BA/6/93
1. In June 1992 I received letters from the Chair of the Young People's Reading and Resources Committee and from the Braille Resources Officer at Linden Lodge School who both requested that BAUK should give serious consideration to bringing the capital sign into full use in United Kingdom braille, though they realized that such a change could not be implemented overnight. They gave reasons for their views which will appear along with others in the next section of this paper, and it was clear from both their letters that they were not writing on behalf of themselves alone, but with the approval of other educators of blind children. I wrote in reply that I would certainly report the matter to BAUK, which I duly did at its July meeting, but that I did not consider it desirable that the publication of British Braille should be further delayed by having to resolve this potentially thorny question. Instead I would circulate a document intended to stimulate public debate on the subject: hence this paper.
2. The decision not normally to show capitals in braille in this country was taken shortly after braille was introduced here in 1868. It was clearly reaffirmed, explicitly or implicitly, at the Edinburgh conference of 1905 when most of the conventions of British braille as we now know it were fixed. In 1932, when Standard English Braille was jointly introduced in this country and in the United States, we did not follow the Americans, who had never abandoned the use of the capital sign, by reintroducing it here. However, it was agreed on both sides of the Atlantic to change the sign from dots 4-6 to dot 6, and to use dots 4-6 instead of dot 6 for the italic sign. This decision was a clear break with established international convention, and a rather sterile controversy continued to surround it in this country at any rate until well into the fifties. It was doubtless influenced by the consideration that the vehicle for highlighting text used throughout the English speaking braille world should be represented by a more substantial braille character. It was in my view a frivolous and divisive mistake, but I am not proposing that we should seek to undo it.
3. When new attempts were made in the early eighties to reduce divergences within English braille, attention naturally focused on the use of the capital sign, which is the principal area of difference between British and American braille as perceived by most ordinary readers of straightforward literary text. At the International Conference on English Braille held in Washington in September 1982, resolution 15 dealt with this issue. In a lengthy preamble acknowledging the existence of divergent opinions on the question it stated that 'some data support the usefulness of the capital sign in promoting smooth and/or fast reading', and went on to urge that 'further investigation should aid in clarifying the usefulness of the capital sign'. While accepting the continued differences in current braille practice, it went on to express the hope 'that braille rule- setting bodies in countries not now using the capital sign will undertake their own or keep abreast of other countries' investigations into the usefulness of the capital sign, and will be open to adopting the capital sign if such investigations support the hypothesis of the capital sign's usefulness'. There was no equivalent clause urging that countries which used the capital sign should be responsive to any evidence that might come to light showing that it had adverse effects on reading or was unnecessary. Nevertheless the resolution was agreed to by all participating countries.
4. At the International Conference on English Literary Braille held at Westfield College, London, in September 1988, resolution 19 dealt with the capital sign. This asked for three things: firstly, 'that a working group, including representatives from BANA and BAUK, conduct a review immediately of the present rules as well as the suggestions made in conference papers on the use of the braille capital sign, and that any recommendations be circulated to ICCOELB members for approval, and [secondly] that in the meantime no additional functions shall be assigned to dot 6 in English literary braille'. Finally it asked that 'reference be made in the Restatement to the rules for use of the capital sign contained in [the American codebook].' The first part of this resolution was implemented immediately after the conference by the setting up of Study Group G on Composition Signs, which of course include the capital sign. The third part was implemented when British Braille appeared at the end of 1992 with an extended set of rules governing the use of the capital sign, though these are not copied directly from the American codebook, and are at variance with it in a number of respects. As regards the second part, a British amendment to add after 'functions' the words 'which conflict with its use as the capital sign' was not accepted by the conference. The amendment was mainly designed to accommodate possible developments in advanced contracted braille. At this conference I also agreed to write a paper to launch a public debate on the subject in the UK, which I am now doing.
5. The current position is as stated in British Braille at the beginning of rule 5.2.1: 'Capital signs are not generally used in the United Kingdom. When they are, the following rules apply'. It is thus possible for those who wish to use the capital sign to find guidance in British Braille as regards the rules that should be applied, and there is nothing which forbids anyone to do so. Some educational establishments already employ the sign in braille produced for their own internal use, and some small producers will do it on request. But none of the major publishing houses does so; nor is this situation likely to change in the near future unless BAUK gives them a green light. Therefore most readers in this country rarely see the capital sign in use, except in books borrowed from countries where the representation of capital letters in braille is the norm. This now includes most major producers of English.
