A Fresh Look at the Sign System of the Braille Code

ERIC P. HAMP & HILDA CATON, Ed.D.

Mr. Hamp is Robert Maynard Hutchins Distinguished Service Professor of Linguistics and Behavioral Sciences, University of Chicago, and linguistic consultant, Patterns: The Primary Braille Reading Program, American Printing House for the Blind. Dr. Caton is project director, Patterns: The Primary Braille Reading Program, and an associate professor, Department of Special Education, University of Louisville.

Dr. Caton, Depatment of Special Education, School of Education, University of Louisville, Louisville, KY 40292

Abstract: An analysis of English braille as a written code from a linguistic viewpoint was undertaken as one of the steps in the development of Patterns: The Primary Braille Reading Program. The analysis led to a regrouping and fresh description of the many elements of English braille, grade 2, with intention to facilitate both the teaching and the learning of reading braille.

There is no dearth of documentation to support the assertion that much of the difficulty young visually handicapped children encounter in learning to read can be ascribed to the fact that the materials used have been transcribed into braille from print editions. The problems made by this practice have been studied and described in detail (cf. Ashcroft, 1960; Bleiberg, 1970; Caton, 1979; Lowenfeld, Abel, & Hatlen, 1969; Nolan & Kederis, 1969).

In an attempt to overcome some of these problems, a primary braille reading program has been developed at the American Printing House for the Blind. Development of this program, Patterns: The Primary Braille Reading Program (Caton, Pester, & Bradley, 1980), began with an analysis of all available research related to the braille code, including the studies cited above.

However, as this analysis was conducted, it became apparent that the information gained was not adequate for the development of the most effective materials for teaching or learning braille reading. One major problem was that the existing categories, or groupings, of the various elements of the code were confusing to both teachers and students-primarily because the descriptive language used was inadequate. For example, terms like "contraction," "sign," and "symbol," are used interchangeably to describe whole-word contractions, part-word contractions, short forms, alphabet word lower-cell words, upper-cell words, dot-five words, whole-cell part words, etc. Many other examples of the confusion caused by existing groupings and descriptions of the braille code could be presented if space permitted; but the point is that it was not possible to describe the braille code clearly to the students who would use it to learn to read. Another problem was that no teaching material existed that reflected the internal characteristics of the braille code, and teachers were forced to use materials reflecting the characteristics of print. That is, all sequencing of vocabulary, reading skills, and teaching strategies was based entirely on the principles of print reading.

For these reasons, it was necessary to conduct an internal analysis of the braille code that would result in a regrouping and fresh description of its various elements, thus facilitating both the teaching and the learning of braille reading. The new descriptions and groupings allowed teachers to discuss and describe the various elements of the code in a clear, precise manner, so that their students were not confused. In addition, the analysis clarified the internal characteristics of the code, so that teachers could devise more effective teaching strategies.

The authors wish to emphasize that the information provided through the research mentioned above is relevant and significant, and, in fact, was used to a large extent in sequencing the vocabulary and skills in Patterns: The Primary Braille Reading Program. We maintain, however, that the earlier work did not sufficiently recognize the truly specific characteristics of the braille code; and that the internal analysis discussed in this paper does so.

In order to present our results in a clear concise way, we begin by describing total process involved in the development of the reading program; first discussing the problems addressed in the design of the course material and in the internal analysis itself.

The entire study was undertaken from a linguistic viewpoint. The authors wish to emphasize that the analysis and explication of braille throughout the paper is done without appeal to the characteristics and setting of sighted print-an intrusion irrelevant to our analysis and too often invoked.

Toward An Internal Analysis

It has long been recognized that a major problem in the teaching of braille reading arises from differences between the representations of natural spoken language in print and in the braille code. These differences have tended to be depicted in terms of additions or subtractions; that is, people have been occupied with contractions as being somehow shorter by subtraction from the full print form, and the like. Attention has also been focused on the particularities and tactile difficulties of shapes of the dot configurations within cells. This is clearly important in itself, yet has little to do with the functions which have been allotted to various character groups.

