Berkeley's Theory of Signification

Armstrong, Robert L. (Robert Lawrence), 1926-

Journal of the History of Philosophy, Volume 7, Number 2, April

1969, pp. 163-176 (Article)

Published by The Johns Hopkins University Press

For additional information about this article

http://muse.jhu.edu/journals/hph/summary/v007/7.2armstrong.html

Access Provided by UFSC-Universidade Federal de Santa Catarina at 11/25/11 10:27AM GMT

 Berkeley's Theory
of Signification
ROBERT L. A R M S T R O N G

THE PURPOSE OF "THIS PAPER is to examine Berkeley's theory of signification and

the application of it in his metaphysics. It is my hope that attention to the theory
of signification will afford a fresh view of his immaterialist philosophy and make
some of his more puzzling arguments easier to understand.
The discussion of abstract ideas in the introductory chapter of the Treatise
Concerning the Principles of Human Knowledge I will serve to introduce the theory
of signification. Berkeley maintains that the belief in abstract ideas is the chief
cause of our difficulties in the area of "metaphysical knowledge" and accordingly
devotes his first chapter to an attack on this belief. He begins by denying that the
human faculties are designed by nature only "'for the support and comfort of life,
and not to penetrate into the inward essence and constitution of things" (Intro.,
par. 2). Man's mind is finite, and thereby, he admits, it becomes confused when
trying to consider infinite matters. But he insists that the fault is not with the
faculties themselves but with "the wrong use we make of them" (par. 3). If we have
an appetite for metaphysical knowledge, he suggests, then God must have furnished
us with the means of satisfying it- The difficulties that have "blocked up the way to
knowledge" are "entirely owing to ourselves . . . . we have raised a dust and then
complain we cannot see" (par. 3). The chief cause of these difficulties is the belief
that

the mind hath a power of framing abstract ideas or notions of things . . . . These are
in a more especial manner thought to be the object of those sciences which go by the
name of logic and metaphysics, and of all that which passes under the notion of the
most abstracted and sublime learning, in all which one shall scarce find any question
handled in such a manner as does not suppose their existence in the mind, and that
it is well acquainted with them. (par. 6)

The source of the trouble, Berkeley argues, is the mistaken notion that the mind
has a power to form abstract ideas which go beyond particular sense images and
yet are supposed to be true to the nature of external things. For example, we see

The Works of George Berkeley, Bishop o] Cloyne, eds. A. A. Luce and T. E. Icssop,
9 vols. (London and Edinburgh: Nelson, 1949), II, 19-113.
164 HISTORY OF PHILOSOPHY

objects extended and colored and, observing what is common to them all, frame the
abstract ideas of color and extension. We then procede to consider or think about
color and extension entirely apart from particular objects. Or, to take a more
concrete example, the mind having observed "Peter, James and John resemble each
other in certain common agreements of shape; and other quahties" it leaves out the
details wherein they differ and forms an abstract idea of mankind (par. 9). But
Berkeley insists that he "cannot by any effort of thought conceive the abstract idea
above described" (par. 10). He claims that he can abstract in the sense of imagining
particular qualities separated from others, but only if it is possible that they may
actually exist separated from the others. Thus he can imagine a hand without a
body, but he cannot imagine a hand without shape or color. So he denies that he
can abstract from one another "those qualities which it is impossible should exist
so separated; or that I can frame a general notion by abstracting from particulars
in the manner aforesaid" (par. 10).
Berkeley is arguing against Locke who held that most of our knowledge of
things (if not all of it) is in the form of complex abstract ideas which constitute
nominal essences. The mind is supposed to abstract a common property from the
ideas of sense and produce an abstract general idea of that property. This theory
is supported by common sense introspection and is similar to scholastic epistemo-
logical theory, though Locke disregards the scholastic terminology. Innocuous as
this theory seems, Berkeley is determined to discredit it and establish an alternative
theory based on signification. He will allow only particular ideas which are
particular images or sense impressions. The idea of a hand must have color and
shape, that is, it must be a particular image. The problem of generality or univer-
sality is met by allowing general words: "a word becomes general by being made
the sign not of an abstract general idea, but of several particular ideas, any of
which it indifferently suggests to the mind" (par. 11). This is the alternative theory
which is supposed to eliminate all abstract general ideas and make do with partic-
ular ideas and general words. We have a general word "mankind" but the idea
si~ified by the word is not an abstract general one but, rather, an indefinite
number of particular ideas of particular men that one has known. Abstract general
ideas are not merely surplus entities that need not be postulated but they simply
do not exist, and to suppose they do is to mistake what goes on in the mind. A
good deal could be said in favor of the traditional theory of abstract general ideas
as opposed to Berkeley's theory but this is not to the point. We want to notice, in
the first place, that Berkeley's alternative theory relies on a theory of signification,
and, in the second place, we want to bring out his reasons for so strongly opposing
the seemingly innocuous doctrine of abstract ideas.
Looking to his reasons for opposing the abstract idea doctrine, we must bear in
mind his deep concern with the role of religion and morality in society and his
announced purpose of bringing about the "utter destruction" of atheism and
skepticism? Like Ralph Cudworth before him, Berkeley felt that the doctrine of