6. The distinction between small and capital letters is a print feature which can have semantic as well as pictorial significance. Not to show capitals in braille is to deprive the blind reader of information which is directly available to his sighted colleague. Some people would go so far as to regard this as a form of censorship. The reader should be left to decide for himself what use, if any, he wishes to make of this information.
7. There has been a steady drift of English speaking countries towards capitalization. Canada and New Zealand adopted it many years ago, and in Australia and South Africa capitals now predominate. All these countries once followed the British model. There is no example of a country which, having once embarked upon capitalization, even if initially only on a tentative or limited basis, has subsequently gone on to reject it.
8. The use and recognition of capital letters has now become an attainment target in the national curriculum. Knowledge of how to use capitals correctly is to be regarded as a criterion of literacy in the same way as the ability to spell and punctuate accurately also are.
9. Knowledge of capitalization is necessary in the acquisition of keyboarding (including typing) skills, which many blind youngsters will require for both educational and vocational purposes. Learning the rules needs to be reinforced by constant confrontation with the capital sign in use, if people are to apply them correctly and spontaneously. Not all uses of capital letters in print are obvious: for example, the use of internal capitals in some commercial names, such as 'WordPerfect', where both the W and the P are capital; or conversely, the use of small initial letters for words which were once proper names, to show that they have now become part of the common language, such as braille and moon types, which are now commonly written with a small b or m. Blind people will only become familiar with these conventions through seeing them in what they read, and failure to reproduce them accurately in their written communications with sighted people may be viewed by some as an indication of a more general inadequacy.
10. To always or never show capitals in braille would both be simple choices, but it is conceded by everyone that the second of these choices is simply not available. Not only are capitals regularly shown in technical codes, most foreign language codes, and unit abbreviations, but there are some purely literary situations where British Braille recognizes the need to show them occasionally. They may be needed, for example, to make a semantic distinction which could not easily or clearly be made in any other way (rule 5.2.2); to distinguish between small and capital Roman numerals (rule 7.3.7); to show a change to ordinary type for emphasis in the middle of an italicized passage (rule 5.4.4 (b)); and they are regularly used in brailling typists' exam papers (rule 5.2.4). Such occasional use of capitals makes them seem alien and conspicuous in a way that is not related to the function they perform. Moreover, there are some distinctions which are not considered by British Braille to merit special treatment: for instance, the use of print capitals to mark different levels of heading, or to show the personification of abstract nouns, especially in poetry and heightened descriptive prose. Evidently there is no clear line to be drawn between cases where the information conveyed by capitals is essential rather than desirable or merely marginal. Would it not be simpler and therefore better if producers of braille did not have to make such judgments at all?
