One of the many ways in which the Internet affects language (Stein 2006) is the creation of new genres, or at least the transformation of old genres into new garb and functions. Through its technical possibilities and the new communicational situations enabled by it, the Internet evolves genres that have the potential to redefine and change social structure and networks (Bergs 2006), including the power structure, rules of cooperation and norms for language use in the domains. The issue of genres in the Internet is only now coming into focus of linguistic research (Kwasnik & Crowston 2005, especially the contribution by Askehave & Nielsen). One of those new genres is the company website. It is clear that companies would not pass up on the opportunity offered by the Internet to advertise themselves world-wide for very little money. Embedded in a more general interest in genre theory, especially as it is confronted by interesting challenges by the advent of the Internet, websites as a new genre was examined in the course of a study on language on the Internet. This focus has been promoted by an independent interest in the quality of the English found in these websites. It was discovered that there were indeed language issues, and very serious ones, but there was above all a genre issue.

By way of background information it should be mentioned that the results of the analysis were reported on in the shape of a ranking of banks in an economists’ daily, the Düsseldorf-based “Handelsblatt”, and caused quite a bit of a rumpus. Some banks were very upset and in fact angry at being labelled “insufficient” in the eye of the public. It actually turned out that the ranking corresponded very closely to the financial performance of the banks, with the Swiss banks top and the German banks bottom, with some of the worse German banks in the meantime being bought up by other national banks. It was only privately-owned German banks, like Sal. Oppenheimer, whose financial performance was, iconically, in step with the linguistic quality of the website.

The language matter cannot be pursued here any further, but there is clearly an issue with the perception of the quality of mastery of English in Germany, and very certainly so with the higher ranks of German businesses (not the middle and upper middle management).

A couple of remarks on the quality of English are in order before turning to the genre-issue, not only since this was the original motivation for doing the analysis in the first place, but there is also a medium-related aspect. Here are some representative examples for the quality of the English of the bottom-end websites: one bank extols its ability to “truly comprehend what you have to say”, another offers a “selection trading ideas with reliable settlement” and also “advise [sic] in relation to the structure, timetable and organization of the launch”, a third bank asks itself “Does a ‘super cycle’ await?”.

Interesting though these examples of very public English grammar may be, and their effect on business prospects, more important in the present context is another aspect. In no other medium are grammar mistakes as salient and totally unforgiving perceptually as on the screen. Language on the screen carries a much higher perceptual information load than in spoken and written language. On the screen, given that there is very little time devoted to actual reading (cf. below), first impressions are paramount. It is still medially written language in the shopping window of the world and damage is instantly done on a world-wide level. There are no second chances: the potential reader will momentarily opt out and not revert to the website, except, perhaps, for having linguistic fun. Here is a clear medium-related factor in the sphere of banking.

The juxtaposition of two excerpts from two very differently-rated bank websites will make it intuitively clear which website is the better one:

“With total assets of more than € 290 billion, WestLB AG is one of the leading providers of financial services in Germany. Three mutual strengthening pillars underpin the business model pursued by WestLB which is an international commercial bank as well as a competence center for the saving banks and midsized companies. WestLB is hence a financial partner for major corporate clients and financial institutions, mid-sized corporates, banks and insurance companies, public sector clients and saving banks.”

“You want to manage your investment yourself, or you may prefer to have them managed by a qualified specialist. You also value personal advice and tips that come fresh from the market to help you with important investment decisions.”

The most badly evaluated texts were simply paper texts put on the Internet. For the most part they represent simply medially – and stylistically - translated versions. These texts were not composed specifically for the purpose of leading a communicative life in the Internet. Most bad websites actually look like corpses in their new media environment. The problem boils down to an opposition between paper text and screen text. There are lot of theoretical issues involved when it comes to talking about “text”. We will simply ignore them here and set up two categories “paper text” and “hypertext”, or screen text.

