62 
how it comes to pass that a conditional act leads to an un- 
conditional.” Let it he observed that conditional and uncon- 
ditional are here used in two different senses. 
40. With respect to formal inference or deduction^ agree with 
Dr. Newman that the mind generally proceeds from premise 
to conclusion without a direct consciousness of the connecting 
link. It reasons, not secundum artem, but intuitively. Ai 
ficial systems, such as formal logic, are intended not to mcrease 
our mental powers, which are given us by nature, and are 
perfected by practice, but to guard us against the mistakes 
which we are liable. After having established a proposition by 
a course of reasoning, the knowledge of formal logic enable us 
to ascertain if we have fallen into error, and when, where, and 
how the error has been committed. . 
41. Dr. Newman points out with great force the various 
dangers to which deductive reasoning is liable. -But these are 
not so much in the process itself as in its accessories. But when 
he infers that the conclusion, at best, can be only probable, I am 
unable to discover how this follows from his premisses. Suie y 
the conclusions of geometry are characterized by certainty. 
As long as our reasonings embrace a simple conception only, 
as space or quantity, we use the same term unmodified i m 
meaning in our principles, premisses, and conclusions. B 
in all other subjects of thought, a number of conceptions, 
some of them indefinite, enter into the terms. Hence t e 
danger, in long courses of reason, of confusing the terms m 
the premisses and the conclusion. Against this the only thing 
which avails is the gift of a clear head. When reasoning con- 
fines itself to the use of symbols, its conclusions are free from 
some of this liability to error ; but the process is useless unless 
we can translate the symbols into notional or real conceptions. 
Dr. Newman maintains that for the purpose of avoiding erroi 
reasoners are obliged to contract their conceptions, so 
as to render them more and more inadequate to represent 
external realities; and consequently that we can only 
arrive at probable truth by a process of deductive reason- 
ing I cannot admit this in the unqualified manner in 
which Dr. Newman puts it. If I saw a triangular piece of 
ground, I should be quite sure that two of its sides were 
fonger than the third. I dispute not. that our processes of 
reasoning are liable to many imperfections, our judgments are 
imperfect; actual things have a vast complexity compared 
with our conceptions of them. No doubt it would be very 
desirable if our faculties were more perfect. But still, i± we 
use all the aid which scientific processes afford against the 
intrusion of error, and test them again and again, our de- 