Less obvious is the fundamental difference in internal structure between many aspects and subsections of the braille code and those of a full font of type. A further important point to be stressed is that of the nature and status of relations, in an abstract sense, between braille and print. That is to say, there are clear and partly simple relations that hold between print and the braille code - for, after all, the braille code was devised on the basis of print, or writing; but frequently the learner of braille is not already in possession of a knowledge of print or writing.

Problem Areas

After reflecting on such questions, we have arrived at the conclusion that two large and essential problems confront us at this time:

l. The present lack of a thorough and relatively abstract internal analysis of the braille code stands in the way of our ability to confront effectively and confidently many of the major decisions involved in designing adequate learning materials. Although we do not minimize the high intellectual fascination that such a formal analysis holds, we have been impelled to procede by practical concerns; the following analysis reflects this, and the ordering of elements is, to a considerable extent, influenced by our sense of practicality in teaching and in course design.

2. In scrutinizing materials and practices employed to date in teaching braille, and even those explanatory materials that authoritatively present the braille (English Braille American Edition 1959), we are impressed by the prevailing tendency to analyze the code and explicate problems via elements, processes, assumptions, and customs based on visual experience. Even though all concerned are well aware of the situation of the blind student, and particularly of a young person who has never had the opportunity to be exposed to print, much of the discussion and explanation is carried on as if all members of the dialogue had a degree of experience with the rudiments of printed or written English. So, for example, in materials of discussion directed to a blind student, the braille unit that conveys the dental spirant sound at the beginning of a word such as thin is usually talked about as though it were obvious that it should be spelled in conventional English with the letters t and h. Of course, no teacher of braille thinks of the word thin as starting out with the two sounds t and h; and by mentioning this letter combination to a person who has never seen it, the teacher can give an initial impression only of irrelevancy at best.

We propose to deal with these problem areas by presenting, as compactly as possible, the results of a fresh and completely internal analysis of the braille code.

In this analysis the characteristics of visual print are never used as internal elements of the system to be analyzed. Rather, the features of visual print are segregated as a separate system external to the braille code; that is, as a set of elements with which the braille elements may be contrasted, and for which relations may later be defined between the two. Our analysis both characterizes the internal braille code and states the relations between braille and print. We shall return shortly to a more precise statement of how this mode of analysis is to be conducted.

Major Steps in Ordering

To keep clearly in view our ultimate practical aim in light of the problems discerned above, we propose that the goals for the devising of teaching materials should follow these major steps:

1. A relatively complete internal analysis of the braille code. We say "relatively complete" in order to isolate major issues; to see in which direction one must move; and, finally, so as not to overload the analysis and presentation with endless detail or rarely occurring, nonessential elements. We restrict the analysis to the generally used major part of literary braille, and content ourselves with giving as examples only representative or crucially interesting instances for most of the classes discussed. A fairly complete list of the resulting categorization is given in Box 1.

2. Design of the most direct and reasoned route toward mastery of the elements and their combinations, and of a compact and maximally simple notation to carry out such discussion. This means that all teaching materials were ordered for the presentation of all elements for which a principled ordering can be determined. This is not simply a question of leaving nothing to chance; it is a matter of proceeding logically from the known to the less known.

3. Introduction of visual elements, including print and writing. (As mentioned earlier, these elements must be clearly discriminated for the purposes of the analysis; we are not here concerned with the problem of the correct introduction of these visual elements as a goal of the learning and teaching process.)

4. A careful consideration to be given, finally, to the correct phasing and insertion of the activities specified under numbers 2 and 3: in short, the effective teaching of braille and the appropriate teaching of print. (Again, this task falls beyond our present purpose.)

As we have just observed, numbers 3 and 4 fall outside the scope of the present study. Number 2 depends crucially on number 1, and the publication Patterns: The Primary Braille Reading Program represents the first essay toward fulfilling the objectives of number 2. The following discussion, therefore, addresses itself to the objectives of number I only.

Braille and Print Writing:

Their Mutual Relation

If no English speaker used any mode of representation for English other than braille, the analysis of braille would not differ essentially from graphic analyses of one or another writing system now or formerly in use in the world. Of course details would differ. The writing and printing of English is essentially an alphabetic enterprise. The mixed representation that braille employs consists of symbols for letters, syllables, and whole words. Consider these examples: the letter b in the braille word boy; the syllable er in the braille word exercise; and the whole braille word people. An entire world of English speakers using only braille would present us with the straightforward task of correlating the braille system, internally formulated, with the grammatical and semantic facts of the English language. Thus this job would be analogous to, and only more complex than, that of correlating conventional English spelling with the way our words are put together. But braille users do not constitute the entire English speaking population; nor is it expected that those who use braille will fail to master conventional print English also.

