Hola, Leo.
    Difícilmente puedo entender tu largo mensaje de abajo en inglés. ¿Me lo
puedes reenviar en alguna auxilengua o en alguna lengua territorial
    Gracias, de Alejandro Javier.

----- Original Message -----

Coming back to Auxlang after some years, I note that much of the recent
activity has focused on new IALs that are almost totally Romance in origin.
This is not totally new, since words from Greek, Latin, and Romance
languages play a major role in the vocabulary of most a posteriori IALs.
Latin words have often spread around the world.

The proportion of Greek roots also rises as we look to the "International
Science Vocabulary" (ISV.) While there is much discussion of the degree to
which Romance and related Greek words should be used in IALs, I see little
systematic comment on how to evaluate individual roots from these origins.

This, to me, is the important factor.

Each word is its own problem.

Efforts to create a credible a posteriori IAL would be rendered very
difficult IMO if any effort were made to limit the number of Latin roots
to(say) the proportion of Romance speakers in the current world population.
It would be even more ridiculous to limit the number of Greek words to be
used,by the population of Greece and Cyprus.

Many words of Greek, Latin and other Romance origins are very widely used
beyond those populations. Examine languages like Uzbek and Indonesian, for
example. And more importantly, there is often no other alternative roots
from other origins that are nearly as widely known. It is poor judgement to
attack any specific Latin or Greek based word/root on the grounds that its
incorporation in the IAL is a "Eurocentric" act. Itcould be, if it is
thoughtlessly done. But not necessarily.

Any easy-to-learn global IAL will have IMO many such roots.But this does not
mean that it is logical for us to go back to Dr. Peano and put our thumbs in
our Latin dictionaries to find our words. Nor does it mean it is logical to
copy Gode and survey little more than a stack of Anglo-Romance dictionaries
(with a bit of attention to whether the same words also appear in German and
Russian). Such methods are not the logical way to find the most familiar and
widely used roots for an international language.

It is possible to look, word item by word item, at the vocabularies of ALL
the languages of the world and see what roots are most widely distributed,
taking these as candidates for IAL inclusion. Then, of course, these
candidate words/roots have be evaluated word-by-word in terms of how well
they fit into the system set up for the IAL, its grammar, its usages, its
writing system, its phonetics, its morphemic structure, et al.

Each candidate root would only be chosen if it were sufficiently distinct
from other vocabulary items already chosen, not similar to any obscenity or
offensive matter in any major language, etc. Roots that passed these tests
would be chosen. Some that were not quite ideal might be slightly modified
and then chosen. (A touch of Volapukization!)

How then would each Greek or Romance root best be evaluated (along with
allother potential candidate roots)?

Examples to follow largely cite IALA's "Interlingua." This is not to attack
IL, but only because that is the best documented of the highly Romance IALs.

1) Distribution:
How widespread is the particular root?

If it is in all the WENSA languages, that is a good beginning, but not
necessarily sufficient for a global IAL. WENSA is my term for Western
Europe, N&S America, and Australia (its languages: Fr. Eng. Sp. Port.)

If the root/word is also in German and/or Russian that is much better
evidence that it is a very good candidate. That is the distribution I call

But should it be in most of these and be ALSO found in one or more of the
major languages of Afro-Asia -- say Japanese, Hindi, Turkish, Arabic,
Bengali, or Indonesian -- then it would take a lot of convincing IMO to make
the point that some other root should be accepted in its place. But there
still may be reasons.

There are, incidentally, quite a few Latinate words that have spread this
far, especially those that have gone to Asia via English, French, Dutch,
Portuguese, or the ISV. Needless to say, on the other hand, if a Latin root
is not even in all the Romance languages, our system should assure that we
see if there is a root word from some other language family that IS
widespread. There may very likely be one or even more -- yet most Romance
IAL projects never take such a step -- as a matter of design they simply
take something from one of the Romance languages as a default.

Many will take by default a root that exists only in Classical Latin. This
is knee-jerk attitude is IMO hardly defensible for anyone designing a
language for global use. It is what I would call Eurocentrism, since it
arbitrarily leaves much of the world out of all consideration.

