OF UNIFORMITIES OF COEXISTENCE NOT DEPENDENT ON CAUSATION.
§ 1. The order of the occurrence of phenomena in time, is either
successive or simultaneous; the uniformities, therefore, which obtain in
their occurrence, are either uniformities of succession or of
coexistence. Uniformities of succession are all comprehended under the
law of causation and its consequences. Every phenomenon has a cause,
which it invariably follows; and from this are derived other invariable
sequences among the successive stages of the same effect, as well as
between the effects resulting from causes which invariably succeed one
another.
In the same manner with these derivative uniformities of succession, a
great variety of uniformities of coexistence also take their rise.
Coordinate effects of the same cause naturally coexist with one another.
High water at any point on the earth's surface, and high water at the
point diametrically opposite to it, are effects uniformly simultaneous,
resulting from the direction in which the combined attractions of the
sun and moon act upon the waters of the ocean. An eclipse of the sun to
us, and an eclipse of the earth to a spectator situated in the moon, are
in like manner phenomena invariably coexistent; and their coexistence
can equally be deduced from the laws of their production.
It is an obvious question, therefore, whether all the uniformities of
coexistence among phenomena may not be accounted for in this manner. And
it cannot be doubted that between phenomena which are themselves
effects, the coexistences must necessarily depend on the causes of those
phenomena. If they are effects immediately or remotely of the same
cause, they cannot coexist except by virtue of some laws or properties
of that cause: if they are effects of different causes, they cannot
coexist unless it be because their causes coexist; and the uniformity of
coexistence, if such there be, between the effects, proves that those
particular causes, within the limits of our observation, have uniformly
been coexistent.
§ 2. But these same considerations compel us to recognise that there
must be one class of coexistences which cannot depend on causation; the
coexistences between the ultimate properties of things--those properties
which are the causes of all phenomena, but are not themselves caused by
any phenomenon, and a cause for which could only be sought by ascending
to the origin of all things. Yet among these ultimate properties there
are not only coexistences, but uniformities of coexistence. General
propositions may be, and are, formed, which assert that whenever certain
properties are found, certain others are found along with them. We
perceive an object; say, for instance, water. We recognise it to be
water, of course by certain of its properties. Having recognised it, we
are able to affirm of it innumerable other properties; which we could
not do unless it were a general truth, a law or uniformity in nature,
that the set of properties by which we identify the substance as water,
always have those other properties conjoined with them.
In a former place,[31] it has been explained in some detail what is
meant by the Kinds of objects; those classes which differ from one
another not by a limited and definite, but by an indefinite and unknown,
number of distinctions. To this we have now to add, that every
proposition by which anything is asserted of a Kind, affirms an
uniformity of coexistence. Since we know nothing of Kinds but their
properties, the Kind, to us, _is_ the set of properties by which it is
identified, and which must of course be sufficient to distinguish it
from every other kind.[32] In affirming anything, therefore, of a Kind,
we are affirming something to be uniformly coexistent with the
properties by which the kind is recognised; and that is the sole meaning
of the assertion.
Among the uniformities of coexistence which exist in nature, may hence
be numbered all the properties of Kinds. The whole of these, however,
are not independent of causation, but only a portion of them. Some are
ultimate properties, others derivative; of some, no cause can be
assigned, but others are manifestly dependent on causes. Thus, pure
atmospheric air is a Kind, and one of its most unequivocal properties is
its gaseous form: this property, however, has for its cause the presence
of a certain quantity of latent heat; and if that heat could be taken
away (as has been done from so many gases in Faraday's experiments), the
gaseous form would doubtless disappear, together with numerous other
properties which depend on, or are caused by, that property.