= For a statement of this purpose see his preface to the Three Dialogues Between Hylas
and Philonous.
 B E R K E L E Y ON SIGNIFICATION 165

materialism, the belief that only matter in motion was real in nature, was a threat
to religion, because it tended to make the Deity either unnecessary or at least
extremely remote in the context of our understanding of the natural world. Thus
Berkeley's ambition was to devise a philosophy which gave prominent place to the
Deity not only as a creator but as the immanent cause of human knowledge. To do
this effectively he first had to discredit materialism, and his attack on the doctrine
of abstract ideas is directly related to the attack on materialism, since abstract
ideas are one of the chief means to knowledge of independent material things.
He argues that there is no such thing as matter and that the belief in it is a philo-
sophical mistake which stems from the belief that the mind has a power to form
abstract ideas.
What Berkeley regarded as his most important discovery is that what is sensed
is mind-connected or mind-conditioned. Ideas of sense occur in minds, as he often
says, though the preposition "in" usually suggests a meaning of physically located
in the head to those who naturally presuppose materialism. Things, which are col-
lections of ideas, are where they appear to be, but their characteristics are strictly
relative to the nature of the perceiving mind. To suppose things could exist
independent of human perception and still have exactly the same characteristics
they have when they are perceived he regards as absurd, repugnant, and contra-
dictory. Analysis of this common-sense belief leads to skepticism as to the nature
of the external cause of our sense ideas.' But, Berkeley argues, skepticism can be
avoided if the materialistic hypothesis is completely discarded and we stop think-
ing of things as having an unperceivable and unknowable material substratum
which is somehow at the source of the actual perception.
With the two-world theory--a world of sense ideas and a world of the real
external causes of these ideas---our knowledge of the external world is problematical
and theoretical, because our certain knowledge is limited to the world of sense
perception. Assuming empiricism with sense observation as the model for certainty,
the real world, that which is independent of human experience, can only be known
by inference and abstraction. Thus Berkeley's strategy is to deny the legitimacy
of the kind of abstraction that provides the rational justification for a belief in a
material world independent of human experience. Though his arguments against
this kind of abstraction have considerable force, he cannot hope to carry the issue
without providing an alternative theory of general knowledge to replace the rejected
abstract general ideas. Berkeley recognizes the necessity for generality in philo-
sophical knowledge and provides for it by making a particular idea "represent
or stand for all other particular ideas of the same sort (Intro., par. 12). There are no
general ideas but only general terms. There are only particular ideas which are
particular sense images---extreme nominalism and extreme empiricism (perhaps
H. H. Price's term, "Imagism," describes it best 9 Particular ideas signify similar

a See Locke's Essay Concerning Human Understanding for such an analysis, partic-
ularly bks. III and IV.
9 See Thinking and Experience (London: Hutchinson's Univ. Library, 1953).
166 HISTORY OF PHILOSOPHY

particular ideas, and only the word is general. No abstract general ideas of common
properties are necessary, since the sign serves the function of making inference
beyond the particular idea possible. Only particular ideas are real, but general
knowledge is possible since a particular idea signifies any member of a class of
similar particular ideas.
The point that Berkeley does not emphasize in this theory is that it is the
similarity of the particular ideas that is recognized and makes possible their col-
lection in one class. It is by virtue of this similarity that they become signs for one
another. Further, the similarity that each member of such a class of ideas bears to
other members of that class is prior to the fact they they signify one another.
Berkeley wishes to emphasize signification at the expense of similarity, but he is
unable to get around the fact that generality is based upon recognition of similarity.
He admits that the ideas signified by the general word must be of the same sort,
and this means that there must be a similarity or resemblance among them. If some
qualitative similarity obtains among a group of ideas and the mind is able to
recognize this similarity, then this recognition constitutes a kind of mental sepa-
ration of this quality from the other qualities with which it is associated. But this
mental separation is essentially the same kind of process as the abstracting of the
traditional theory described by Locke where the mind frames general ideas of
common properties. An idea of a common property is abstract because it is formed
by a process of mental separation (being one of many qualities discernible in a
thing), and it is general because it is recognized in many otherwise different things.
Thus, Berkeley's alternative theory of general ideas depends upon the recognition
of qualitative similarity just as surely as the theory it is designed to replace.
If recognition of similarity among different things or ideas is necessary for
Berkeley's theory of general ideas, then what is so wrong with saying that one has
an idea of that quality which is reco~ized as similar or resembling in different
things? If one, for example, knows what red is, one can pick out all the red objects
in a large group of things and say, "All these are red.'" According to Locke and the
traditional theory, we know what red is and recognize it when we see it, because
we have an idea of red in mind. And this idea is both abstract and general. Thus
we are able to examine a group of things ignoring shape and texture and other
sensory qualities and attend only to color and succeed in picking out only the red
things. This attending only to color constitutes a kind of mental separation of one
quality from many others, and it would seem that the fact of such a mental sepa-
ration or abstraction would be adequate grounds for maintaining that we have an
idea of that quality. Such an idea would be abstract and general though it need not
necessarily be a sharp particular image. The fact of recognition would seem to
necessitate an idea of the recognized quality and yet Berkeley is committed to the
position that such recognition either does not occur (which is false) or occurs
without an idea of what is recognized (which is absurd or at least very difficult to
understand). We are supposed to recognize similar things, because they are signs
for one another and not because we have an abstract general idea of the recognized
quality. The contention that one can substitute indifferently one object for any
 B E R K E L E Y ON SIGNIFICATION 167