This fact imposes a separate dimension on the analysis, with a corresponding task for analyst and teacher; a dimension arising from two principal facts of language use that have impinged on the character and function of braille.

  1. Historically, of course, braille is not independent of print English. It has been devised over time, with print English preceding chronologically and lurking in the background as a partial model.
  2. Braille will, moreover, be learned by students who wish to convert with maximum efficiency that knowledge and skill into a competence with English print, so that they can type.

If our analysis of braille is to take these two aspects into consideration, the resulting formulation of the relations of the internal braille system with the structure of print English will be complex in a special way; that is, it will distinguish more classes of elements and functions than if only the braille system or only the structure of print were being considered.

At every stage of this analytical process, we must ask ourselves:

  1. What forms are distinguished by what elements and what combintions in terms of braille shapes and braille units.
  2. How each of these forms is correlated with, that is, equivalent to, not equivalent to, or partially equivalent to, elements and features and combinations of print English.

We see immediately from this that there are two important aspects to every determination that we make for a braille element: the configuration of the element itself (so and so many cells consisting of such and such arrangements of dots), and the element or elements of print English with which this braille shape is found to be correlated.

In the theory of signs as elaborated in the technical literature of linguistics, such a relation is known as a "sign relation," and the combination of the signed (the braille configuration in this case) and the signified (the print English) is known as a sign. To invoke an analogue, this value, or function, stands in relation to the signifier much as a meaning does in relation to its signifier that we call a word. Thus, what we are calling value here is much akin to the notion of meaning; but, for the present, we shall not use the latter term in this sense, lest we cause confusion with a different type of sign relation; namely, that of linguistic semantics, with which we are not here concerned.

This consistent correlation of print values with the distinguished braille elements may be called for convenience a value system, or sign system, in the sense just specified. It is in such an analysis, as well as in the details of its execution, that our approach differs fundamentally and to the greatest extent from earlier formulations and presentations of the braille code.

The following internal analysis presents a relatively detailed description of this consistent correlation of print values with the particular braille elements corresponding to them. The presentation consists of a set of new terms and the regrouping of all elements of English braille, grade 2, as well as the new descriptions of these elements previously discussed.

The Internal Analysis: A New Approach to the Braille Code

To begin our internal analysis, it is necessary first to distinguish the following primes of Braille. For clarity, we have restricted the meaning of each prime to a single sense.

Box 1: Outline of Braille Terms-Examples

1. Letters

a. Alphabetic letters (26)

1). Nonalphabetic letters (12)

(1) 0-9

decimal point

fraction bar

(2) Other braille Units with abstract letterlike function

accent sign

apostrophe

asterisk

ellipsis

hyphen or dash-when used to indicate missing letters in words

2. Grams

a. Phonograms

ally (Sally) ed (red)

ance (dance) en (pen)

and (sand) ence (fence)

ar (car) er (certain)

ation (nation) ever (several)

bb (rubber) ff (duffle)

ble (table) for (forest)

cc (occur) ful (awful)

ch (chair) gg (suggest)

com (come) gh (ghost)

con (contrary) here (adhere)

dd (paddle) in (pin)

dis (dispel) ing (sing)

ea (read) ity (city)

b. Morphograms

after (afterlife) ful (wonderful)

ally (mathematically) here (cohere)

ance (avoidance) in (indecent)

and (multiplicand) ing (singing)

ar (secular) ity (rarity)

ation (admiration) less (useless)

be (befriend) ment (ornament, monument)

com (commiserate) ness (openness, oneness)

con (confuse) sion (aversion, confusion)

dis (disengage) some (loathsome)

ed (rubbed) through (throughout, throughway)

en (encephalogram) tion (reaction, prediction)

ence v(providence) th (seventh)

er (Zipper) there (therefore)

