An Etiological Approach to Accessibility
Return to Accessibility Criteria
A response to Steve Tyler's Paper: A Reconsideration of Accessibility - do we know what it means?
Date: 25/04/2008
Article
Steve Tyler's constructively challenging contribution to the discussion of usability and accessibility [i] asks us, explicitly and implicitly, to consider the mathematics of the terminology: is accessibility a sub-set of usability; or is it a 'super-set' which would imply the reverse; or are they on a spectrum with each at one end? The last proposition is, frankly, confusing because it simply wants to evade the issue by saying that x could be a sub set of y or vice versa. He concedes this when allowing that a rigid sub-setting or nesting of one class inside the other would be to give the demarcation a sharp definition which may be theoretically attractive but which is practically unhelpful.
My provisional starting point is to draw a distinction arising from his closely argued Section 8 on Examples and Discussion. Let us take his example of a person who knows no French being confronted with a printed text in that language. My starting point would be that the distinction we need to make is between an ignorance of French as the result of chance or indifference and an ignorance of French resulting from an incapacity. In other words, although social policy does not like this kind of distinction - because it raises the issue of whether, for example, heavy smokers, warned of the dangers of their habit, should be treated free of charge by the National Health Service or whether the Social Services should devote large amounts of resource to clients who make no effort to change their underlying situation - it might be a good starting point because it has one major point in its favour: it allows us to make a statement about choice and independence of action. The person who by chance or indifference knows no French has the power to change that situation, all things being equal; she can alter her priorities and, say, watch less television and spend time at an evening class. Talking of which, it might be as well here to dispose of the economic argument; if she cannot afford a French class this does not mean that her lack of choice arises from accessibility in the sense I want to use the word. The class is inaccessible to her because of her poverty but to use this word, as Tyler suggests, to cover all kinds of barriers to enjoying goods and services, simply confuses the issue. What he is saying, in effect, is that all people who can choose to change their circumstances in order to access goods or services suffer from a lack of accessibility if they do not exercise the choice. What I am saying is the precise reverse: that accessibility issues only arise where there is no choice. Yet even this is not precisely the case. The extent to which people can choose to redistribute their meagre income to learn French or make a lifestyle sacrifice in order to move to a less geographically isolated location on the one hand is of a different order from the choice a person has about blindness or blackness on the other but the choice in the first case may be limited and so the distinction is not quite so clear cut in practice as it is in principle. What defines the kind of inaccessibility we are considering is the degree to which the person has no choice in respect of his condition.
To return to the main line of the argument, my starting point is to consider why people are the way they are rather than to consider in what way they have problems relating to goods and services. After all, we are becoming increasingly aware in our multi cultural society of people who arrive in the United Kingdom with very limited access to goods and services because of their lack of English but we know from experience that there is a whole array of responses to this from enthusiastic learning to downright hostility. This, in turn, raises the interesting social question of the extent to which public resources should be used for multi lingual public information except for those who are unable to choose to learn English because of some innate incapacity; and behind this we are back with the argument about smoking and health services.
If we start with the innate incapacity to interact with goods and services in their 'normative' mode, i.e. in the mode they are generally utilised, we will immediately have to ask ourselves what should be done about this and who should do it? Before answering that question directly, I want to introduce the idea of the creative or authorial intention. If an organisation designs and produces cups to enable liquid to be transported but these are used for, say, juggling or bookends, this does not negate the initial intention. This is a central idea because I want to use the terms usable and accessible to describe the extent to which the end user realises the creative intention. If a few people use cups for juggling or for book-ends, all well and good, but the overwhelming majority of cups will be used to transport liquid. Of course there are occasional massive subversions where the majority of sales of an item are accounted for by a purpose which the author could not have imagined but the idea is a relatively simple one.
Provisionally, then, I am placing usability and accessibility, and their respective negatives, in the same container. On the basis of the example of the French text I can say: if a blind person and a non French understanding person are confronted with an identical piece of printed French text they are equally incapable of realising the Author's initial intention; but they have a very different set of options.
If we consider this example further, we will have to enter the complex world of economics and social justice. The person who knows no French can choose to attend a class; a person who cannot access print under any circumstances could choose from a number of options from resigning himself to being excluded from the text to commissioning and paying for the production of a braille copy, with the intermediate position being, perhaps, persuading a friend to read the text to him over a number of evenings at no charge. This immediately raises a socio-economic question about the extent to which the individual should use his own resources to pay for his individual realisation of the author's intentions and the extent to which this is recognised as so onerous that the obligation is public and collective. I will return to this later.
I immediately grant that the experience of the braille reading French reader is not identical to that of the print reading French reader and I think this point is often overlooked; at the very least the aesthetic experience is different. So let me next graduate to a discussion of images. Setting aside, for the moment, the cost of rendering images into a linguistic form for blind people, any such rendition can only make the image partially accessible and its authorial intention usable for the blind person. Equally, and this is critical, the same image might well be wholly inaccessible or unusable for people with certain kinds of autism. If we think about this a little further, the authorial intention of a line drawing of a road layout may be almost completely accessible and usable via a linguistic rendering whereas a congenitally blind person will never understand the iconic significance of the smile of the Mona Lisa; she will simply know that it is iconic.
