METAP 
the laws of their pure ufe according to principles. It is 
certainly difficult, and requires fome refolution in the 
reader, to become acquainted with a fyltem which affumes 
nothing as a foundation but reafon itfelf, and which 
mult therefore endeavour to develope knowledge from 
its original feeds, without refting upon any faCt of ex¬ 
perience. Thefe Prolegomena however are merely pre¬ 
paratory exercifes, rather denoting what is to be done in 
order to ereCt a fcience, fuppofing it poffible, than enter¬ 
ing upon the fcience itfelf. They mud: therefore red: 
upon lbmething already known, from which we may pro¬ 
ceed with confidence, in afeending to its fource, at pre- 
fent concealed, but the difeovery of which we expert not 
only to explain to us wffiat we already know, but at the 
fame time to afford other knowledge ariling from the 
fame fource. The method to be adopted in thefe Prole¬ 
gomena, as preparatory to future Metaphylics, will con- 
fequently be Analytical. 
Though we cannot boaft of the exiftence of Metaphy¬ 
sics as a Science, yet it is Hill a fart that we really poffefs 
fome pure Jynthetical knowledge it priori : namely, pure 
Mathematics, and pure Natural Philofophy; for both 
thefe fciences contain pofitions which mud: always be ad¬ 
mitted as apodirtically certain either by mere reafon or 
by the univerlal content of experience, and yet at the 
fame time as pofitions independent of experience. 
How is Knowledge from pure Reafon poffible ? 
We have already fhown the difference between Ana¬ 
lytical and Synthetical Judgments. The podibility of 
Analytical Pojitions is eafily conceived, becaufe it grounds 
itfelf entirely upon the principle of contradiction. The 
poffibility of Synthetical pofitions a pojleriori, that is, of 
fuch as are derived from experience, needs no particular 
explanation; for experience itfelf is nothing but a continual 
Jynthefzing of apprehenfions. There only remains there¬ 
fore Synthetical pojitions a priori, whofe poffibility requires 
to be invelligated, fince it muft red: upon fome principle 
different from that of contradiction. 
As the method w'hich we at prefent purfue is analyti¬ 
cal, we fhall begin by obferving that fynthetical pure ra¬ 
tional knowdedge, really exifts. We need not therefore 
prove its poffibility, but w r e muft invefligate the ground 
of this poffibility: that is, enquire how > it is poffible; in 
order, from the principle of its poffibility, to determine, 
not only the conditions, but the extent and limits, of the 
ufe of this knowledge. The true problem, exprefied 
with precifion, and upon which every thing depends, is 
therefore, 
How are Synthetical Pofitions a priori poffible ? 
I have above exprefied this pofition fomewhat differ¬ 
ently ; namely, as a queftion regarding the Knowledge of 
pure Reafon, which I was enabled to do without injury 
to our prefent inveftigation, becaufe, as we have here 
only to do with Metaphysics and their fource, we mud: 
always bear in mind that, whenever we fpeak of Know¬ 
ledge from pure reafon, we do not mean Analytical, but 
always Synthetical, Knowledge. 
[Remark .—As Knowledge advances, it mud happen 
that many expreffions which have exifted from the in¬ 
fancy of fcience, and have become claffical, will be found 
inadequate; and that a new and more appropriate ufe of 
terms will at firft be in danger of being confounded with 
the old, and lb mifunderftood. The Analytical -method, 
as oppofed to the Synthetical, is fomething quite different 
from a whole of analytical pojitions ; it only means that we 
fet out from what we feek as if it were given, and afeend 
to the conditions under which alone it is poffible. In 
this method we frequently make ufe of nothing but /i/w- 
theticalpofitions, as the mathematical analyfis fully evinces; 
and it might be termed the Regreffive method, as diftin- 
guifhed from the Synthetical or Progreffive. The term 
Analytic is applied alfo to the chief part of Logic ; and 
there it dignifies the Logic of Truth, in oppoiition to 
H Y S I C S. 215 
Dialectic. Whether the parts of Knowledge with which 
it is concerned are Analytical or Synthetical, is not taken 
into confideration.] 