In regard to all substances which are chemical compounds, and which
therefore may be regarded as products of the juxtaposition of substances
different in Kind from themselves, there is considerable reason to
presume that the specific properties of the compound are consequent, as
effects, on some of the properties of the elements, though little
progress has yet been made in tracing any invariable relation between
the latter and the former. Still more strongly will a similar
presumption exist, when the object itself, as in the case of organized
beings, is no primeval agent, but an effect, which depends on a cause or
causes for its very existence. The Kinds therefore which are called in
chemistry simple substances, or elementary natural agents, are the only
ones, any of whose properties can with certainty be considered
ultimate; and of these the ultimate properties are probably much more
numerous that we at present recognise, since every successful instance
of the resolution of the properties of their compounds into simpler
laws, generally leads to the recognition of properties in the elements
distinct from any previously known. The resolution of the laws of the
heavenly motions, established the previously unknown ultimate property
of a mutual attraction between all bodies: the resolution, so far as it
has yet proceeded, of the laws of crystallization, of chemical
composition, electricity, magnetism, &c., points to various polarities,
ultimately inherent in the particles of which bodies are composed; the
comparative atomic weights of different kinds of bodies were ascertained
by resolving, into more general laws, the uniformities observed in the
proportions in which substances combine with one another; and so forth.
Thus although every resolution of a complex uniformity into simpler and
more elementary laws has an apparent tendency to diminish the number of
the ultimate properties, and really does remove many properties from the
list; yet, (since the result of this simplifying process is to trace up
an ever greater variety of different effects to the same agents,) the
further we advance in this direction, the greater number of distinct
properties we are forced to recognise in one and the same object: the
coexistences of which properties must accordingly be ranked among the
ultimate generalities of nature.
§ 3. There are, therefore, only two kinds of propositions which assert
uniformity of coexistence between properties. Either the properties
depend on causes, or they do not. If they do, the proposition which
affirms them to be coexistent is a derivative law of coexistence between
effects, and until resolved into the laws of causation on which it
depends, is an empirical law, and to be tried by the principles of
induction to which such laws are amenable. If, on the other hand, the
properties do not depend on causes, but are ultimate properties; then if
it be true that they invariably coexist, they must all be ultimate
properties of one and the same Kind; and it is of these only that the
coexistences can be classed as a peculiar sort of laws of nature.
When we affirm that all crows are black, or that all negroes have woolly
hair, we assert an uniformity of coexistence. We assert that the
property of blackness, or of having woolly hair, invariably coexists
with the properties which, in common language, or in the scientific
classification that we adopt, are taken to constitute the class crow, or
the class negro. Now, supposing blackness to be an ultimate property of
black objects, or woolly hair an ultimate property of the animals which
possess it; supposing that these properties are not results of
causation, are not connected with antecedent phenomena by any law; then
if all crows are black, and all negroes have woolly hair, these must be
ultimate properties of the Kind _crow_, or _negro_, or of some Kind
which includes them. If, on the contrary, blackness or woolly hair be an
effect depending on causes, these general propositions are manifestly
empirical laws; and all that has already been said respecting that class
of generalizations may be applied without modification to these.
Now, we have seen that in the case of all compounds--of all things, in
short, except the elementary substances and primary powers of
nature--the presumption is, that the properties do really depend upon
causes; and it is impossible in any case whatever to be certain that
they do not. We therefore should not be safe in claiming for any
generalization respecting the coexistence of properties, a degree of
certainty to which, if the properties should happen to be the result of
causes, it would have no claim. A generalization respecting coexistence,
or in other words respecting the properties of Kinds, may be an ultimate
truth, but it may, also, be merely a derivative one; and since, if so,
it is one of those derivative laws which are neither laws of causation,
nor have been resolved into the laws of causation on which they depend,
it can possess no higher degree of evidence than belongs to an empirical
law.
§ 4. This conclusion will be confirmed by the consideration of one
great deficiency, which precludes the application to the ultimate
uniformities of coexistence, of a system of rigorous scientific
induction, such as the uniformities in the succession of phenomena have
been found to admit of. The basis of such a system is wanting: there is
no general axiom, standing in the same relation to the uniformities of
coexistence as the law of causation does to those of succession. The
Methods of Induction applicable to the ascertainment of causes and
effects, are grounded on the principle that everything which has a
beginning must have some cause or other; that among the circumstances
which actually existed at the time of its commencement, there is
certainly some one combination, on which the effect in question is
unconditionally consequent, and on the repetition of which it would
certainly again recur. But in an inquiry whether some kind (as crow)
universally possesses a certain property (as blackness), there is no
room for any assumption analogous to this. We have no previous certainty
that the property must have something which constantly coexists with it;
must have an invariable coexistent, in the same manner as an event must
have an invariable antecedent. When we feel pain, we must be in some
circumstances under which if exactly repeated we should always feel
pain. But when we are conscious of blackness, it does not follow that
there is something else present of which blackness is a constant
accompaniment. There is, therefore, no room for elimination; no Method
of Agreement or Difference, or of Concomitant Variations (which is but a
modification either of the Method of Agreement or of the Method of
Difference). We cannot conclude that the blackness we see in crows must
be an invariable property of crows, merely because there is nothing else
present of which it can be an invariable property. We therefore inquire
into the truth of a proposition like "All crows are black," under the
same disadvantage as if, in our inquiries into causation, we were
compelled to let in, as one of the possibilities, that the effect may in
that particular instance have arisen without any cause at all.
