The value of linguistics

To what extent can knowledge of linguistics and theories of human language be of value to the technical author?

Introduction

Writing technical documents is a practical problem a large element of which revolves around bringing the language of the document in line with the language of the target audience. Linguistics is the science of language, so it could be expected that linguistics could help technical authors write documents. This essay examines the issues involved trying to use the academic theories of linguistics in a practical context.

After putting forward my view of the main challenge facing technical authors I look at two scenarios: technical authors learning linguistics to apply it to their work; and linguists applying their expertise to the problems of technical authors. I raise some questions about how universally useful a knowledge of the general theories of linguistics is to technical authors, and argue that linguists studying technical authorship would be more effective way of bringing the theory into practice.

What is a technical author?

As Noel Williams writes [Williams (2000) pp. 7-8], technical writing is writing with a purpose. To satisfy that purpose the range of interpretation of the writing must be constrained to those that meet it. Unfortunately, language is ambiguous; there is no document that cannot be misinterpreted by or be completely incomprehensible to someone somewhere.

Technical authors get around the ambiguity in language by aiming their writing at a target audience. They can then use their knowledge of the target audience to predict how the audience will interpret their work (assuming the authors know, or find out, enough about the target audience to make accurate predictions). The authors then accept responsibility for the interpretation put on their writing by the target audience. If the target audience misinterprets a document then it is a problem with the document not the audience.

Two people reading the same text may interpret it differently. This can be a benefit allowing a single document to fill a number of roles, as in the scenario described by Peter Medway where an architect distributes a document to all the participants of a building project and

the text in fact means different things to each [reader] because the intertextual connections giving it meaning are different for each individual. For Joe's boss the Site Instruction to lower the ceiling relates to the conversation in which Joe asked for his opinion, and reflects his judgement on what should happen to the ceiling. For Luc, the site supervisor, the document relates to a series of conversations with Joe during the latter's site visit, and to the vital distinction, maintained by unspoken agreement, between problems that had to be dealt with "on the record", e.g. through a Site Instruction, and others that the two of them would resolve by informal, unrecorded deals, such as - though it was not put so starkly into words - "You don't charge for widening that chimney hole in the concrete slab and I'll overlook the missing window seal". One meaning of the Site Instruction for Luc is thus, "This is one of the problems we had to deal with by the book". That meaning will not, however, be activated for an auditor who eventually allocates the cost of the alteration. [Medway (1996), p. 32]

More often the multiple interpretations by the target audience are a problem to be overcome by the author. For example, a document aimed at both experts and novices would need to use technical terms in the same exacting way as the experts, but in such a way that novices unfamiliar with the technical definitions can still understand them. In this case the author would work towards getting the experts' and the novices' interpretation as close as possible.

For a document to succeed completely all the interpretations of the target audience need to meet the purpose of the document. The task facing a technical author can be looked at as being one of expressing the message of the document in the language (or languages) of the target audience so that all its interpretations are appropriate. Linguistics, being the scientific study of languages, would seem to be a good place for the author to look for help.

The value of linguistic expertise to authors

An introductory book such as John Lyons' "Language and Linguistics" [Lyons (1981)] gives an account of the concepts of linguistics. These concepts capture the insights into languages of the people who built up the theories and explicitly express the rules and structures that are implicit in our use of language. So, in the terms of Hartley and Williams [(2000) p. 24], they make our unconscious knowledge of language conscious. The two most obvious applicable areas brought consciousness this way that are grammar and semantics.

Grammar looks at the way that components of a sentences are constructed; it gives the author tools to analyse the structure of their writing. Technical authors need to choose the grammatical structures that best suit the purposes of their documents and to create grammatical rules in their work. Using illustrations, headings, bulleted lists, numbered lists, and changes in typefaces all involve setting up grammatical conventions within a document (and applying them consistently).

The way that I characterised the tasks facing technical authors in terms of interpretations puts emphasis on semantics. Semantics looks at meaning in language, and to interpret a text is to give it meaning. If the semantics of a document are wrong then it will not be interpreted correctly, and so it will fail in its purpose. If the grammar is wrong then the document will fail if the grammatical problems compromise its semantics, but if the semantics remain intact then the document may still meet its purpose (although the problems with the grammar would make it a less than ideal solution).

It is essential for a technical author to have a tight control of the semantics and grammar of his documents, but he can be (and many are) adept at this without knowing what the linguists have to say about it. Just how much benefit would a technical author gain from the insights in linguistics after reading a book like Lyons'?

