of thought, both of which may be said to have arisen in
Locke one through Berkeley to Hume, the other through
Hartley to the Mills. Its theory of knowledge is that know-
ledge is explicable from the elements of sense-experience
united through the bonds (laws) of association, such con-
nexions being made within the life-experience of the
Knowledge is thus an individual construction,
and is a compound resulting from the fusion, under certain
laws, of sense-elements. It is the product of sense and
association. An Associationist must maintain that there
is nothing in the mind that could not be developed by the
individual for himself. He may be helped to his special
associations by others, but he could do it all for himself.
This is the purest form of Experientialism. Locke himself
was an Associationist, not explicitly but by implication.
Associationists have not worked out a consistent Theory of
Knowledge, but they do make a real attempt to begin at the
XIL] Elements of General Philosophy. 113
Locke and Berkeley.
Locke's ideas of sense and of the construction of knowledge
are, as we have seen, very crude ; nevertheless he first opened
the question of ^^.psychological origin of knowledge. Berkeley,
Locke's immediate successor, marks a distinct advance along
this line. He began a definite psychological inquiry, while
he also took a philosophical position in regard to the know-
ledge of matter, which is at least more circumspect than that
of Locke. He based his philosophy on his psychology ; yet
he was not set philosophising because he was a psychologist,
but because, as a theologian, he wished to get rid of the,
to him, pernicious effects of Materialism. Thenceforward
philosophy and psychology really began to have a separate
history. Berkeley got away from Locke's notion of the
five senses as barely passive ; and further, he began that
definite reference to a principle or principles of intellectual
synthesis without which it is hopeless to explain knowledge.
Associationism is traced to him though he does not use
the word. His theory of knowledge bears more especially
on our third problem the perception of an external world.
Hume not only carried out further Locke's theory of
knowledge, but put the question into such a shape as to
rouse the strongest opposition and so bring about a great
advance in thought. In regard to the cognition of extension,
Hume is behind Berkeley and not superior to Locke. But
he was beyond both in his statement of the formal principles
of knowledge. He proceeds wholly upon Locke's individual-
istic view that there is nothing in the developed knowledge
of any mind which is not explicable from the (incidental)
ii4 Elements of General Philosophy. [LECT.
experience of that mind ; and expresses this (by a modi-
fication of Locke's language) in the oft-repeated formula,
that whenever we 'really' have any idea there is some
assignable impression from which it is derived of which
it is the copy. By thus distinguishing idea from impression,
he gives greater precision to the psychological data which
he assumes in common with Locke. But further, when
Locke, in order to account for the developed complex of
knowledge, is content to assume faculties of ' abstracting,'
'compounding' and the like, Hume formulates definite
principles of association under which the synthesis takes
place : (i) Contiguity, (2) Similarity, (3) Association of
Cause and Effect. He does not work out the last principle
at all, nor the two others at all fully. But not in regard to
these can we gauge the importance of Hume. There are
two facts in cognition that he set himself to account for
knowledge of substance and knowledge of causation. He
was led to the question of cause from the prominence in
modern science of the inquiry, ' What is the cause of what ? '
Berkeley already and the Cartesians before him (e. g. Male-
branche) had seen that what science was concerned with was
the establishment of uniformity in phenomena. But Hume
went so far as to say, that if any phenomenon is by us con-
nected with any other phenomenon in Nature, it is because
of the customary sequence of experience. A subjective bond is
thereby established and thai is all, although through 'custom'
one phenomenon comes to be considered as the objective
' cause ' of the other. Thus he decries knowledge, at least
from the Rationalist point of view. While his Treatise of
Human Nature contains an almost complete theory of know-
ledge, while he vaguely but distinctly recognises intellectual
elaboration of sense-data arranged by 'Abstraction,' he
XIL] Elements of General Philosophy. 115
stunned the philosophic mind of the century by showing
that all previous investigation had, so to speak, led up to
a dead wall that Locke's Experientialism, logically carried
out, landed philosophy in scepticism. Besides his Individual-
ism, his Particularism (i. e. that everything complex or
general has to be made out of particular elements) is very
pronounced as put in the formula which he is constantly
referring to : 'All ideas which are different are separable 1
(i. e. have somehow to be brought together if they appear in
one mature consciousness as conjoined).