To overlook this grand distinction was, as it seems to me, the capital
error in Bacon's view of inductive philosophy. The principle of
elimination, that great logical instrument which he had the immense
merit of first bringing into general use, he deemed applicable in the
same sense, and in as unqualified a manner, to the investigation of the
coexistences, as to that of the successions of phenomena. He seems to
have thought that as every event has a cause, or invariable antecedent,
so every property of an object has an invariable coexistent, which he
called its Form: and the examples he chiefly selected for the
application and illustration of his method, were inquiries into such
Forms; attempts to determine in what else all those objects resembled,
which agreed in some one general property, as hardness or softness,
dryness or moistness, heat or coldness. Such inquiries could lead to no
result. The objects seldom have any such circumstances in common. They
usually agree in the one point inquired into, and in nothing else. A
great proportion of the properties which, so far as we can conjecture,
are the likeliest to be really ultimate, would seem to be inherently
properties of many different Kinds of things, not allied in any other
respect. And as for the properties which, being effects of causes, we
are able to give some account of, they have generally nothing to do with
the ultimate resemblances or diversities in the objects themselves, but
depend on some outward circumstances, under the influence of which any
objects whatever are capable of manifesting those properties; as is
emphatically the case with those favourite subjects of Bacon's
scientific inquiries, hotness and coldness; as well as with hardness and
softness, solidity and fluidity, and many other conspicuous qualities.
In the absence, then, of any universal law of coexistence, similar to
the universal law of causation which regulates sequence, we are thrown
back upon the unscientific induction of the ancients, _per enumerationem
simplicem, ubi non reperitur instantia contradictoria_. The reason we
have for believing that all crows are black, is simply that we have seen
and heard of many black crows, and never one of any other colour. It
remains to be considered how far this evidence can reach, and how we are
to measure its strength in any given case.
§ 5. It sometimes happens that a mere change in the mode of verbally
enunciating a question, though nothing is really added to the meaning
expressed, is of itself a considerable step towards its solution. This,
I think, happens in the present instance. The degree of certainty of any
generalization which rests on no other evidence than the agreement, so
far as it goes, of all past observation, is but another phrase for the
degree of improbability that an exception, if any existed, could have
hitherto remained unobserved. The reason for believing that all crows
are black, is measured by the improbability that crows of any other
colour should have existed to the present time without our being aware
of it. Let us state the question in this last mode, and consider what is
implied in the supposition that there may be crows which are not black,
and under what conditions we can be justified in regarding this as
incredible.
If there really exist crows which are not black, one of two things must
be the fact. Either the circumstance of blackness, in all crows hitherto
observed, must be, as it were, an accident, not connected with any
distinction of Kind; or if it be a property of Kind, the crows which are
not black must be a new Kind, a Kind hitherto overlooked, though coming
under the same general description by which crows have hitherto been
characterized. The first supposition would be proved true if we were to
discover casually a white crow among black ones, or if it were found
that black crows sometimes turn white. The second would be shown to be
the fact if in Australia or Central Africa a species or a race of white
or grey crows were found to exist.
§ 6. The former of these suppositions necessarily implies that the
colour is an effect of causation. If blackness, in the crows in which it
has been observed, be not a property of Kind, but can be present or
absent without any difference generally in the properties of the object;
then it is not an ultimate fact in the individuals themselves, but is
certainly dependent on a cause. There are, no doubt, many properties
which vary from individual to individual of the same Kind, even the
same _infima species_, or lowest Kind. Some flowers may be either white
or red, without differing in any other respect. But these properties are
not ultimate; they depend on causes. So far as the properties of a thing
belong to its own nature, and do not arise from some cause extrinsic to
it, they are always the same in the same Kind. Take, for instance, all
simple substances and elementary powers; the only things of which we are
certain that some at least of their properties are really ultimate.