The rest of the world DOES exist.

All peoples do count, and thus the acceptance or rejection of a Latin root
(like any other root) might logically rise or fall on the basis on whether
it is to be found in languages like Bengali, Malay, Persian, Tagalog and/or
Zulu. In some cases, it may be borrowed; in the case of other IndoEropean
languages, it may have been there since the dawn of history. Yet it is my
impression that few IAL designers even look to see whether a word does
appear beyond the WENSA and ENANSA vocabulary distribution zones! (There are
many other wide non-European distributions that I will not go into here.)

Before taking a little known Classical Latin word as default, it would be
far more logical to take a root from Chinese, Hindi, or Arabic -- assuming
that the root chosen fits well into the remainder of the IAL, phonetically,
etc. In the past I have noted that many Auxlang IAL-makers were offended by
such a thought. They found the very idea repugnant! The result will not be
pure, the system will have been bastardized. I suggest that such people try
to fine a "pure" natural language (if they can) and promote it.

Dr. Z. has been criticized for gathering his roots from all over Europe, yet
he must have done something right to have followers even today, including
native speakers of his language. So a whole can be made up from parts from
many origins, as natural languages like English, Japanese, and Uzbek also

Widespread "internationalized" Greek and Latin roots -- Yes; but constricted
ones that are in narrow use and not widely used -- No.

2) Phonetic relationships:
Does the Greek or Romance term fit into the general phonetic system of
theIAL as designed?

Problems would include consonant clusters that are not allowed, or roots
that show too great a similarity to other roots already chosen. Consider,for
example: IALA Interlingua has "strata" for road, and "strato" for layer.
Many IALs would not allow such initial clusters as "str-." Furthermore, the
point can be made that the two words are unnecessarily similar in sound.
IALA Interlingua also has words spelled mnemo-, chthonie, pneuma, etc. which
would not fit most preordained concepts of acceptable IAL phonetics.

The fact that classical Latin lacked a SH or CH sound, presents something of
a problem here, since proper names and borrowings from other cultures often
would seem to require such sounds in an IAL. If a highly Latin vocabulary
has no letter for such sounds, how to talk of Shanxi province,
shish-kebob,Shona, or borsch? Even Novial floundered IMO on this issue.

Greek and Latin roots -- Yes; but those that are unpronouncable by the vast
majority of mankind -- No

3) Morphemic Relationships.
Is the Romance root confusable with other combinations of Romance roots? Can
morphemes be clearly identified?

Many common Romance words seem (to the outsider) to be made up of elements
that are not relevant (i. e., they are not composed of the morphemes that
they appear to contain, or they have lost any significant morphemic
relationship to those elements.)In many cases, "prefixes" and "suffixes" (or
what appear to be such) may not be at all reliable. The prefix "in-" found
in hundreds of words like" incorporate" has a meaning related to the idea
"in/into." But the same prefix in thousands of words like "incompatible" is
a negative. There are many in both sets in IL. (Occidental tried marking one
with an accent mark to distinguish). Note that a word like "expectorate"
does not derive from "expect." Or "detest" from "de-" and "test." The word
"contempt" seems made of "con-" and "tempt" -- yet they give hardly any clue
to the meaning of the word.

There are thousands of common Greek and Latinate words that are "internally
defective" in this way. And there are other confusions needlessly created.
How many people can define the sciences of "pedology" and "nosology." Such
things are candidates for clarification or avoidance.

Greek and Latin words in many cases -- Yes; but internally confusing ones --

4) Semantic Content:
Are semantic relationships between potential roots clear and reasonably
logical? Is there unnecessary dissonance?

In many Romance IALs, the root libr- may be used for a "pound" in weight,
for various units of currency, for a "scale" or a "balance," for the
astrological sign "Libra," for a "book," and even for the adjective "free."
An IAL that is so seemingly ill-planned is not likely to appeal to (say) a
Chinese or Arab.

At the other extreme, IALA Interlingua has (for example) a basic root
"psest-" meaning "rubbed away." This root, while distinctive, seems so rare
in use and so unnecessary (given other Interlingua roots with that can
express that meaning) as to be open to serious question. There are hundreds
of such word roots in Interlingua.

