?6S MORAL PHILOSOPH Y. 
however it requires omnifcience to afcertain,) then only- 
do we derive the adlion from a Will good in itfelf, and 
diftinguifli fuch a Will from every other that is more or 
lefs affedred by foreign incentives. 
As to the queftion, What is an abfolutely good Will? 
it cannot at prefent be anfwered. It is like the queftion, 
What is an object? Living as we do in experience, we 
perfedtly underftand that objects correfponcl with our 
reprefentations; but it requires fome refledlion to com¬ 
prehend that the Categories produce the notion of 
Oeject (or objectivity) in our reprefentations. In the 
fame manner we know alfo how to diltinguifli a Will ab¬ 
solutely and in every reJpeSl good in itfelf, from every other 
mill which is only good for a particular purpofe. We 
eafily perceive alfo, that the conception of an abfoluteh) 
good Will agrees with that of a pure Will; fince both 
thefe conceptions are only determined negatively. Every 
other Will is accompanied by fomething empirical; name¬ 
ly, by the reprefentation of the pleafure to be obtained 
by it; and fuch a Will can only be called good fo far as 
it is adapted to this end. A Will good in itfelf is there¬ 
fore a Pure Will. 
We fliall not at prefent explain what is meant by an 
abfolutely good will; but proceed to treat of the concep¬ 
tion of Duty in which it is implied. 
We eafily perceive that it is not the fame thing whe¬ 
ther an aCtion be merely conformable to duty, or be done 
for the fake of Duty. An aCtion may certainly anfwer 
the Demands of Duty, and yet it may be a queftion 
whether it be done from a principle of Duty. It may 
have fprung from inclination, and yet be conformable to 
Duty. But deprive the mind thus prompted of the in¬ 
clination which ftimulates it, and the aCtion will not be 
performed. To relieve the diftrefted is a duty; but many 
minds are fo tenderly difpofcd, that they themfelves fuffer 
on witnefting the fufferings of others. Deprive them of this 
accidental difpofition, and the benevolent aClion will re¬ 
main unperformed. But let us conceive a man hardened 
by his own misfortunes, who trulls and perhaps requires 
that others fhould poll'efs the fame ftrength of mind to 
overcome or to endure the evils of life which he himfelf 
pofiefles. He does not fympathife in the fufferings of 
others, but let us imagine he aflifts them becaufe he con- 
fiders it his Duty. Here the aCtion arifes from Duty, and 
is not merely conformable to it. 
It is evident, therefore, that the conception of a pure 
Will, or a Will good in itfelf, coincides with that of a 
Will aCfing on the principle of Duty. 
It is alfo evident, that the conception of Duty agrees 
with that of the necefity of an action. It implies that an 
aCfioft muft be performed, without the lead regard to fen- 
fible incitements, and whether we are inclined to the 
aClion or not. This conception contains alfo a reference 
to a law which muft be obeyed. For, if aCtions are to 
be performed independently of all fenfual incitement, 
pure Reafon muft contain a law which prefcribes fuch 
aCtions. But, if the matter of the law (the reprefentation 
of what is to be obtained by fulfilling it) do not deter¬ 
mine the will, we muft afcribe the determination folely 
to its form; that is, to its legality. 
But this circumftance points out to us the Law itfelf. 
For, if there exifts a law of fuch a nature as to determine 
the Will generally, without particular motives for this 
or that individual; this law muft be valid for all ratio¬ 
nal beings. It is the Categories which give objec¬ 
tivity to our reprefentations, and eftabiilh them as Know¬ 
ledge. 
A Practical Rule, i. e. a pofition direCling that fome¬ 
thing fliall be done, we call an imperative. Thefe impe¬ 
ratives may be diftinguiftied like reprefentations, either 
as Jubjeliive or objective. Subjective imperatives prelcribe 
certain aCtions tor the attainment of particular ends. 