This is an empirically testable question that could be answered by taking two similar groups of technical authors and training one group in linguistics to see if they became more effective technical authors than the control group. It is impossible to give a definite authoritative answer to the question without studying how authors put the theory into practice. I can, however, make an educated guess that overall the linguistically trained group would improve as their new conscious understanding allowed them to see into what had previously been blind spots in their performance.

However, if such an experiment were carried out I would expect the degree of improvement to vary considerably from individual to individual. Some authors would improve significantly, some would improve marginally or not at all (because their own insights were already adequate or they had trouble applying the theory) and the practice of a few authors would actually deteriorate (as they tried to apply the theories inappropriately).

The value of linguistics to the profession of technical authorship

Individual authors do not have to know the details of linguistic theories to benefit from them. They can learn from the work of others, and an analysis of technical authorship carried out by a skilled linguist can pick out elements of the theories that are of particular use to authors. For example, Agnes Kukulska-Hulme's communicative approach to user interface and documentation design is "based on concepts from the field of linguistics" [Kukulska-Hulme (1999) p. 12]. She argues in the introduction that linguistics is an essential tool for developing good user documentation for computer systems, saying

The most constructive stance we can take in designing the user interface is to regard all computer users as language learners, since every new application environment creates new meanings for familiar terms and introduces new terms and concepts. Considering what appears on the screen and in the user documentation as a "new language" for the user can make one question beliefs and assumptions about what the user will or will not understand. It also introduces a productive aspect to language: User's certainly need to understand, but they also need to be able to produce language appropriate to the application in order to get the most from its use. [Kukulska-Hulme (1999) p. 4]

The book goes on to describe the communicative approach, using examples from the interfaces and documentation of commercial software to illustrate the problems that can arise when systems and documents are designed with insufficient attention paid to their language.

Books aimed squarely at applying linguistics to technical authorship are hard to find, putting limits on the usefulness of linguistics to technical authors. The alternative described above, where an author goes straight to the source and learns about linguistics in general and then applies it to his writing, is far from ideal. I am an experienced academic but Lyons (1981) was still a difficult book for me (and I had no way of telling which parts I could safely skim over). It is also twenty years old, so it does not cover recent developments. If I rely on my own expertise then, at the very best, I can only apply basic amateur linguistics to my writing.

Conclusion

It is essential for technical authors to understand the nature of language. Linguistics provides one way to achieve this, but it is not the only one. An intuitive grasp of how people use and interpret language can be just as effective. Intuition is easy to apply in practice, but it is difficult to develop. In contrast, linguistic theories are much harder to apply to real life writing but knowledge of them is easier to develop (you just read more books and articles).

The extent to which knowledge of linguistics and theories of human language would be of value to an individual technical author depends on the author in question. For the theories to be useful they need to find their way into the work of the author, and the ability to apply abstract theory to concrete situations is a specialist skill in its own right. Authors will vary in how well they understand linguistics and in how easily they can apply this understanding to their work.

Having individuals adept at linguistics and theories of human language who apply their skills to the problems of technical authorship is important for technical authors as a body of professionals. If we want to develop our expertise as authors then linguistics has a lot to offer. A work like Kukulska-Hulme's (1999) reduces the theory-practice gap and in so doing makes it more likely that practising authors will benefit from linguistics. An additional benefit of works by trained linguists about technical authorship is that they can draw upon a much wider range of linguistic theory than would be practical for most technical authors to learn.

In general I would advise technical authors to learn some general linguistics (though unless they were particularly academic I wouldn't advise them to start with Lyons (1981)) but not to worry if some of the details do end up going over their head, and I would encourage any that find the theories stimulating and useful to share their experience so that others benefit as well.

Bibliography

Hartley, P. and N. Williams (2000) MA Technical Authorship module 2.1: Approaches to communication and linguistics, Sheffield Hallam University

Kukulska-Hulme, A. (1999) Language and Communication: Essential Concepts for User Interface and Documentation Design, Oxford University Press

Lyons, J. (1981) Language and Linguistics: An Introduction, Cambridge University Press

Medway, P. (1996) "Writing, speaking, drawing: The distribution of meaning in architects' communication," in Sharples and van der Geest (eds) (1996), pp. 25-42

Sharples M. and T. van der Geest (eds) (1996) The New Writing Environment: Writers at Work in a World of Technology, Springer-Verlag

Williams, N. (2000) MA Technical Authorship module 1.1: The writer's tasks, Sheffield Hallam University