A word is in order, though, on the notion of “hypertext”. It can refer to the links themselves, as the meta-information (normally set off by a special color), or it can refer to any linear coherent stretch of text that appears on the screen, and it can refer to any text that is individually composed by the use of links. We will refer to hypertext as a mixture of notions two and three.

The problem, then, is that the bad texts are paper texts conceptually, i.e., the way they are composed. Paper texts are “read”, hypertexts are more scanned. They are processed much differently. There is, in hypertexts, much less density of information in terms of linguistic morphemes. They are scanned and must contain clear visible pointers for scanning, such as uses of space (Morkes & Nielsen 1997). Scannability refers to the space of within one screen page. Any scrolling is dispreferred. In terms of proportions of space and text, no more than one third of a screen should be devoted to text. There is much more visible macrostructure marking, in terms of graphic elements such as spaces, and much less linguistic macrostructure signalling, such as text cohesion markers (discussed in, e.g., Halliday & Hasan 1976). It is all geared to the eye, and should not have to be unearthed by the reader with great cognitive burden and interpretive meaning archeology. In paper text, words and morphemes carry all the meaning, which is what we have been educated and socialized to attend to, as part of the acquisition of paper literacy. Here it is visual cues and visualized structure that decide if we want to follow up on links and define our own way into the hypertextual architecture.

The badly ranked websites are conceptually and prototypically paper texts. Their syntax weighs tons. The reader has to do a lot of work, and will never stay online on a website. In promotional texts, life should be made easy for readers. Part of the promotional effect is inviting the reader to engage in and to engage the text. Hypertext has great potential of interactively engaging the reader. There is, in the badly ranked texts, no reader involvement either emotionally and rhetorically, or interactively in the hypertext sense. The reader is not invited to check out links, to the extent that there are links. The reader is not encouraged to stay on. With paper text, the reader might skip the first pages and take a look at a later page. On the Internet, first impressions, and the entry page, are paramount for engaging readers or for scaring them away. Again, the Internet it is a very unforgiving system.

Obviously, as a new language medium evolves new genres, they are striving to discover an optimal language form and content. In the evolution of written language, an often-described effect was the so-called oral residue (Ong 1965) – forms of language that were carried over from the spoken medium, and which is not suited to the new communicative, written setting. I believe what we observe in the present situation is the transfer of written strategies to a new medium, where the written language properties of paper text are very different from screen text, which, as I said, is not read in the sense of paper text. A simple medial translation will not do. This applies to several aspects. For one, the function of genres tends to become more hybrid and less clear-cut than with traditional spoken and written media. Without trying to engage in a full analysis here, the function of the website sits uneasily somewhere between a commercial, technical description of the product and a more formal report on facts. It is likely that all Internet genres – classic example: the blog (McNeill 2005) – are undergoing a process of eking out an identity between a pedigree of genres in the traditional media and the technical constraints of the new medium, which is itself in a process of permanent technical evolution. Clearly, one of the dominant functions of the discourse is the same, promotional discourse, but the company website is surely a new professional genre, the rules of which are in the process of formation.

Bolter (1991: 40f) has argued that we have seen new hard structures arise, written language and paper text triggering the Gutenberg revolution, and now the Internet, with digital language and hypertext, triggering another hard structural revolution. Language in these media can be seen as the soft structures that go with the hard structures. The development of the soft structures always lag behind evolutionally. The case discussed here involves a soft structure left-over from the earlier medium, a situation that appears typical for a number of professional new genres. Schmitz (2002: 182f) points out parallel examples from the rise of newspapers and news shows, which in their early stages carried over features from their predecessor media. He also points out that the processing of text and information in the new medium, the Internet in the case at hand, requires a transition which website designers must take into account: from linear, one-by-one processing as in paper reading to a web-type processing that may well put a strain on processability.

