WOKDS. 
ped, and the words of the new language be- 
come substitutes for, and suggest directly 
and immediately objects and ideas; also 
clusters of other words in the same lan- 
guage. 
In learning a new language it is nuicii 
easier to translate from it into the native 
one, than back again ; just as young chil- 
dren are much better able to understand 
the expressions of others, than to express 
their own conceptions. And the reason is 
the same in both cases. Young children 
learn at first to go from the words of others, 
and those who learn a new language from 
the words of that language, to the tilings 
signified. And the reverse of this, viz. to go 
from the things signified to the words, must 
be difficult for a time from the nature of 
successive associations. It is to be added 
here, that the nature and connections of 
the things signified, often determine the 
import of sentences though their gramma- 
tical analysis is not understood ; and that 
we suppose the person who attempts to 
translate from a new language is sufficiently 
expert in passing from ffie things signified 
to the corresponding words of his own lan- 
guage. The power of association is every 
where conspicuous in these remarks. 
3. It follows also from the foregoing 
reasoning, that persons who speak the same 
language cannot always mean the same 
things by the same words, but must some- 
times mistake each others meaning. This 
confusion and uncertainty arises from the 
different associations transferred upon the 
same words by the difference in the acci- 
dents and events of our lives. It is, how- 
ever, much more common in discourses con- 
cerning abstract matters, where the terms 
stand for collections of other terms, some- 
times at the pleasure of the speaker or 
writer, than in the common and necessary 
affairs of life; for here frequent use, and 
the constancy of the phenomena of nature, 
intended to be expressed by words, have 
rendered their sense determinate and cer- 
tain. Howev’er, it seems possible, and 
even not very difficult, for two truly can- 
did and intelligent persons to understand 
each other upon any subject. 
That we may enter more particularly 
into the causes of this confusion, and con- 
*equeiitly be the better enabled to prevent 
it, let us consider words according to the 
four classes above mentioned. 
Now, mistakes will happen in words of 
the first class, viz. such as have ideas only, 
where the persons have associated these 
words with different impressions. And 
the , method to rectify any mistake of this 
kind, is for each person to show with wliat 
actual impressions he lias associated the 
word in question. But mistakes here are 
not conmioD. ' 
In words of the second class, viz. such as 
have both ideas and definitions, it often 
happens that one person’s knowledge is 
much more full than another’s, and conse- 
quently his idea and definition much more 
extensive. This must cause a misapprehen- 
sion on one side, which yet may be easily 
rectified by recuriing to the definition. It 
happens also sometimes in words of this 
class, that a man’s ideas are not always suit- 
able to his definition ; that is, are not the 
same with those which the words of the 
definition would excite. If then this (>cr- 
son should pretend, or even design, to rea- 
son from his definition, and yet reason from 
his idea, misapprehension will arise in the 
hearer wlio supposes him to reason from his 
definition merely. 
In words of the third class, which have 
definitions only and no immediate ideas, 
mistakes generally arise through want of 
fixed definitions being mutually acknow- 
ledged and kept to. However, as imper- 
fect fluctuating, ideas that have little rela- 
tion to the definitions, are often apt to ad- 
here to tlie words of this class, mistakes 
must arise from this cause also. 
As to the words of the fourth class, or 
those which have neither ideas nor defini- 
tions, it is easy to ascertain their use by 
inserting them in sentences where import is 
known and acknowledged ; this being the 
method in which children learn to decypher 
them; so that mistakes could notarise in 
the words of this class did we use moderate 
care and cgndour. And, indeed, since chil- 
dren learn the uses of words most evidently 
without having any data, any fixed point at 
all, it is to be hoped tliat philosophers and 
candid persons may learn at least to under- 
staudone another with facility and certainty; 
and get to the very bottom of the connec- 
tion between words and ideas. 
4. When words have acquired any con- 
siderable power of exciting pleasant or pain- 
ful feelings, by being often associated with 
such things as do this, they m-ay transfer a 
part of their pleasures and pains upon indif- 
ferent tilings, by being at other times often 
associated with such. This is one of the 
principal sources of the several factitious 
pleasures and pains of human life. Tims, to 
give an instance from childhood; the word* 
