Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1.1 9/4/83; site scc.UUCP
Path: utzoo!watmath!clyde!akgua!sdcsvax!sdcrdcf!hplabs!pesnta!scc!steiny
From: steiny@scc.UUCP (Don Steiny)
Newsgroups: net.nlang
Subject: Re: Any new irregular verbs?
Message-ID: <168@scc.UUCP>
Date: Mon, 21-May-84 19:27:21 EDT
Article-I.D.: scc.168
Posted: Mon May 21 19:27:21 1984
Date-Received: Wed, 23-May-84 19:38:59 EDT
References: <49700001@hpfcra.UUCP>
Organization: Santa Cruz Computer, Aptos, Calif.
Lines: 77

***

	Irregular verbs are fossels.  They verbs contain the
vowel gradations of Proto Indo-European.  They are not
abbrevations.  The only way that a new one could occur
is by leveling with an existing paradigm.  For instance,

	"sing"
	"sang"
	"sung"

	"bring"
	"brought"
	"brung"

Children often make this mistake.

	English only has two verb tenses, present and past.
If a new verb is introduced in English, its past tense is
formed by adding either a "t" sound or a "d" sound depending
on the the final sound of the verb stem.  For instance,
"grok" in Heinlein's "Stranger in a Strange Land."  The past
tense is "groked".  

	The intuition that irregular verbs are commonly used words
is correct.  That is why they remain irregular.  English does not
mark the distinction between subject an object on the nouns except in
the case of pronouns.   We need to say "she" in the subject position and
"her" in the object position.  "her saw she" is NOT English.  In Old
English EVERY noun was marked as subject or object.  English changed so
that this relationship was indicated by word order, but the inflections
remained on pronouns for exactly the reason that they are so commonly used.

	Suppose we say that there are n verbs in English.
For each verb we must have a present and a past.  In the worst
case there is a third ending, the preterate.  For instance:

	"eat"
	"ate"
	"eaten"

	Even in the simpler case of "bring" - "brought", we must have
two distinct words stored for each verb.  Therefore we must store
at least 2*n distinct words.  The figure is closer to 3*n.

	On the other hand, adding a single ending to indicate past
or preterate means that we will have exactly n+1 items to store.
Though such a blatant analogy to computers might be overkill,
many of us who has tried to learn vocabularies of a new language 
find the idea of a single rule that would cut the number of words we
had to memorize in third so appealing that we might be tempted to call
it "simpler".   No one knows "why" language changes.
By whose standards should we judge the "simplicity" of a language.
Russian has an incredibly complex inflectional system, basically,
their language went the direction of more inflections.  There is
prefered word order in Russian, but in some cases, any word order will
suffice.  That means that if you are going to learn Russian you will have
to learn many endings that are used in many circumstances.  Many English
speakers find this difficult.  On the other hand a Russian learning English
might be frustrated because there is no indication of subject/object
relationship on the nouns, no future tense, and so on.  The "simplist"
language is the one you learned as a child.

	English is a strange and wonderful language.  Surprisingly
its syntax is closer to Chinese than it is to its Indo-European
relatives.   Many of its irregularities are historical records.
They mark the history of the people that used the language and
the marks are often records of events that happened long before
English was a gleam in Proto-Germanic's eye.  

	A good introduction to the history of the English language
is Pyles': "The Origns and Development of the English Language."

				Don Steiny
				Personetics