V.

A DISCUSSION OF A CERTAIN TYPE OF

NEGATIVE PROPOSITION.
BY EAPHAEL DEMOS.

IN this article, I propose to discuss such negative propositions as are particular, that is, propositions like " I will not
attend Prof. B.'s lecture," or, " Leather is not as cheap this
year as the year before," or, " Mr. Smith was not feeling
well last week ". I shall take no account of general negative
propositions, that is, of propositions like, " What is white is
not black," or, " No man shall escape death ". Again, of
particular negatives, I propose to select for discussion
primarily simple negatives, postponing consideration of
double or ' n-ple ' negatives to the end of this paper. In
dealing with such particular simple negative propositions,
I shall speak simply of " negative propositions," avoiding
the longer and more awkward expression, and my aim will
be to discover the definition of the negative propositions of
the sort in question, that is to say, their general form and
their relation to the total field of propositions. I propose to
lead up to the constructive contribution of this paper by
stating what, in my view, the negative proposition is not.
(a) A negative proposition is not dependent upon the
cognitive subject in its definition ; consequently, it is perfectly objective, not merely as a proposition, but in its
character as negative as well. I refer here to that view
which defines a proposition to be negative in so far as the
cognitive subject denies it, and positive in so far as the latter
affirms it, and, more generally, endeavours to account for
this difference between propositions in terms of a difference
in cognitive attitudes. Such a view is untenable because a
specific correspondence such. as it presupposes between
character of proposition and character of attitude does not
exist. Moreover, any work on symbolic logic contains many
propositions as members of deductive systems. There, no
attitude seems to be involved, and yet the propositions are
not devoid of their peculiar character as positive or negative.
I shall not further labour my point because thinkers nowadays seem to be practically agreed on its validity.

DISCUSSION OF A TYPE OF NEGATIVE PROPOSITION.

189

(6) The negative proposition must not be taken at its

face value. The reason is as follows: At its face value, the
negative proposition appears to be co-ordinate with the positive proposition but different from it in form, and to correspond, in the world of external objects, to a fact different
in type from a positive fact. Thus, the negative proposition,
" Mr. X is not at home," seems to be formally different from
ihe positive proposition, " Mr. X is at home," and in so far as
true, to imply the corresponding negative fact " that X is not
at home ". Hence a view which adopted appearances would
have to add to the world of positive propositions a new class
of propositions which are negative, and to the world of positive facts, a new class of negative facts. Now, the reason
why such a view must not be entertained is the empirical
consideration that strictly negative facts are nowhere to be
met with in experience, and that any knowledge of a negative nature seems to be derived from perception of a positive
kind. I once undertook a fairly systematic interrogation on
the matter among intelligent acquaintances who had not given
previous thought to the subject and hence were least apt to
be biased in their reply, and they were practically unanimous
in their testimony that they had never encountered a negative
fact and that every case of knowledge expressed through a
negative proposition was in reality of a positive nature, in
a fashion which they were unable to comprehend. Consequently, unless the verdict of experience is to be flouted, the
conception of negative facts must be rejected, or at least not
adopted before the resources at the command of the conception of positive facts and propositions are given a fair chance
to account for the situation.
The above discussion yields the following positive result:
Granting that there are no negative facts, then, in so far as
a negative proposition is asserted of fact at all, the term of
reference must be the world of positive facts. Hence, appearances must be discarded and a special interpretation
given to the negative proposition, which shall allow of such
reference. The clue for such procedure lies in the distinctive
element in the negative proposition, namely, the element
symbolised by the word " not ".
(c) But here again a certain reservation must be made.
The interpretation suggested must not take the form of
regarding " not" as a qualification of the predicate of the
negative proposition, e.g., of defining a proposition like " X
is not white " to be really " X is not-white ". I have in
mind the general view which makes the peculiarity of the
negative proposition appear to be a peculiarity of the predi-