other in the class of resembling things is supposed to make the theory of abstract
general ideas unnecessary. However, when such a contention is made it would seem
to be strongly countered by the reply that we may substitute indifferently because
we recognize a similar quality. And the recognition of that quality is possible
because we have an idea or notion of it in mind but not necessarily a particular
image. Berkeley admits the natural similarity and recognition of the quality, but
denies that this is done by virtue of an abstract general idea. Recognition, according
to Berkeley, is possible by virtue of a capacity or disposition of the mind which
does not require an idea of what is reco~ized. This peculiar position--to say the
least--is necessary because he insists that all ideas be particular. The recognized
similar quality is not particular because it is experienced as part of the perception
of a particular thing. There could be no idea of the quality itself pure and simple,
because, he argues, it does not exist in nature separated from other qualities. And
since it does not exist in nature separated from other qualities, it cannot do so in
mind. Persuasive as Berkeley's argument may be here, natural qualitative similarity
among sensory ideas and its recognition by the mind still seem to require a kind of
abstraction that Berkeley is loath to admit.
If Berkeley's attack upon the theory of abstract ideas is anywhere near as
ineffective as I make it out here, the puzzling question remains as to just why such
an attack was necessary. As we have already noted, his reason was that abstract
ideas make possible a way of rationally justifying a belief in the existence of an
external material substance. We need to look more closely at the role that abstrac-
tion plays in arguments for the existence of an external material substance. Though
the following argument is not explicitly discussed by Berkeley I will try to show
that his theory of vision and his doctrine of ideas are designed to counter it.
According to the traditional theory of abstract ideas, we see similar colors in
many differently shaped and textured things and form an idea, an abstract general
idea, of that color. Such a process can also occur when the idea or quality is one
which is experienced by different senses. For example, we see the shape of a table
and we can also feel the shape of the table ff we wish. The idea of its shape, then,
is derived from two different senses, sight and touch. Now, suppose it is pointed
out that many of our sensory ideas are peculiar to one sense, such as smells--the
fragrance of a flower is never tasted, seen, felt, or heard--it is only smelled. Since
all smells are peculiar to the one sense, it is not unreasonable to suspect that their
character may be due, in whole or in part, to the peculiar nature of the sensory
apparatus or, more generally, the mind. At the very least the mind or sensory
apparatus is a necessary condition for their occurrence and their specific qualities.
The question then arises as to just what exists externally and independently of the
human mind and sensory apparatus. Any idea or quality which is known by means
of only one sense may reasonably be suspected of deriving its specific character
not from the external world but from the character of the mind or sensory
apparatus. Such ideas or qualities constitute a rather large class, the so-called
secondary qualities of Locke (which were also claimed to be subjective by Des-
cartes, Hobbes, Galileo--and Democritus). Those qualities which are known by
168 HISTORY OF PHILOSOPHY