11. British public opinion is known to be opposed to the general introduction of the capital sign in this country. The standpoint of educators is therefore exceptional, and the influential position they occupy should not allow their views to outweigh those of other braille user groups. Two recent surveys have been conducted into the state of British public opinion with regard to the braille code. The first was carried out by Dr John Gill as part of the Birmingham/Warwick University Study Of Braille Contractions published in 1980. Gill's question 28 asked: 'Would you like the capital sign to be used wherever a capital letter occurs in the ink print?' Of his 301 respondents, 64 (31%) said yes, 215 (72%) said no, leaving 22 (7%) who gave no clear answer. The second survey was conducted by BAUK in the summer of 1986. The full results of this survey have never been published, but all responses and other relevant documents are held in an archive at Birmingham University. About 1550 people answered the BAUK questionnaire, but just over 300 of these were either not resident in the United Kingdom or read braille only by sight. In three papers presented to BAUK I analysed and discussed the responses of the remaining 1217 to most of the non-profile questions. BAUK's question 49 asked for people's attitude to the use of the capital sign, and gave them five choices: A to use the capital sign wherever a capital letter occurs in print; B to use it only in all schoolbooks; C to use it for proper names etc, but not just to mark the first word of a sentence; D to use it only when necessary to clear up confusion; E never to use it. In my analysis I amalgamated the D and E responses, since the latter is not a realistic choice, and very few opted for it anyhow. Of my 1217 respondents, 193 (16%) answered A; 149 (12%) answered B; 171 (14%) answered C; and 622 (51%) DE. Thus my A response is 5 percentage points down on Gill's, but clearly the extra options provided by BAUK revealed a greater diversity of significant opinion than Gill was able to elicit: 82 (7%) of my respondents gave no clear answer. Since only 51% of my respondents favoured the DE option, and the aggregate of ABC responses was 513 (42%), the outcome might be considered to be quite close as between those who favoured an extension to the use of the capital sign and those who favoured the status quo. However, BAUK's questionnaire only asked for first preference votes, and did not seek to ascertain what other options (if any) people would find acceptable if they could not have their first choice. It would seem reasonable to infer from Gill's findings that most of the 26% of my respondents who favoured an intermediate position (options BC) would have come down on the side of the status quo if forced to choose between that and full capitalization, unless there was a shift in public opinion between the time when Gill conducted his survey (1977) and the circulation of the BAUK questionnaire (1986). In this connection I was able to show in my third paper that the answers to question 49 (and indeed to some other questions) were age related. I divided my respondents into four age groups: 24-, 25-44, 45-64, 65+. The percentages in each of these four groups who made a DE response were respectively 43, 45, 50, 51; the percentages making an A response were 35, 18, 12, 9. It is clear that, while there is a slight tendency for support for the status quo to go up with age, there is a massive increase in support for full capitalization as people get younger. I argued in my third paper that this is part of a more general drift towards following print more closely among younger respondents. Clearly, if this trend with regard to capitalization has continued in the years since the BAUK questionnaire was issued, then the present state of public opinion may be much more evenly split.
12. Capitalization in braille, unlike print, occupies extra character spaces. If we were able to consider an 8-dot literary code, this would not be the case. But as things stand, the increase in the number of characters due to the capital sign is more than 4.4% according to the Kederis Count, conducted in the US in 1965, but still in my view the most comprehensive and reliable estimate of the frequency of braille punctuation and composition signs so far made. (There has been other attempts to measure the extra space created by the capital sign, one reported in a paper by Darleen Bogart written for the Westfield Conference in 1988, which gives a rather lower figure, and another published by me in 1985 in a comparison between grades 1 and 2 and four trial codes, which gives a figure much closer to that of Kederis.) It seems clear that the type of material used caused some variation in the result. Kederis used straightforward prose texts suitable for both children and adults. A 4.4% increase in characters, if evenly distributed throughout a text, should be expected to cause a small but significant loss in reading speed. It might also have a slight adverse effect on production costs.
13. The capital sign is perceived as clutter by a good many braille readers. (It is only fair to add that punctuation and composition signs generally, in so far as they appear not to convey direct meaning, are quite often perceived in this way.) It is interesting to note that in my experience some of the strongest opposition to showing capitals in braille comes from people who lost their sight as older children or adults, and who therefore know how capital letters are used in print, but whose fluency in reading braille is no more than average, and who therefore resent the presence of too many characters which get in the way of their comprehension of the text. But this impression of clutter is not simply due to the number of extra characters which the capital sign involves, but is also related to their distribution and the way in which they can adjoin certain other characters. For example, the need to use the double capital sign before each of a succession of words printed in block capitals is burdensome to the reader, especially in headings, where it is liable to increase the number of lines occupied. Or again, studies have shown that right hand characters tend to create perceptual problems in reading and that dot 6 is the one most easily missed. According to Kederis and his colleagues, of the 255 braille elements (more or less equivalent to what we call signs) which they counted, the capital sign ranks sixth in frequency, and is the only one of their top ten elements which is not an alphabetic character. Dot 6 is the seventh commonest character, and the next commonest right hand character is dot 5, which ranks thirtieth. The capital sign accounts for 63.7% of all right hand characters, and for 97.45% of all composition sign characters! Finally, the number of capital signs is equal to 68.4% of the total number of punctuation sign characters. It is particularly unwelcome when the capital sign occurs in contact with the italic, letter or accent signs, and the frequency of this phenomenon depends on what rules are in operation. I tend to favour clarity of expression as against economy of signs, but that would be a matter for future negotiation among different jurisdictions. Finally, it is known that perceptual problems are caused when the capital sign adjoins certain other characters: eg dot 6 followed by the letter A can be misread as dots 3-4; dot 6 lower H as dot 5 H; and so on.
