting and then a
consistency check. At the very end, run Spellcheck to catch any typing errors you may have missed
(or introduced yourself while editing!).
The third step might be to try a combined edit in which you correct errors of all types as you come
to them. Move back and forth in the text as necessary. When you’ve finished, put the result aside.
Then, a few days later, correct the same text using separate edits. Compare the results. Look in
particular for the following in your combined edit:

• errors you missed;

• wasted work: copyediting changes that were overridden by stylistic or structural changes;
• errors you introduced.
Alternatively, work with a partner. You do a combined edit while your partner does separate edits.
Exchange copies of your edited texts and compare the results.

Practice
Exercise: Copyediting non-native writers
Your instructor will give you a document written in English by someone who is not a native
speaker of English, but whose native language you know. Find and correct linguistic errors.
3. Copyediting

In this chapter, we’ll look at copyediting under five headings: ‘house style’; spelling; syntax and idiom;
punctuation; correct usage. Checking for consistency, which is also an aspect of copyediting.
Copyediting may be defined as checking and correcting a document to bring it into conformance with pre-set
rules. The second last word of the sentence you are now reading must be ‘says’, not ‘say’, because there is a
rule in the grammar of standard written English that says so. (Several forms of spoken English omit this –s.) In
the case of correct usage, the rules to be enforced are controversial, and involve matters of authority, ideology
and tradition. In the case of punctuation, the rules are often not clear-cut. The sections dealing with these two
topics are therefore rather lengthy.
Copyediting requires close attention to small details; you can’t do it properly if your mind is on other things.
Sometimes you may find it a relief from the more demanding (less clear-cut) aspects of writing and translating
work, and sometimes you may get satisfaction from those copyediting decisions that do require some thought.
But unless you derive pleasure from correcting other people’s errors, or creating ‘order’ out of untidiness, you
may find this necessary task somewhat unattractive.
Ultimately, you may discover that you can combine copyediting with stylistic, structural and content editing,
but while you are still learning, you should do it separately. Perhaps try thinking of it as a game: How many
mistakes can I find? Can I score better than last time?
Copyediting is line-by-line, ‘micro-level’ work. It is therefore done after the author and editor have
completed ‘macro-level’ changes to the content and structure of the text. There is no point copyediting a
paragraph which will later be deleted.
Copyeditors also check certain typographical and layout features, especially for consistency: Are all
paragraphs indented? Are all headings bolded? However, some of these features are really a matter of stylistic
editing or structural editing. For example, italics are commonly used to indicate to the reader that mental stress
should be placed on a particular word. This is a matter of readability, and more specifically, smoothing—a
stylistic matter which is dealt with in Chapter 4.
Similarly, headings may be underlined and indented as a way of signaling the structure of an argument to
readers.

Terminology notes. The term copyediting is used by some editors to include stylistic editing. Indeed, some
editors use copyediting to include fact-checking as well as any other tasks which are performed on a ‘line-
by-line’ basis. These are all ‘micro-editing’ tasks, as opposed to such ‘macro-editing’ tasks as rearranging
the order of presentation of topics in a document.
Where British and American terminology differ, this book uses the American term: ‘period’ instead of
‘full stop’; ‘parenthesis’ instead of ‘bracket’; ‘typo’ instead of ‘literal’.

3.1 House style

The word ‘style’ is unfortunate in that style sheets deal with mechanical matters, whereas ‘stylistic editing’, to
be discussed in the next chapter, refers to matters which are very far from mechanical. Here are some of the
instructions from the style sheet prescribed by the editors of the Routledge journal Translation Studies:

• In general spell out numbers under 100; but use numerals for measurements (e.g. 12km) and ages (e.g. 10
years old). Insert a comma for both thousands and tens of thousands (e.g. 1,000 and 20,000).
• Keep capitalization to a minimum. When possible, use lower case for government, church, state, party,
volume etc.; north, south, etc. are only capitalized if used as part of a recognized place name, e.g. Western
Australia, South Africa; use lower case for general terms, e.g. western France, south-west of Berlin. Books
or films referred to in body of text have capital letter on all main words.
• Either UK spelling (but ‘z’ rather than ‘s’, e.g. ‘modernization’ not ‘modernisation’) or US spelling can be
used, as long as it is consistent. UK punctuation conventions will be applied throughout, for consistency.
 The work of an editor 18
In addition to (or instead of) a style sheet, editors will often direct writers to follow a particular published style
manual or guide, such as the Chicago Manual of Style. These manuals may be hundreds of pages long. A style
manual gives instructions on a wide variety of matters, including spelling (advertise or advertize?),
capitalization, hyphenation, numerals (eight days or 8 days?), Latin or English plurals (fungi or funguses?),
acronyms, use of italicization and bolding, presentation of quotations, footnotes and reference works, treatment
of place names (Montreal or Montréal?), transliteration of names from languages that use a different script,
what if anything to do about non-gender-neutral language, and much more.
Sometimes style manuals give a choice of approach, and simply demand consistency (e.g. spell numbers up
to nine, then use figures from 10; or spell up to ninety-nine, then use figures from 100). Note, by the way, that if
you follow the first of these two rules about numbers blindly, you may end up writing sentences like:

There was one case of 11 people in a car and 12 cases of nine in a car.