Colour is generally esteemed the most variable of all properties: yet we
do not find that sulphur is sometimes yellow and sometimes white, or
that it varies in colour at all, except so far as colour is the effect
of some extrinsic cause, as of the sort of light thrown upon it, the
mechanical arrangement of the particles, (as after fusion) &c. We do not
find that iron is sometimes fluid and sometimes solid at the same
temperature; gold sometimes malleable and sometimes brittle; that
hydrogen will sometimes combine with oxygen and sometimes not; or the
like. If from simple substances we pass to any of their definite
compounds, as water, lime, or sulphuric acid, there is the same
constancy in their properties. When properties vary from individual to
individual, it is either in the case of miscellaneous aggregations, such
as atmospheric air or rock, composed of heterogeneous substances, and
not constituting or belonging to any real Kind,[33] or it is in the case
of organic beings. In them, indeed, there is variability in a high
degree. Animals of the same species and race, human beings of the same
age, sex, and country, will be most different, for example, in face and
figure. But organized beings (from the extreme complication of the laws
by which they are regulated) being more eminently modifiable, that is,
liable to be influenced by a greater number and variety of causes, than
any other phenomena whatever; having also themselves had a beginning,
and therefore a cause; there is reason to believe that none of their
properties are ultimate, but all of them derivative, and produced by
causation. And the presumption is confirmed, by the fact that the
properties which vary from one individual to another, also generally
vary more or less at different times in the same individual; which
variation, like any other event, supposes a cause, and implies,
consequently, that the properties are not independent of causation.
If, therefore, blackness be merely accidental in crows, and capable of
varying while the Kind remains the same, its presence or absence is
doubtless no ultimate fact, but the effect of some unknown cause: and in
that case the universality of the experience that all crows are black is
sufficient proof of a common cause, and establishes the generalization
as an empirical law. Since there are innumerable instances in the
affirmative, and hitherto none at all in the negative, the causes on
which the property depends must exist everywhere in the limits of the
observations which have been made; and the proposition may be received
as universal within those limits, and with the allowable degree of
extension to adjacent cases.
§ 7. If, in the second place, the property, in the instances in which it
has been observed, is not an effect of causation, it is a property of
Kind; and in that case the generalization can only be set aside by the
discovery of a new Kind of crow. That, however, a peculiar Kind, not
hitherto discovered, should exist in nature, is a supposition so often
realized, that it cannot be considered at all improbable. We have
nothing to authorize us in attempting to limit the Kinds of things which
exist in nature. The only unlikelihood would be that a new Kind should
be discovered in localities which there was previously reason to believe
had been thoroughly explored; and even this improbability depends on the
degree of conspicuousness of the difference between the newly-discovered
Kind and all others, since new Kinds of minerals, plants, and even
animals, previously overlooked or confounded with known species, are
still continually detected in the most frequented situations. On this
second ground, therefore, as well as on the first, the observed
uniformity of coexistence can only hold good as an empirical law, within
the limits not only of actual observation, but of an observation as
accurate as the nature of the case required. And hence it is that (as
remarked in an early chapter of the present Book) we so often give up
generalizations of this class at the first summons. If any credible
witness stated that he had seen a white crow, under circumstances which
made it not incredible that it should have escaped notice previously, we
should give full credence to the statement.
It appears, then, that the uniformities which obtain in the coexistence
of phenomena,--those which we have reason to consider as ultimate, no
less than those which arise from the laws of causes yet undetected--are
entitled to reception only as empirical laws; are not to be presumed
true except within the limits of time, place, and circumstance, in which
the observations were made, or except in cases strictly adjacent.