14. Because of inherent differences between tactile and visual scripts the capital sign cannot perform the same function in braille that capital letters perform in print. Capital letters stand out conspicuously, so that it is possible, for example, for a sighted person on request to locate the opening words of a sentence within a prescribed area of text with great speed. By contrast braille cells are of uniform size, and characters on a page cannot be picked out individually by touch anything like so quickly. All that the capital sign can do, therefore, is to indicate to the braille reader that there is a capital letter in print; but the extra emphasis which this conveys has to be inferred rather than directly perceived. (The same kind of argument could also be used in relation to other type change indicators.)
15. For people with no sight who are able to use braille, and for some others, print is a secondary script, needed only to enable such people to communicate with their sighted colleagues. For some people, and in some situations, this communication is of course of vital importance. But should one have to accept the kinds of inconvenience outlined above into one's own primary script simply for the sake of making it a more accurate replica of someone else's?
16. From the end of the eighteenth century until over half way through this one there has been in this country a generally agreed standard of correct usage with regard to capitalization that applied in nearly all cases; indeed the national curriculum may be seeking to recapture this state of affairs. But both before and since then things have been somewhat different. There are of course still major areas of general agreement: you should capitalize the first word of a sentence, the pronoun 'I', possessives and pronouns relating to God, and of course proper names, provided you can get people to agree on what constitutes a proper name. But all these rules can be easily learned at the time one first needs to acquire keyboarding skills. But nowadays people may capitalize, or fail to, for all kinds of special purposes. Small letters may be used for bylines, and capitals to mark words belonging to a category to which the author attaches a particular importance in his discourse. Modern verse and experimental prose exhibit an even greater diversity of practice. No wonder literate sighted people quite often seek my advice on whether or not to capitalize, and are exasperated at the frequency with which I tell them that in my view it is optional. We cannot therefore surely any longer defend the use of the capital sign as a tool to inculcate a universally agreed standard of correct writing, but to acquaint braille readers with the increasingly wide range of variation that is acceptable in print. But when English spelling was fluid in the sixteenth and seventeenth centuries, the printer of one of my books would no more have thought it worth his while to copy the idiosyncrasies of my personal spelling into his published text than to try to imitate my handwriting, unless I had been exceptionally insistent, eg as a participant in the controversy which then raged over whether there should be a universal standard of correct spelling, and if so what. Is there not a lesson for us here today?
17. There would be some costs involved in going over to full capitalization, in terms of retraining transcribers and modifying computer translation programs. Are we sure that the benefits would outweigh these costs?
18. There has been a long tradition of minimal use of the capital sign in this country without any apparent major ill effects. Although a number of English speaking countries have moved away from this tradition, there are countries in other language groups which have not. (Russia, Italy and Turkey are I think still in this category.) Is there any empirical evidence of a higher error rate among braille readers with regard to the correct use of capital letters than with regard to punctuation and/or spelling, or as compared with tape or print readers?
19. (A) The simplest option would be to take no action, leaving people free to apply the rules of capitalization given in British Braille if they wanted to, but knowing that outside school environments braille producers, especially the major ones, would be most unlikely to do anything of the kind. We would then need to explain to teachers why we considered no change desirable.
20. (B) The next simplest option to implement would be to go over to complete capitalization immediately, or rather, as soon as practicable. It would then be necessary to explain to braille users why we had decided to make this change without any consultation or public debate, knowing as we do that the best empirical evidence available suggests that public opinion would still be preponderantly hostile.
21. (C) To openly acknowledge a dual standard of correctness, and state explicitly that BAUK accepts that braille with or without the capital sign is correct. At first sight this might not seem in any way different from the status quo; but with appropriate publicity, we could give positive encouragement to producers to experiment with full capitalization in some or all of their publications. Such a course of action might best be pursued at the end of a public debate.