c. Logogram

(1) Letter word

as just so

but knowledge that

can like us

do more very

every not will

from people it

go quite you

have rather

(2) wordlet

about always cannot ever know

above and character father herself

according be child first himself

across because children for itself

after before conceive friend thyself

afternoon behind conceiving good myself

afterward below could great yourself

again beneath day had oneself

against beside deceive here ourselves

almost between deceiving him themselves

already beyond declare his yourselves

also blind declaring immediate

although braille either in

altogether by enough its

3. Modulations

a.Punctuation

(1) look back

colon period

comma question mark

exclamation point semicolon

(2) enclose

bracket or brace (in pairs)

comma (in pairs)

parenthesis (in pairs)

quotation marks, single (in pairs)

quotation marks, double (in pairs)

(3) link

bar long dash

hyphen bracket or brace (one)

dash

b. Register

capital sign, single letter sign

capital sign, double number sign

italic sign, single termination sign

italic sign, double

Cell. This term has been used in more than one sense in the literature. Here cell is defined as an abstract space twice as high as it is wide, in which there arc six positions arranged in three rows and two columns, in which dots may appear.

Shape. We defined as a shape a single configuration made up of one to six dots and occupying a single cell. Note that, so long as a shape is defined in this fashion, it does not yet have any necessary value or mean-ing, and therefore is not a sign in the sense used in the sign theory alluded to above.

Dot. A dot is defined as the element of which shapes in a cell are composed. The dots of a braille shape occur physically as bumps or bosses.

Braille unit. We now require a term for any shape or shapes taken together in correlation with its (or their) value or meaning. It would be convenient and nonoffensive to ordinary English if the word character could be used for this concept, especially since that is the established English term for the element of Chinese writing that closely resembles in function this braille element. But unfortuntely this term has already been employed for some time by users of braille in a deceptively similar but ambiguous sense; and so, unless users of braille wish to alter their terminology, it will be necessary to find some other term for this pivotal concept in the present analysis. For this purpose the term braille unit has been proposed by us.

A single braille unit may consist of one or more shapes; as, for example, go = 1 shape, tion = 2 shapes, etc. Braille units fall into three major types-letters, modulations, and grams.

1. Letters. Letters are either alphabetic, such as b, or nonalphabetic, such as 2.

a. Alphabetic letters (or letters propper) have print-alphabetic values; that is to say, these braille units match occurrences of ordinary letters in print. For example, the shape has the value of i.

b. Nonalpbabetic letters comprise (1) numbers, the decimal point, and the fraction bar, and (2) certain other braille units with abstract letterlike segmental function, such as the asterisk and the apostrophe. The numbers, of course, take the number sign, and then the shapes with the resulting number value may be thought of as letters of a numerical alphabet consisting of twelve shapes (counting the decimal and the fraction bar) and spelling number words. The asterisk may be viewed as a rather odd unpronounceable letter, and the apostrophe often has the implied meaning "letter left out." The reason for classing these two as if they were letters or numbers is that, like conventional letters, they have sequential segmental properties in linear order, and the characters they match in the print text consist regularly of single segmental print shapes. 2. Modulations. Modulations are of two rather different sorts: punctuation, as for example, the question mark; and register, as for example, italics. What these have in common is that they "do things" to, that is, have effects on, other elements-the segmental elements in the chain.

a. Punctuation, in fact, has print values that are, themselves, sequential in position; but these braille units differ from others in having domains of effect extending at times to considerable distances to the right and left of their sequential position. Some punctuation looks back; examples are the period and the non-Spanish exclamation point. Other punctuation encloses; examples are the hyphen and the dash. Those that look back have as a domain of their force what has gone before; those that enclose both warn us of their application and close their doman; those that link affect things on both sides. b. Register is the term applied to those braille units that include what have traditionally been called composition signs; these units look forward, and may also automatically specify where the scope or domain terminates. Examples are capital, italic, letter and number signs. These elements always have the effect of modifying the basic segmental values of what follows; thus, they change the dress of some elements, such as lower case into capitals, or change what we think of as type style, such as italic, or change letters into numbers, or change the abstract reading of an element, such as the letter sign. Registers have the unique property of finding no separate segmental counterpart in print.

