[205] Further, the Dogmatists take great pride in their systematic treatment of definitions, which they include in the logical division of their Philosophical System, as they call it. So come and let us now make a few observations on definitions.

Now while the Dogmatists hold that definitions have many uses, you will probably find that these fall under two main heads which, as they say, include all their necessary uses;

[206] for, as they explain, definitions are necessary in all cases either for apprehension ¹ or for instruction. If, then, we shall show that they are of use for neither of these purposes, we shall, I think, bring to naught all the labour so vainly spent on them by the Dogmatists.

[207] So then, without preliminary, if, on the one hand, the man who knows not the object of definition is unable to define the object unknown to him, while, on the other hand, the man who knows and proceeds to define has not apprehended the object from its definition but has put together his definition to fit the object already apprehended, then the definition is not necessary for the apprehension of objects. And since, if we propose to define absolutely all things, we shall define nothing, because of the regress ad infinitum; while if we allow that some things are apprehended even without definitions, we are declaring that definitions are not necessary for apprehension, seeing that we are able to apprehend all things apart from definitions in the same way as the undefined objects were apprehended,

[208] —then we shall either define absolutely nothing or we shall declare that definitions are not necessary.

And for these reasons they are not necessary for instruction either, as we shall discover. For just as the man who first perceived the object perceived it apart from definition, so likewise the man who receives instruction about it can be instructed without definition.

[209] Moreover, they judge the definitions by the objects defined and declare those definitions to be faulty which include any attributes not belonging either to all or to some of the objects defined. Hence, whenever one states that man is ‟a rational immortal animal” or ‟a rational mortal literary animal,” whereas no man is immortal, and some are not literary, such a definition they say is faulty.

[210] And it may be also that the definitions do not admit of judgement owing to the infinity of the particulars by which they ought to be judged; and consequently they will not convey apprehension and instruction regarding the objects whereby they are judged, which evidently have been known beforehand, if at all, and apprehended beforehand.

And how could it be other than absurd to assert that definitions are of use for apprehension or instruction or elucidation of any kind, when they involve us in such a fog of uncertainty?

[211] Thus, for instance, to take a ridiculous case, suppose that one wished to ask someone whether he had met a man riding a horse and leading a dog and put the question in this form—‟O rational mortal animal, receptive of intelligence and science, have you met with an animal capable of laughter, with broad nails and receptive of political science, ² a with his (posterior) hemispheres seated on a mortal animal capable of neighing, and leading a four-footed animal capable of barking?”—how would one be otherwise than ridiculous, in thus reducing the man to speechlessness concerning so familiar an object because of one’s definitions?

So then we must declare that, so far as we may judge by this, the definition is useless,

[212] whether it be described as ‟a statement which by a brief reminder brings us to a conception of the objects which underlie the terms,”—as is plain (is it not?) from what we have said just a moment ago,—or as ‟a statement declaratory of the essence,” ³ or what you like. For in fact, in their desire to propound a definition of the definition they plunge into an endless controversy which I now pass over, because of the plan of my present treatise, although it seems to overthrow definitions.

So what I have said about definitions is enough for the present.