where form does not match meaning: the number of people should be either ‘11…9’ or ‘eleven…nine’.
Style manuals are published by governments, newspapers, university presses and editors’ associations. A few
are listed at the end of the chapter. You may find it useful to compare manuals for English with manuals for
your other languages, noting differences in matters such as comma use.
Style manuals and style sheets help create a distinctive institutional voice and visual image for a publication
—a ‘house style’. They also create a degree of consistency in journals, magazines and collections of articles,
where several different authors are contributing to a single issue or book. Once the contribution arrives, it is up
to the copyeditor to check that the instructions have been followed.
In multilingual countries or institutions, style manuals may give instructions on how to handle wordings in
another language that are being reproduced. Unfortunately, writers are usually left to their own devices when
they decide to write a few words of their own in another language or translate a quotation. In Canada, English-
speaking journalists frequently write short phrases in French but don’t bother checking the gender of nouns or
the positioning of French accent marks, and editors fail to make corrections. As for translations produced by
journalists, these tend to be extremely literal to the point of being unidiomatic. The British newspaper The
Guardian once quoted France’s president as saying, about Brexit, that what was needed was a ‘retirement
agreement’, which an editor should have corrected to ‘withdrawal agreement’.

3.2 Spelling and typing errors

Why should a text be correctly spelled and be free of typing errors? Even asking this question may seem odd.
The need for correct spelling was drummed into us at elementary school, and we may never have given a
moment’s thought to its rationale.
Spelling errors are bad because of the effect on the reader. Misspellings and typing errors produce a very bad
impression. They suggest that the author and editor are sloppy thinkers, and that the publisher tolerates
carelessness. As a result, readers may lose confidence in the actual content of the work. Of course, it does not
follow logically that if there are spelling errors, there must also be errors in the facts or arguments presented,
but subconsciously at least, that is what readers will suspect. Misspellings and typos are also distracting, and
therefore they slow down the reading process. Finally, typos can directly affect meaning, both when keys get
pressed in the wrong order (have you read about the artist who fearlessly attacked scared cows?) and when the
wrong word is transmitted from the mind to the fingers: there is a big difference between adopting a plan and
adapting one, between having an aptitude and having an attitude.
The work of searching for misspellings and typos is greatly facilitated by the Spellcheck utilities included
with word processing software. If you are not already in the habit of using Spellcheck, then get into the habit
immediately. There can be no excuse for this type of mistake. Run Spellcheck after all other changes have been
made.
Spellcheckers do have weaknesses and pitfalls. One of the most notable weaknesses is proper names; you
will need to independently verify that the names of people and places are properly spelled. Consider this
sentence from a funding proposal:
 The work of an editor 19
Our health centre worked in partnership with Merck Frost.

Since ‘frost’ is a correctly spelled English word, it would be easy to let this sentence slip by unchecked. In fact,
the correct spelling of the former name of this pharmaceutical corporation is Merck Frosst. You could easily
find this spelling by entering ‘Merck Frost’ in your web browser.
One aspect of English spelling is highly variable. Which is correct: lifestyle, lifestyle or lifestyle? The answer
is: all three, depending on which dictionary you consult. Also, usage may vary with the field; for example, the
Canadian Government’s terminology bank Termium says that ‘caseworker’ (one word) is correct in the field of
social services but ‘case worker’ (two words) is correct for the person who works with inmates in a
penitentiary. If you are doing freelance editing for a corporation or government ministry, documents on the
subject of your text may reveal your client’s practices regarding common compounds.
If your style sheet prescribes a particular dictionary, then the compounding problem will often be solved.
However compounding is a highly productive process in English; that is, writers can make up new compounds
at the drop of a hat, and these will not appear in your prescribed dictionary. The easiest principle to follow here
is consistency: make your choice for each compound, and then make sure you stick to it throughout the text.
You can also try using Google to investigate the relative frequency of the two-word versus one-word treatment
of a compound. See Chapter 7 for a discussion of problems in the use of Google for such purposes.
In general, there is a progression over time from open spelling (two words) when a compound is first
introduced, through hyphenation, to solid spelling (one word) as a compound becomes established in the
language. The Americans tend to move through this progression more quickly than the British. Hyphens are
less common in U.S. English; words written with a hyphen in Britain will tend to be written solid in the United
States or (less often) as two words. (There is also a trend, more advanced in the U.S., towards omitting the
hyphen in prefixed words like ‘coordinate’, ‘cooperate’ and ‘preeminent’).
A final point, concerning another use of hyphens: if your style sheet calls for breaking long words at the ends
of lines, note that American practice is to break at phonetically natural points (‘trium-phant’) whereas British
practice tends to draw on morphological considerations (‘triumph-ant’). Check to see which principle the
automatic hyphenator in your word processor follows. Note that some automatic hyphenating utilities may
produce wrong or even bizarre results (bat-hroom).

3.3 Syntax and idiom

If the text you are editing is written by a reasonably well-educated native speaker of English, and is not a
translation, there is a good chance that it will be syntactically correct and idiomatic. That is, it will not contain
sentences like

He washes frequently his teeth, sometimes after every dining.

in which the adverb ‘frequently’ is in a position it cannot occupy, the word combination ‘wash teeth’ is
unidiomatic, and the word ‘dining’ is used in a meaning it does not have. These are errors of a kind which
native speakers normally do not make. However, there are several exceptions:

1. People attempting to write in fields with which they are not familiar may have problems with the specialized
phraseology of that field. Similarly, if you are just beginning to edit in a field with which you are not yet
familiar, you must be careful not to replace the customary phrasings of that field with more universal ones.
For example, when editing a work in the field of meteorology, you might come across the term ‘summer
severe weather’ and you might be tempted to normalize the word order to ‘severe summer weather’. That
would be a mistake; the phrase is correct as it stands, ‘severe weather’ being a defined concept in this field.
When a severe weather event occurs in summer, it is ‘summer severe winter’; when it occurs in winter, it is
‘winter severe weather’.
2. Since the advent of word processors, mechanical slips during composition often create serious errors in
sentence structure.
 The work of an editor 20
(a) There may be word missing (or an one unwanted extra word) if during self-editing the writer
pressed the delete key once too often (or not often enough). Did you notice the missing word and
the extra word in the previous sentence?
(b) Cut-and-paste or click-and-drag operations, during which a passage is moved within a document or
pasted in from another document, often produce imperfect transitions between the pasted passage
and what surrounds it. The structure of the pasted portion may not fit into the sentence properly, or
there may be a word missing at the boundary of the pasted portion, or there may be an extra word—
commonly a double word. (Spellcheckers catch double words, but be careful not to automatically
correct the sequence ‘had had’; ‘he had had a bad time’ may be an incorrect doubling, or it may be
correct if the sentence calls for the pluperfect tense of ‘have’.)
(c) Partly amended sentences such as the following are now common:
It would be appropriate for computational terminology researchers would do well to investigate the
potential usefulness of existing knowledge-engineering technology.
The writer decided to add ‘would do well’ but forgot to delete ‘it would be appropriate for’. In the days of
typewriters, such sentences were hardly ever produced. Changing the structure of a sentence once it was
down was a very time-consuming (and messy) operation. As a result, people either spent more timing
planning their sentences, or else they made changes in handwriting during a separate self-editing phase,
when their attention was on the sentence as a whole. (Then someone else—a typist—would prepare an
entirely fresh copy.) Nowadays, it is very easy to make changes as you write with a computing device, and
there is a tendency to focus only on the bit you are changing.
3. Writers sometimes make present-tense verbs agree in number with the nearest noun whose combination with
the verb makes sense:
The legacy of the social service cutbacks of previous governments remain with us.
4. The mind sometimes retrieves the wrong word from the mental store:
Bank machines, photocopiers and central heating are a few examples from an almost infinite list of
technologies and products that are an inexhaustible feature of modern life.
Here, ‘inexhaustible’ (perhaps triggered by ‘infinite’ earlier in the sentence) does not make much sense; the
writer may have meant ‘indelible’ (in the sense of ‘permanent’) or ‘irreducible’. Words that sound like the
word having the intended meaning (‘inedible’ in the case ‘indelible’) may also be retrieved. And sometimes
two words or phrases are retrieved at once:
Beyond a question of a doubt, this enhanced our cynicism in parliament as an effective instrument of
government.
Here ‘beyond any question’ and ‘beyond a shadow of a doubt’ have been retrieved together.
5. When people are translating into their native language, they often write ungrammatical and especially
unidiomatic sentences, under the influence of the source text. When the source language is one whose
vocabulary includes many cognates of words in their native language (e.g. any Romance or Germanic
language in the case of translation into English), translators may use words in meanings they do not have
(‘he was invited to give a conference’ for French ‘conférence’, which often means ‘lecture’ or ‘talk’). Such
unidiomatic usages may also appear in the original writing of people who work in a multilingual
environment. If the readers also operate in such an environment, there may be no problem. But if they do
not, then the editor must take action.
These then are the syntax and idiom problems found in the work of well-educated native speakers. But
you may also find yourself having to edit writing by people who are not well educated or not native
speakers.

Terminology note: The word idiomatic is used in this book to cover a variety of phenomena which are
sometimes distinguished: collocations such as ‘brush one’s teeth’, prepositional idioms such as ‘depend
 The work of an editor 21
on’ and phrasal verbs such as ‘put up with’; set phrases such as ‘not on your life’; or clichés such as
‘please be advised that’. Similarly, expressions like ‘wash one’s teeth’ or ‘depend from’ are all
described as unidiomatic. The term idiomatic is also sometimes used in a broader sense to refer to ‘the
way we say things in our language’, that is, to refer to stylistic/ rhetorical preferences such as the
English preference for the plural rather than the singular in generic statements (‘students must obtain a
mark of C in order to pass’ rather than ‘the student must…’). Here copyediting shades into stylistic
editing.