§ 8. We have seen in the last chapter that there is a point of
generality at which empirical laws become as certain as laws of nature,
or rather, at which there is no longer any distinction between empirical
laws and laws of nature. As empirical laws approach this point, in other
words, as they rise in their degree of generality, they become more
certain; their universality may be more strongly relied on. For, in the
first place, if they are results of causation (which, even in the class
of uniformities treated of in the present chapter, we never can be
certain that they are not) the more general they are, the greater is
proved to be the space over which the necessary collocations prevail,
and within which no causes exist capable of counteracting the unknown
causes on which the empirical law depends. To say that anything is an
invariable property of some very limited class of objects, is to say
that it invariably accompanies some very numerous and complex group of
distinguishing properties; which, if causation be at all concerned in
the matter, argues a combination of many causes, and therefore a great
liability to counteraction; while the comparatively narrow range of the
observations renders it impossible to predict to what extent unknown
counteracting causes may be distributed throughout nature. But when a
generalization has been found to hold good of a very large proportion of
all things whatever, it is already proved that nearly all the causes
which exist in nature have no power over it; that very few changes in
the combination of causes can effect it; since the greater number of
possible combinations must have already existed in some one or other of
the instances in which it has been found true. If, therefore, any
empirical law is a result of causation, the more general it is, the more
it may be depended on. And even if it be no result of causation, but an
ultimate coexistence, the more general it is, the greater amount of
experience it is derived from, and the greater therefore is the
probability that if exceptions had existed, some would already have
presented themselves.
For these reasons, it requires much more evidence to establish an
exception to one of the more general empirical laws than to the more
special ones. We should not have any difficulty in believing that there
might be a new Kind of crow; or a new kind of bird resembling a crow in
the properties hitherto considered distinctive of that Kind. But it
would require stronger proof to convince us of the existence of a Kind
of crow having properties at variance with any generally recognised
universal property of birds; and a still higher degree if the properties
conflict with any recognised universal property of animals. And this is
conformable to the mode of judgment recommended by the common sense and
general practice of mankind, who are more incredulous as to any
novelties in nature, according to the degree of generality of the
experience which these novelties seem to contradict.
§ 9. Still, however, even these greater generalizations, which embrace
comprehensive Kinds, containing under them a great number and variety of
_infimæ species_, are only empirical laws, resting on induction by
simple enumeration merely, and not on any process of elimination, a
process wholly inapplicable to this sort of case. Such generalizations,
therefore, ought to be grounded on an examination of all the _infimæ
species_ comprehended in them, and not of a portion only. We cannot
conclude (where causation is not concerned), because a proposition is
true of a number of things resembling one another only in being animals,
that it is therefore true of all animals. If, indeed, anything be true
of species which differ more from one another than either differs from a
third, (especially if that third species occupies in most of its known
properties a position between the two former,) there is some probability
that the same thing will also be true of that intermediate species; for
it is often, though by no means universally, found, that there is a sort
of parallelism in the properties of different Kinds, and that their
degree of unlikeness in one respect bears some proportion to their
unlikeness in others. We see this parallelism in the properties of the
different metals; in those of sulphur, phosphorus, and carbon; of
chlorine, iodine, and bromine; in the natural orders of plants and
animals, &c. But there are innumerable anomalies and exceptions to this
sort of conformity; if indeed the conformity itself be anything but an
anomaly and an exception in nature.
Universal propositions, therefore, respecting the properties of superior
Kinds, unless grounded on proved or presumed connexion by causation,
ought not to be hazarded except after separately examining every known
sub-kind included in the larger Kind. And even then such generalizations
must be held in readiness to be given up on the occurrence of some new
anomaly, which, when the uniformity is not derived from causation, can
never, even in the case of the most general of these empirical laws, be
considered very improbable. Thus all the universal propositions which it
has been attempted to lay down respecting simple substances, or
concerning any of the classes which have been formed among simple
substances, (and the attempt has been often made,) have, with the
progress of experience, either faded into inanity, or been proved to be
erroneous; and each Kind of simple substance remains with its own
collection of properties apart from the rest, saving a certain
parallelism with a few other Kinds, the most similar to itself. In
organized beings, indeed, there are abundance of propositions
ascertained to be universally true of superior genera, to many of which
the discovery hereafter of any exceptions must be regarded as extremely
improbable. But these, as already observed, are, we have every reason to
believe, properties dependent on causation.
Uniformities of coexistence, then, not only when they are consequences
of laws of succession, but also when they are ultimate truths, must be
ranked, for the purpose of logic, among empirical laws; and are amenable
in every respect to the same rules with those unresolved uniformities
which are known to be dependent on causation.