‘impressions and ideas’ have to undergo, if their association is to
yield the result which Hume requires--if through it the succession
of feelings is to become a knowledge of things and their relations.
Such a result was required as the only means of maintaining together
the two characteristic positions of Locke’s philosophy; that,
namely, the only world we can know is the world of ‘ideas,’ and
that thought cannot originate ideas. Those relations, which Locke
had inconsistently treated at once as intellectual superinductions
and as ultimate conditions of reality, must be dealt with by one
of two methods. They must be reduced to impressions where that
could plausibly be done: where it could not, it must be admitted
that we have no ideas of them, but only ‘tendencies to suppose’
that we have such, arising from the association, through ‘natural
relations,’ of the ideas that we have. So dexterously does Hume work
the former method that, of all the ‘philosophical relations’ which
he recognizes, only Identity and Causation remain to be disposed of
by the latter; and if the other relations--resemblance, time and
space, proportion in quantity and degree in quality--could really
be admitted as data of sense, there would at least be a possible
basis for those ‘tendencies to suppose’ which, in the absence of any
corresponding ideas, the terms ‘Identity’ and ‘Causation’ must be
taken to represent. But, as we have shown, they can only be claimed
for sense, if sense is so far one with thought--one not by conversion
of thought into sense but by taking of sense into thought--as that
Hume’s favourite appeals to sense against the reality of intelligible
relations become unmeaning. They may be ‘impressions,’ there may
be ‘impressions of them,’ but only if we deny of the impression
what Hume asserts of it, and assert of it what he denies--only
if we understand by ‘impression’ not an ‘internal and perishing
existence;’ not that which, if other than taste, colour, sound, smell
or touch, must be a ‘passion or emotion ‘; _not_ that which carries
no reference to an object other than itself, and which must _either_
be single _or_ compound; but something permanent and constituted
by permanently coexisting parts; something that may ‘be conjoined
with’ any feeling, because it is none; that always carries with it a
reference to a subject which it is not but of which it is a quality;
and that is both many and one, since ‘in its simplicity it contains
many different resemblances and relations.’
282. In the account just adduced of vacuum, the effect of that double
dealing with ‘impressions,’ which we shall have to trace at large in
Hume’s explanation of our language about Causation and Identity, is
already exhibited in little. Just as, after the idea of pure space
has been excluded because not a copy of any possible impression, we
yet find an ‘idea,’ only differing from it in name, introduced as
the basis of that tendency to suppose which is to take the place of
the excluded idea, so we shall find ideas of relation in the way
of Identity and Causation--ideas which according to Hume we have
not--presupposed as the source of those ‘propensities to feign’ which
he accounts for the appearance of our having them.
Knowledge of relation in way of Identity and Causation excluded by
Locke’s definition of knowledge.
283. The primary characteristic of these relations according to Hume,
which they share with those of space and time, and which in fact
vitiates that definition of ‘philosophical relation,’ as depending
on comparison, which he adopts, is that they ‘depend not on the
ideas compared together, but may be changed without any change in
the ideas’. [1] It follows that they are not objects of knowledge,
according to the definition of knowledge which Hume inherited,
as ‘the perception of agreement or disagreement between ideas.’
A partial recognition of this consequence in regard to cause and
effect we found in Locke’s suspicion that a science of nature was
impossible--impossible because, however often a certain ‘idea of
quality and substance’ may have followed or accompanied another, such
sequence or accompaniment never amounts to agreement or ‘necessary
connexion’ between the ideas, and therefore never can warrant a
general assertion, but only the particular one, that the ideas in
question have so many times occurred in such an order. ‘Matters of
fact,’ however, which no more consist in agreement of ideas than
does causation, are by Locke treated without scruple as matter of
knowledge when they can be regarded as relations between present
sensations. Thus the ‘particular experiment’ in Physics constitutes
knowledge--the knowledge, for instance, that a piece of gold is now
dissolved in aqua regia; and when ‘I myself see a man walk on the
ice, it is knowledge.’ In such cases it does not occur to him to ask,
either what are the ideas that agree or how much of the experiment is
a present sensation. [2] Nor does Hume commonly carry his analysis
further. After admitting that the relations called ‘identity and
situation in time and place’ do not depend on the nature of the
ideas related, he proceeds: ‘When both the objects are present to
the senses along with the relation, we call _this_ perception rather
than reasoning; nor is there in this case any exercise of the thought
or any action, properly speaking, but a mere passive admission of
the impressions through the organs of sensation. According to this
way of thinking, we ought not to receive as reasoning any of the
observations we may make concerning _identity_ and the _relations_ of
_time_ and _place_; since in none of them the mind can go beyond what
is immediately present to the senses, either to discover the real
existence or the relations of objects’. [3]
[1] P. 372. [Book I, part III., sec. I.]