3.Grams. These are of three kinds: phonograms, such as the ance in dance; morphograms, such as the ance in reliance; and logograms, such as the words rather, the, friend, mother, and immediate. The distinction between grams and the two preceding units is that, unlike modulations, they are segmental in value; but, unlike letters, they have no single counterparts in a type font. Because of this last property, these are the braille units that later will give rise to bidirectional problems in writing and spelling.

a. A phonogram is a braille unit having a phonetic value that would be written in print by more than one alphabetic symbol. Phonograms are either one-shaped, like th, ch, gh, the ing in sing, the ea in read, the ed in bed, and the ar in bar, or multishaped like the ation in nation, the ound in sound, the ong in long, the ence in fence, the ity in pity, the ness in Tennessee, and the less in bless. b. A morphogram is a braille unit having the value of an element in a word, such as an inflectional ending, prefix, or suffix. Examples are the s in words, the ing in looking, the ed in looked, the ance in avoidance, the ation in admiration, and the in in inconsistent. Note that the shape(s) that make up ing, ed, ance, etc. appear as phonograms or morphograms, depending on their function in words, that is, their "value." c. A logogram is a braille unit made up of one or more shapes having the value of an English word (conventionally, a chain of letters between spaces) with either limited reflection or no reflection at all of phonetic values. There are two principal configurations of logograms, single-shape and multishape. Single shapes comprise letter words and wordlets. A letter word is a logogram that has a shape the same as that of a letter. Examples are but, can, do, rather. Wordlets comprise all other logograms carrying a word value. Examples are and, the, shall, still, there, ought, young, those, enough, cannot, paid, declare, was, to. It should be observed that logograms do not lose their status as such when they are derived by affixes. Thus spirit remains a logogram, in this case a compound-letter word(let), when it occurs as part of the derived adjective spiritual. Such a definitional provision avoids the need for pedantically encumbering the analysis of renaming hosts of wordlets "morphograms" because they may form parts of longer derived words.

Summary

As we have stated, the internal analysis as described in this paper was used for regrouping the various elements of the braille code and devising new descriptions of these elements. This was done to eliminate the many conflicting and confusing terms and categories previously used by teachers and students, and to provide them with a new system consisting of a relatively small number of categories, or groups, and clear, precise, linguistically based descriptions of all the elements of English braille, grade 2.

In addition, we have presented a discussion of the braille code, based on linguistic principles, which is intended to clarify the internal characteristics of the code. Specifically, it is intended to dispel the notion that the braille code and the principles of braille reading are in a way analogous to the print code and the principles of print reading.

We believe that the results of this analysis can be used to:

  1. Provide teachers with an outline of braille terms that will enable them to describe and discuss any element of the braille code in a manner easily understood by children.
  2. Emphasize that the teaching of braille reading and print reading are not analogous, and promote an understanding of the internal characteristics of the braille code, thus providing teachers with more effective strategies for teaching reading to children who use braille as their primary medium.

References

Ashcroft, S. C. (1960). Errors in oral reading of braille at elementary grade levels. Unpublished doctoral dissertation, University of Illinois.

Bleiberg, R. (1970). Is there a need for specially designed reading series for beginning blind readers? New Outlook for the Blind, 64, 135-138.

Cardinale, J., & Cline, C. (1973). Methods and procedures of braille reading. AFB Research Bulletin, 171-183.

Caton, H. R. (1970). Institute report.- Primary braille reading. Louisville, KY: American Printing House for the Blind.

Caton, H. R. (1979). A primary reading program for beginning braille readers. Journal of Visual Impairment& Blindness, 73, 309-313.

Caton, H. R., Pester, E., & Bradley, E. J. (1980). Patterns.- The primary braille reading program. Louisville, KY: American Printing House for the Blind.

Chall, J. S. (1967). Learning to read.. The great debate. New York: McGraw-Hill.

deSaussure, F. (1959). Course in general linguistics. New York: McGraw-Hill.

English Braille, American Edition. (1959) American Printing House for the Blind. Louisville, KY: 1972.

Lowenfcld, B., Abel, G. L., & Hatien, P H. (1969) Blind children learn to read. Springfield, II,: Charles C Thomas.

Nolan, C. Y, & Kederis, C.3. (1969). Perceptual factors in braille word recognition. American Foundation for the Blind, Research Series, No. 20. New York: American Foundation for the Blind.