[From the May 2000, Vol. 47, No. 2, issue of Technical Communication, the journal of the Society for Technical Communication; reprinted with permission of STC.]
What is plain language? Actually, defining it is not unlike defining information design. Ask 10 people and you'll get 10 different answers. Yet just as with information design, there is a common thread.
For example, one definition states that plain language is "language that reflects the interests and needs of the reader and consumer rather than the legal, bureaucratic, or technological interests of the writer or of the organization that the writer represents" (Steinberg 1991, p. 7). Martin Cutts, research director of the Plain Language Commission in the United Kingdom, defines plain language as "The writing and setting out of essential information in a way that gives a cooperative, motivated person a good chance of understanding the document at the first reading, and in the same sense that the writer meant it to be understood" (1998, p. 3).
Plain language also has in common with information design both a broad and a narrow definition (see Ginny Redish's commentary in this issue). Some definitions, such as Cutts' above, suggest the broader goal of plain language that involves both writing and "setting out" language so that the reader understands it. Other definitions refer more to the origin of the term plain "language" or plain "English." (In this article, the more generic term "plain language" is used unless a cited work specifically refers to plain English.) For example, Berry notes that the "goal of the plain-language movement is to produce language (particularly written English) which is clear, straightforward expression, using only as many words as are necessary, and which avoids obscurity, inflated vocabulary and convoluted sentence construction" (1995, p. 48). This latter definition and others like it are common for those, such as Berry, who approach plain language from the goal of producing plain language in legal documents.
Understanding plain language is more than a philosophical discussion though, for on 1 June 1998, U.S. President Bill Clinton issued a memorandum to the heads of U.S. federal executive departments and agencies directing them to begin using "plain language" to make government "more responsive, accessible, and understandable in its communications with the public. . . . Plain language saves the Government and the private sector time, effort, and money" (Clinton 1998).
THE ORIGINS OF PLAIN LANGUAGE
Two good resources for the history of plain language are Redish (1985) and Schriver (1997). Both credit Stuart Chase as an original plain language proponent in the U.S.—in The power of words (1953), Chase complained about "gobbledygook" in texts. In 1971, the National Council of Teachers of English in the U.S. formed the Public Doublespeak Committee (Penman 1993). In 1972, U.S. President Richard Nixon created plain language momentum when he decreed that the "Federal Register be written in 'layman's terms'" (Dorney 1988). Industry soon followed. In 1973, Citibank converted a promissory note to plain language, a change that "brought great prestige to Citibank, which was seen as a leader in improving consumer relations" (Williams 1999, p. 3).
The next major event in the U.S. history of plain language occurred in 1978, when U.S. President Jimmy Carter issued Executive Orders 12,044 and 12,174. These were intended to make government regulations cost-effective and easy to understand by those who were required to comply with them. In 1981, U.S. President Ronald Reagan rescinded those orders. Nevertheless, many continued their efforts to simplify documents; by 1991, eight states had passed statutes related to plain language (Schriver 1997).
The plain-language movement has also been active outside the U.S. "In 1982, the British government issued a White Paper (a policy statement) ordering departments for the first time to count their forms, abolish unnecessary ones, clarify the rest, and report their progress annually to the prime minister" (Cutts 1995, p. 6). In the foreword to a book by the Plain English Campaign, a private company in the U.K., Chrissie Maher notes that they have "attacked unclear legal language for the last fifteen years" (1996).
Proponents of plain language have also been active in Australia since 1976 and in Canada since 1988 (Schriver 1997; Berry 1995). Other countries with plain language efforts include Sweden, South Africa, and New Zealand (Baldwin 1999).
WHY PLAIN LANGUAGE?
Proponents assert that documents created using plain language techniques are effective in a number of ways. A recent plain language resource (Baldwin 1999) lists the following reasons:
- Readers understand documents better.
- Readers prefer plain language.
- Readers locate information faster.
- Documents are easier to update.
- It is easier to train people.
- Documents are more cost-effective.
