In a Very Important Sense There Are More Than 26 Letters in the English Alphabet
Rick Garlikov

I would like to give an explanation for why we should consider the English alphabet as having 34 letters (or possibly 36, or even 39) instead of the usual 26 it is said to have.  There are two exceptions to my explanation, and I will point them out after I give the explanation, which is that upper and lower case versions of letters that do not look at all alike should be considered two different letters even though they have the same letter name and even though they have the same sounds in words.  I’ll explain that more fully shortly. 

But first, in this chart

A a

B b

C c

D d

E e

F f

G g

H h

I i

J j

K k

L l

M m

N n

O o

P p

Q q

R r

S s

T t

U u

V v

W w

X x

Y y

Z z


the pairs in the aqua color boxes have very different looking lower-case shapes from their upper-case ones.  For example "G" does not look at all like "g" and "R" does not look at all like "r".  There are eight such pairs.  In the white boxes, the upper- and lower-case letters look identical, or nearly identical, except for size.1 

The three yellow boxes could be considered to have lower- and upper-case letters that resemble each other sufficiently to be fairly readily recognized as being the same letter (i.e., the letter with the same name) or they might be considered different enough from each other to be different letters that sound the same and have the same name.  The pairs in the gray boxes resemble each other more than the ones in the aqua boxes do, but less than the ones in the yellow boxes – the “b” is one curved stroke short of the “B”, and the “h” is one straight stroke short of the “H”).  The letters in the green boxes seem to me to be something of a cross between those in the yellow boxes and those in the white ones -- fairly similar between lower and upper case, but where the dot above the rest of the letter indicates lower case in the same way that size difference does in the letters in the white boxes; the bottom portion of the lower case "i" and the lower case "j" are a smaller size "I" and "J" respectively, but the dot both designates that and also disguises it for someone just learning to read letters.  It might be that students who have difficulty learning to recognize and read letters could be helped by giving them print that has all upper case letters first or all lower case letters first, so that they can learn just the 26 letters instead of having to learn to recognize the variations found in the opposite case.  Once they become more comfortable reading in one case, they can practice reading in the other case, and then finally practice normal print with the two cases combined.  I don't know whether that might actually help struggling readers or not, but theoretically it might.2

I mention all this because it seems to me it has great bearing on the ease or difficulty of learning to read and on reading for people who have difficulty recognizing, remembering, or distinguishing shapes, particularly abstract symbols such as letters.  The more different symbols you have to learn in order to be able to read a language, the more difficult it is for some people, particularly in a language like English, which is not particularly phonetic because some letters have different sounds (such as ‘hard’ and soft “c’s” and “g”s) or where vowels have long and short or other sounds, depending on other parts of the word, and where different combinations of letters have different possible sounds.  For example, George Bernard Shaw pointed out that “ghoti” would spell the word “fish” if you pronounced the “gh” as in “rough”, the “o” as in “women”, and the “ti” as in “nation.

The Two Exceptions To This Idea
1) There are (only) 26 letter "names" -- the names that we say when we spell words out loud or when we recite the alphabet -- names which could be written out, if we wanted to, as "ay, bee, see, dee, eee, eff, gee," etc.  In that sense there is only one letter "gee", whether it is written as "G" or as "g".  It is only in the written or visually symbolized sense that there are two forms of it (or more, if different font styles of it are radically different looking").  In that sense then, there are only 26 letters in the English alphabet, regardless of how many different appearances or shapes those letters are represented by visually.

2) Similarly, the different visual appearances of the 26 letters have the same sounds in the same words or contexts where they appear, although, as pointed out, since English is not phonetic in the sense of each letter having only one sound and always being pronounced the same, the number of sounds letters have in English are far, far more than 26, and are often different in different words.  As one comedienne pointed out on a recent TV show, in the word "eight" the only letter that seems to fit and make sense is the "t". Or others have talked about the "silent p" in "psychology" or the "silent k" in "knowledge", which makes you wonder whether the word "acknowledge" shouldn't be thought of as "ack - nowledge" rather than "ac - knowledge"....

But the point is that those who know how to read well generally never knew, or have forgotten, how complex reading is -- in that it is a visual representation of speech, which is a complex process involving all kinds of intricate sound/symbol correspondence that is something of an approximation to how to write what words sound like.  If you imagine trying to spell what thunder sounds like so someone would recognize by reading it what sound you are trying to convey, or if you imagine trying to spell the sound of a catcher catching a fastball or the sound of laughter or applause, you would see that capturing sounds with letters is no simple, obvious, or automatic task. 