two or more senses, then, are free of the suspicion of being entirely relative to the
peculiar nature of a single sense and become candidates for the status of inde-
pendent and external existence. Since we see things moving and feel things moving,
and since seeing and touching are such entirely different experiences, then we have
some rather convincing evidence that motion is a quality not dependent upon any
one sense and, hence, is a quality of the real external world. Motion, as such, is
not peculiar to any one sense, and since peculiarity to one sense was the
ground for suspecting qualities of not existing independently, it is free of this
suspicion. Similar analysis applies to most of the other so-called primary qualities
and the picture of the external physical world emerges as many differently shaped
and variously sized objects capable of movement. (This analysis, however, does not
apply to the primary quality of solidity which is experienced through only one
sense.)
Though this argument for an independent external world is hardly conclusive
it nevertheless has some weight. Berkeley, at least, was evidently impressed by it
and went to considerable effort to refute it in his Essay Towards A N e w Theory o[
Vision. After going through the same arguments he later used in the Introduction to
the Principles, he considers the question of "whether the particular extensions,
figures, and motions, perceived by sight be of the same kind, with the particular
extensions, figures, and motions, perceived by touch (par. 127)? He immediately
answers this question by laying down the following proposition: "'The extension,
figures, and motions perceived by sight are specifically distinct from the ideas of
touch, called by the same names, nor is there any such thing as one idea or kind
of idea common to both senses" (par. 127). He then supports this proposition with
the argument that a blind man "would not at first reception of his sight think the
things he saw were of the same nature with the objects of touch, or had anything
in common with them, but that they were a new set of ideas perceived in a new
m a n n e r . . . " (par. 128). He also argues that "there is no other immediate object
of sight besides light and colors." And, "it is therefore a direct consequence that
there is no idea common to both senses" (par. 129). Because, obviously, color
and light cannot be touched but must be seen.
The apparent visual perception of solid objects and the discernment of distance
are accounted for, in Berkeley's view, by experience and learning, that is, by the
correlation of our ideas of sight (light and colors) with the utterly different ideas of
touch. We see, for example, a patch of colored light, move toward it, and it
becomes larger, until we finally have ideas of touch, smell and taste. After adequate
repetition and instruction we learn to give the visual, the tangible, the olfactory,
and the gustatory idea the same name "orange." This is not because we are dealing
with one thing, Berkeley insists, but because these entirely different kinds of ideas
are associated (they are in the same place, so to speak). "Visible figures," Berkeley
maintains, "are the marks of tangible figures" (par. 140), and they "represent

a In A New Theory of Vision and Other Writings, intro. A. D. Lindsay (London:

J. M. Dent, 1910).
 B E R K E L E Y ON SIGNIFICATION 169

tangible figures, much after the same manner that written words do sounds"
(par. 142). There is no more similarity, he argues, between tangible objects and
visible objects than there is between an object and a word that signifies it. And,
more generally expressed, he concludes that

the proper objects of vision constitute a universal language of the Author of nature,
whereby we are instructed how to regulate our actions, in order to attain those things
that are necessary to the preservation and well-being of our bodies, as also to avoid
whatever may be hurtful and destructive of them. (par. 147)

There are, of course, many ditticulties in this theory which need not be discussed
here. It is necessary, however, that we realize that this theory will aocount for
appearances in a general way even though we may find the common sense belief
in one material object as the cause of the different ideas of sense more plausible.
The notion of mutual signification of associated ideas replaces the notion of one
substance. Berkeley's observation that we perceive only ideas and never the sup-
posed material substance hardly shakes the common sense conviction that there is
one object experienced by different senses. It should be granted to Berkeley, how-
ever, that this conviction is justified by an inference--habitual, natural and irresist-
ible--but, nevertheless, an inference. The purpose of his alternative theory of
associated ideas which signify one another is to demonstrate that this inference
is not strictly necessary--since another account will do the job.
Turning to another phase of Berkeley's attack on materialism we find him
arguing that even if there is such an independent material substance, it could not
cause our ideas of sense because it is supposed to be passive and inert. As such it
is incapable of activity and therefore incapable of causing anything, since causing
is certainly a kind of activity. This is a strange argument, but Berkeley relied on it
heavily, and it needs to be clarified. From a materialistic point of view, that of
Hobbes for example, matter is eminently suited for the role of causal agent, because
bodies in motion are understood as the very model of causal agent. That is, motion
and cause are presupposed to be the same thing--with matter being that which is
in motion. But from an alternative metaphysical position (the "idealistic" or, better
the "mentalistic"), the model for causal agent is willing. When I lift my arm, it is in
motion, to be sure, but the cause is my act of willing to raise my arm--this is
conscious intention, a mental activity. Hence the mentalist interpretation of cause
is that since the willing is prior to the physical motion, it is the willing which is the
real cause or activity and not the motion. And since matter is not supposed to
possess the mental characteristic of being able to will, it follows that matter is
passive and inert in the sense of being unable actively to cause anything.
Berkeley, of course, adopts the mentalist interpretation and one might suppose
that the position that matter is passive and inert is a straw man invented by Ber-
keley. This is not the case, however, as the proposition that matter is passive and
inert was a tenet of Newtonian Theism. Newton himself held that matter was dead,
inert, and lifeless, and men like Richard Bentley and William Whiston expounded
170 H I S T O R Y OF P H I L O S O P H Y