Thefe are hypothetical imperatives; fince it is only on 
the prefuppofition of thefe ends that they become prafiii- 
cal precepts. They are of two kinds; Imperatives of 
Adclrefs , and Imperatives of Prudence. Both are known 
analytically, becaufe, in the reprefentation of the com¬ 
plete Will to an end, that of the aCtion which leads to 
this end is implied. The former regard intentions which 
the fubjeCl may have; as, for inftance, to bifelt a fraight ' 
line: the latter refer to an intention which every one muft 
have; namely, his own- happincjs. As Prudence is re¬ 
quired in the Individual in order to choofe jthe beft 
means of rendering- himfelf happy, we may term thefe 
Imperatives of Prudence. 
But, if there be a Law which does not determine the 
will by fomething to be obtained by it, but merely by 
its own form, and which therefore muft be confidered by 
every man’s Reafon as a Law; this is an objective Law, 
and may be called the Categorical imperative. It 
cannot be known, however, analytically : for, as it con¬ 
tains no prefuppofition of an intention which determines 
the Will, no conception exifts from which the praCtical 
precaution could be deduced. The fynthetical judg¬ 
ments a priori upon which experience refts, are derived 
from the principles of the poftibility of experience; but 
from what principle can the poftibility of the Catego¬ 
rical imperative be derived? 
In order to know' whether a hypothetical imperative ap¬ 
plies to an individual, we muft be acquainted with his 
difpofition, and know that he has a certain end in view. 
But, as to the Categorical imperative, I know a 
priori that it is valid for every one’s Reafon. Such an 
objeltive law, therefore, refers to this very criterion ; 
namely, that reafon itfelf acknowledges it as a Law ; 
this is the criterion of univerfal validity. Confequently 
the Categorical Imperative is no other than this : Alt 
according to the Maxim of which thou canjl will that it be 
a univerfal law. 
All imperatives exprefs an ought. This ought, how¬ 
ever, in the hypothetical imperative, is itfelf hypotheti¬ 
cal, and refts on the prefuppofition of a certain inten¬ 
tion; but the ought of the Categorical Imperative is 
Categorical without reference to any intention whatever. 
Yet, as this ought indicates the relation of an objedlive 
Law to a Will, which has to overcome fubjeltive hin¬ 
drances in order to obey it; the word imperative does 
not apply to a will of itfelf, conformable to the Law, that 
is, to a facred will. 
The queftion, howthe Categorical Imperative is poftible? 
is the fame as, how pure Reafon can be practical? We 
do not enquire how pure Reafon can exercife coercion 
upon the Will, but only how it can make a demand, that 
is, how- a Categorical Imperative can be thought which 
cannot be known analytically. The conception of a Hy¬ 
pothetical Imperative is comprehenfible, becaufe it is 
accompanied by a prefuppofition ; for whoever wills an 
end, muft alfo will the adlion which is the means to this 
end. But in the Categorical Imperative there is no end, 
for the fake of which the adlion is willed : its poftibility 
therefore cannot be conceived in the fame way. To de¬ 
rive it from experience, and to pretend thence to fiiow 
its poftibility, would be a ufelefs undertaking, becaufe 
we cannot be certain, in any fingle inftance, whether an 
action has been the refult of the Categorical or of the 
Hypothetical imperative. The poftibility of the Categori¬ 
cal Imperative muft therefore be entirely derived a priori. 
In the Hypothetical Imperative the Will is connected 
with the Deed in luch a manner that the conception of 
the latter is confidered a conllituent part of the former: 
for to will an objedl completely, is alfo to will the adlion 
that produces the objedl. In the Categorical Imperative 
I alfo connedl the will with the adlion, but do not con- 
fider the latter as a neceftary part of it; for then the im¬ 
perative would be hypothetical. We may fay, every Hy¬ 
pothetical Imperative exprefles the neceflity of an adlion, 
in reference to an end required: the Categorical Impe¬ 
rative, on the other hand, exprefles {imply the neceftity 
of an adtion, without reference to an end. 
It is to be remarked, that the end of the adlion de¬ 
manded by a hypothetical imperative is always fubjeltive. 
I will for my own felf or fubjedt this or that end; confe- 
2 quently. 