In writing about the Citibank promissory note mentioned above, Cheryl Stephens notes that cost-saving was a motive:
Citibank had spent a lot of time in Small Claims court trying to collect on their promissory notes. It had also spent a lot of time training staff to answer consumer questions about their complicated forms and contracts. After the adoption of the plain language note and other plain language forms, there was a measurable savings in staff training time [and] in the reduction in small claims lawsuits. And a substantial increase in market-share. And the wording of the new form has not been challenged in court. (Williams 1999, p. 2).
Kimble (1996, 1997) cites a number of projects showing the benefits of plain language techniques. One study cited a project for the U.S. Department of Veteran Affairs in which a sample letter was revised. Benefits counselors estimated that 750 copies of the original letter had been sent in one year with over 1,100 calls as a result. After the letter was revised, 710 copies were sent with just under 200 calls as a result. Similarly, after Allen-Bradley revised their documentation using plain language techniques, their phone center call volume reduced from 50 calls a day to 2 calls a month (Jereb 1991).
Perhaps more importantly, plain language is credited with increased comprehension as well as being preferred by readers. For example, using his own guidelines, Cutts revised a document (1993) and later tested it against the original. The result was that 87 percent of the law students tested preferred the revision. More importantly, students using the revised version performed better on 9 out of 12 questions (Cutts 1998).
CRITICISM OF PLAIN LANGUAGE
However, plain language has been the target of considerable criticism. In this article, I will review some of this criticism while examining past and current plain language literature. In doing so, I will avoid for the sake of brevity many resources cited by plain language proponents (for example, Strunk and White's Elements of style) and concentrate on those specifically about plain language.
What is plain language anyway?
In response to plain language criticism, Baldwin asks "But what are they criticizing? There is no single, world-standard definition" (Baldwin 1999, p. 17). This in and of itself has been a long-time problem for the movement. In a critique of plain language, Penman cites Charrow's 1979 work What is plain English anyway? and notes that since then "the movement has, if anything, become even more varied in its understandings of what is plain English" (1993, p. 122).
For example, as mentioned earlier, Kimble cited a study regarding the benefits of a revised letter for use by the U.S. Department of Veteran Affairs (VA). The implication is that plain language techniques were used to revise the letter. But were they? In reviewing the original citation, the author does not mention plain language explicitly, but notes that she
. . . produced research-based guidelines for VA, building on work in psycholinguistics and cognitive psychology, document design, and reading and information processing theory. The guidelines cover audience analysis, organization, document design, style, syntax, supplements, and graphics. Furthermore, I specified that the process of developing letters had to include iterative cycles of drafting, review, testing with representative readers, and revision based on the testing. (Daniels 1995)
So, was this plain language?
The answer is yes if you consider the above to be an example of a "reader-oriented" approach to plain language. This is the second of three "tendencies" to plain language identified by Coe and cited by Penman, which include text-based, reader-oriented, and collaborative. Text-based approaches place "the major focus on the document, not the reader per se" (1993, p. 122). Penman notes that neither he nor Coe had come across any examples of the last approach, which involves having "representative readers involved from the beginning [being] actively allowed to determine what is written and how it is written to suit them as readers" (1993, p. 126).
Current information design practice would appear to have us placed rather firmly in the middle category as well, with occasional calls for movement into the third category. But there appears to be no real text-based counterpart in the information design/document design world (in fact, some might argue that document design actually moves away from the text-based focus of "traditional" technical writing).
Plain language is just about text—shortening it, dumbing it down.
The majority of plain-language resources do not advocate shortening and dumbing down documents. In fact, many plain language proponents seem to share a similar respect for the user with their information design counterparts. For example, one resource notes, plain language ". . . does not mean always using simple words at the expense of the most accurate words or writing whole documents in kindergarten language" (Cutts 1995, p. 3).
This same philosophy can also be found in plain language resources for lawyers.