190

BAPHAEL DEMOS:

cate and is thus enabled to define the class of negative propositions as simply a subdivision in the class of positive
propositions, a subdivision namely of all those positive propositions that contain a " not "-predicate. My objections to
this view are as follows: First, the negative element bears
upon the grammatical subject almost as often as it does
upon the grammatical predicate of the proposition. I may
assert that God will not provide because I believe that there
is no God, as well as because I believe that He is nonprovident. Secondly and more important, a large number
of propositions, and specifically relational negative propositions like " X is not to the right of Y," cannot be said to
have any predicate at all.
We have thus far seen that (a) a negative proposition is
an objective entity, in all respects independent of the cognitive subject; (6) it must not be taken at its face value, but
must be so interpreted as to exhibit itself as referring to the
world of positive facts; (c) the interpretation must bear
upon the element " not " in the proposition but must not
take the form of regarding " not " as a qualification of the
predicate in the proposition. In embarking now upon the
positive task of finding out which is the true interpretation
of the negative proposition, I shall follow out the clue already indicated. Evidently " not " is a qualification, not
of any individual element in the negative proposition, but of
the whole content of it.
Thus the statement, " X is not
dead," is really of the form " not (X is dead)," and any
negative proposition is of the form " not-p " where p is the
entire content of the proposition apart from "not," and
" not-p " is a function of p in terms of " not ". Inasmuch
as this content of the negative proposition is positive, any
negative proposition may be regarded as a negative function
of some particular positive proposition.
What is the nature of the modification effected in terms
of " not "? The reply is that " not " is a relational modification of p, and means "opposite of" or "contrary of".
Thus, " John is not at home," or " not (John is at home),"
means " an opposite of (John is at home)," and the statement,
" I believe that John is not at home," is really the statement,
" I believe a contrary of (John is at home) ". To explain,
every relational term is the source of some qualifying expression ; thus, the relation of " begetting " is the source of the
qualifying expression " father ". Now, there is a certain
relation among propositions which, in accord with traditional
usage, I shall call the relation of opposition or of contrariety
or of inconsistency, and which gives rise to the qualifying

DISCUSSION OF A TYPE OF NEGATIVE PROPOSITION.

191

expression " opposite," or " contrary," or " inconsistent

with". The word " not " is precisely a symbol for this
qualifying predicate, and " not-p" means " opposite, or
contrary, of p ". The relation of opposition is such that
if p opposes q, p and q are not both true" (at least one of
them is false). This must not be taken as a definition, for
it makes use of the notion " not " which, I said, is equivalent
to the notion "opposite ". In fact, opposition seems epistemologically to be a primitive notion. The relation of opposition must be, moreover, distinguished from the relation
of contradiction ; the fact that contradiction obtains among
two propositions implies not only that not both are true but
also that not both are false.
I shall now suggest that negative propositions in the form
"not-p,"meaning "opposite of p,"are descriptions of
some positive proposition. However, before I elaborate this
point, I shall proceed to analyse the concept of description
as such, making use of Mr. Eussell's contribution to the
subject (see MIND, N.S., vol. xiv., pp. 477-493, article " On
Denoting"; also "Principia Mathematica," pp. 31-33, 6688, 181-216). (a) Descriptions are all expressions which
are of the form " a so-and-so" or "the so-and-so". Any
entity enters into relations or possesses properties which are
said to describe it, and expressions of the above form are
called descriptions in so far as they constitute mention of
some relation or property of some object. Thus, the expression " servant of Y " is a description of X, through mention
of the relation of service which X sustains towards Y. (b) A
description is definite when it is satisfied by one object only,
and ambiguous when satisfied by more than one object. Thus
" present president of the U.S." is a definite description, and
" Harvard student" an ambiguous description, (c) It is important to note that the object described is not a constituent
of the description. The description is of the form " some
object which is so-and-so," where we have a variable (" some
object") and a function ("so-and-so "), but not the actual
object, (d) Now, descriptions are instruments of reference
to some particular object to which they apply whether definitely or ambiguously. When I speak of the servant of Y,
saying, let us suppose, that he is sick, it is to X that I refer,
and I mean that X is sick. But in so far as, in reference by
description, the object described is not a constituent of the
description, the object is not a datum to the cognitive subject
referring, and in a proposition in which reference is made to
an object by description, the object is not a constituent of
the proposition about it. Any example will make this

192

RAPHAEL DEMOS:

obvious. By means of the description, " the winner of the

next Marathon race," I am referring to some object, but I
am not acquainted with it Again, in the proposition, " The
servant of Y is sick," reference is made to X, but X is not a
constituent of the proposition. Thus, in terms of description,
reference to objects is achieved without the occurrence of the
latter as data in the complex of reference. Now, if " mention "
(or " statement") is denned to mean reference to an object
such that the object referred to is a constituent in the complex of reference, then reference by description may be said
to be reference without mention of the object.
It is as a descriptive phrase in the above sense that the
reader is invited to regard the negative proposition, and
more particularly as an ambiguous description of some
positive proposition in terms of its opposition to some
other positive proposition. It has been already pointed
out that between two given propositions the relation of
opposition may obtain, (a) Now, a negative proposition,
such as " not-p," that is, " an opposite of p," is a description
of some positive proposition q, in terms of its opposition to
p, just as " a servant of Y " is a description of X in terms of
X's serving Y. More specifically, a negative proposition like
" John is not at home," that is, " not (John is at home),"
that is, " an opposite of (John is at home)," constitutes a
description of a positive proposition, like " John is at the
shop," or " John is in the fields," in terms of its opposition
to the content " John is at home ". (6) Inasmuch as there
may be several propositions contrary to a given proposition,
a negative proposition interpreted to mean " an opposite, or,
a contrary of p " is to be regarded as an ambiguous description, (c) In a description, any negative proposition is an
instrument of reference to some particular positive proposition to which it applies, and in any complex involving a
negative proposition reference is made to some positive
proposition of which the former is a description. E.g.,
when I say, " I agree that this is not all mine." i.e., "1 I
agree with a contrary of the proposition ' This is all mine, "
I refer, say, to the proposition. " Some of this is yours," and
I really mean " I agree that some of this is yours ". (d) It
must be pointed out next that reference to a positive proposition in terms of the negative proposition describing it
is achieved without having the former as a datum to the
subject referring, or, in general, without having it as a
constituent in the complex of reference; in a word, it is
reference without mention. E.g., in saying, " I believe that
John is not at home," I am referring to the positive propo-

DISCUSSION OP A TYPE OF NEGATIVE PHOPOSITION.

193

sition, say, " John is at school " (as that which I believe),
but I do not state it. Thus, though negative assertion is in
every case reference to some positive proposition and, hence,
in a sense is assertion of something positive, no analysis of
negative assertion could yield a trace of a positive proposition. In this respect, we may characterise negative assertion as always positive in reference but never positive in
content.
(e) There is still one more point to be made about the
structure of the negative proposition. A descriptive phrase
is incomplete as it appears, and needs to be supplemented
by an assertion of the existence of the object described if
all its meaning is to be made explicit. The exclamation,
" Eain ! " is really of the form, " It is raining," or, " There
is rain," and the proposition, " I saw the servant of Y," is
" There is one who is the servant of Y and I saw him ".
That existence is thus implicitly asserted is evidenced by
the fact that exception may be taken to the above proposition by the retort, " But Y has no servant," or, " There
exists no servant of Y ". Hence, negative propositions are
incompletely stated and, in order to have their whole meaning made explicit, must be supplemented in their stated
content by assertion of the truth of the proposition described
truth being to propositions what existence is to things.
Corresponding to the exclamation of the simple phrase we
have the assertion of the negative proposition, and just as
" Rain ! " is really " Bain exists " (" There is rain "), so
" nct-p " is really " not-p is true," or " an opposite of p is
true," or " some proposition is true which is a contrary of
p ". In general, in negative assertion I am referring descriptively to that proposition which is true. Thus, when
I say that " John is not at home," I have reference to where,
as a matter of fact, John is, that is, to the true proposition
about John, and my statement is " An opposite of ' John is
at home ' is true," or, " The true proposition (the truth) is
an opposite of ' John is at home ' ". In actual usage, it is
understood that in assertion one is dealing with the proposition which is true ; consequently the statement to that effect
is omitted and only the description of q, namely as " not-p "
is given. Thus, instead of "A contrary of ' X is white' is
true," we have only "A contrary of ' X is white ' " or, " not
(X is white)," or " X is not white ".
We are now ready to give a summary definition of negative
propositions of the sort we are discussing. A particular and
simple negative proposition is of the form " not-p is true,"
where p is any positive proposition, and " not " means " an
13

194

RAPHAEL DEM08:

opposite or a contrary of ". As such, a regative proposition

constitutes a description of some true positive proposition in
terms of the relation of opposition which the latter sustains
to some other positive proposition.
The following example from actual usage wili exhibit, in
concrete fashion, all that the above definition signifies.
Suppose you ask me where John is and suppose 1 reply
" John is not at home," what is it that I convey in my
reply ? In asking me where John is, you are asking for the
truth about John, i.e., for a true proposition as to John's
whereabouts. Now, I know that John is at the store, i.e.,
I know that the true proposition which you are looking for
is, " John is at the store ". I may reply to your question
simply by stating this proposition, or again I may refer to
it indirectly, that is, I may describe the truth. Actually, I
choose the latter alternative and reply by describing the true
proposition. The true proposition " John is at the store " is
in fact a contrary of " John is at home," and hence may be
described as a proposition which is a contrary of the latter.
Thus, in reply to your question as to the truth about John,
I furnish the statement " The true proposition, or the truth
as to John's whereabouts, is a contrary of the proposition,
" John is at home ". However, as it is understood that I
am referring to the true proposition, I make no mention of
that, and in my reply I give its description only, i.e., I state
" a contrary of ' John is at home,' " or " not (John is at
home)," or " John is not at home ".
Substantially, the above definition of simple negative propositions applies to double and ' n-ple ' negatives as well;
the latter, too, are descriptions of positive propositions which
are true, in terms of what they oppose. There is this difference, however, that whereas simple negatives are functions
of a positive content, double and other negatives are functions of a negative content, such that any negative proposition in the n-th power is a function of a content which is
negative in the (n-l)th power.
Through the definition of negative propositions just offered,
the world of positive objects is re-established as the ultimate
term of reference in all assertions of a particular nature.
Negative propositions refer to positive propositions and
positive propositions in their turn assert positive facts. In
both cases there is reference to the latter, but in the first
case the reference is indirect, and in the second direct.
From this angle, a negative proposition may be defined as
;a referent to a referent or a description of a description.
Negative knowledge may be defined as knowledge of a true

DISCUSSION OP A TYPE OP NEGATIVE PROPOSITION.

195

positive proposition by description in terms of its opposition

to some other proposition. This type of propositional knowledge is one in which the cognitive subject does not apprehend
the proposition to which he is referring. On being informed
that X is not over twelve years old, I come into possession
of knowledge which is positive in reference, inasmuch as it
is about the positive proposition concerning X's age, namely
knowledge that the truth about X's age is not that X is over
twelve, or " that X is over twelve is contrary to the truth,"
but it is negative in content because I do not know what
X's age is. It is thus knowledge about the positive proposition and not knowledge of it.
I have already hinted at my indebtedness to certain
notions and definitions of which Mr. Russell has made use
in his treatment of classes and descriptive phrases in general.
I have tried to apply to negative propositions the treatment
which Mr. Russell has applied to simple descriptive phrases
or incomplete symbols. After all, my problem is identical
with his. He found himself confronted with the fact that
to accept descriptive phrases as significant in their given
form, would be to people the world of things with the apparent objects of such self-contradictory and fantastic descriptions as " round-square," " centaur," etc. I was faced
with the fact that to accept negative propositions at their
face value would be to people the world of objects with
negative facts, a type of objects which experience fails to
disclose. Mr. Russell solved his problem, partly, by declaring
descriptive phrases to be devoid of meaning in their apparent
form, and their apparent object to be consequently nothing.
I, too, by viewing the negative proposition as an incomplete
symbol, have been led to declare it meaningless in its apparent form, and its apparent objectthe negative factto
be nothing. The parallelism in the further treatment on
the one hand of simple phrases by Mr. Russell, and on the
other of negative propositions by myself, such that the
former are supplemented by an assertion of existence, and
the latter by an assertion of truth, is obvious.
To recapitulate : in this article I have stated, first, that a
particular simple negative proposition is an objective entity
whose peculiarity as negative is not dependent upon the
mind's attitude towards it. I then argued that the negative
proposition cannot be construed in the form which it apparently possesses, inasmuch as such construction would
make it formally different from positive propositions and
would endow it with purely negative objects, which, after
all, are not to be found in experience. I concluded that

1 9 6 RAPHAEL DEMOS : A TYPE OP NEGATIVE PBOPOSITION.

some special interpretation must be given to the negative

proposition, and proceeded to show that its negative element
is a modification not of any distinct constituent in the proposition but of the whole content. Thus any negative proposition is a modification, in terms of " not," of the rest of
its content,and, since the latter is positive,a modification of some positive proposition. I stated the meaning of
" not " to be " opposite,"a relational qualification in terms
of the familiar relation of opposition or contrariety among
positive propositions,and hence the meaning of the whole
proposition " not-p " to be " opposite of p ". I argued that,
so stated, a negative proposition is an ambiguous description
of some positive proposition, and that completely stated it is
of the form " an opposite of p is true," or " some q is true
which is an opposite of p ". Thus I defined a particular
simple negative proposition as an ambiguous description of
some true positive in terms of the latter's opposing a certain
positive proposition, such that, in terms of the former, reference is achieved to the latter. I explained that negative
assertion or knowledge is reference to (or knowledge of) a
true positive by description, and hence must be characterised
as positive in reference but not in content, inasmuch as the
proposition referred to is not a constituent of the complex of
assertion or knowledge.