While the structuring of information is constrained, at least initially, by our processing structures that have evolved to serve the purposes of traditional reading, it has been argued (Tufte 2003) that, on the next higher level beyond processing, the medium may well have an effect on how we interpret and integrate information in ways of which we are not yet aware. Discussing power point demonstrations, he argues the hierarchical type of presentation predetermines and constrains ways of thinking and cognition.

If this paper took a brief look at how the new hard structure necessitates a new structure of presentation, it would be an intriguing idea to explore the – slightly worrying – idea to which extent the hardware also constrains our cognitive interpretation and deductive processes.

Finally, to revert to matters more narrowly linguistic, it would be interesting to see how e-journals change the character of the articles submitted and published from paper text properties to hypertext properties, involving all the (text-) linguistic trappings of hypertext. Editorial gate-keeping is bound to be in for some problems. Should the style of writing in which the articles in the present collection, and in fact in e-journals generally, be more like traditional paper writing, or will they be more like hypertext writing, such as set out in style manuals such as Lackerbauer (2003)? Will the effect of writing on the Internet be a stylistic convergence (in English only?) across cultures, affected by the high technicity of the medium? It is not the only aspect under which the new medium, in an evolutionary process, invites new language forms, genres, registers and medial varieties. Another paper in the present collection (Felden 2006) points to another way in which new technical possibilities of knowledge exploitation in the Internet, in order to optimally conducted, invites new styles of writing.

Askehave, Inger & Anne Ellerup Nielsen (2005). Digital genres: a challenge to traditional genre theory. Information Technology & People 18(2): 120-141.

Bergs, Alexander (2006). Analyzing online communication from a social network point of view: questions, problems, perspectives. Language@Internet SV1-3. urn:nbn:de:0009-7-3712 (http://www.languageatinternet.de).

Bolter, Jay David (1991). Writing Space. The Computer. Hypertext, and the History of Writing. Hillsdale, New Jersey: Lawrence Erlbaum.

Felden, Carsten (2006). Integrating Structured and Unstructured Data in a Business-Intelligence-System. Language@Internet SV15.-urn:nbn:de:0009-7-3738 (http://www.languageatinternet.de).

Halliday, Michael & Ruqaiya Hasan (1976). Cohesion in English. London: Longman.

Kwasnik, Barbara & Kevin Crowston (eds.) (2005). Information Technology and People. 18(2) Special Issue: Genres of Digital Documents.

Lackerbauer, Ingo (2003). Handbuch für Online-Texter und Online-Redakteure. Berlin, Heidelberg: Springer.

McNeill, Laurie (2005). Genre Under Construction: The Diary on the Internet. Language@Internet 1/2005. urn:nbn:de:0009-7-1203 ( http://www.languageatinternet.de/articles/120).

Morkes, John & Jakob Nielsen (1997). Conscise, SCANNABLE, and Objective. How to Write for the Web. Available: http://www.useit.com/papers/webwriting/writing.html. [Last access 6 February 2006].

Ong, Walter J. (1965). Oral Residue in Tudor Prose Style. Publications of the Modern Language Association 80(3): 145-154.

Schmitz, Ulrich (2002). Qualität und Reklame im WWW. Zur Gestaltung von Internet-Auftritten deutscher Wirschaftsunternehmen. In Thimm, Caja (ed.) Unternehmenskommunikation offline/online. Wandelprozesse interner und externer Kommunikation durch neue Medien. Frankfurt am Main: Peter Lang. 164-184.

Stein, Dieter (2006). Language on the Internet. In Brown, Edward K. (ed.) Encyclopedia of Language & Linguistics. Amsterdam: Elsevier.

Thimm, Caja (ed.) (2002). Unternehmenskommunikation offline/online. Wandelprozesse interner und externer Kommunikation durch neue Medien. Frankfurt am Main: Peter Lang.

Tufte, Edward R. (2003). The cognitive style of PowerPoint. Cheshire, CT: Graphics Press.

Submitted: 05.10.2005

Review results sent out: 07.01.2006

Resubmitted: 30.01.2006

Accepted: 02.02.2006