theologies including this tenet.' Briefly, matter had to be passive and inert to lend
force to the argument that God must exist as the active first cause or creator of
the universe. Any emphasis upon an active or creative power in matter detracted
from the force of the argument that God must be the first cause or creator.
Consequently most theologians and natural philosophers of Berkeley's time were
not only familiar with the doctrine that matter is passive and inert but accepted
it. Further, the Newtonians were forced to emphasize the passive and inert
character of matter in opposition to Leibnitz who maintained that force or activity
was the essence of matter.
Berkeley assumed, along with most of his contemporaries, that there is an
external cause of the ideas of sense and that matter is passive and inert (if it exists)
and therefore cannot be the cause of our ideas. Further, he had to presuppose---or
sometimes also argue--that willing, activity and cause are all the same kind of
thing. It then follows that matter cannot be the cause of our ideas of sense. And
with the further premise (the Cartesian presupposition of the era) that the external
cause must be either matter or spirit (mind), it follows that the external cause of our
ideas of sense is a spirit or mind. H e also argues that it is unnecessary to suppose
an intermediary, matter, between the active cause (God) and the ideas of sense in
the minds of men. Such a supposition involves a contradiction since matter is
passive and, therefore, could not be the cause of anything. Perhaps the obvious
weak link in this argument is the premise that cause must be a species of active
willing, but, as implausible or far-fetched as this proposition is to the materialistic
imagination, it must be allowed provisionally if one is to understand Berkeley.
Allowing this mentalistic conception of cause we can outline Berkeley's theory:
mind A (God) produces (wills or causes) an idea of sense (the sensory image) in
mind B (human perceiver). Of course, we have no model for this kind of event in
our ordinary experience, but we do have a model for causing or producing an idea
in our own minds. We can, for example, recollect or imagine a sensory idea by an
act of will--this is the true model for causing according to Berkeley. Since these
ideas of imagination or memory are qualitatively similar to the ideas of sense, they
must, he argues (Principles, pars. 25-33), have a similar cause. The cause of ideas
of sense is not in our own minds but external to us and more powerful than mere
h u m a n minds, since ideas of sense are more vivid, enduring, and coherent than the
ideas of the human imagination. As to how this external powerful mind produces
an idea in the human mind, we are offered the metaphor of language and
signification.
It is this use Berkeley makes of signification in developing his metaphysical
theory that is our primary concern. Sitmifieation is a mentalistic relation rather than
a physical relation in the sense that its model is the designation of a word to stand
for an idea or thing--a conscious act of a human mind. Theoretically, the mind

6 See the unpubl, diss. (Univ. of Calif., Berkeley) by Robert H. Hurlbutt, "Newtonian
Theism in Eighteenth-Century England," for an excellent account of these theological
positions.
 B E R K E L E Y ON SIGNIFICATION 171

arbitrarily associates a convenient sound or series of sounds with some idea or

thing so that the sound will stand for or substitute for the idea or thing. Thus
signification is the basis for communication between minds which is accomplished
by a system of signs, a language. The metaphysical analogy that Berkeley offers is
that the ideas of sense, what we see and feel as things, are like signs or sentences
in a language which we understand. Sensory experience is supposed to be like
understanding sentences in a familiar language. In the case of ordinary conversation,
someone (some other mind) is trying to communicate with us by means of a com-
mon language. Just so, Berkeley maintains, the author of nature communicates
with us using a language or system of signs (the ideas of touch, sight, sound, etc.).
The universality of these natural signs among mankind is accounted for by the
power and benevolence of the external mind, the author of nature. Different men
experience similar ideas of sense in similar situations, because the cause of these,
the powerful external mind, is coherent and consistent. It is as if the author of
nature were eternally making the same speech for the benefit of other limited or
finite minds. In the case of human minds, signification is a theoretically arbitrary
process, but in the case of the author of nature, consistency and coherence are
maintained because of the perfection and benevolence of this all-powerful spirit.
If this outline of the broad metaphysical theory is clear enough we are in a
position to look more closely at the nature of signification. In the New Theory of
Vision, Berkeley gives the signification theory only a limited application, applying
it to the relation between ideas of sight and tangible objects. The idea of sight
is a sign that indicates a tangible object in some direction and at some distance.
In this work he limits himself to the problem of vision and does not emphasize his
identification of tangible objects with ideas of tangible objects. He treats tangible
objects as terminal entities, that is, as real objects whose legitimacy is not question-
ed. In the later Principles and Dialogues, however, it is almost immediately evident
that the five kinds of ideas which correspond to the five senses are theoretically
equal in status. A large part of the first Dialogue is devoted to arguments intended
to prove that the sense of touch is just as relative as the other senses. If the sense of
touch has the same status as the other senses and if the ideas of sight are signs for
the ideas of touch, then the general application of the signification theory is strongly
suggested. The ideas of touch, in their turn, may serve as signs for ideas of sight
or sound or taste, etc. Any of the five kinds of ideas of sense may s i c i l y any of the
other ideas of sense, this mutual signification constituting the "language of the
Creator," as he ealis it in the New Theory of Vision.
This general application of the theory of signification comes out explicitly in the
Principles. In Principle L X V Berkeley says that "the fire which I see is not the cause
of the pain I suffer upon my approaching it, but the mark that forewarns me of it."
And, "in like manner, the noise that I hear is not the effect of this or that motion
or collision of the ambient bodies, but the sign thereof." Thus signification takes
the place of causation. If matter is not the cause, as Berkeley maintains, then he
must provide an alternative explanation of the cause. Berkeley's alternative, of
course, is the author of nature, or God, a natural alternative to the religious mind
172 H I S T O R Y OF P H I L O S O P H Y

of the late seventeenth or early eighteenth century. Berkeley goes further than this,
however, and seeks to provide a theory of how the author of nature produces ideas
in human minds. This is done in a manner analogous to the way men use language.
Thus he says in Principle L X V I :