Syntactic change and variability

Syntax and idiom are not eternal; they change as the generations pass. As a result, what older editors see as a
clear error may be perfectly acceptable to educated younger speakers. Once innovations have begun to spread,
editors have to decide whether they are now acceptable or still have to be edited out. A recent syntactic
innovation is that in some versions of standard written English, the pronoun ‘which’ is no longer acceptable in
restrictive relative clauses. Editors will automatically change ‘the term bank which we use’ to ‘the term bank
that we use’. A common type of innovation in English is the conversion of adjectives and nouns to verbs:
‘pockets of downtown that are resurging as fashionable addresses’; ‘an escaped convict upheaves the lives of a
businessman and his wife’. These could be one-off innovations by the writer, but you may want to check the
most recent edition of your dictionary. Some publications insist on removing anything that is not recognized in
an authoritative dictionary (unless it’s in a quotation); others take a more relaxed attitude and allow
considerable room for authors to innovate.
Quite apart from innovation, notions of what is acceptable always vary somewhat from person to person. A
writer may have used a structure or word combination which you find odd or impossible; but that does not
mean it is wrong. For example, the pronoun ‘whom’ is slowly disappearing from the written language; many
find ‘the man who we met on the bus’ perfectly acceptable; others do not. Some people find the use of ‘whose’
with inanimate antecedent (‘the book whose cover was so striking’) perfectly acceptable; others do not. I once
discovered, through a Google search, that one can say, ‘underhand deal’; previously I thought it had to be
‘underhanded’. It seems that the two forms vary with geography, with most hits for ‘underhand’ coming from
the UK, Africa, India and Pakistan, and most hits for ‘underhanded’ coming from the United States and
Canada.
A further danger with using yourself as an authority is that you may end up introducing your own personal
linguistic idiosyncrasies into a text. A few years ago, I discovered to my surprise that the expression ‘she
favours her right arm’ does not at all mean ‘she tends to use her right arm rather than her left’ but rather the
exact opposite: ‘she avoids putting too much strain on her right arm, by using her left arm instead’; her right
arm is ‘favoured’ by being given a rest! A further discovery: I had always thought that ‘fulsome praise’ means
abundant praise, but recently I discovered that for many people it means excessive praise.
A related problem is that your mental lexicon may have changed over the years under the influence of
translating from this or that source language. As a result, the meanings you attribute to certain words and
phrases of your first language may be rather different from their meanings in the minds of non-translators.
It’s always important to keep in mind that you are working on someone else’s text. You must not replace
their voice with yours. This is especially a danger with older editors. You may not care for certain recent
innovations, such as the use of ‘demographic’ as a noun (‘the problems of marketing to a younger
demographic’), but that is not a reason to prevent others from using them.
Acceptable syntax also varies with genre. For example, in cookbooks, ellipsis of articles and pronouns is
common: ‘Slice onions. Then saute on high heat’ rather than ‘Slice the onions. Then sauté them on high heat’.
Such ellipsis is also often accepted in lists of points, which may diverge in other ways as well from the rules
that apply to continuous prose (e.g. no capital letter at the beginning or period at the end).
To check idiomaticity, consult the combinatory dictionaries listed at the end of this chapter. For example, if
the text you are editing has the phrase ‘sorry choice’, and you are uncertain whether this is an idiomatic word
combination, the Benson dictionary will confirm that it is (you will learn that a choice can be bad, sorry, wrong,
careful, difficult, good, happy, intelligent, judicious, wise, random or wide). The dictionaries by Wood and by
 The work of an editor 22
Cowie & Mackin are entirely devoted to prepositional idioms, which often pose problems even for educated
native speakers (report to or for work? compared to or with Paris?). To check the syntactic structure used with a
word, try the Collins Cobuild English Dictionary, which always gives a full-sentence example for each sense of
a word; it’s available free online at www.collinsdictionary.com. To check whether a word has a particular
sense, or can be used in a certain syntactic structure, or in combination with a certain other word, you can also
try on-line concordances, some of which are free or have free versions, such as the American National Corpus
at www.anc.org or the British National Corpus and Corpus of Contemporary American English at
corpus.byu.edu. Be aware however that a corpus does not tell you what is ‘correct’, only what is actually in use
by native speakers and writers. On grammar-checking software; on concordances.
A final type of variability that may be important for editors in some geographical areas is dialect differences.
Standard languages are historically based in certain local forms of speech as opposed to other local forms,
which are deemed ‘nonstandard’. As a result, some people’s natural syntax and idiom may be unacceptable in
writing, and children are taught to avoid using them. Thus, to many native speakers of English in Ireland, the
following are all perfectly natural sentences, inherent in their language: ‘Is this car belonging to you?’, ‘They
were after leaving a gym across the street’, ‘It does be colder at nights’. Even if such syntax is not currently
acceptable in formal writing, that could change. In a given geographical area, people who use a non-standard
dialect in their everyday speech may want to start using that dialect in books and newspapers (and not just in
quotations for ‘local colour’). At that point, editors have to decide what to do.

3.4 Punctuation
Punctuation, in a narrow sense, includes the familiar marks: the comma, the period, the quotation mark, the
dash and so on. In a broader sense, punctuation includes a variety of other indicators that provide guidance to
readers: the space between words, the indentation introducing a new paragraph, the capital letter that begins a
sentence.

didyouknowthatatonetimetextswereunpunctuatedthere
werenodemarcationsbetweenwordssentencespartsofsentencesorparagraphs.

A few aspects of punctuation (paragraph divisions, some uses of the comma) are really stylistic or structural
matters and will be considered in the next two chapters.
Some very common errors are opening a parenthetical remark but forgetting the closing parenthesis, and
inconsistent punctuation in point-form presentation.
The rules governing punctuation are not as clear-cut as those governing spelling. The British and U.S. rules
differ somewhat, for example in the positioning of closing quotation marks. Also, while sentences inside
paragraphs have a highly standardized punctuation regime (initial capital; period or question mark or
exclamation mark at the end), words in other parts of texts do not. Section headings, points in lists, captions of
graphics and column titles on tables may take a wide variety of regimes: all keywords capitalized, first word
capitalized, or no words capitalized; various punctuation marks or no punctuation mark at the end.
Most uses of the English comma are not bound by rules at all. Using commas well calls for thought. There
are two main types of variation:

• Some writers use many commas, others use few.

• Some writers use commas to mark speech features such as pauses and emphasis, others to mark the
boundaries of syntactic structures.

Regarding the second of these differences, it seems that there are three principles upon which comma use in
English has historically been based:

(A) When writing a sentence, use commas to indicate where someone should pause when reading the text aloud.
(B) Place commas at the boundaries of the syntactic constituents of the sentence.
 The work of an editor 23
(C) Imagine the sentence being spoken, then place commas to reflect your mental pauses or emphases (which
may or may not occur at syntactic boundaries).

