[ Pobierz całość w formacie PDF ]
stationary condition of industry, it becomes valueless in face of the in-
creasing complexity of the modern economic world; and that under such
conditions, at any rate, it must give way to the method of specific expe-
rience.123 We have already indicated reasons why the reverse is nearer
the truth. It is needless to say that as the forces in operation become
more numerous and their modes of interaction more intricate, and as
changes in the general conditions of industry succeed one another more
rapidly, the problems of economics become more difficult whatever
method may be employed for their solution. The deductive method needs
to be applied with the greater caution, and the apparatus of facts with
which it has to be supplemented increases in importance. But, on the
other hand, facts by themselves leave us the more helpless, and mere
empiricism is the more misleading. Compare, for example, the investi-
gation of the connexion between changes in gold supply and prices, or
between changes in gold supply and the rate of discount, under simple
and comparatively stationary industrial conditions, with the same in-
vestigation under complex industrial conditions, where credit and bank-
ing are fully developed, and where prices, and the condition of the loan
market, are liable to alterations from an indefinite number of causes. If
a mere a posteriori examination of statistics, without any appeal to
deductive argument, might in the former case be of some value, it would
certainly in the latter case be almost worse than useless.124
There is a special reason why no attempt should be made to ignore
or disguise the weakness of pure induction in complex economic enqui-
ries. The prevalence of a low type of inductive reasoning in the treat-
ment of economic questions is one of the most fertile sources of eco-
100/John Neville Keynes
nomic fallacy; and, however legitimate the employment of the inductive
method may be under certain conditions, there can be no doubt that this
method is liable to serious abuse.
Chapter VII
On The Deductive Method in Political Economy
1. Nature of the deductive method. In so far as the method of spe-
cific experience fails to afford reliable knowledge of economic laws,
recourse must be had to a method, whose essence consists in the pre-
liminary determination of the principal forces in operation, and the de-
duction of their consequences under various conditions. For an a poste-
riori argument depending entirely upon the examination of concrete facts
in all the complexity of their actual presentation, is substituted an a
priori argument depending upon knowledge of the general characteris-
tics displayed by men in their economic dealings one with another. The
problem of the deductive method, says Mill, is to find the law of an
effect from the laws of the different tendencies of which it is the joint
result. The method in its complete form consists of three steps. It is
necessary, first, to determine what are the principal forces in operation,
and the laws in accordance with which they
operate. Next comes the purely deductive stage, in which are in-
ferred the consequences that will ensue from the operation of these forces
under given conditions. Lastly, by a comparison of what has been in-
ferred with what can be directly observed to occur, an opportunity is
afforded for testing the correctness and practical adequacy of the two
preceding steps, and for the suggestion of necessary qualifications. It
will be observed that only one of these three steps namely, the middle
one -is strictly speaking deductive. The so-called deductive method in
its complete form is thus seen to be not an exclusively deductive method.
It may more accurately be described as a method which, whilst pre-
dominantly deductive, is still aided and controlled by induction. This
point will be further brought out in what follows, but it seems desirable
to call attention to it at the outset.
2. The application of the term hypothetical to economic sci-
ence. Political economy, in having recourse to the deductive method,
is usually described as essentially hypothetical in character. This de-
scription of the science needs, however, to be carefully explained and
guarded, as there is some danger of confusion of thought in regard to the
implications contained in it.
The Scope and Method of Political Economy/101
All laws of causation may be said to be hypothetical, in so far as
they merely assert that given causes will in the absence of counteract-
ing causes produce certain effects. As a matter of fact, in the instances
that actually occur of the operation of a given cause, counteracting causes
sometimes will and sometimes will not be present; and, therefore, laws
of causation are to be regarded as statements of tendencies only. It fol-
lows that all sciences of causation, and pre-eminently sciences employ-
ing the deductive method including political economy and astronomy
contain a hypothetical element.
