[90] (1) Let us also beware of another great cause of confusion, which prevents the understanding from reflecting on itself. (2) Sometimes, while making no distinction between the imagination and the intellect, we think that what we more readily imagine is clearer to us; and also we think that what we imagine we understand. (3) Thus, we put first that which should be last: the true order of progression is reversed, and no legitimate conclusion is drawn. [91] [91e] (1) Now, in order at length to pass on to the second part of this method, I shall first set forth the object aimed at, and next the means for its attainment. (2) The object aimed at is the acquisition of clear and distinct ideas, such as are produced by the pure intellect, and not by chance physical motions. (3) In order that all ideas may be reduced to unity, we shall endeavor so to associate and arrange them that our mind may, as far as possible, reflect subjectively the reality of nature, both as a whole and as parts. [92] (1) As for the first point, it is necessary (as we have said) for our purpose that everything should be conceived, either solely through its essence, or through its proximate cause. (2) If the thing be self-existent, or, as is commonly said, the cause of itself, it must be understood through its essence only; if it be not self-existent, but requires a cause for its existence, it must be understood through its proximate cause. (3) For, in reality, the knowledge, [92f] of an effect is nothing else than the acquisition of more perfect knowledge of its cause. [93] (1) Therefore, we may never, while we are concerned with inquiries into actual things, draw any conclusion from abstractions; we shall be extremely careful not to confound that which is only in the understanding with that which is in the thing itself. (2) The best basis for drawing a conclusion will be either some particular affirmative essence, or a true and legitimate definition. (93:3) For the understanding cannot descend from universal axioms by themselves to particular things, since axioms are of infinite extent, and do not determine the understanding to contemplate one particular thing more than another. [94] (1) Thus the true method of discovery is to form thoughts from some given definition. (2) This process will be the more fruitful and easy in proportion as the thing given be better defined. (3) Wherefore, the cardinal point of all this second part of method consists in the knowledge of the conditions of good definition, and the means of finding them. (4) I will first treat of the conditions of definition.