First, it should be admitted that this idea would
make learning to spell correctly more difficult and
time consuming because the spelling rules would
tend to be even more arbitrary, phonetically, than
the standard rules of English spelling. However it
would make actual reading by an experienced reader
much faster becuase the eye could identify each
distinctive word shape more quickly.
First change: A special font would be designed
which has no upper/lower case distinction. In the
place of capital letters would be alternate lower
case letter forms which would be distinctly
different from the traditional lower case letter
forms. In addition, the lower case letters would
be modified to make them more visually distinctive
from each other.
Second change: A systematic set of alternate
spellings would be defined for each sound. For
example, it could be declared that the letters
"j" and "z" when they immediately follow "t",
"c", "s" or 'P' have the same effect as the
letter "h". Thus the "sh" sound could be written
"sh", "sj", or "sz". Also, given that there are
two alternate lower case letter forms for each
letter which replace the upper case letter forms,
that sound could also be written "Sh", "sH",
"SH", "Sj", "sJ", "SJ", "Sz", "sZ", "SZ" for a
total of 12 different ways to write "sh".
Similar rules would be introduced for vowel
sounds like "ate", "ait", "ayt", "ajt", etc.
Third change: The "correct" or canonical spelling
of a word would be choosen based on making that
word as visually distinct from similar words as
possible. Thus "fish" and "dish" might be spelled
"fiSz" and "dIsj" (remembering that what appears
here as Roman upper case letters are actually
alternate lower case forms in the new font.) The
general rule would be that two words must differ
from each other by two letter forms at the very
least. "them", "then", "they" might become "tHem",
"TjEn", "tzeY", for example.
Spellings would not be canonized in isolation, but
as part of a set of similar words with due regard
to how each word in the set is to be spelled. Thus,
in deciding on the canonical spellings for the
4-letter set { bail fail hail jail mail nail rail
sail tail wail veil gale male pale tale } it would
have to be insured that each word had the minimum
of two shape distinctions from every other word
in the set. for example:
bail fAil haIl jAIle mayL nAjL RAeLE... etc.
Thus no two written word shapes could be easily
confused even when their sounds are very similar:
"far", "fur", "fear", "fair", "fire" ->
"far", "phuR", "FeJr", "Pjair", "pziRE"
--gary
Fascinating concept. The guiding principle being maximizing the
distinctions among word-recognition strategies, or some such, I
presume?
This is a new idea to me, in the conlang universe. Never thought
of it before. Cool!
:)
Yes, that's the idea. We normally read by recognizing the whole word as a
unit anyway. Why not make it easier to recognize that word shape?
>
> This is a new idea to me, in the conlang universe. Never thought
> of it before. Cool!
>
> :)
>
The idea kind of grew out of another idea to design a conlang with an
alphabet that was strictly visual, where each letter had no assigned sound,
but wsa used only for its shape.f That idea turned out not to be very
practical since there would be no way to figure out the sound of an
unfamiliar word. This new system retains the phonetic values of the letters
and so it is still possible to figure out the sound of an unfamiliar printed
word.
--gary
I think that words written similarly are likely to be mistaken for each other
when the difference involves letters that are similar in form, e.g. n and m, or
some vowels that look similar, especially in sloppy handwriting (a and o, i and
e, or a and u). If the writer forgets to cross a t, it may be mistaken for an
l. To reduce the risk, several letters could be redesigned to be more
distinctive. Fonts using these new designs could be used to display existing
text files and web pages without any need for conversion or memorization of new
spellings, since it's a matter of getting used to the shapes of some letters
being somewhat different from their current shapes. Our brains are already
somewhat used to seeing some letters with variant shapes, such as cursive or
the different ways that a's and g's are written in some fonts, and serif vs.
sans serif, and capitals vs. lowercase.
Interesting idea. Certainly much less ambitious than redesigning the whole
system of English spelling.
--gary