Hume's contemporary, Hartley, was independent of him,
but a follower of Locke. lie was the first lo formulate the
law of Contiguous Association as accounting sufficiently,
without other laws of association, for intellectual synthesis.
Berkeley did not formulate any such laws ; Hume did, as
we have seen, but he did not apply them. When later
Associationists (the Mills and Professor Bain) faced the
problem of knowledge, they worked with reference to Hartley
and not to Hume's laws of association. Hartley was the
first who distinctly asked how a multitude of sensations,
which for us are discretes, come to be fused, or to coalesce
into that coherent appearance of an object with a variety
of qualities which expresses what our experience really is.
It is, he said, by this one associative principle. Thinkers
before him, from Aristotle onwards, had used association
only in accounting for the imaginative life or representative
experience. Hartley was the first to employ it in explaining
the synthesis of sensations. He did not give a complete
exposition of this theory, or analyse sufficiently the elements
of sense, but he fiist started the Associations! method.
n6 Elements of General Philosophy. [LECT.
Thomas Brown was a strong Associationist, thinking with
ultimate relation to Locke, but with modifications due to
the influence of the French Sensationalists, Destutt de Tracy
and others. They first laid hold decisively on ' muscular
sense,' a discovery of great importance in philosophic theories
of extension. To this subject Brown's lectures were largely
devoted, and to it we shall return. Brown used Hartley's
theory of association most earnestly, but was repelled by the
latter's introduction of the physiological theory of vibrations.
/. 5. Mill.
It is John Stuart Mill and Professor Bain who, as inheritors
of the Sensationalist tradition of the eighteenth century, have
set up the formulated theory of knowledge, bolh psychological
and philosophical, known as Associationism. The latter
gi/es better data for a true theory, especially in regard to
external perception; the former is the better systematiser.
In my judgment their Associationism, while it is an approxi-
mation to a theory of knowledge, comes evidently short.
However important are the factors brought out by Mill, he
just fails to solve the problem. He declares that a number
of the subjective experiences, had by an individual human
being, become for him aggregated according to certain laws
(of association), and that these aggregated appearances can
come to assume the form of knowledge for the individual
and since it is knowledge to be objective or valid for all.
But it is just this last point that he does not account for.
Our knowledge, as I have said, is a coherent system of fact
and relation held in common by me and equally by others.
This objectivity is the distinctive constituent of knowledge,
yet Mill never satisfactorily accounts for it never gets out
xil.] Elements of General Philosophy. 117
of the charmed circle, the sphere of the subjective. No
doubt this is the right way to begin, but it is the wrong way
to end if we want to give an account of knowledge as the
common property of all men. Mill never gets off psycho-
logical ground. Now I am in sympathy with Associationism
as psychology only. Mill's psychology is rather defective.
He borrows from Professor Bain without comprehending
him properly. However, Mill's shortcomings in framing
a philosophical theory of knowledge do not detract from his
great philosophical merit in his theory of general knowledge,
viz. his logic. It is as a logician that he is effective, rather
than as an epistemologist not that I always go with him in
his logic. In this he gives an account of knowledge in a
constructive spirit that is very different from the destructive
spirit of Hume. Living in a scientific age, Mill attempted
to set up a fundamental theory of positive science involved
in all the special sciences. But he does not explain how we
come to know the world as consisting of a number of things,
of bodies and minds. He works from the phenomenal point
of view and from that of individual experience. He tries
to show how the individual experiences of the mind can
become associated so as to enable one man to ask another
to accept them as valid.
