I am 63 years old with the reading age of a slow 4 year old. But I’m not stupid. I can tell that everything around me—the pictograms and ideograms, people’s faces, their clothes, the wonderful decorations on their pottery, the bright pinks, blues and oranges of their houses, the soil they work—is suffused with rich meanings which they read as easily as they breath, unaware of the great intelligence and skill they have in doing so. But I cannot do so. To me it is one vast terra incognito, beyond my ken. I am in Mexico City.

I am here after spending a week in the ancient and mysterious city of Cholula with my colleagues at Centro de Estudios Avanzados de Diseño (CEAD)—www.ceadmex.org—working on guidelines to help Mexican pharmaceutical companies redesign medicine labelling so that Mexicans literate in Spanish can read and use them appropriately.

In truth, I have done very little. I am like the catalyst in a chemical reaction. Along with other members of the Medicine Lablelling Group (www.medicinelabellinggroup.org) I help the process along. But the ingredients that make up the new guidelines, like the ingredients that make up a new substance, are the result of the work of others. In this case Maria Gonzalez de Cosio, Alejandro Lo Celso, their colleagues at CEAD, and the many Mexicans who participated in helping to test new prototype designs that are serving as a examples in these emerging guidelines. Indeed this whole project is a rare example of collaborative design. The project has been funded and steered jointly by the Mexican government regulator and the Mexican over the counter pharmaceutical industry, with a wide variety of interest groups who have converged around the common purpose of making labels easily usable by Mexicans literate in Spanish.

I emphasise ‘literate in Spanish’ because it is so clearly not the only, nor necessarily the most important, form of literacy in this many-layered mixture of highly sophisticated ways of life.

It is all too easy to take a narrow and disrespectful view of other people’s literacy; all too easy to think of an absence of skill in one limited form of literacy as a deficiency that must be remedied by appropriate treatment; all too easy to see resistance to taking up a new and alien form of literacy as a symptom of laziness or, worse, stupidity; and all too easy to say that because of laziness and stupidity, there is no point in even trying. Not far behind this is the view that the ‘illiterate’ poor are both lazy and stupid. Even to this illiterate outsider, with the understanding of a slow 4 year old, those easy ways of thinking seem profoundly wrong. Yet these are some of the persistent political currents of thinking that swirl around this project.

Even as the designers in this project conduct rigorous observations of people using labels, give meticulous attention to typography and wording, and prepare guidelines for others to follow, these and many other currents of thought swirl around the project, pulling the work this way and that.

What is most heartening is that the dominant current influencing everyone working directly on the project, including those funding it, is a simple common desire to help people avoid harming themselves inadvertently and to help them use medicines appropriately to reduce harm. So far nobody has wavered from this central collective purpose.

If successful in Mexico, this project has the capacity to benefit the whole of Latin America. But only if it remains mindful of its place in the scheme of things. To this illiterate, this project looks like a fragile rope bridge across a deep chasm of incomprehension between different forms of life. It would not take much to break it, and all the good work would then be lost, dashed against the uncaring rocks and swept away by the raging silence below. But, with caution and bravery the risky crossing could reduce distress and save lives.

Serendipity is a wonderful thing. Just after I published my last blog on the dangers of plain language I got a newsletter from a plain language advocate who proudly announced the makeover of an ‘informed consent form’ You can see the supporting article and the before and after makeovers at:

http://www.firstclinical.com/resources/journal/0605/0605.html.

This is typical of the well-meaning attempts by plain Language advocates to demonstrate how plain Language ‘makes a difference’. Full marks for effort!

But what does it really demonstrate? Not a great deal. The authors candidly admit:
“The forms have not been tested for effectiveness, so the changes may or may not accomplish the above objectives”.

Now let’s put this in context. This is a form used to get people to participate in clinical trials. Why do scientists do clinical trials? They do them in order to find out whether or not a drug does what they want it to do, without any adverse reactions. They test people before, during and after the trials, and they are not allowed to market the drug unless they provide EVIDENCE that the drug works appropriately.

But here we have a document untested before or after, with no account of any potential adverse reactions such as misunderstandings or inappropriate behaviour.

Yet the rhetoric, full of confidence, tells us the document has been improved:
“Not only is the text easier to read, but the contents are more useful”.
And again:
“Color elements help the reader’s eye move through the form. The colors are bright but somewhat subdued to avoid invoking inappropriate emotional reactions”.
Interesting, even plausible opinions. But evidence? Hardly!
So, on the one hand we have scientific procedures and stringent requirements for evidence, and on the other hand rhetoric.
I do not doubt that this has been an interesting exercise. But, like so much of the work of Plain Language advocates, it relies on a confident rhetoric and a claim to self-evident truths rather than external evidence.

