> [mailto:[log in to unmask]] On Behalf Of Steve Rice
> By definition, a mnemonic is "a device or code" "assisting or
> intended to assist  memory" (Merriam-Webster's Collegiate
Dictionary, 11 ed.)

Right, but to assist in memorization it would have to have some sort
of connection to the learning existing knowledge base of knowledge.
Even initial-letter recognition has some value, but not as much as
having a whole word that resembles one already known.

> The idea of "b" for "biology"
> >involved somewhat unrelated memorization exercises like those
> >typically found in those "improve your memory" courses where
> some sort
> >of connection is made.  I don't doubt that this makes the
> words easier
> >to remember but that could probably be done with any word if you
> >wanted to reach far enough.  I'm thinking more of mnemonic hooks
are
> >are more intutive, where the words bears some resemblance to words
> >already known.  A direct shot, rather than having to pass through
two
> >or three steps.
>
> But once you know the schema, you have such a direct shot. It's like
requiring
> all learners to first pick up a "language" with no grammar  and a
handful of
> words that will then provide the kind of mnemonic you're talking
about.

"Once you know the schema" is the key phrase here.  Until/Unless the
schema is learned there isn't any real value to the system.  It could
be argued that once you learn *any* schema you are on the way
regardless of where it comes from.


> The schematic approach isn't a bad one though, except
> >that I having all relational words start with "m" means they'll
sound
> >too much alike.
>
> As I said, there are other ways of handling the method, and I
> wouldn't use this
> one myself. However, the point can be countered pretty easily:

> ...

> 2. I never said that the relational words would all start
> with "m"; I said their
> first consonant would be "m." So if they start with a vowel
> (or perhaps glides
> are considered honorary non-consonants), we could have ama
> and yome as  well as muta. That should break it up a bit.

It could, but this is really not a memorization aid, but more a way of
hinting at a word's meaning for those who are unfamilar.  "Well, it
has an M so it must be some type of ...".


> 3. Something like this already occurs in natlangs, as in the
> nasal-oriented
> words snore, sneeze, snoot, shnozz (minor variation), etc.
> These don't get confused. Perhaps the problem is in how the division
is made.

These often occur because they are derived from the same word, though
these words are also long enough to be distinct. They are also whole
words, not words made up of components.  I would assume that a
schematic system would contain some really small pieces that would be
put together in some methodical way.

FWIW: I toyed with some of these ideas in the initial stages of
Sasxsek's development.  There are some remnants of it still there like
negative forms all beginning with "n-", indefinite/impersonal forms
with "w-".


> 4. A minor variation of this method would use the final consonant.
That would
> be just counterintuitive enough, I think, that although you  might
know there
> was a system consciously, your subconscious wouldn't perceive it and
call it a
> fake. Though some rhymes could arise, they could either be minimized
or
> banned (I wouldn't recommend the latter).

I see the E-o/Ido POS markers as being the beginnings of a schematic
system.  The -o means basically "thing" (in a very broad sense), the
"-i" means something like "action", "-a" meaning something like
"quality", etc.  A more schematic system would just have more of these
markers.  Instead of "noun", it would break nouns down into classes
like maybe "person", "animal", "thing", "place", "emotion", etc.


> >I'm not a fan of SSM.  It tends to strip words of their unique
shapes
> >that make them easier to identify, and I have yet to see where it
has
> >any realy value since the learner of even an SSM language will
still
> >have to train his/her mind to parse the morphemes according to
those
> >rules.
>
> That assumes that they have unique shapes. If the morphology  of a
system
> boils down to CV(n) syllables (easily pronounced), some resolution
method is
> called for. Besides, if we're dealing with an a priori
> system, we aren't borrowing
> basic words, so there's no "unique" pre-existing shape anyway.

In an a priori system, just about anything would work so you could
easily put in just about any morphology.  The CV(N) system works well
for the largest numbers, but again it's a fairly bland shape.  There
are shapes that are still pretty easy like maybe C(G)V(N) which give a
little more shape to the words because you can have a few different
variations like CV, CVN, CGV, CGVN.


> As to the difficulty of segregation rules, I agree that  they're
often too
> complex. But if every morpheme begins and ends with a vowel, and a
glide, for
> example, is inserted for compounds, then any VV sequence will be a
word
> juncture. How hard to hear and parse is that? (And again, there are
other
> ways of doing it.)

I've seen a couple of languages that try the V-V for a word boundary.
That happens to be the one I find the worst.  Listen to Spanish
speakers, who like to run bordering vowels together,and you'll quickly
hear why.


> >I do have some philosophical and a priori ideas like this I'm
toying
> >with, but I wouldn't propose them as auxlangs.  I don't know  of
any of
> >these languages to gain much support, but I have noticed a bit of
> >interest in Sona lately from some of Rick Harrison's posts on other
> >lists.
>
> I'm currently working on a very different textbook for Sona.
> I got stuck a time
> or two, but then I decided to abandon the usual model and do
> something a bit
> more appealing (and politically incorrect). It's mostly
> written longhand at the
> moment, but I'll be typing it up properly as time permits.
> I'm going for about 10-
> 12 lessons plus an odd kind of reader.

I'm thinking about something sort of like Sona, but with a different
schema and lexicon.  Lexically, I'm thinking about maybe just some
type of shorthand English, which would probably have some sort of
mnemonic value like you mentioned before with the B for "bio-" idea,
though more along the lines of single syllable parts like "zo" (short
for "zoo-") to mark animals.