[2] Above, §§ 122 & 123.
[3] P. 376. [Book I, part III., sec. II.]
Inference a transition from an object perceived or remembered to one
that is not so.
284. This passage points out the way which Hume’s doctrine of
causation was to follow. That in any case ‘the mind should go beyond
a present feeling, either to discover the real existence or the
relations of objects’ other than present feelings, was what he could
not consistently admit. In the judgment of causation, however, it
seems to do so. ‘From the existence or action of one object,’ seen
or remembered, it seems to be assured of the existence or action
of another, not seen or remembered, on the ground of a necessary
connection between the two. [1] It is such assurance that is reckoned
to constitute reasoning in the distinctive sense of the term, as
different at once from the analysis of complex ideas and the simple
succession of ideas--such reasoning as, in the language of a later
philosophy, can yield synthetic propositions. What Hume has to do,
then, is to explain this ‘assurance’ away by showing that it is
not essentially different from that judgment of relation in time
and place which, because the related objects are ‘present to the
senses along with the relation,’ is called ‘perception rather than
reasoning,’ and to which no ‘exercise of the thought’ is necessary,
but a ‘mere passive admission of impressions through the organs of
sensation.’ Nor, for the assimilation of reasoning to perception,
is anything further needed than a reference to the connection
of ideas with impressions and of the ideas of imagination with
those of memory, as originally stated by Hume. When both of the
objects compared are present to the senses, we call the comparison
perception; when neither, or only one, is so present, we call it
reasoning. But the difference between the object that is present
to sense, and that which is not, is merely the difference between
impression and idea, which again is merely the difference between
the more and the less lively feeling. [2] To feeling, whether with
more or with less vivacity, every object, whether of perception
or reasoning, must alike be present. Is it then a sufficient
account of the matter, according to Hume, to say that when we are
conscious of contiguity and succession between objects of which
both are impressions we call it perception; but that when both
objects are ideas, or one an impression and the other an idea, we
call it reasoning? Not quite so. Suppose that I ‘have seen that
species of object we call flame, and have afterwards felt that
species of sensation we call heat.’ If I afterwards remembered the
succession of the feeling upon the sight, both objects (according
to Hume’s original usage of terms [3]) would be ideas as distinct
from the impressions; or, if upon seeing the flame I remembered the
previous experience of heat, one object would be an idea; but we
should not reckon it a case of reasoning. ‘In all cases wherein we
reason concerning objects, there is only one either perceived or
_remembered_, and the other is supplied in conformity to our past
experience’--supplied by the only other faculty than memory that can
‘supply an idea,’ viz. imagination. [4]
[1] Pp. 376, 384. [Book I, part III., secs. II. and IV.]
[2] Pp. 327, 375. [Book I, part I., sec. VII. and part III., sec.
III.]
[3] Above, par. 195.
[4] Pp. 384, 388. [Book I, part III., secs. IV. and V.]
Relation of cause and effect the same as this transition.
285. This being the only account of ‘inference from the known to
the unknown,’ which Hume could consistently admit, his view of the
relation of cause and effect must be adjusted to it. It could not
be other than a relation either between impression and impression,
or between impression and idea, or between idea and idea; and all
these relations are equally between feelings that we experience.
Thus, instead of being the ‘objective basis’ on which inference from
the known to the unknown rests, it is itself the inference; or, more
properly, it and the inference alike disappear into a particular
sort of transition from feeling to feeling. The problem, then, is to
account for its seeming to be other than this. ‘There is nothing in
any objects to persuade us that they are always _remote_ or always
_contiguous_; and when from experience and observation we discover
that the relation in this particular is invariable, we always
conclude that there is some secret _cause_ which separates or unites
them’. [1] It would seem, then, that the relation of cause and effect
is something which we infer from experience, from the connection of
impressions and ideas, but which is not itself impression or idea.
