Re: Most developed conlang

Among the candidate definitions of 'word' in Anna-Maria Di Sciullo and Edwin
Williams's (1987) _On the Definition of Word_ is what they call 'listeme',
namely something that has to be listed because its properties are not wholly
predictable from its parts. IMO 'listeme' comes closest to the definition most
relevant for assessing vocab size as an index of a conlang's degree of
development. Each listeme must be individually created by the conlanger (-- I
set aside ghastly automated processes).

--And.

Dirk Elzinga, On 20/04/2007 20:58:
> On 4/20/07, MorphemeAddict@wmconnect.com <MorphemeAddict@...>
> wrote:
>> In a message dated 4/20/2007 6:43:52 AM Central Daylight Time,
>> theiling@ABSINT.COM writes:
>>
>> [snip]
>>
>> > To get back to a more productive level of conversation, how would you
>> > classify actor words in '-er', like 'speaker'?  '-er' is a quite
>> > productive derivation ending, more so than '-ment'.  If you decide
>> > those are one word, then what is the required productivity level you
>> > start to unify two 'words' into one?
>> >
>>
>> "-er" can be applied to just about any English verb.  So words in "-er"
>> shouldn't count as separate words from the verb.
>>
>> > Then what about 'speak' vs. 'speaking'?  One word or two? ('-ing' is
>> > almost universally productive, I think)
>> >
>> Same with "-ing", only more so.
>>
>> > 'thorough' vs. 'thoroughly'?
>> >    -> -ly can almost universally be used
>> >
>> Same: only one word with a derivative.
>>
>> > 'move' vs. 'moved'?
>> >    -> '-(e)d' is probably less universal than '-er'
>> >
>> Taking "-(e)d" to be a pass tense morpheme and/or past participle
>> ending, and
>> thus including forms such as "was", "went", "gone", I believe these
>> words are
>> regular derivatives and should not count as separate words.
>>
>>
>> > 'hot' vs. 'dog' vs. 'hotdog'?
>> >    -> non-predictable composition of meaning, so a new word?
>> >
>> This, on the other hand, should definitely be counted as separate and
>> distinct words, largely because the meaning of "hotdog" is not
>> derivable from "hot" +
>> "dog".
>>
>>
>> > Or is your measure the distinction of inflection vs. derivation?
>> > Is this useful for languages like Inuktitut?
>> >
>> I've been using 'derivation' to include inflection.  Inflected forms
>> should
>> not count as separate words.  Completely predictable derivations
>> should also
>> not count as separate words.  But derivatives which can't be predicted
>> should
>> count as separate words, because part of the information contained in the
>> derivative that is not contained in the base word is the very fact
>> that such an
>> unpredictable derivative can be made.
>> I don't know about Inuktitut.
>>
>> On the other hand, if pairs such as "develop/development" are
>> considered to
>> be one basic word, then the number of words of the pair's language may be
>> significantly lower than if they're considered separate words.
>>
>> stevo
>>
>> > **Henrik
>
> Two things occurred to me in watching this exchange. First, stevo
> seems to be assuming a "dual-route" model of morphological processing,
> and second, by his criterion of complete predictability, all of
> Henrik's examples count as processes which create new words.
>
> First, since the term 'dual-route' may not be familiar to all, let me
> explain briefly. One definition of the lexicon assumes that it
> contains only the information that is unpredictable (this isn't the
> only definition, but I'll use it since it seems to advantage stevo's
> claims); anything that is predictable is provided by rule. Regular
> word formation, by definition, is predictable, so words formed in this
> way will not be included in the lexicon. However, if the word
> formation process is unpredictable, the result of applying it to a
> base must be listed separately in the lexicon, thus creating a new
> word. Hence, 'dual-route': one route is the regular application of a
> rule of grammar creating a word "on the fly", the other route is
> accessing a word already existing in lexical memory.
>
> The dual-route model is not universally accepted, however. There are
> many linguists (including me) who believe that all morphological
> processing (and probably all phonological processing) is done on the
> basis of whole words used as analogical models or exemplars for novel
> forms (thus the "single-route" of lexical access). That is, we know
> that '-er' is a suffix because we recognize it on a large number of
> words--not because it is provided by a rule of grammar.
>
> Second, I can find examples of irregular or unpredictable usages of
> each of the suffixes Henrik mentions. Just to give one example: the
> suffix '-er' is used to form agentive nouns from verbs; thus 'runner'
> is "one who runs". But it can also be used to form nouns denoting
> instruments; thus 'blender' is "an instrument with which one blends".
> 65 years ago the word 'computer' was understood as an agentive noun
> "one who computes", and large companies which depended on numerical
> analysis hired many people to perform numerical calculations. Now we
> understand the word 'computer' as an instrumental noun. So the result
> of adding this suffix to a stem is not entirely predictable, and thus
> words formed by it must be listed separately in the lexicon.
>
> It won't do to say "Yeah, but I mean the 'productive' use of the
> suffix" without defining your criteria for productivity. Even then,
> it's not an "either-or" proposition; you will find word formation
> processes arrayed on a continuum from highly productive to highly
> unproductive. Dividing the continuum at *any* point will be arbitrary.
>
> Dirk
>

Reply