Even as an inquiry of positive science Mill's work is
de r ective. From one point of view his positive theory may
be called no less sceptical than that of Hume. Jevons's
Principles of Science is more complete though still less
Professor Bain has been the most important contributor
to psychology in England in this century. His pre-eminence
extends over the whole field of psychology as distinct from
u8 Elements of General Philosophy. [LECT.
philosophy. Towards the general theory of knowledge he
does not contribute any advance on Mill and the Associationists
generally. lie works from the individual point of view. He
makes but little attempt to apply the laws of association to
cognition as such. He does not ask, e. g. how we can
explain the concieteness of an object on the principles of
association, although he gives a careful statement of those
laws. Yet he posits an element of personal initiative for the
explanation of developed consciousness; he tacitly denies
the tabula rasa hypothesis. In the mature consciousness he
finds an element not derived from the sense-experience of
the individual because he considers mental life in connexion
with the nervous system. It is recognised that the individual
comes into the world organised up to a certain point ; and <
this fact, taken into account on the bodily side, has correspond-
ing to it a certain pre-determination of conscious life.
The ' Common Sense ' School.
Reid, Stewart and Hamilton put forth their epistemological
view in antithesis to Hume's theory of knowledge. The
first declared that, while sense was of account for knowledge,
knowledge could not be explained out of the elements
assumed by the Associationist doctrines. So he fell back
on other assumptions. What struck him in the general
theory of knowledge, as distinct from the special problem
of the cognition of an external world, was the community
of knowledge was the fact that while there is more th.m
sense in knowledge, this ' more ' is had by all, cultivated or
uncultivated, young or old. This he attributed to the sub-
jective factor of common sense. Now common sense in
psychology is a name for organic or general sensation 1 .
1 V. Element* of Pathology, p. 62. ED.
xii.] Elements of General Philosophy. 119
In popular parlance it is the faculty of ready judgment,
mother wit. Reid employed it thus : We are so con-
stituted that we interpret our experience alike. When we
are affected through our senses, we refer those sensible
impressions to a thing or substance of which they are
qualities, by a fundamental principle of judgment or
common sense. If we interrogate consciousness we reach
this ultimate and objectively valid principle, beyond which
we cannot reason.
This was a valuable idea, but Reid's method was hap-
hazard, his assertions too readily made, his elementary
principles too easily found. His ' common sense ' expresses
rather the result, than the means, of the determination of our
impressions. It was a kind of revival of the old doctrine of
innate ideas, although accompanied by a much more elabo-
rate analysis of knowledge than any preceding Rationalists
had given. We may not agree with him, nevertheless his
system was an advance on Locke and Hume, if only because
it made other thinkers more circumspect.
Dugald Stewart carried on the doctrine on the same lines.
Knowledge could not be explained without the assumption
of certain fundamental principles of belief which determine
the objective validity of knowledge.
Reid, Stewart and Hamilton are the three typical ex-
ponents of faculty-psychology. The term ' faculty ' is very
crudely used by the first two, but definitely by the last.
Hamilton, while he justifies his own use of the word by saying
that it is merely a way of massing together a number of mental
phenomena, points out, as against his predecessors, that the
discrepancies in their use of it show a want of principle
120 Elements of General Philosophy. [LECT.
and are essentially indeterminate. Reid, e.g. is redundant in
making two distinct powers of Conception and Abstraction.
He and Stewart pretend to fulfil the whole function of psycho-
logy, viz. explanation, whereas they only describe. For the
only scientific mode of explanation is the bringing phenomena
under laws. Explaining facts by faculties is essentially un-
scientific, for we must ascribe a quasi-independence to these
faculties. Even Hamilton, in spite of his having guarded
himself, falls into using the word as if for so many mutually
independent powers, as though as some one has said he
were dealing with European Powers. Psychology, as a rule,
begins where Reid and Stewart leave off. Still for Hamilton
I claim a certain amount of exemption from blame. He
is guided, moreover, as to much of his scheme by a
scientific principle : he goes from simple to complex. The
most salient feature in his classification is that each faculty is
explicable from the preceding. His scheme is better than
a mere string of beads. But in it psychology and philosophy
become hopelessly confused.
His scheme divides intellect into six faculties, in which we
find a close correspondence with our own arrangement :
(i) Presentative ') External . . . Perception.
,, (6) Internal. .
(5) Elaborative or Discursive . . . Conception, Thought.
(6) Regulative 1 .