And it would _seem_ further, that, as we infer such an unexperienced
relation, so likewise we make inferences from it. In regard to
identity ‘we readily suppose an object may continue individually the
same, though several times absent from and present to the senses;
and ascribe to it an identity, notwithstanding the interruption of
the perception, whenever we conclude that if we had kept our hand
or eye constantly upon it, it would have conveyed an invariable and
uninterrupted perception. But this conclusion beyond the impressions
of our senses can be founded only on the connection of _cause and
effect_; nor can we otherwise have any security that the object is
not changed upon us, however much the new object may resemble that
which was formerly present to the senses.’
[1] P. 376. [Book I, part III., sec. II.]
Yet seems other than this. How this appearance is to be explained.
286. This relation which, going beyond our actual experience, we seem
to infer as the explanation of invariable contiguity in place or time
of certain impressions, and from which again we seem to infer the
identity of an object of which the perception has been interrupted,
is what we call necessary connection. It is their supposed necessary
connection which distinguishes objects related as cause and effect
from those related merely in the way of contiguity and succession,
[1] and it is a like supposition that leads us to infer what we do
not see or remember from what we do. If then the reduction of thought
and the intelligible world to feeling was to be made good, this
supposition, not being an impression of sense or a copy of such, must
be shown to be an ‘impression of reflection,’ according to Hume’s
sense of the term, _i.e._ a tendency of the soul, analogous to desire
and aversion, hope and fear, derived from impressions of sense but
not copied from them; [2] and the inference which it determines
must be shown to be the work of imagination, as affected by such
impression of reflection. This in brief is the purport of Hume’s
doctrine of causation.
[1] P. 376. [Book I, part III., sec. II.]
[2] Above, par. 195.
Inference, resting on supposition of necessary connection, to be
explained before that connection.
287. After his manner, however, he will go about with his reader. The
supposed ‘objective basis’ of knowledge is to be made to disappear,
but in such a way that no one shall miss it. So dexterously, indeed,
is this done, that perhaps to this day the ordinary student of Hume
is scarcely conscious of the disappearance. Hume merely announces to
begin with that he will ‘postpone the direct survey of this question
concerning the nature of necessary connection,’ and deal first with
these other two questions, viz. (1) ‘For what reason we pronounce it
_necessary_ that everything whose existence has a beginning, should
also have a cause?’ and (2) ‘Why we conclude that such particular
causes must _necessarily_ have such particular effects; and what is
the nature of that _inference_ we draw from the one to the other, and
of the _belief_ we repose in it?’ That is to say, he will consider
the inference from cause or effect, before he considers cause and
effect as a relation between objects, on which the inference is
supposed to depend. Meanwhile necessary connection, as a relation
between objects, is naturally supposed in some sense or other to
survive. In _what_ sense, the reader expects to find when these two
preliminary questions have been answered. But when they have been
answered, necessary connection, as a relation between objects, turns
out to have vanished.
Account of the inference given by Locke and Clarke rejected.
288. With the first of the above questions Hume only concerns
himself so far as to show that we cannot know either intuitively or
demonstratively, in Locke’s sense of the words, that ‘everything
whose existence has a beginning also has a cause.’ Locke’s own
argument for the necessity of causation--that ‘something cannot be
produced by nothing’--as well as Clarke’s--that ‘if anything wanted a
cause it would produce itself, _i.e._ exist before it existed’--are
merely different ways, as Hume shows, of assuming the point in
question. ‘If everything must have a cause, it follows that upon
exclusion of other causes we must accept of the object itself, or
of nothing, as causes. But ’tis the very point in question, whether
everything must have a cause or not’. [1] On that point, according
to Locke’s own showing, there can be no certainty, intuitive or
demonstrative; for between the idea of beginning to exist and the
idea of cause there is clearly no agreement, mediate or immediate.