Approach (A) was historically first. Until a couple of hundred years ago, most literate people did not read
silently. Aside from documents such as records (tax lists, property titles and so on), writing was a sort of script
for reading aloud, either to oneself or to others. Punctuation indicated places to breathe, or to pause for
rhetorical effect. Commas, colons and periods seem to have indicated increasing lengths of pause.
During the late 18th century and throughout the 19th century, approach (B) was widely advocated, though
the older rhetorical tradition never died out. In this approach, thought to be suited to silent reading at great
speed, commas help the eye by picking out syntactic structure, and thus clarify meaning. Finally, during the
20th century, approach (C) grew in importance, though it has not yet displaced the syntactic principle. The
upshot is that people often use a combination of approaches (B) and (C).
Here’s a very simple example of the difference between the two approaches:

approach (B):
Marilyn was the best translator available and, as soon as she returned from holiday, she was chosen to head
up the prestigious project.

approach (C):
Marilyn was the best translator available, and as soon as she returned from holiday, she was chosen to head
up a prestigious project.

In the first sentence, the commas visually mark off ‘as soon as she returned from holiday’ as a clause
interrupting the conjoined structure ‘Mary was…and…she was…’. In the second sentence, the comma reflects
how someone might have mentally imagined speaking the sentence. A further point of interest here: if you ever
had occasion to read the first of these two sentences aloud, you might revert to approach (A) and use the
commas as indicators of when to pause, or perhaps lower the voice. However, this would be a case of
‘pronouncing the commas’, as opposed to using them to reflect a prior imagined speaking, for the position after
‘and’ is just not a natural place to pause during speech.
An important point about approach (C) is that sometimes the addition of a comma to indicate a mental pause
has the effect of adding attitudinal meaning. Consider:

He was apparently willing to support you.

He was, apparently, willing to support you.

The second sentence expresses a bit of surprise, or casts doubt either on ‘his’ motive for supporting ‘you’ or on
whether ‘he’ really was willing, as ‘you’ may have alleged.
More generally, the choice of a comma rather than some other punctuation mark (or no mark) can be used to
reflect varying degrees of some attitudinal feature:

I went to his house and I found him there.

I went to his house, and I found him there.
I went to his house. And I found him there.

As we move from the first sentence to the third, an increasing degree of surprise at ‘his’ presence ‘there’ is
expressed. Now, according to some versions of punctuation rules (perhaps those you learned at school), the last
two of the above sentences are impermissible. However, if you rigidly exclude sentences beginning with ‘and’,
you will not be able to obtain the effect achieved in the third sentence. Indeed, you will often find that if you
follow the most rigid version of rules, in any area of language, you will reduce the number of semantic options
open to you. Worse than that, if you try to implement rigid rules with word processor tools, you may create a
disaster. One editor decided that the word ‘however’ must always be followed by a comma and implemented
 The work of an editor 24
this decision using the Search & Replace All function. The result, in one passage, was a sentence which began
‘However, much you enjoy translating…’.
A final important point about the two conflicting principles (B) and (C): some uses of the comma to reflect
mental pauses are still quite strictly prohibited despite the rise of approach (C). If you are editing the work of
people with relatively poor education in the standard written language, you may find such sentences as the
following, from a report written by a health and safety officer:

The beeping of the alarm at an interval of thirty seconds or a minute, is a warning you should attend to. It
means the batteries are dying, you need to replace them with fresh batteries.

The first sentence has a comma functioning to separate the subject of the sentence from the predicate. Although
people do often pause at the subject-predicate boundary in speech, this use of commas ceased to be permissible
in the written language during the 19th century. The second sentence has a comma where there should be a
period or semi-colon. This usage is particularly common in the writing of less well-educated people. A written
sentence is not a natural unit corresponding to a structure of the spoken language, and as a result it takes
children some time to learn where to place periods. Some people never succeed, and you may find yourself
having to correct their errors.
Turning now to the second type of variation in comma use, let’s look briefly at heavy versus light
punctuation. The heavy punctuation of the 19th century was associated with the use of commas to mark
grammatical boundaries. Over the course of the 20th century, punctuation became lighter, especially in the U.S.
This was partly because sentences became shorter; obviously, short sentences usually do not need as many
internal boundary markers as long ones. But in addition, commas became optional at many boundaries. In the
lightest use, a comma will only appear when absolutely necessary to avoid misunderstanding. Four of the six
commas in this paragraph you are now reading could be eliminated.
When you are not sure whether to use a comma, do not agonize. Avoid the situation Oscar Wilde describes:
‘I was working on the proof of one of my poems all morning and took out a comma. In the afternoon, I put it
back.’ Instead, follow this handy rule of thumb: If in doubt, leave it out.