Some people think that because plain language is simple, it must be simplistic—a kind of baby-talk. [But simple] in this sense doesn't mean simplistic. It means straightforward, clear, precise. . . . What is appropriate in one context may be inappropriate in another. And it takes time to develop the necessary sensitivity to the problems of your readers. (Asprey 1991, pp. 11.12)
The U.S. Securities and Exchange Commission notes that plain English ". . . does not mean deleting complex information to make the document easier to understand. For investors to make informed decisions, disclosure documents must impart complex information. Using plain English assures the orderly and clear presentation of complex information so that investors have the best possible chance of understanding it" (1998, p. 5).
Some plain-language resources do instruct writers to shorten sentences. However, even when such instructions are given, there can be an interesting information design parallel. For example, Wydick directs lawyers to "omit surplus words" by removing "glue" words and concentrating on the "working" words. The example given shortens the sentence "A trial by jury was requested by the defendant" to "The defendant requested a jury trial" (Wydick 1994, p. 8). Wydick's working words are strikingly similar to Edward Tufte's concept of data density.
Plain language doesn't concern itself with visual design.
Some plain-language resources do tend to be predominantly about writing, particularly those directed at lawyers. Plain English for lawyers provides seven chapters covering topics that are predominantly text oriented (Wydick 1994). In Plain language pleadings, the main focus is on plain language in specific instances such as pleadings, legislation, and forms (Wilson 1996). In Legal writing, writers are admonished to follow seven text-based rules, such as 'choose clarity' (Mellinkoff 1982). And unfortunately, President Clinton's memo (1998) is a target of this criticism, since it notes that plain language documents, while having "easy-to-read design features" use:
- Common, everyday words, except for necessary technical terms
- "You" and other pronouns
- The active voice
- Short sentences
In some plain-language resources, there is at least a mention of visual design. In The plain English approach to business writing, Bailey notes that his "audiences usually consider layout to be the most important topic I cover." His Chapter 4 is titled "Layout: Adding visual impact," and while it, like some other plain-language resources, is short on specific typographic and visual design instruction, it does cover the basic "white space" instructions in terms of paragraph length, headings, and lists (Bailey 1990, p. 37). Writing user-friendly documents (no date), the resource available from U.S. Vice President Al Gore's Plain Language Action Network (part of the U.S. National Partnership for Reinventing Government) likewise covers these topics somewhat sparingly.
However, many plain-language resources do address the visual aspects of document design in considerable detail. For example, the Document Design Center's Guidelines for document designers (Felker and others 1981) is considered one of plain language's foundational documents. Written in 1981, this impressive resource includes two sections related to visual design: one on typographic principles, the other on graphic principles.
A more recent (and graphically sophisticated) plain language resource is the 1998 Plain English handbook from the U.S. Securities and Exchange Commission (SEC), a must-read for those interested in plain language. While it is targeted primarily at those who must adhere to SEC regulations, this resource contains a wealth of information. In particular, it includes some strong statements about the importance of design:
A plain English document reflects thoughtful design choices. . . . In a plain English document, design serves the goal of communicating the information as clearly as possible. (p. 37)
Finally, a new plain language resource goes a step further and incorporates concepts from Edward Tufte and Robert Horn to emphasize "visual language" in the form of "information design displays." In fact, the author recommends that writing and editing begin only after questions about document design ("defining the look, navigation features; deciding where and how to use visual language; creating the headings") and information display ("creating the tables, charts, infographics, and other graphical elements") are answered (Baldwin 1999, p. 21). Baldwin suggests that most writers have been conditioned to use a "piles of paragraphs" approach. In other words, we still create documents as if our only tool is the typewriter.
Plain language uses readability formulas of questionable validity.