They are not similar feelings, they are not quantities that can be
measured against each other, and to these alone can the definition
of knowledge and reasoning, which Hume retained, apply. There thus
disappears that last remnant of ‘knowledge’ in regard to nature which
Locke had allowed to survive--the knowledge that there is a necessary
connection, though one which we cannot find out. [2]
[1] P. 382. [Book I, part III., sec. III.]
[2] cf. Locke IV. 3, 29, and Introduc, par. 121.
Three points to be explained in the inference according to Hume.
289. Having thus shown, as he conceives, what the true answer to the
first of the above questions is not, Hume proceeds to show what it
is by answering the second. ‘Since it is not from knowledge or any
scientific reasoning that we derive the opinion of the necessity
of a cause to every new production,’ it must be from experience;
[1] and every general opinion derived from experience is merely the
summary of a multitude of particular ones. Accordingly when it has
been explained why we infer particular causes from particular effects
(and _vice versâ_), the inference from every event to a cause will
have explained itself. Now ‘all our arguments concerning causes
and effects consist both of an impression of the memory or senses,
and of the idea of that existence which produces the object of the
impression or is produced by it. Here, therefore, we have three
things to explain, viz. _first_, the original impression; _secondly_,
the transition to the idea of the connected cause or effect;
_thirdly_, the nature and qualities of that idea.’ [2]
[1] P. 383. [Book I, part III., sec. III.]
[2] P. 385. [Book I, part III., sec. V.]
_a_. The original impression from which the transition is made, and
_b_. The transition to inferred idea
290. As to the original impression we must notice that there is a
certain inconsistency with Hume’s previous usage of terms in speaking
of an _impression_ of memory at all. [1] This, however, will be
excused when we reflect that according to him impression and idea
only differ in liveliness, and that he is consistent in claiming for
the ideas of memory, not indeed the maximum, but a high degree of
vivacity, superior to that which belongs to ideas of imagination. All
that can be said, then, of that ‘original impression,’ whether of the
memory or senses, which is necessary to any ‘reasoning from cause or
effect,’ is that it is highly vivacious. That the transition from it
to the ‘idea of the connected cause or effect’ is not determined by
reason, has already been settled. It could only be so determined,
according to the received account of reason, if there were some
agreement in respect of quantity or quality between the idea of cause
and that of the effect, to be ascertained by the interposition of
other ideas. [2] But when we examine any particular objects that we
hold to be related as cause and effect, _e.g._ the sight of flame and
the feeling of heat, we find no such agreement. What we _do_ find
is their ‘constant conjunction’ in experience, and ‘conjunction’ is
equivalent to that ‘contiguity in time and place,’ which has already
been pointed out as one of those ‘natural relations’ which act as
‘principles of union’ between ideas. [3] Because the impression of
flame has always been found to be followed by the impression of heat,
the idea of flame always suggests the idea of heat. It is simple
custom then that determines the transition from the one to the other,
or renders ‘necessary’ the connection between them. In order that the
transition, however, may constitute an inference from cause to effect
(or _vice versâ_), one of the two objects thus naturally related,
but not both, must be presented as an impression. If both were
impressions it would be a case of ‘sensation, not reasoning;’ if both
were ideas, no belief would attend the transition. This brings us to
the question as to the ‘nature and qualities’ of the inferred idea.
[1] Above, par. 195.
[2] Cf. Locke IV. 17, 2.
[3] Above, par. 206.
_c_. The qualities of this idea.
291. ‘’Tis evident that all reasonings from causes or effects
terminate in conclusions concerning matter of fact, _i.e._ concerning
the existence of objects or of their qualities’; [1] in other words,
in belief. If this meant a new idea, an idea that we have not
previously had, it would follow that inference could really carry us
beyond sense, that there could be an idea not copied from any prior
impression. But according to Hume it does not mean this. ‘The idea
of existence is the very same with the idea of what we conceive to
be existent;’ [2] and not only so, ‘the _belief_ of existence joins
no new ideas to those which compose the idea of the object. When I
think of God, when I think of him as existent, and when I believe him
to be existent, my idea of him neither increases nor diminishes’.
[3] In what then lies the difference between incredulity and belief;
between an ‘idea assented to,’ or an object believed to exist, and
a fictitious object or idea from which we dissent? The answer is,
‘not in the parts or composition of the idea, but in the manner
of conceiving it,’ which must be understood to mean the manner of
‘feeling’ it; and this difference is further explained to lie in ‘the
superior force, or vivacity, or steadiness’ with which it is felt.’