3.5 Usage
Copyeditors are widely expected to make texts conform to something variously called ‘correct usage’, ‘good
grammar’, ‘correct English’ or ‘proper English’. This is something quite different from the problems of Syntax
and Idiom discussed earlier. There, the task was to make sure the text conforms to rules which are inherent in
the spoken language, and don’t need to be stated or taught to children (e.g. the possible positions in a sentence
of an adverb like ‘frequently’). Occasionally people fail to observe these rules (for example in long sentences
with complicated structures, or when translating) but there is no debate about them; as soon as an error is
pointed out, people immediately recognize it as an error. No native speaker, of any educational level, thinks ‘he
washes frequently his teeth’ is acceptable English.
Correct usage, on the contrary, is a matter of debate. It is overtly prescribed in publications by various
‘authorities’ as well as in angry letters to the editor by private individuals. These prescriptivists, as I will call
them, condemn certain usages as wrong, but many people do not agree and simply ignore the various
prescriptions in their own writing.
Webster’s Dictionary of English Usage defines usage as ‘a collection of opinions about what English
grammar is or should be, about the propriety of using certain words and phrases, and about the social status of
those who use certain words and constructions’. These opinions are voiced with a view to standardization, that
is, the elimination of variants. If some people write ‘it’s me’ and others write ‘it’s I’, only one—in this
approach—can be right; the other must be proscribed. It’s worth noting that this idea—there is only one right
way—is not as widely accepted in the English-speaking world as it is in some other language communities. A
common view among English-speaking writers is that one should certainly consider all opinions regarding a
point of usage, but each person should then decide for themselves what is best.
 The work of an editor 25
Now in every speech community, variants are constantly appearing in the spoken language. People in one
geographical area start to pronounce a word differently; members of the younger generation start to give a word
a slightly changed meaning. Obviously, there are limits to such variation if communication is to be maintained.
As a result, every language community has a process, operating below the level of consciousness, whereby
some variant usages are rejected, and others accepted. However, greater variation can be tolerated in speech
than in writing. Written language needs a higher level of standardization so that texts written at one time and
place will be understandable at other times and places, possibly by readers not known to the author.
The question is: what degree of standardization should be enforced and on what principle should a proposed
standard be accepted or rejected? More specifically for our concerns in this book: what should be the attitude of
editors to matters of correct usage?

Consider the following sentences and ask yourself whether you would make any corrections in them:

(1) If everybody minded their own business, the world would go round a good deal faster than it does.
(2) A flock of birds were alighting here and there around the field.
(3) Hopefully this text will be translated by tomorrow.
(4) The volume can be increased by turning the blue knob.
(5) Their mission is to boldly translate what no one has translated before.

There is nothing in any of these sentences that violates any syntactic rule inherent in the English language. Yet
they all contain features that continue to be condemned in angry letters to the editor. According to some people:

In (1), their is wrong because everybody is singular; it has to be his.

In (2), were is wrong because flock is singular; it has to be was.
In (3), hopefully cannot be used as a disjunctive adverb; it’s always a manner adverb, as in ‘he looked at me
hopefully’; the sentence should read ‘it is to be hoped that this text…’.
In (4) the subject of turning must be the same as the subject of can, but it is not the volume that will be turning
the knob; the sentence must be reworded to ‘You can increase the volume by turning…’ or ‘The volume can
be increased if you turn…’.
In (5) boldly must be moved because it is ‘splitting’ the infinitive ‘to translate’.

For an editor, the first thing to notice about all these complaints is that they have little to do with successful
communication. None of these sentences are hard to read and none will be misunderstood.
A second point worth noticing is that prescriptions sometimes mask ideological agendas. Consider sentence
(1). Those who demand ‘everybody minded his own business’ instead of ‘their own business’ claim that this is
a matter of grammar (‘everybody’ is grammatically singular), but there is obviously an ideological agenda at
work as well—a resistance to gender-neutral language. In fact, the use of their as a gender-neutral pronoun goes
back to the 14th century when the singular antecedent is indefinite, as in sentence (1), or generic (‘if the student
wishes to receive their grade sooner…’). This usage was not proscribed until the late 18th century. Recently,
‘they’ has also come to be used with definite singular antecedents either to avoid attributing gender (‘The
translator I asked to work on this report said they wouldn’t have time’) or because the person in question does
not wish to be referred to with gendered pronouns. You may have noticed that ‘they’ and ‘their’ are used with
singular antecedents throughout this book. For more on this, see the Wikipedia article ‘Singular they’.
The 18th century was a time when many notions of correct usage were first formulated, and Latin was often
used as the model for what proper English should be. This is the origin for example of the rule prohibiting so-
called split infinitives (see sentence 5 above). If a Latin sentence containing an infinitive is turned into English:

Nec quicquam est philosophia, si interpretari velis, quam studium sapientiae. (Cicero)
Philosophy is nothing other—if you wanted to translate—than the study of wisdom.
 The work of an editor 26
the part of the English corresponding to the italicized Latin infinitive has two words (to translate).
Grammarians therefore decided, taking Latin as a universally valid model, that in English the infinitive is two
words long (‘to X’). Since obviously no adverb can be placed in the middle of the Latin infinitive, it ‘followed’
that no adverb should be placed between the two parts of the English infinitive. Expressions like ‘to boldly go’
were proscribed, even though they had been in use for centuries in written English. Split infinitives have in fact
never ceased to be in widespread use; most people simply ignore the proscription, probably because it has no
bearing whatsoever on the successful communication of ideas. Moreover, many sentences read awkwardly if
the adverb is moved from its position between to and the verb: ‘You can choose to cooperate always with
colleagues inside and also outside your work unit’ (‘you can always choose to cooperate…’ is not awkward, but
it has a different meaning). Overly zealous avoidance of split infinitives can even create ambiguity: ‘He asked
us clearly to underline the main points’; this could mean either ‘ask clearly’ or ‘underline clearly’.
Prescriptions sometimes actually create ‘incorrect’ usage through a process known as hypercorrection. This
occurs in particular when they are taught in primary and secondary school classrooms, but not fully understood.
You may recall being told not to write ‘Gwendolyn and me translated this text together’; it should be
‘Gwendolyn and I…’ because ‘I’ is the proper form for the subject of a finite verb. Many people have taken in
the injunction itself (don’t use ‘Gwendolyn and me’) but not the explanation. As a result, one now frequently
comes across sentences such as ‘This text was translated by Gwendolyn and I’. The ‘correct’ usage is in fact
‘Gwendolyn and me’ because ‘me’ is the correct form for the object of a preposition; you wouldn’t say ‘…
translated by I’.
Not only do prescriptions sometimes create error, and not only do they have next to nothing to do with
effective communication, but also, they may actually hinder communication, by reducing the semantic options
available to writers. Consider the rule that requires present-tense verbs to agree in number with their subject.
Purveyors of correctness insist on a very rigid application of this rule.
They prescribe ‘A flock of birds was alighting’ and rule out ‘were alighting’. This makes it impossible (without
expanding the sentence) to distinguish two different situations: the ducks all alighted together at one spot
(‘flock…was alighting’) as opposed to the situation where some alighted here and others there, at different
times (‘birds were alighting’). If as editor you change ‘were alighting’ to ‘was alighting’, you may well be
preventing the writer from saying what they want to say. More generally, usage ‘rules’ can become a crutch for
editors. It is so much easier to mechanically apply pseudo-rules like ‘never start a sentence with a conjunction’
than to ask whether starting a particular sentence with ‘but’ is communicatively effective.
Another criticism one can make of the prescriptivists is their arbitrariness. For example, they rule out the use
of ‘hopefully’ as a disjunctive adverb—see sentence (3) above—but they do not criticize other such adverbs.
They have nothing to say about a sentence like ‘Frankly, this text will not be translated by tomorrow’. Yet the
sentences are exactly parallel in meaning: I tell you hopefully/frankly that this text…
The prescriptivists also distinguish themselves by not being there when you need them. They complain about
usages which do not impede effective communication but fail to complain about usages which do impede it. For
example, they do not draw attention to a use of ‘may’ which is often ambiguous, even in context: ‘Helicopters
may be used to fly heart attack victims to hospital’ can mean either that it is permitted to so use the helicopters
or that it is possible that they will be so used.
A final criticism is that sometimes prescriptivists do manage to pick out a point that really can lead to
misunderstanding, but their recommendations are not helpful. An example is the position of the word ‘only’.
The written sentence ‘His condition can only be alleviated by surgery’ is ambiguous; it can mean either that his
condition can be alleviated but not cured by surgery, or it can mean that the alleviation can be accomplished
through surgery but not by any other means. In speech, this distinction is made by placing stress on alleviated
for the former meaning and on surgery for the latter. The prescriptivists correctly say that in writing, ambiguity
can be avoided if ‘only’ is always placed directly before the expression it modifies: ‘only be alleviated’ for the
first meaning, ‘only by surgery’ for the second. The problem is that if we followed this rule all the time, we
would be forced to write awkward and unnatural sentences; instead of ‘I only wanted to talk to her’, we would
have to write ‘I wanted only to talk to her’. There is simply no easy way to avoid ambiguity with ‘only’; you
need to think about the possibility of misinterpretation every time.
 The work of an editor 27
Now prescriptivists often say that a certain usage should be followed because it was observed by the best
writers of the past. Such references to writers of the past lend a patina of objectivity to their claims, but in
reality, the prescriptivists do not do any research to determine the usage of ‘the best writers’. Sentence (1), for
example, is by Lewis Carroll—surely a good writer—and many usages condemned by prescriptivists can be
found in Milton and Shakespeare. In practice, the ‘best writers’ turn out to be those who follow the critic’s
prescriptions!
Why do some people get angry about what they perceive as incorrect usage? For some, the motivation is
social liberalism; they believe that if the children of poorly educated parents, or parents educated outside the
English-speaking world, could learn a certain version of Standard English usage, this would help pave the way
for their social advancement. Indeed, it may have been a political concern to eliminate differences among
immigrants and among social classes that originally led to a much greater interest in prescriptive grammar in
the United States than in Britain. There continues to be much greater resistance in the U.S. to the idea that
dictionaries and grammars simply describe the language. There is a demand— both from the linguistically
insecure and from the self-appointed saviours of the language—that such publications serve as sources of
authority, that they prescribe what is right. Quite different are British authorities such as Henry Fowler and
Ernest Gowers, who tend to take a relatively moderate and reasoned approach; they do not rule out split
infinitives, for example. They tend to be more focused on effective communication than on correctness.
American linguistic conservatives like John Simon, on the other hand, tend to ban certain usages outright and
fail to give reasons for their prescriptions; such and such a usage is just wrong, indeed barbaric, and shame on
you for not knowing so! There is often a strong moralizing tone in their writings, suggesting that incorrect
usage is on a par with sexual permissiveness and other conservative bugbears. English, in this view, is not
merely changing; it is in decline and needs to be saved. Linguistic conservatives are motivated by various
combinations of snobbery (any cultured person would know that you don’t start a sentence with ‘but’) and
despair that the younger generation is not emulating the older.
All these criticisms of the prescriptivists are not meant to suggest that there are no problems standing in the
way of effective communication. Of course, such problems exist; indeed, that is why editors are needed. As we
saw in Chapter 1, writing lends itself much more than speech to misunderstanding. The problem with the
prescriptivists is that they generally do not draw attention to the problems that hinder effective communication.
In the next chapter, we’ll be looking at features of writing which really do cause readers problems, features
which prescriptivists practically never mention, such as poor inter-sentence connections.
Does all this mean that editors can ignore the prescriptivists? Definitely not. That is because many people
think ‘correct’ usage is important and they expect editors to serve as sources of authority, defending the
language against ‘incorrect’ usage. Also, many readers of your edited text will be displeased by ‘incorrect’
usages. They may very well make the condemned errors themselves, in their own writing, but they believe in
the idea of maintaining the standard.
How far should you go, as an editor, in enforcing ‘correct’ usage? Since the various published authorities
often do not agree on particular points of usage, you will need to adopt an approach to each contentious point.
Sometimes your employer’s style guide, or a senior editor, will decide the matter for you, but more often you
will have to decide yourself. You must bear in mind that if you adopt a conservative position, you risk being
branded as out of touch with the younger generation, with current social movements, and other sources of
linguistic innovation. On the other hand, if you adopt a more liberal position, you risk annoying conservative
readers and being branded as an agent of declining standards. You won’t be able to satisfy everyone.
A point to bear in mind in this regard is that translators and editors, by virtue of their self-image as ‘servants’,
or by virtue of demands made on them to be ‘language guardians’, probably have a tendency to lean
unconsciously toward a conservative approach to usage. A special effort will be needed if you want to
counteract this and take a more liberal or even innovating approach to language when appropriate.
One possibly comforting thought is that as the number of people who write in English as their second
language increases, editors may become less fussy about correctness because these writers, being members of
other cultures, will not have any particular allegiance to traditions of correctness; they will be concerned only
with communicative effectiveness. This will of course also be true of the constantly increasing number of
readers of English who are not native speakers. They will in all likelihood never have heard of split infinitives
 The work of an editor 28
and be blissfully unaware of their incorrectness. On the other hand, the situation may be quite different with
those non-native users of English who have spent long years studying the language and have achieved a very
high level of mastery. They may have been taught a rather rigid and conservative version of English and may be
shocked at the ‘laxness’ of many native users. As a result, they may provide added support for native-speaker
traditionalists.
To make usage decisions, rely on sources whose judgments are based on actual investigations of what
appears to be acceptable and what not. If you are wondering, for example, whether ‘they substituted x with y’ is
acceptable, Webster’s Dictionary of English Usage will tell you that ‘substitute with’ is standard but that one
may wish to avoid it because of the potential for negative reaction. The Canadian Oxford Dictionary is
somewhat more negative, saying that this is a disputed usage and should be avoided in standard English: use
‘they substituted y for x’ or ‘they replaced x with y’. The New Oxford Dictionary of English, on the other hand,
says that despite the potential for confusion, ‘substitute with’ is well established, especially in some scientific
contexts, and though still disapproved of by traditionalists, is now generally regarded as part of normal standard
English. This suggests that an editor who wishes to appeal to either a traditionalist or a Canadian audience will
avoid ‘substitute with’, but that otherwise a writer’s ‘substitute with’ need not be altered.