Hence it is evident, that those things which, under the notion ol a cause co-operating
or concurring to the production o] effects, are altogether inexplicable, and run us into
great absurdities, may be very naturally explained, and have a proper and obvious
use assigned them, when they are considered only as marks or signs for our information.
[And it is the searching alter, and endeavouring to understand those signs (this lan-
guage, if I may so call it) instituted by the author of nature, that ought to be the
employment of the natural philosopher, and not the pretending to explain things by
corporeal causes; which doctrine seems to have too much estranged the minds of men
from that active principle, that supreme and wise spirit, 'in whom we live, move, and
have our being.']

The theory of signification has two levels or dimensions: one is the relation of
the ideas of the different senses to one another, and the other is the relation of the
idea of sense to its external cause. In the case of the relation of the ideas of one
sense to those of another sense (this might be called the "epistemological dimen-
sion"), Berkeley argues very forcibly that there is no qualitative similarity between
the different ideas. They are simply signs for one another and are associated in
our experience; they often appear to be clustered in the same place at the same
time. In the case of the relation between an idea of sense and its external cause,
the active principle or "author of nature" (this might be called the "ontological
dimension"), there is likewise not supposed to be any qualitative similarity. The
cause of the idea is supposed to be an active spirit and, as such, must be completely
different in character from the idea which it produces in the mind of man. This
difference, as Berkeley frequently emphasizes, is that spirit or mind is an active
principle, while the ideas of sense are passive and inert.
But this complete disparateness of the idea and its external cause that the theory
of signification requires is hardly compatible with Berkeley's repeated assurances
that things exist in the mind of God when they are not perceived by other minds.
As Philonous says toward the end of the third Dialogue, "Do I not acknowledge
a two fold state of things, the one ectypal or natural, the other archetypal and
eternal?" The term "ectypal" suggests a reproduction or copy of a prototype or
archetype in the mind of the author of nature. In this case the ectype, or idea
perceived by the mind of man, is more than a mere sign, since it, Presumably, bears
qualitative similarity to the original archetype in the mind of God. When Berkeley
takes up this position, appearing to condone the belief that the idea of sense is a
copy or poor reproduction of a real object existing eternally in the mind of God,
he reverts to a representational theory of perception. The novelty and the force
of his esse-is-percipi.doctrine are that it is a presentational theory of perception.
One could simply point out this inconsistency and observe that Berkeley's reversion
to a representational theory was a response to critics who quote Scripture and that
 B E R K E L E Y ON SIGNIFICATION 173

it occurs principally in his later works. This may be an oversimplification, and even
if it is not and Berkeley did change his mind, I should nevertheless argue that his
true position is the presentational theory based on signification and active cause.
He expounds it constantly in his major works and only reverts to the represen-
tational theory in the early stages of an argument or when hard-pressed by argu-
ments based on Scripture which he is bound to respect. When he makes a con-
cession to the representational position he usually accompanies it with qualifications
which lead back to his true position. In Principle 3, for example, in speaking of the
table he writes on, he says that when he is out of his study, it exists, "meaning
thereby that if I was in my study I might perceive it, or that some other spirit
actually does perceive it." The second alternative, that some other spirit actually
perceives it, is the simple-minded theory for "the vulgar." The first alternative, the
hypothetical interpretation, is consistent with the signification theory and might be
reworded in this fashion: to say that a table exists in the study when nobody is
perceiving it is to say that an active spirit exists (in the study and every place else
also) who will produce in my mind the ideas of sense which constitute the real
table when I enter the study. With respect to the archetypes or perfect copies of
all ideas which are supposed to be in the mind of God, Berkeley maintains that
they could be nothing like ours, since to have sense ideas implies imperfection.
Also, we could have no certain idea or notion as to what is in the "eternal mind"
anyway. Hence there need not be any qualitative similarity between the "archetype"
in the mind of God and the "ectype" which the author of nature causes us to have.
Thus Berkeley's true theory, or serious theory, does not necessarily imply that there
be qualitative similarity between our ideas of sense and their external cause.
Berkeley requires signification to do so much work that it is ditficult to see how
its supposedly simple character can be maintained. There appear to be at least
three different kinds of signification: signification among ideas of the same sense,
signification among ideas of different senses, and signification between ideas and
their external cause, the author of nature. The first kind of signification, as when
an idea of vision signifies another idea of vision, may be called "signification within
categories." This is the case when the visual orange, the fruit that one may eat,
signifies a woman's dress which is orange-colored; the two objects "have the same
color" as we should ordinarily say. Any one of a class of visual ideas may signify
the whole class, ideas of "'the same kind." This signification is possible, and
necessary, because, I should emphasize where Berkeley does not, the ideas are
qualitatively similar. Thus Berkeley maintains that we reco~ize resemblances
among our ideas, but he denies that we may have an idea of the resemblance itself.
Such an idea would be abstract and general, whereas his position of extreme
nominalism is that all ideas are particular (particular images). In the case of
signification between an idea of one sense and the idea of another sense, "signifi-
cation across categories" I should call it, there is supposed to be absolutely no
qualitative similarity. The orange we see is supposed to be utterly different in
character from the orange we taste or smell, the one is merely the sign of the other.
The association of the two ideas is theoretically arbitrary. That is, the red visual
174 H I S T O R Y OF P H I L O S O P H Y