1 I am not disposed to reject the prominence given to (a\ apart
from (31 and (4). Decidedly some retain well, but cannot at
will reproduce equally well. I could rather object to separating (3)
and (4). The fifth is the most instructive to study. I commend his
emphatic use of the word 'thought' as meaning re-representative
XIL] Elements of General Philosophy. 121
Hamilton confuses Psychology and Philosophy.
Now here in faculties (2) to (5) Hamilton is on psycho-
logical ground ; in (i) and (6) he trespasses on philosophy.
For instance, his first faculty he defines as that by which we
have (a) consciousness of objects, (3) consciousness of self.
This is more than we undertook to find in intellection ; it is
cognition in the fullest sense. Under the guise of psychology he
is already dealing with the problem of knowledge. Now it is
hardly fair to speak as though Hamilton professed to give us
a work on psychology, when for his title he has Metaphysics.
But we must charge him with not making the necessary
distinction, any more than Professor Bain does in another
direction, between psychology and philosophy. Here he
certainly does not pass gradually from simple to complex.
And the matter is made worse by the use of the apparently
very simple term Presentative. He over-simplifies in one
way, over-complicates in another. He himself, when in a
psychological mood, sees that Presentation is but a starting-
point. I deny ( i ) that we can start from perception of object
and self, (2) that there is purely presentative intellection.
The profit to the reader in those lectures on the first faculty
lies in the historical information; otherwise there is much
that is confusing and inconsistent. It was not a fortunate
Then as to the sixth. Till this is exercised, till the results
of the other five have been operated upon, regulated, by it,
intellection only, and have sought to establish in the traditions of
English psychology this usage, brought in first by Hamilton from
Kant. 'Discursive' too is a valuable old term, first showing the
function of thought as a 'ranging over' in order to bring together.
He calls this faculty also ' understanding,' as opposed to reason or
ratio, his sixth faculty.
122 Elements of General Philosophy. [LECT.
you have not, according to Hamilton, got knowledge. Not
professedly does he here pass again over to philosophy ; he
thinks it is all psychology. Yet he himself denies that this
is a faculty in the same sense as the others. He calls it by
a Latin name, as though English were not good enough for
it the locus principiorum nest or aggregate of principles
which have to be made manifest as involved in knowledge.
What does he mean by this Regulative Faculty, or the
Reason ? ' Regulative ' is a term he borrowed from Kant,
though not exactly the Kantian usage along with it. He
did not use it as I do to describe the function of such
philosophical doctrines as Logic or Ethics, his generic
term for such functioning being Nomology (as distinct from
Phenomenology). By ' Regulative ' he meant ordering or
interpreting or conditioning. Certain principles constitute
so many forms or conditions under which what we perceive,
remember, think, &c. comes to be held as knowledge. For
instance, by the action of the principle of Substance we
interpret what is presented in consciousness as qualities
cohering in a substance. And again, the flow of our
representations does not give us cognition till they are
ordered by the principle of Causality as effects of certain
causes. Not content herewith, he endeavours to reduce all
principles to one the principle of the Conditioned.
Note how he had already begged the sixth faculty to
expound the first.
We have now seen what the Common Sense school found
wanting in the Associationist doctrine, and how they sought
to supply it. In connexion herewith they tend to use belief
as being the foundation of knowledge, those fundamental
XII.] Elements of General Philosophy 123
principles of Common Sense or Reason being held in the
mind in the form of belief.
No student will lose his time if he study Hamilton. What-
ever his faults, his work is unsurpassed for instructive,
stimulative value. He really and consciously exhausted
intellect no less than is done in Mr. Spencer's scheme and
my own. Whereas with the classifications of Reid and
Stewart we might ask why they stop where they do.
For LECTURE XIII read :
Mill, Logic, Bk. II, ch. v. vi ' Of Demonstration and Necessary
THE NATURE OF KNOWLEDGE. CRITICAL PHILOSOPHY.
