Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site sdchema.UUCP
Path: utzoo!linus!philabs!sdcsvax!sdchema!donn
From: donn@sdchema.UUCP (Donn Seeley)
Newsgroups: net.nlang
Subject: Re: Creole languages and double negatives
Message-ID: <780@sdchema.UUCP>
Date: Thu, 4-Aug-83 02:00:19 EDT
Article-I.D.: sdchema.780
Posted: Thu Aug  4 02:00:19 1983
Date-Received: Thu, 4-Aug-83 23:45:54 EDT
References: <512@ihuxr.UUCP> <2377@rochester.UUCP>
Organization: UC San Diego Chemistry Dept. NIH Research Resource
Lines: 103

These comments on "double negatives" are interesting:

	[R]ecently, I have tried to explain the logical interpretation
	of double negatives to my children, with no success. This is in
	line with the "Creole" authors' assertion that children readily
	accept the natural features of language (found in Creole), but
	have difficulty with "unnatural" features of formal language.

	I guess I've finally reached a higher level of innocence from
	which I can regard "I didn't do nothing" with complete
	equanimity. Perhaps one can regard the double negative as a
	syntax requiring "matching logical sense", just as the "neither
	... nor" formulation. Its variance with Boolean logic needn't
	be a problem.

			Lew Mammel, Jr. ihuxr!lew

It seems to me that this sort of data indicates that negatives don't
conform strictly to the logical notion of 'not'.  You might view
redundant negation as being a form of emphasis, for example; one
psycholinguistics text that I own remarks that:

	It appears that children often use multiple negatives to
	enphasize the negative character of what they are saying.  It
	is interesting that the majority of the world's languages
	regard this use of multiple negation for emphasis as perfectly
	acceptable, even though English does not.  [Foss & Hakes,
	PSYCHOLINGUISTICS, p. 258]

Under this hypothesis, multiple negatives succeed only in making a
clause 'more negative', rather than applying more layers of negative
predicates to the interpretation.  This doesn't strike me as
particularly revealing, though; there is an intuitive relation between
the negative marker 'not' and some other word in the sentence which is
not caught by this explanation.  Why do we hear 'I don't do nothing'
instead of 'I don't not do something' or 'I don't not do nothing'?
There is a restriction on the 'redundant' negation that goes beyond
simple emphasis.  What is more, there are phenomena in standard English
that seem to be related to multiple negation which don't receive a
common explanation under the 'emphasis' hypothesis.  I'm thinking of
sentence pairs like 'I did something' vs. 'I didn't do anything'; the
morphemes 'some' and 'any' alternate, with 'any' often limited to
'negation' environments (notice you don't say *'I did anything').  It
might be possible to maintain that 'no' is a nonstandard variant of
'any'.

There has been speculation in linguistics that negation markers like
'not' are marking some kind of scope rather than simply indicating that
in the logical form of the sentence there is a 'negation predicate'
applied to the clause.  I'm not going to pretend to be an expert on the
subject but there are some interesting proposals that deal with this
that I thought I might mention...

Consider some data from Italian which are presented by Luigi Rizzi in
his recent book ISSUES IN ITALIAN SYNTAX.  Apparently Italian REQUIRES
'double negation' in many situations:

	Mario non ha  visto nessuno.
	Mario not has seen  nobody
	'Mario has seen nobody.' (11a, p121)

	Mario non ha  fatto niente.
	Mario not has done  nothing
	'Mario has done nothing.' (11b, p121)

Curiously, the presence of 'non' is necessary if the negative pronoun
follows but not if it precedes; for example:

	Nessuno ha  visto Mario.
	Nobody  has seen  Mario
	'Nobody has seen Mario.' (12a, p121)

	Con  nessuno ho       parlato!
	With nobody  have-1sg spoken
	'With NOBODY have I spoken!' (13, p121)

In fact in the latter cases 'non' must be absent.

Rizzi proposes a rule of logical interpretation which assumes that a
quantified noun phrase may be bound at the position of 'non'.  For
certain theoretical reasons, the subject position may not contain a
remotely bound variable in logical form, hence 'non' does not appear
with negated subjects.  If English double negation works the same way,
you would expect to see that

	'Mario hasn't seen nobody.'
	(for 'Mario hasn't seen anybody.')

is okay, but not

	*'Nobody hasn't seen Mario.'
	(for 'Nobody has seen Mario.')

(Well, what do you expect from modern linguistics, anyway?)  At any
rate I hope it is at least plausible to the gentle reader that negation
markers may actually serve as indicators of quantifier scope in some
situations, and that negative indefinite pronouns may be bound
variables.  I can produce even more complicated explanations upon
request, but keep in mind that things like this helped motivate me to
quit linguistics...

Donn Seeley  UCSD Chemistry Dept. RRCF  ucbvax!sdcsvax!sdchema!donn
       (ex-) UCSD Linguistics Dept.     sdamos!donn@nprdc