Now if you believe that reading basically involves phonetically sounding out words, then you ought to be able to read the following story, and you ought to be able to read it quite easily. After all, this is how things look to your children when you want them to read by means of phonics. By the way, if you cannot understand it, just read it out loud to someone who is not looking at the printed words. They will tell you what you are reading to them even though you won't know what you are reading. And they won't understand how you can possibly NOT know what you are reading.

Ladle Read Rotten Hut

Wants open a dime, depend add ark and chanted farce, tear vase a would cut her widow eye fanned a ladle gull bide an aim off Fred rotten hut. Wand Ada ladle gull smothers headed waste I'm far hurt ago tour gram mudder souse widow bass cat fool off me tend utters tougher gram auto heifer herd inner. Ladle read rotten hut vase knot toot airy or fuller hound Honda weigh to grammars souse. Butter ladle gull too curt I'm end vase topped buy a be Gary wool few wand it too nowhere sheer gong. "Two migraine mud errs souse toot acre this bass cat ford inner...." 

If you have tried to understand what that paragraph says, but cannot do it, here is what you are reading written in its normal way.

And even for capturing the sounds of words printed in the normal way through normal spelling and typing them out with letters, it is not an easy task for someone to read what you type who has more than normal difficulty learning to identify letters by their shapes.  (And I say "type" to avoid the pitfalls of deciphering handwriting, which makes reading even more difficult generally.)   And there are at least eight letters in English that have two totally different shapes when typed or printed, so instead of learning to read eight letter shapes, one has to learn 16 letter shapes to recognize the eight letters, besides having to learn the other letters of the alphabet also.  So there are more than 26 characters one has to learn in order to be able to read English.  That is not an easy task for someone who has trouble with shape recognition and distinguishing relatively minor differences in shapes.

Footnote 1 for a bit more completeness for those who will find it interesting, but not necessary for the main point of this essay for those who will not find it interesting:

I have used Times New Roman font for this essay and chart because it has serifs, which are the small, decorative strokes (or “feet”) at the ends of lines that fonts without serifs do not have, as illustrated in the upper case “F” and the lower case “A” here, with the serifs circled in red:


The serifs make a difference in the cases of the letters I and L in that it is sometimes very difficult to distinguish upper case “I” from lower case “L” in type fonts without serifs (called “sans serif” fonts, the French word sans meaning without).  It can be difficult in small print even with serifs, as here in Times New Roman “I l”, but is extremely difficult in sans serif fonts like Arial: "I l", particularly out of context or if you don’t know the word. For example, compare Plato’s "Iliad" with "Iliad".   

[The issue of recognizing letters in different font styles, particularly vastly different font styles, is beyond the scope of this article and beyond my level of knowledge.  See for example, Douglas Hofstadter’s “Metafont, Metamathematics, and Metaphysics” at , or his comments about fonts in his interesting book, Metamagical Themas, an anagram honoring Martin Gardner’s long-running column in Scientific American that was named Mathematical Games.Hofstadter was intrigued by what minimal strokes would let us identify letters – or what might be considered the essence of any letter that a font would have to have in order to be readable.] (Return to text.)

Footnote 2:

Some languages have additional variety, depending, for example, on where a letter appears in a word.  For example in Hebrew, the letter called Nun, which is equivalent to the English letter "n", whose name if written out would look like "En":

נ Nun

Nun (Pronounced both noon and nun), is pronounced “n” like the English letter N. It can have a dagesh (that is, a dot placed in the center of the letter), which would look like this נּ but the pronunciation remains the same. Nun has a final form for when it’s found at the end of a word, which looks like this: ןThat is like if English had two different letter shapes that both were pronounced with the 'n' sound and had the same letter name, but where one version was used to write words like "number" or "announce" and the other one was used in words that ended with a slightly different 'n' sound like "soon" or "begin".  The word "noon" would then be written with two different symbols, one for the beginning "n" and one for the ending "n". (Return to text.)

Little Red Riding Hood

Once upon a time, deep in a dark enchanted forest, there was a woodcutter with a wife and a little girl by the name of Red Riding Hood. One day the little girl's mother said it was time for her to go to her grandmother's house with a basket full of meat and other stuff for grandma to have for her dinner. Little Red Riding Hood was not to tarry or fool around on the way to grandma's house. But the little girl took her time and was stopped by a big hairy wolf who wanted to know where she were going. "To my grandmother's house to take her this basket for dinner...."

For more about the intricacy of what reading involves, see "Reading as Children Do".

(Return to text.)