cherry and the round, smooth, (tactile) cherry are two ideas associated in nature,
being contiguous in time and space, but we could easily imagine these ideas being
separated. Their association is not really arbitrary because the author of nature
presumably intended that they be associated in the human mind. The association is
something simply given in nature which could theoretically be otherwise so far as
the finite mind of man is concerned. Thus these two kinds of signification are
different in character: in the one case, there is qualitative similarity and, in the
other case, there is no similarity, only contiguous association in apparent space
and time. The difference between these two kinds of signification was either not
noticed or, more likely, deliberately underemphasized by Berkeley.
There is, further, a third kind of signification which may be called "meta-
physical" or "ontological" in relation to the first two kinds which are epistemo-
logical. It is an entirely different kind of relation than the other two. On the one
hand, both signification within categories and signification among categories are
relations among ideas. On the other hand, the relation between ideas in the human
mind and the cause of these ideas, the author of nature, is a relation between ideas
and active spirit as cause. There are not supposed to be anything like passive, inert
ideas in the mind of this perfectly active spirit. The mind of the author of nature
is unavailable to human inspection and beyond our full comprehension, so the
supposed archetypal character of ideas in the divine mind is a matter Of speculation.
On the premise that the divine mind is like human minds, Berkeley maintains that
the principle characteristic of the divine mind is activity which is like human
willing. And using the metaphor of language, the objects in nature are the language
of the author of nature, his way of communicating with other minds or spirits.
Thus the relation between ideas of sense and their external cause is a combination
of sit,nification and active willing. We may, on the one hand, describe the three
relations (within categories, among categories, and the metaphysical relation
between idea and its cause) as three kinds of sit,nification. Or on the other hand, and
perhaps more cogently, we may describe them as combinations of signification with
three quite different other relations, qualitative similarity, spatial and temporal asso-
ciation, and active willing.' Thus signification within categories is signification based
upon qualitative similarity or resemblance. The orange dress signifies the orange
in the fruit bowl, because they resemble one another with respect to color. Sit,nifi-
cation among categories is signification based upon customary association or
contiguity in space or time. Thus the visual idea of the red cherry signifies the idea
of the round smooth fruit, because the two different ideas of vision and touch are
associated in space, that is, they appear in the same place. Metaphysical or
ontological signification is based upon the causal agency of a will or mind. Thus the
visual ideas of a sunset signify the intention of the author of nature that human
minds experience beauty. Or, a simpler example, the sight of fresh water signifies
that if we perform the appropriate bodily actions, we may quench our thirst.*

' Notice that these three relations correspond to David Hume's three principles of
association, resemblance, contiguity and causality; see his Enquiry Concerning Human
Understanding, sec. 3.
8 I omit Berkeley's account of mathematical and logical signs, since this kind of
 B E R K E L E Y ON S I G N I F I C A T I O N 175

