356 



ON HUMAN UNDERSTANDING. 



glanced at it already,* and may perhaps return to it hereafter, but I shall post- 

 pone it for the present, that we may hasten with due speed to the goal before 

 us. Allow me, however, before we quit it, to observe that words bear pre- 

 cisely the same relation to ideas that ideas do to objects; for as ideas are the 

 mere signs of objects, so words are the mere signs of ideas ; and hence that 

 every rule which applies to the variety, precision, and arrangement of our 

 ideas, applies with equal force to the variety, precision, and arrangement of 

 our words ; and that without a clear and determinate meaning to the latter, 

 we can no more have a clear and determinate apprehension of the former 

 than we can have of a person's features by a confused or unlike picture. And 

 hence the importance of attending to our vocabulary ; of minutely measuring 

 and weighing the terms we make use of, so as to adjust them exactly to the 

 measure and weight of our ideas, must be obvious at the first glance; as it 

 must be also that the more exact and copious a language is found, the more 

 clear and comprehensive must be the general knowledge of the nation to 

 which it belongs. 



But ideas and words, though the materials of which knowledge is con- 

 structed, and without which it cannot among mankind be constructed at all, 

 are no more knowledge itself than the bricks and mortar of a house are the 

 house itself. Both, as I have indeed hinted at already, must be collected in 

 sufficient abundance, compared with each other, duly assorted, arranged, and 

 united together, before the proper building can be produced ; and we have yet, 

 therefore, to contemplate the most important part of the subject before us, 

 and that to which the preceding parts are subservient — the general nature of 

 knowledge, its kinds, degrees, and reality. 



Knowledge may be defined the perception of truth, or, in the language of 

 Aristotle, the science of truth : and, consequently, he who acquires know- 

 ledge perceives or acquires truth. But what is truth 1 This is a question 

 which has been asked for ages: the particular answer, however, must neces- 

 sarily depend upon the particular subject to which it refers. We are now 

 considering general truth, which may be defined the connexion and agreement,^ 

 or repugnancy and disagreement, of our ideas. 



This definition requires some attention ; but when it is thoroughly compre- 

 hended, it will be found to apply to truths of every kind, in the arts, physics, 

 and morals, as well as in metaphysics ; for the law of adjustment, of con- 

 nexion and disconnexion, of congruity and incongruity, it refers to, is a 

 universal law or constitution of nature, and hence must hold equally every 

 where. Thus, in a building, where the diff'erent parts of which it consists per- 

 fectly agree, the lines accurately correspond, and the dependencies fit and are 

 proportioned to each other, every part is true to every part, and the whole is 

 true to itself. 



So in working a mathematical problem, or determining a fact from cir- 

 cumstantial evidence, every separate link or idea that constitutes a part of 

 the general chain, must have its proper connexion or agreement with the link 

 or idea that lies next to it, as well above as below : for it is these connexions or 

 agreements between one idea and another that constitute the proofs, and a 

 failure in anyone destroys our knowledge upon the subject; or, in other 

 words, prevents us from perceiving its truth. 



It sometimes happens that we are able to discover at once this agreement 

 or disagreement, this connexion or repugnancy, in the ideas that are presented 

 to us ; and in such case our knowledge is instantaneous, and constitutes what 

 we call INTUITION or intuitive knowledge. But it happens far more generally 

 that the agreement or disagreement is by no means obvious ; and we are 

 obliged, as in the case of circumstantial evidence, to look out for some inter- 

 mediate idea, which the schools denominate a rnedius terminus, by which the 

 separate ideas may be united. To make this research is the peculiar province 

 of the discursive faculty of reason; and hence the information thus obtained 

 is called rational knowledge. 



Let us take a brief view of both these. When I atfirm that white is not 



* Series ii Lecture viii. ix. x. 