Greek and Latin roots in many case -- Yes; but ambiguous and needlessly
overlapping ones -- No

5) Regularity
Is the candidate root or word from Latin capable of being treated in the
same consistent manner (grammatically) as other words being accepted into
the IAL?

Many Latinate words are far from consistently formed even in respect to each
other. For example, the abstract noun derived from a verb will often -- but
not always -- end in something like -tion in IALA Interlingua.
This shows the type of inconsistencies that may exist:
ager   --- action
arguer --- argumento
bombardar --- bombardamento
favorar --- favor
flecter --- flexion
litigar --- litigation
ligar   --- ligatura
rolar   --- rolamento
usar    --- uso

Importing Latin roots into a IAL should avoid (if at all possible) bringing
in all such legacies of inconsistency from the history of the Romance
languages. Such things have no place in a language designed for global use.
Greek and Romance roots in many cases -- Yes; but irregular subsets of words
based on long-forgotten hisory -- No

6) Word Length:
Is the candidate word of an appropriate length for the likely frequency of
its use?

A five or six syllable Latinate word should not be selected for a very
common word in the IAL. Nor should a one or two syllable word be wasted on a
rare concept or proper noun.

IALA Interlingua has many six syllable words. On the other hand, a short
word, "panna" means "an automotive breakdown" -- despite the fact that "pan"
is "bread" and the forms of "panar" cover the verb "to bread" -- with "pana"
being its present tense. Also a word "panno" is "cloth or diaper," and then
"pan-" is used as a prefix to mean "all." So many usages should not be
crowded into so short a root. The adding of "panna" was IMO particularly

Greek and Romance roots in many cases -- Yes; but those that violate Zipf's
law -- No

7) Number of speakers.
Is the candidate word from Romance actually the most familiar root for this
semantic concept -- worldwide?

In many cases a Romance root, while very common in Europe (and in part of
WENSA) may not be the most widespread root for that meaning in the larger
picture worldwide. One of the most common cases is where there is an
English-origin word that has entered many languages, such as Hindi, Bengali,
Japanese, and a variety of other languages large and small, and is of course
known by speakers of English itself -- and by many who studied some English.

I have run into Auxlangers that did not like my use of the word
"worldwide" -- but in my opinion if you wish to create or support an IAL,
you will have to think in those terms.

8) Nature of the word itself, including places of usage.
Are Romance speakers those who are most likely to be using this individual

We must think of "place of use" as part of the frequency formula. The
concept of "dharma" is more commonly expressed, and "roti" more commonly
eaten, in India than in Iceland. I got into a lot of trouble long ago on
Auxlang with some Esperantists by suggesting that the word "chopsticks"
might most logically come from the Mandarin word for chopsticks rather than
being a compound of European words: "mang^obastonoj." (It would be shorter,
perhaps more like "quaidso" or "kwaize")

How many times is the word being spoken or written by a human being in any
given day -- and where? I believe we must think these things out. Esperanto
allows a kimono to remain a kimono, but how would Dr. Peano find that word
in his Latin dictionary? Many IALs accept "exotic" words from around the
world, but as a marginalized gesture, and as "exotic." But this idea should
apply to less "exotic" things as well.

9) Other
Does the acceptance of this particular root best serve the overall goals of
the IAL?

For example, if a purely Latinate root is a candidate, but so is a rather
more widely distributed IndoEupopean root, the overall goal of a global IAL
might best be served by accepting the IE form.

Thus the form of a root like "brater" for brother might be selected over
"frater" from Latin, despite the widespread use of the Latinate form. The
compelling factor could be that the form with "b-" is closer to the original
IE and thus resonates with the Germanic, Slavic, Baltic, and Irano-Indian
branches of IE -- not just with the Romance. These can be very subjective
factors, but given the wide distribution of Romance forms otherwise, it may
be wise to give the narrow decisions to the alternative sourcing. The
overall result then will be more encompassing and that much fairer to the

Should we accept appropriate Greek and Latin roots -- Yes; should we do so
invariably? -- No.

Best regards, LEO