[4] We are thus brought to the further question, how it is that this
‘superior vivacity’ belongs to the inferred idea when we ‘reason’
from cause to effect or from effect to cause. The answer here is
that the ‘impression of the memory or senses,’ which in virtue of
a ‘natural relation’ suggests the idea, also ‘communicates to it a
share of its force or vivacity.’
[1] P. 394. [Book I, part III., sec. VII.]
[2] P. 370. [Book I, part II., sec. VI.]
[3] P. 395. [Book I, part III., sec. II.]
[4] P. 398 [Book I, part III., sec. VII.]. Cf. above, par. 170, for
the corresponding view in Berkeley.
It results that necessary connection is an impression of reflection,
_i.e._, a propensity to the transition described.
292. Thus it appears that in order to the conclusion that any
particular cause must have any particular effect, there is needed
first the presence of an impression, and secondly the joint action
of those two ‘principles of union among ideas,’ resemblance and
contiguity. In virtue of the former principle the given impression
calls up the image of a like impression previously experienced,
which again in virtue of the latter calls up the image of its usual
attendant, and the liveliness of the given impression so communicates
itself to the recalled ideas as to constitute belief in their
existence. If this is the true account of the matter, the question
as to the nature of necessary connexion has answered itself. ‘The
necessary connexion betwixt causes and effects is the foundation of
our inference from one to the other. The foundation of the inference
is the transition arising from the accustomed union. These are
therefore the same’. [1] We may thus understand how it is that there
seems to be an idea of such connexion to which no impression of the
senses, or (to use an equivalent phrase of Hume’s) no ‘quality in
objects’ corresponds. If the first presentation of two objects, of
which one is cause, the other effect, (_i.e._ of which we afterwards
come to consider one the cause, the other the effect) gives no idea
of a connexion between them, as it clearly does not, neither can
it do so however often repeated. It would not do so, unless the
repetition ‘either discovered or produced something new’ in the
objects; and it does neither. But it does ‘produce a new impression
in the mind.’ After observing a ‘constant conjunction of the objects,
and an uninterrupted resemblance of their relations of contiguity
and succession, we immediately feel a determination of the mind to
pass from one of the objects to its usual attendant, and to conceive
it in a stronger light on account of that relation.’ It is of this
‘internal impression,’ this ‘propensity which custom produces,’ that
the idea of necessary connexion is the copy. [2]
[1] P. 460. [Book I, part III., sec. XIV.]
[2] Pp. 457-460. [Book I, part III., sec. XIV.]
The transition not to anything beyond sense.
293. The sequence of ideas, which this propensity determines, clearly
does not involve any inference ‘beyond sense,’ ‘from the known to the
unknown,’ ‘from instances of which we have had experience, to those
of which we have had none,’ any more than does any other ‘recurrence
of an idea’--which, as we have seen, merely means, according to Hume,
the return of a feeling at a lower level of intensity after it has
been felt at a higher. The idea which we speak of as an inferred
cause or effect is only an ‘instance of which we have no experience’
in the sense of being _numerically different_ from the similar
ideas, whose previous constant association with an impression like
the given one, determines the ‘inference;’ but in the same sense
the ‘impression’ which I now feel on putting my hand to the fire
is different from the impressions previously felt under the same
circumstances, and I do not for that reason speak of this impression
as an instance of which I have had no experience. Thus Hume, though
retaining the received phraseology in reference to the ‘conclusion
from any particular cause to any particular effect’--phraseology
which implies that prior to the inference the object inferred is in
some sense unknown or unexperienced--yet deprives it of meaning by a
doctrine which makes inference, as he himself puts it, ‘a species of
sensation,’ ‘an unintelligible instinct of our souls,’ ‘more properly
an act of the sensitive than of the cogitative part of our natures’
[1]--which in fact leaves no ‘part of our natures’ to be cogitative
at all.
[1] Pp. 404, 475, and 471. [Book I, part III., sec. VIII., part IV.
sec. I. and part III., sec. XVI.]
Nor determined by any objective relation.