As with potentially dangerous treatments, we need more than the skills of a snake oil merchant to persuade us that a cure is at hand by using plain language.

So chaps, good effort, but next time do the job properly. And, by the way, have a look at some of the research already done on improving informed consent forms. Some of it comes with EVIDENCE.

If I thought it would do any good, I would advocate that every so-called ‘plain language’ document should have a warning.

WARNING!
THIS DOCUMENT CAN MISLEAD

But warnings, like many aspects of communication—including ‘plain’ language—are not all they are cracked up to be (see our recent review of boxed warnings). The simple fact, with far from simple consequences, is that communication is messy and non-predictable (not just unpredictable). Any attempt to reduce communication to a few simple rules fails. Plain Language advocates offer us a few simple rules. They are good rules and in the main, well intentioned. But neither of these are a necessary or sufficient basis for good communication.

Here is a cautionary tale. In 1983 the Australian Federal Government introduced a Plain Language policy, and plain English has been specified in Federal legislation since then. It would now be the case that most government public documents in Australia, at federal, state, and local government levels, are written in ‘plain’ language. The Australian style manual—the authoritative guide on writing for our public servants—advises public servants to use plain language, which it describes cautiously as ‘the standard register’ (p 53). We should, therefore, be living in something approaching plain language heaven, where communication between government and citizen is clear, understandable, and simple. How strange that we are not!

For example, one of the most disliked documents in Australia must be the TaxPack: a comprehensive guide and form for completing your annual tax return (I must confess to having had some part in developing the basic idea. But it is my equivalent of Oppenheimer’s bomb). It is written in ‘plain’ language to the highest standard. It is well laid out and has many good typographical and design features. Yet every year three quarters of the Australian tax-paying public choose not to use it, and instead part with their own money to employ a tax consultant to do their annual tax return for them, spending about one billion dollars in the process. Something is obviously not quite right.

Here’s another example, from our recent work with one of our Corporate Members—a state government organization. This organization wants to redesign a form that has to be filled in by every family in that state. As is our normal practice, we tested the existing form before starting work on redesign. Not one of the people who participated in our testing could fill in the form correctly. Yet it is written in so-called ‘plain’ language and, like the tax pack, has many good typographical and design features.

These are not isolated examples.

I was prompted to write this particular blog because Plain Language advocates in the USA are all of a dither with excitement; there is legislation before the USA Congress requiring all government bodies to use plain language, no doubt in the belief that this will lead in due course to plain language heaven.

Why is plain language heaven as unlikely in the USA as it has been in Australia? I could give a long and complex explanation, but the basics are simple enough. There is no secret to good communication. Good communication grows out of conversation between people with a mutual interest in the same thing. Our bureaucrats in Australia, like those in most countries, do not know how to conduct conversations with citizens, are certainly not interested in doing so through mutual interest, and, importantly, are not required to do so. Even when they seem to be attempting to do so through such devices as surveys and focus groups, it is, like most of what they do in the area of communication, a token exercise in which they can be seen to be doing something without actually doing it. As one public servant explained to me once, when I suggested some real conversations with citizens: ‘You have to understand, our concern is with process, not outcomes.’

And this is why plain language is positively dangerous.

Our research, and that of many others show that when people see a document in plain language—a token simulation of a conversational style—they feel confident that they will be able to understand it. But again, as the research shows, this is not necessarily the case. When they don’t, and are aware of their failure, they are more inclined to blame themselves than the document; after all, it is in plain language! And this is where the DANGER lies. People are lulled into a totally false sense of confidence when faced with ‘plain’ language, as are the critics of governments’ communication with citizens. At least with closely-set 6 point gobbledigook you knew someone was trying to pull the wool over your eyes, or worse. With so-called plain language, it looks as if power and control have been put to one side in favour of open and reciprocal dialogue. Sadly, the Plain Language advocates who believe that are simply naïve.

My preference is for an agreed level of performance for government documents that is backed by evidence, not some token gesture that is based on a particular style of writing. We know from our own research that this is not only achievable but capable of being implemented in practical regulation of communication.

I think the intentions of many advocates of Plain Language are honourable, as are the people who want to warn us about potential dangers, but the road to hell (to paraphrase someone who is not Samuel Johnson) is clearly signposted with warnings in plain language.

Snooks & Co 2002
Style manual for authors, editors and printers 6th Edition
Australia: John Wiley and Sons
“For many kinds of government and other information documents, a neutral style of communication is needed, one that puts neither distance nor undue familiarity into the relationship with readers. This is the role of the standard register which makes no assumptions about its readers, and informs them fully so that they can independently on that information”. (p 52)

The Style Guide goes on to say:
“Plain English roughly equates to the standard register…and plain English has been specified in federal legislation since 1983″. (p 53)