22. (D) To introduce full capitalization into educational books only, however these would be defined. This is the option favoured by my teacher correspondents, though they see it as a first step towards universal capitalization. Indeed I myself believe, on the basis of what has happened in other countries, that the adoption of any intermediate option would be the inevitable prelude to universal capitalization. My teachers suggested that we should start with national curriculum core study texts and children's literature. This would have what might be seen as the merit of insulating older braillists from the change for longest, given that opposition to change appears to be strongest among the 65+ age group. The teachers also recognized that, even if option B were to be implemented immediately, it would be a considerable time before most of the books actually being read would contain the capital sign and therefore reinforce the learning of its correct use. We have larger existing book stocks than most other English speaking countries.
23. (E) To use the capital sign only in cases where the presence of a capital letter in print would not otherwise be obvious, however these would be defined. We might, for example, not use it for the first word of a sentence or for the pronoun 'I', but always elsewhere. However, if the first word would have had to be capitalized for some other reason even if it had occurred in some other part of the sentence, eg because it was a proper name, would we then capitalize it for consistency with other instances of the word, or would we leave it unmarked like other first words of sentences?
24. (F) We could phase in capitalization gradually in some other way, with a declared intention to make it universal within a specified period of time. We could, for example, produce capitals for customers on request. It would of course be possible to generate braille copies with or without capitals from the same computer file. However, not all braille is produced in this way.
25. I do not think we should allow our decision on the capital sign to be influenced by our attitude towards the Unified Braille Code project (UBC). If we believe that the adoption of UBC is not in the British interest, we can still decide to go over to the capital sign if we wish without seeing this as the first step towards the inevitable acceptance of a package which is not beneficial to us. On the other hand, if a form of UBC is evolved which we feel we can subscribe to, we can still, if we wish, opt out of the full use of the capital sign, as we have up to now in Standard English Braille.
26. The rules governing the use of the capital sign in UBC are of course likely to be significantly different from those in British Braille. But, as with other things, we should in any case in my view seek to negotiate with BANA over existing differences in rules and other matters relating to the capital sign. These should include its use in contact with other composition signs, especially the letter sign; its status as a lower sign; and the significance to be allocated to the triple capital sign. However, if in the meantime we were to decide to adopt option B with all speed, the rules given in British Braille are adequate to enable texts to be transcribed. Updating British Braille should in any case prove to be much more straightforward than it used to be.
27. I have tried not to let my personal views affect my presentation of the case on either side. Having written this paper, I intend to play a low key role in the forthcoming debate. I have tried to put both sets of arguments with equal force and fairness, and I have aimed to incorporate answers to each side in the arguments of the other. I would particularly not like it to be concluded that the extra space I have devoted to the arguments against change is a measure of their superiority in my judgement. This came about because of the amount of statistical material which I deployed in this section, and because I wanted to float some ideas which I thought might be relatively unfamiliar to some readers. On the other hand, the directness and straightforwardness of the arguments for change will carry conviction with many people who will see many of the arguments on the other side as convoluted or sophistical. I do not believe that if capitalization was now already established in this country, people would expend such efforts to persuade us to abandon it. However, there is one thing I would like to make clear. I intend to use my casting vote, if it is necessary to do so to break a tie, in favour of no change rather than some change, on the ground that, if BAUK is to spearhead a campaign to turn round public opinion on this issue, it must at least have a clear majority within itself for doing this. On the other hand, if it were to be a choice between complete change and only some change, I would cast my vote in favour of the former, on the ground that, once we have decided to go ahead with change, we should bite the bullet and not prolong confusion and uncertainty, not to mention controversy, through a staged transition. The clear cut options ABC are in my view to be preferred to the intermediate ones DEF.
28. We now have to decide: (a) whether to choose one of the six options listed above immediately, or whether, as I think we should, to conduct some kind of public consultation exercise; (b) if the latter, whether to circulate this paper more widely (in an abridged form, no doubt) to launch the public debate, or to do so in some other way; (c) what other action, if any, needs to be taken in connection with the capital sign question.
Bill Poole - Chairman, BAUK