Practice
In copyediting, there are a great many different kinds of error to catch, and you may find it difficult to attend to
them all at once. In particular, you may find it hard to pay attention to errors that affect individual words or
short phrases and at the same time to pay attention to errors that affect larger units of the text. For example, if
you are attending to very small units, you may not notice that a lengthy parenthetical expression has no closing
parenthesis, or that some paragraphs in the text are indented whereas others are not. Sometimes your attention
may be so focused on individual words that you do not even notice errors such as ‘funds to assist towns rebuild
their sewers’, where ‘help’ was changed to ‘assist’ but the needed accompanying change in syntax was not
made (‘in rebuilding’). This problem of what you are attending to affects all types of editing (and revision), not
just copyediting.
You may find it easier, at first, to work through a text twice: once paying attention to micro-level problems
and once to macro-level problems. Some of the exercises suggested below go even further: you will be asked to
copyedit for just one feature, such as specific punctuation marks, ignoring all other problems. If you are a
student who missed many errors on your first marked assignments, go through the (presumably short) text
many times: once for punctuation, once for layout, once for inter-sentence connections, and so on.
Later, when you are practicing ‘full’ copyediting (that is, for all types of error), count the number of
problems of each type which you missed: typos, inconsistency of format, closing parentheses and so on. It may
be that mere awareness of what you are missing will help; subconsciously, you will start paying more attention
to that type of problem. Otherwise, if you continue to miss a significant number of errors, you should make a
practice of going through a text more than once.
Regarding the speed with which you move through the text, your instructor will give you some time-limited
exercises to do in class. However, you may also find it useful to experiment with speed at home. For example,
before you prepare the final version of an assignment (one that will not be graded), work very quickly through
the first half of a text and much more slowly through the second half. Then, when the class goes over the text,
see whether you caught more errors when working slowly.
A tip on micro-editing: you may find it useful to place a ruler or sheet of paper under the line you are
working on. This will direct your attention to the words on that line and ensure that your eye does not skip
lines. By the way, it is much easier for the eye to miss problems if you work on screen (see Chapter 9.3), so for
now, do all your copyediting work on paper.
Exercises can be speeded up if you simply underline places in the text where a change is needed, without
actually making the change. Remember that the difficult thing in editing is finding the problems. Correcting
copyediting problems, once you have found them, is usually fairly easy.
 The work of an editor 29
Exercise 1. Following style sheets
Using the style sheet your instructor gives you, find (but don’t correct) the features of the practice text that
deviate from the style sheet.

Exercise 2. Punctuation – commas

Your instructor will give you a text from which all the commas have been stripped. Add just those commas
which are necessary for clarity.
You will then receive a text that does contain commas. Remove all those not necessary for clarity. If
necessary, reword sentences if you feel there are too many commas.

Exercise 3. Spelling – spotting the errors

Your instructor will give you a text containing spelling and typing errors that would not be detected by
Spellcheck. Find them but don’t correct them.

Exercise 4. Usage
Many usage authorities require the so-called ‘serial comma’ (use of a comma before the final ‘and’ or ‘or’
of a list, as in ‘height, width, or depth’. Others disapprove it, while still others allow or recommend it under
certain circumstances. Read the Wikipedia article ‘Serial comma’, which lists the views of a considerable
number of style guides. What is your opinion?