Upon the folution of this queftion, the very exiftence 
of Metaphyfics muft either ftand or fall. Whatever may 
be aflerted, and whatever arguments may be adduced in 
Metaphyfics, unlefs this queftion be fatisfaftorily an- 
fwered, the world may juftly fay that, Philofophy is a 
vain and empty pretenfion to wifdom. 
The anfwer to this great queftion is as indifpenfable as 
it is difficult. The principal reafon, why no attempt has 
hitherto been made to anlwer it is, that the poffibility of 
fuch a queftion was never before conceived. Indeed a 
fatisfartory anfwer to this one queftion, would have re¬ 
quired a far more laborious ftudy, than the moft exten- 
five work on Metaphyfics ever did, though on its firft ap¬ 
pearance it might promife immortality to its author. 
Every intelligent reader, who carefuily contemplates this 
problem in all its confequences, will be led at firft, by its 
difficulty, to confider it as infolvable ; and indeed, but for 
the fart that we have fuch pure fynthetical knowledge 
a priori, as a queftion utterly without meaning. This 
artually happened to Hume, though he did not reprefent 
it to himfelf univerfally, as is here done, and indeed muft 
be done, if the anfwer is to be decifive for Metaphyfics 
generally. How is it poffible, faid this penetrating man, 
that, when a conception is given to me, I can go beyond 
it, and connert another conception with it, which forms 
no part of the former, and this in fuch a manner, as if 
the one neceffarily belonged to the other (Caufe and Effert) 
The difficulty of conceiving this, led him to confider it as 
impoffible ; from whence he concluded that Experience 
alone furnifhed fuch connections, and that their fuppofed 
neceffity, or, which is the fame, our knowledge of them 
a priori, is nothing but a long habit of obferving them, 
and therefore a miftaking of j'ubjedlive neceffity for objective. 
If the reader complains of the difficulty and trouble 
that I give him in the folution of this Problem, let him 
make the attempt himfelf to folve it in an ealier manner. 
Perhaps he will then confider himfelf obliged to one who 
has undertaken a work of fuch deep inveftigation, and be 
rather furprifed at the facility with which, confidering the 
nature of the fubjert, it has been poffible to accomplilh it. 
It has however coft me many years labour to folve this 
problem in its whole univerfality, (taking the W'ord in 
the fenfie in which mathematicians ufe it; that is, as im¬ 
plying a fufficiency for all cafes,) and to reprefent it in 
an Analytical form, as is here done. 
Thole, therefore, w'ho cannot fatisfartorily anfwer the 
Queftion, How is Synthetical Knowledge a priori poffible l 
muft not for the future pretend to the title of Metaphyfi- 
cians. But, if they are content to carry on their Philo¬ 
fophy, not as a fcience, but as a mere art of perfuafion,. 
with the view of aiding common fenfe, we cannot in juft 
tice deny,them this occupation. Then, however, they 
will only modeftly aflert the reajbnabletiejs of their belief; 
acknowledging that they are not permitted even to indulge 
a conjecture about that which lies beyond the bounds of 
experience, much lels to pretend to know any thing of it; 
they will merely as it were advife the adoption of what 
they may conceive poffible, and even indifpenfable, for the 
direction of the judgment and the will in common life ; 
that is to fay, not for the purpofes of fpeculation, for this 
they muft refign; but merely as a guide for practice. 
Thus alone will they be entitled to the appellation of ufe- 
ful and wife men, giving up that of Metaphyficians ; for 
the latter aim at being lpeculative philofophers ; and in 
fpeculations concerning judgments a priori, fhallow pro¬ 
babilities, and conjectures, amount to nothing. 
It now appears, that Transcendental Philosophy 
which muft neceffarily precede all Metaphyfics, is nothing 
but the complete folution of the propofed queftion given 
at large, and arranged in lyftematicai order; and that, orior 
to our day, there was no fuch thing as Tranfcendental Phi¬ 
lofophy. Whatever bore that name was only a part of Me¬ 
taphylics 3 