We have therefore reached two provisional definitions: a phenomenon is inaccessible to the extent that its initial authorial intention cannot be realised by the user because of innate incapacity over which no choice can be exercised; and, in turn, accessibility is the extent to which the authorial intention can be rendered through intervention.
We might note that there is an issue here to which we have alluded before; to what extent is the problem of a very old person with a VCR a usability or an accessibility problem? The answer is that it is difficult to make a clear-cut distinction. We know that there is a clear authorial failure to make adequate provision to have her intention fully realised but disentangling the cause of the difficulty with the device is near impossible.
This in turn leads to the proposition that an author, fully aware of the array of an audience, can choose to make a phenomenon more or less usable and/or accessible. He could, for example, make his plot more usable by sticking to simple narrative rather than inter-cutting; and he could make the plot more accessible by including a closed captioning and audio description strand to run in parallel with the drama. This raises a very important cultural issue; rendering a phenomenon more usable may make it less rich and the author needs to be aware of the trade-off in order to optimise the extent to which her intention is realised. Breaking up a complex sentence into shorter sentences may make it more usable but less useful. On the other hand, whether the plot is simple or complex, the parallel strand is a specific provision for rendering the authorial intention in a way which will increase its accessibility or, to put it more accurately, to decrease its inaccessibility. This pair of examples illustrates yet another aspect of the subject: the author can choose a level of richness which takes account of the ability of end users to grasp his intention and in doing so he can deliberately include or exclude whole categories of end user.
Summing up the first part of the argument, mathematical questions would appear to be the wrong kind of questions; the more appropriate kinds of questions would seem to revolve around the etiology of the functional gap between the authorial intention and its reception by the end user. This, in turn, has massive implications for social and economic justice. It might be argued, for example, that collective provision should be made to render gaps as narrow as possible which arise through a lack of choice; and, in turn, if we live in a society which is attuned to that kind of argument, the etiological distinction between accessibility and usability might be more useful than the mathematical distinction. In the end - and not unusually - it is the taxonomy that matters, not the phenomena.
Returning, then, to Tyler's Paper. In considering the way in which people are able to enjoy goods and services, it might be helpful to think in terms of what is peer normative rather than in terms of what is average for a peer group or for a whole population. This does not alter the degree of accessibility of a phenomenon but it might determine the extent to which an investment is made in narrowing the gap between the authorial intention and the user. If we put most emphasis on narrowing gaps for people in their peer normative ecologies we may be able to prioritise resources. We do this in an instinctively enlightened way by accepting that a child's requirement for a brailled textbook is more pressing than an adult's demand for a brailled novel; but behind this ready ranking there are many other issues where we are much less sure. This, in turn, has tended to lead us to think of accessibility as a set of standardised outcomes instead of as a personal and particular tension between an author's intention and an end-user's experience.
We might, then, want to modify our definitions just a little further by saying that accessibility and inaccessibility are peer-normative and personal phenomena rather than technological phenomena. Tyler might have pointed out to advantage that a blind person with an average range of skills might find learning braille impossibly difficult which gives him another facet of inaccessibility over which he has no choice. Some people have problems concentrating on audio; and others find the trade off between the size of print and the string of symbols impossibly non productive. You can make a piece of text as manipulatively pluralist in its rendering as technology will allow and some people will still find it totally inaccessible.
The real interest in the distinction between accessibility and usability in the way in which it is usually couched is a symptom of technocentricity; it poses the question: to what extent is the functionality gap between producers and consumers a phenomenon which can be understood in terms of the producer? This is certainly a good question because it gives salience to the issue of authorial choice in widening or narrowing the usability and accessibility gaps and the number of those who experience them; and the extent to which there is authorial choice will in turn influence the discussion about who pays for greater accessibility, the author or another body. This is not the place to discuss this issue in detail (see WorkBook One) but it does lead to my central point on finance. Our social policy tends to mitigate disadvantages both at the point of its experience and as the result of its etiology. To that extent we would wish to campaign for greater public investment in accessibility compared with usability and we might also expect more and more stringent legislation to achieve accessibility. To that extent, as a driver of public policy, the distinction between usability and accessibility is much more than a matter of academic niceties. Tyler's approach takes us much further than we have been before, well past the simplicities of the alternatively nested phenomena of usability and accessibility; but any final set of definitions should respectively take account of the radical difference between choice and no choice and between potential full realisation of authorial intention and irremediable but vitally partial alternative rendering.
[i] A Reconsideration of Accessibility - do we know what it means? - Donald Broadbent Lecture, The Ergonomics Society 2008 Annual Conference, Nottingham, UK. April 2008
By Steve Tyler
Head of Innovation and Disability Access Services
Access and Innovation Group
Royal National Institute of Blind People