Schriver reports that by the mid-1980s, researchers had "abandoned" plain language studies because of doubts of the efficacy of the approach (1997). One major concern was the reliance on techniques such as readability formulas. Rudolf Flesch's How to write plain English was perhaps the strongest proponent of this method. In it, the average number of words in a sentence and the average number of syllables in a word are related using a scale. The lower the two variables, the higher the "readability" of the document (1979). Plain language proponents such as the Document Design Center, however, were arguing against readability formulas as far back as 1980. In current literature, very few plain-language resources promote the use of this type of readability measure. Those who do mention them do not recommend their use. For example, the SEC takes this approach:
Readability formulas determine how difficult a piece of writing is to read. However, you should be aware of a major flaw in every readability formula. No formula takes into account the content of the document being evaluated. In other words, no formula can tell you if you have conveyed the information clearly. . . . The final test of whether any piece of writing meets its goal of communicating information comes when humans read it. (1998, p. 57)
Plain language tests document at the end of the design process, if at all.
Another criticism about plain language is that, while information/document design has moved forward in support of user-centered design throughout the design process, plain language has not. If plain language proponents test, they do so only at the end of the process. However, if this objection was true at one time, it certainly appears to have changed in recent years.
The resource Plain language online notes that a "crucial feature of plain language is testing the writing to determine whether it adequately conveys to the targeted reader the writer's intentions. . . . This definition of plain language is 'reader-based' and not [a] 'text-based' analysis of a writing style." Plain language online also makes an interesting point that testing the original document may prove helpful and refers to a standard usability text for more information (1996).
In Plain language for lawyers, Asprey notes
You need to begin testing (or at least test once) early in the drafting process before your ideas have become fixed and you've gone too far to turn back. If you test early, you'll be more receptive to suggestions, more open to changing strategy, and have more time to incorporate changes. If you test early, you'll find out early if you have any fundamental misunderstandings about how the document works in practice. (1991, p. 228)
The inability to implement (or get a client to pay for) a fully iterative approach for every project is something that plain language has in common with information design. Perhaps more plain language proponents are pragmatists who are willing to accept that while involving readers at all stages is the ideal solution, it is not practical for every piece of writing that is done. This appears to be particularly true in the legal community (Kimble 1994, 1995).
Plain language is not backed up by research.
The major criticism of plain language is that its guidelines do not have sufficient research to back them up. This essentially translates to "does plain language work?" A complete review of this question is outside the scope of this article and is certainly worthy of a follow-up article. But there are two points to consider in this area.
The first is whether guidelines are based on empirical research. It is true that the majority of plain language resources do not cite research since the majority of them are directed toward the general public. If research is mentioned, it is generally without specific citations. However, of the resources I reviewed, the Document Design Center's Guidelines for document designers has no peer in this area. For each of their 25 guidelines, they provide a section titled "What the research says." One such guideline is the suggestion to "avoid whiz deletions." A whiz deletion is the absence of introductory text for subordinate clauses. The Guidelines offer the comparison between the sentence "The director wants the report which was written by the Home Office." and "The director wants the report written by the Home Office" (Felker and others 1981, pp. 39-40). This guideline was based on direct research done by Charrow and Charrow (1978). In their extensive study of jury instructions, these authors found that whiz deletions made jury instructions harder to understand (Felker and others 1981).
The second issue regarding plain-language guidelines and research is that actual practice does not appear to follow the guidelines. For example, a group of researchers asserted that the Document Design Center's guideline about whiz deletions was not valid, since whiz deletions were a common occurrence in (presumably) well-written documents (Huckin, Curtin, and Graham 1991). This discrepancy between guidelines and practice was also illustrated by van der Waarde's study in Technical communication (1999). A review of 330 documents found that the majority did not follow standard guidelines with regard to typographic dimension (x-height and line spacing). Does this mean that the guidelines themselves are invalid? Perhaps. But among other possible explanations for this finding, van der Waarde considered that "legibility and attractiveness are not the criteria that are most often used in practice" and that criteria such as cost, standardization, or production deadlines might have more impact on document choices. Or as Redish and Rosen suggest, "Real-world documents are compromises" (1991).
Plain language is about inviolate rules.
The last criticism I'll address is the rules versus guidelines issue. With the exception of some older plain language resources such as Flesch (1979) and Mellinkoff (1982), many plain language proponents point out that guidelines are not rules; their observance requires judgment:
- "I say guidelines, not rules" (Cutts 1995, p. 2).