This analysis shows that signification is not the simple relation that Berkeley
makes it out to be. Signification depends upon other relations, either qualitative
similarity, association in space and time, or active will. Yet in his criticism of
abstract ideas Berkeley attempts to substitute signification for qualitative similarity,
thus failing to realize that the signification of this alternative theory of generality
depends upon the recognition of qualitative similarity. His reason for seeking to
discredit qualitative similarity is that it is the basis for the theory of abstraction
used by Locke which makes possible some fairly strong arguments for the existence
of an independent material substance. On a broader historical level this attempt
to substitute signification for qualitative similarity is an attempt to develop a new
metaphysics completely free of the traditional formistic theory of Plato and Aris-
totle. Qualitative similarity is at the core of the formistic theory since it (or the
stronger version of it, identity) is postulated as the relation beween what exists
externally and what is conceived in the mind as knowledge.
This attempt to get free of the formistic metaphysics has been one of the
characteristic themes in the development of British empiricism. Perhaps a brief
summary of the role of this issue will serve to provide a context in which Berkeley's
theory of signification and his metaphysics may be understood as part of a con-
tinuous historical development. In the case of Bacon there was little more than
strong feeling and the intention to be free of the Aristotelian metaphysics. His
conception of knowledge remained essentially formistic. Hobbes, despite his devel-
opment of a materialistic ontology, continued to use the Aristotelian logic and the
formistic way of thinking in his epistemology. He failed to integrate the material-
istic ontology with an essentially formistic epistemology. The Cambridge Platonists,
notably Henry More and Ralph Cudworth, emphasize activity, as Berkeley does,
and defend the formistic metaphysics in their attack on materialism. Locke made
considerable progress toward the solution of Hobbes's problem, that of an adequate
epistemology for the materialistic ontology, but did not fully succeed in exorcising
the ghost of the forrnistic theory.' The haunting remnant of this ghost, so to speak,
is the relation of qualitative similarity which Locke still retained as the relation
between primary qualities in things and the ideas of these qualities in the mind. His
later position, developed in books III and IV of the Essay, constitutes one solution
to the problem: the denial of qualitative similarity between idea and the real
constitution of external things. Locke appeared not to like this position very much,
though he did suggest that signification is the operative relation between the
external thing and the idea in the mind. Berkeley's metaphysics is a development
of this suggestion from Locke.

signification does not bear directly upon his metaphysical theory. He holds that number
is a mentally constituted sign and that as words signify ideas of sense or imagination (hence,
things), numbers signify the quantitative relations between things. His treatment of mathe-
matical and logical signs is not novel except insofar as he demands conceivability on a s e n s e
basis. As for a metaphysical interpretation of logical and mathematical notions, these may be
understood as the way or manner in which the author of nature orders the ideas he presents
t o our minds. Metaphorically, since we understand the language of the author of nature, w e
are capable of understanding its rules of grammar and syntax as well.
' See my "John Locke's 'Doctrine of Signs': A New Metaphysics," Journal of the
History of Ideas, XXVI, 3 (1965), 369-382.
176 HISTORY OF PHILOSOPHY

Assuming a causal relationship between the real constitution of things and the
ideas of sense in the mind, Locke further postulated the relation of sitnaification
between them--idea signifies thing (because thing causes the idea). This position
suggests a general theory whereby the word signifies the idea and the idea signifies
the external thing. Thus the three epistemological entities, so to speak, of word,
idea, and thing are related by the one relation, signification. In the case of the
relation between word and idea, signification has an arbitrary mentalistic character,
as in the model of the stipulative definition where one designates a term to signify
a thing (or idea). In the case of the relation between external cause and the idea, no
such mentalistic or arbitrary character is suggested, though the relation is supposed
to be signification. This latter kind of signification is akin to the natural sign of
Ockham and Hobbes. In emphasizing that the one relation holds between thing,
idea, and word, Locke tends to ignore the difference between the two kinds of
signification, causal or natural signification in one case and arbitrary or mentalistic
sitmificafion in the other. Berkeley's metaphysics may be understood as an attempt
to develop Locke's theory, that is, to exploit the notion that signification is the
operative relation between idea and its external cause as well as between word
and idea. He postulates one kind of signification, the mentalistic kind, in both
cases; as human minds designate terms to stand for things (or ideas), the author
of nature designates ideas of sense (things in nature) to stand for ideas he wishes
to communicate to other minds. Thus the notion of active willing is integrated
with the notion of signification in Berkeley's metaphysics. Signification is something
a mind does; it wills that a sign stand for a thing or idea. Human minds do this when
they stipulate definitions, and the author of nature does this when he communicates to
other minds using the language of nature: what we see, hear, etc. The arbitrary nature
of human signification differs from the ordered signification of the author of nature
as the imperfect, fallible, limited mind differs from the perfect, omnipotent, all-
good, infinite mind of God.
Though this metaphysics of sitmification may be as fantastic to the modern
reader as it was to most of Berkeley's contemporaries, it is interesting and instruc-
tive for a number of reasons. It emerges from a context of continuous epistemo-
logical enquiry through the better part of almost two centuries. Once the formistic
metaphysics was rejected in favor of a materialistic ontology, the question of how
we are to know, in any direct way, the nature of this world of bodies in motion
became acute. Though Locke made a good effort to provide an epistemology com-
patible with materialism, Berkeley maintained that materialism had to be rejected
before a consistent epistemology could be worked out. Assuming Berkeley's alter-
native mentalistic ontology to be unacceptable and his arguments against Locke's
position to be cogent, we are left with a negative result. The attempt to develop
a theory of knowledge suitable for materialism and Newtonian natural philosophy
appears to end in failure. The articulation of this result is to be found in the
work of David Hume though the outline of a pragmatic epistemology was already
worked out by George Berkeley.

The University of West Florida