Re: Dublex (was: Washing-machine words (was: Futurese, Chinese,

Jeffrey:
> Raymond Brown comuni:
> > I can well believe that - but all attempts to reduce meaning to 'essential
> > basics' from the "catalanguage" IALs of the 17th cent to the Dutton's "491
> > root-ideas" of the 20th cent. and contemporary Dublex's 400 primitives,
> > have seemed to me to create more problems than they solve.
>
> No worries, Ray, but that's not at all what I am attempting.  I would never
> group Dublex with those languages, with Ro or similar efforts.  I make no
> claim about these 400 roots reflecting any universal truths or categories.
> I'm looking for an objective, empirical, quantifiable result.  I'm simply
> seeking the 400 roots that will produce the highest percentage of useful
> words from the 160,000 possible two-root compounds.  Someone can objectively
> demonstrate a better 400 roots for compounding simply by creating their own
> system.
How does the Dublex programme go about doing this? For example, are there
different candidate sets of 400? -- E.g. you might start with 1000 and
choose the 400 best. Or you might start with no upper number, but take
words from English and replace them by compounds, recursively replacing
the constituents of compounds by further compounds, until you end up
with 400 'atomic' morphemes. And how is the usefulness of a word
measured? Have you, say, consulted a frequency list of English words?
(BP posted a while back an url to a frequency list based on the 100m
word British National Corpus, which takes the 10000 or so most
frequent lexemes, making distinctions by part of speech but not by
sense; and Michael Rouse once compiled a list based on the whole web
(several years ago, when it was smaller and all in English), but with
items defined by orthographic shape alone.)

--And.

Reply