- "Don't make Plain Language guidelines into rules" (Baldwin 1999, p. 19).
- "As with all the advice in this handbook, feel free to tailor these tips to your schedule, your document, and your budget. . . . Pick and choose the ones that work for you." (SEC 1998).
Redish and Rosen provide an interesting discussion on guidelines. First, they begin with a definition: "A guideline is a suggestion that helps writers achieve the goal of communicating clearly with their readers." They also note that "guidelines are a necessary part of any heuristic" and argue that many writers have essentially internalized guidelines that are used as they write.
The authors interviewed 30 people to find out whether they used guidelines in their writing. Those who were professional writers said that they did not use guidelines at this stage in their career (although many had used them earlier). Those who were recent graduates of technical writing programs had "mixed feelings" about guidelines. Some thought they were useful reminders, while others thought this was information they had already learned in school. The last group consisted of professionals in fields other than writing. For this group, 9 of 10 "reported that they rely on guidelines in their writing" (Redish and Rosen 1991).
The important point here is that guidelines are useful tools for those who write as a secondary activity rather than as their primary profession. And this is exactly the audience for whom plain-language guidelines are usually written.
Plain language today has been and is being informed by the work of information and document designers. Of the resources I reviewed, I would recommend the SEC's Plain English handbook, Asprey's Plain language for lawyers, and Cutts' Plain English guide as worthwhile resources (Guidelines for document designers is no longer in print). Baldwin's Plain language and the document revolution also deserves a look. It takes plain language to a new playing field, some of which is intriguing and some of which is curious. For example, although it provides a lengthy discussion of information displays that hold considerable promise for future documents, it lacks all but a cursory discussion of actual usability testing.
Today's plain language proponents clearly need more contributions from the academic and research organizations that provided much of its foundation. Both the Communications Design Center at Carnegie Mellon University and the Information Design Center (and its predecessor, the Document Design Center) at the American Institutes for Research are no longer in operation, but their principals, Karen Schriver, Ginny Redish, and Susan Kleimann, continue to be very active in both information design and plain language.
What is necessary for plain language to succeed? Redish (1985, p. 136) suggests that we need to:
- Increase awareness of the problems that traditional documents cause.
- Understand what causes the problems.
- Develop ways to solve the problems.
- Apply the solutions.
- Teach others how to apply the solutions.
In nearly 15 years, the essence of the issue remains the same. Our job as information designers should be to stay current with plain language, help inform it, and to make sure that others who are interested in plain language understand its breadth.
The momentum for plain language is definitely growing . . . at least outside of our own field. Recently, the American Bar Association passed a resolution that states ". . . That the American Bar Association urges agencies to use plain language in writing regulations, as a means of promoting the understanding of legal obligations . . ." (1999).
Information designers take very seriously our obligations to users. While some criticisms may validly be laid against various manifestations of the plain language movement, the movement can only benefit from attention and assistance from the information design community.
Clinton, W. 1998. "Memorandum for the heads of executive departments and agencies." http://www.plainlanguage.gov/cites/memo.htm (now found at http://www.plainlanguage.gov/whatisPL/govmandates/memo.cfm)
Huckin, T., E. Curtin, and D. Graham. 1991. "Prescriptive linguistics and plain English: The case of 'whiz deletions.'." In Plain language: Principles and practice. edited by Steinberg, E., pp. 83-92.Detroit, MI: Wayne State University Press.
"Plain Language Online."
Redish, J. C. 1985. "The plain English movement." In The English language today. edited by Greenbaum, S., pp. 125-138.New York, NY: Pergamon Press.
U.S. Securities and Exchange Commission.
1998. A plain English handbook: How to create clear
SEC disclosure documents. http://www.sec.gov/news/handbook.htm
van der Waarde, K. 1999. "Typographic dimensions and conventional wisdom: A discrepancy?." Technical communication 46, no. 1:67-74.