The above may be expressed somewhat differently by saying that
the use of the deductive method in economics is solves, at a certain
stage, a process of abstraction, necessitating a frequent recurrence of
the qualification ceteris paribus. The abstraction is carried furthest in
reasonings where the motive of self-interest is supposed to operate un-
checked in a state of economic freedom; that is, in reasonings which
involve the conception of the economic man. But in all cases where
the deductive method is used, it is present more or less. For in the deduc-
tive investigation of the economic consequences of any particular cir-
cumstance or any particular change, the absence of interfering agencies
and of concurrent but independent changes is of necessity assumed. So
far as other changes themselves result from the one change or the one
circumstance specially under consideration, account must of course be
taken of them; but the distinguishing characteristic of the deductive
method consists in seeking, in the first instance, to effect a mental isola-
tion from the operation of all modifying forces that are not in some way
connected causally with the particular subject of enquiry. The distinc-
tion between dependent and independent changes, here indicated, is of [ Pobierz całość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl rafalstec.xlx.pl
stationary condition of industry, it becomes valueless in face of the in-
creasing complexity of the modern economic world; and that under such
conditions, at any rate, it must give way to the method of specific expe-
rience.123 We have already indicated reasons why the reverse is nearer
the truth. It is needless to say that as the forces in operation become
more numerous and their modes of interaction more intricate, and as
changes in the general conditions of industry succeed one another more
rapidly, the problems of economics become more difficult whatever
method may be employed for their solution. The deductive method needs
to be applied with the greater caution, and the apparatus of facts with
which it has to be supplemented increases in importance. But, on the
other hand, facts by themselves leave us the more helpless, and mere
empiricism is the more misleading. Compare, for example, the investi-
gation of the connexion between changes in gold supply and prices, or
between changes in gold supply and the rate of discount, under simple
and comparatively stationary industrial conditions, with the same in-
vestigation under complex industrial conditions, where credit and bank-
ing are fully developed, and where prices, and the condition of the loan
market, are liable to alterations from an indefinite number of causes. If
a mere a posteriori examination of statistics, without any appeal to
deductive argument, might in the former case be of some value, it would
certainly in the latter case be almost worse than useless.124
There is a special reason why no attempt should be made to ignore
or disguise the weakness of pure induction in complex economic enqui-
ries. The prevalence of a low type of inductive reasoning in the treat-
ment of economic questions is one of the most fertile sources of eco-
100/John Neville Keynes
nomic fallacy; and, however legitimate the employment of the inductive
method may be under certain conditions, there can be no doubt that this
method is liable to serious abuse.
Chapter VII
On The Deductive Method in Political Economy
1. Nature of the deductive method. In so far as the method of spe-
cific experience fails to afford reliable knowledge of economic laws,
recourse must be had to a method, whose essence consists in the pre-
liminary determination of the principal forces in operation, and the de-
duction of their consequences under various conditions. For an a poste-
riori argument depending entirely upon the examination of concrete facts
in all the complexity of their actual presentation, is substituted an a
priori argument depending upon knowledge of the general characteris-
tics displayed by men in their economic dealings one with another. The
problem of the deductive method, says Mill, is to find the law of an
effect from the laws of the different tendencies of which it is the joint
result. The method in its complete form consists of three steps. It is
necessary, first, to determine what are the principal forces in operation,
and the laws in accordance with which they
operate. Next comes the purely deductive stage, in which are in-
ferred the consequences that will ensue from the operation of these forces
under given conditions. Lastly, by a comparison of what has been in-
ferred with what can be directly observed to occur, an opportunity is
afforded for testing the correctness and practical adequacy of the two
preceding steps, and for the suggestion of necessary qualifications. It
will be observed that only one of these three steps namely, the middle
one -is strictly speaking deductive. The so-called deductive method in
its complete form is thus seen to be not an exclusively deductive method.
It may more accurately be described as a method which, whilst pre-
dominantly deductive, is still aided and controlled by induction. This
point will be further brought out in what follows, but it seems desirable
to call attention to it at the outset.
2. The application of the term hypothetical to economic sci-
ence. Political economy, in having recourse to the deductive method,
is usually described as essentially hypothetical in character. This de-
scription of the science needs, however, to be carefully explained and
guarded, as there is some danger of confusion of thought in regard to the
implications contained in it.
The Scope and Method of Political Economy/101
All laws of causation may be said to be hypothetical, in so far as
they merely assert that given causes will in the absence of counteract-
ing causes produce certain effects. As a matter of fact, in the instances
that actually occur of the operation of a given cause, counteracting causes
sometimes will and sometimes will not be present; and, therefore, laws
of causation are to be regarded as statements of tendencies only. It fol-
lows that all sciences of causation, and pre-eminently sciences employ-
ing the deductive method including political economy and astronomy
contain a hypothetical element.
The above may be expressed somewhat differently by saying that
the use of the deductive method in economics is solves, at a certain
stage, a process of abstraction, necessitating a frequent recurrence of
the qualification ceteris paribus. The abstraction is carried furthest in
reasonings where the motive of self-interest is supposed to operate un-
checked in a state of economic freedom; that is, in reasonings which
involve the conception of the economic man. But in all cases where
the deductive method is used, it is present more or less. For in the deduc-
tive investigation of the economic consequences of any particular cir-
cumstance or any particular change, the absence of interfering agencies
and of concurrent but independent changes is of necessity assumed. So
far as other changes themselves result from the one change or the one
circumstance specially under consideration, account must of course be
taken of them; but the distinguishing characteristic of the deductive
method consists in seeking, in the first instance, to effect a mental isola-
tion from the operation of all modifying forces that are not in some way
connected causally with the particular subject of enquiry. The distinc-
tion between dependent and independent changes, here indicated, is of [ Pobierz całość w formacie PDF ]