KANT was struck and even oppressed by the negative result
of Hume's analyses. It seemed to him that, if Hume was
right, no explanation of even the plain facts of science
was possible. He was prepared to accept Hume against the
older doctrines of metaphysics Platonic realism, innate
ideas, and so forth but he felt that there was that in know-
ledge which Hume had not touched that his negation of
knowledge was wrong, in that he had not faced the whole
problem. So he sought in the Krilik of Pure Reason to
work out a positive theory of knowledge and to destroy
scepticism, not by mere dogmatism like Descartes and Leibniz,
but by putting the whole of knowledge on a new footing,
and so to find a via media between the Experientialism of
Locke run out into the scepticism of Hume, and the
Rationalism of Descartes and Leibniz.
Kant's Inquiry into the Constituents of Knowledge.
He said that we must first settle what enters into know-
ledge. That sense is of account for knowledge he takes for
granted. Our knowledge is of sensible things. Not that we
have not moral convictions of something beyond, but know-
Elements of General Philosophy. 125
ledge proper always contains sense-elements. Sense itself
does not explain knowledge. Knowledge is not simply sense
transformed, but a resultant of certain elements a posteriori
(empirically given) wrought up with certain other a priori
A priori and a posteriori.
To these terms, which are to be found in Logic since the
time of Aristotle, Kant gave an epislemological significance.
The logical a priori is cognition of anything on the side of
its conditions, of what it can be shown by the laws of thought
to depend upon; it is knowledge in deductive form. And it
is so called because it can be shown to be dependent, through
the laws of thought or consistency, on what has been already
known or assumed, i. e. on premises. This is the only kind
of conclusion that is absolutely certain. But we can make
other inferences, for which we can never claim absolute
certainty, and yet which are the most important, viz. induc-
tions, or general assertions about facts. Here, except in
Jevons's trivial case of Perfect Induction, the certainty of our
inference is technically open to dispute ; it is only probable.
Such an inference is termed knowledge a posteriori.
Kant uses the terms for the two kinds of factors present in
knowledge. That which comes from sense, without which
no exercise of ' pure ' reason has any validity, is knowledge
a posteriori. But without the a priori factor of ' pure reason '
(reason not derived from experience) working on experience
we cannot get knowledge. For Kant, a priori is a general
name for ' rational ' as opposed to ' empirical ; ' it is what
Leibniz, in correcting Locke, meant by intellecius, or that
which is furnished by the mind's original constitution.
Kant, be it noted, was very vague in his use of ' experience.'
Sometimes it means with him the contribution of sense to
is6 Elements of General Philosophy. [LECT.
knowledge; at other times it stands, not for bare sense-
material, but for sense as ordered and interpreted by a priori
principles in fact for knowledge.
A priori Forms.
Again, just as in Logic a distinction is drawn between
matter and form of thought, so Kant distinguished episte-
mologically between matter and form of cognition generally.
The matter of knowledge is the data of sense; these are
taken up into, or perceived under, ' pure forms.' The ' forms '
of sense are space and time. When I get external sensations
I am so constituted that I order them in space. And I order
all my sensations in time. Space and time are pure forms
of intuition a term which Kant was careful to connect with
sense-perception only, and not with Reason, seeing how
related the words are.
Next, sense-perception, so explained from the conjunction
of matter and pure forms, becomes ready for conceptual know-
ing, i. e. for an orderly scheme or fabric of knowing common
to man and man in other words, objective knowledge.
Objective knowledge does not necessarily refer to objects in
space. Is it a fact that every event has a cause ? If it be
agreed that this is so, here is objective knowledge, although
it does not refer to objects in space. Such knowledge con-
sists of sense-phenomena subsumed or brought under pure
concepts of the understanding or fundamental principles of
judgment, by which Kant did not understand so many 'innate
ideas/ but postulated certain necessary forms of thought.
Universality and Necessity in Knowledge.
For there is a part of our knowledge, there are some of
our cognitions, which are not only universal or objective, but
also necessary. Some judgments assume the form ' S is P,'
XIIL] Elements of General Philosophy. 127
but some that of ' S must be P.' Now no experience can
explain so philosophers said why a ' must be ' is used any
more than it can warrant universal validity. Experience deals