A man who was not a man (a eunuch) threw a stone that was not a stone (a pumice stone) at a bird that was not a bird (a bat) sitting on a twig that was not a twig (a reed).
3 Demonstrative Knowledge and Its Starting Points: Aristotle, Posterior Analytics*
Readers of the preceding extracts may feel inclined to agree with Plato that knowledge is superior to mere opinion, and that it needs to go beyond the particular to some more abstract level of rational justification; but they may also feel sceptical both about the notion of innate sources of knowledge (put forward in the Meno) and also about the sharp contrast (in the Republic) between the visible and the intelligible realms, which seems to downgrade the role of ordinary sensory information as a source of knowledge. The following extract from Aristotle puts pressure on both these Platonic ideas.
In his views on knowledge, Aristotle accepted the Platonic idea that what is known must have a certain stability, and immunity from change and fluctuation. Genuine scientific knowledge, it is asserted in the following extracts from the Posterior Analytics, is of that which ‘cannot be otherwise’; it concerns ‘eternal truths’, not particulars. Aristotle also lays out a normative framework for such knowledge: it must proceed from self-evident premises, or starting points, and it must advance by rigorous logical steps from premises to conclusion. Notice that there are two requirements here. The second, that the conclusion should follow from the premises, is the requirement of deductive validity in argument (‘deductive’ because the conclusions follow inevitably from the premises from which they are logically deduced). Aristotle, in his famous theory of the syllogism, had drawn up a procedure for testing the validity of arguments (a syllogism is a standard pattern of formal valid argument, such as ‘all As are B, all Bs are C, therefore all As are C’). But validity alone does not suffice to produce knowledge. The syllogism ‘All planets are stars, all stars are square, therefore all planets are square’ is perfectly valid – the conclusion follows inescapably from the premises – but it is worthless as a contribution to scientific knowledge, since the premises, or starting points of the argument, are false. So Aristotle insists that what is required for deductive knowledge, in addition to the logical validity of the relevant argument patterns, is that the starting points themselves should be self-evidently true.
Plato, as we have seen (extract 1), believed that the mind has innate knowledge of certain self-evident truths. But Aristotle questions the suggestion that the starting points for knowledge have to be innate, stressing instead the crucial role of sense-perception in providing the raw materials of knowledge. Knowledge must involve going beyond particular instances, and grasping universal truths, but this need not, according to Aristotle, imply the existence of abstract Forms over and above particular objects and groups of objects. Rather, knowledge develops naturally from sense-perception, since the human mind has the capacity for noticing and remembering general similarities which underlie the flux of sensory experience. This faculty for grasping the universal in the particular is called by Aristotle nous or ‘intuition’ (though he does not succeed in making it clear just how the results of intuition are supposed to have the self-evidence and certainty needed to serve as the starting points for scientific demonstration).
All teaching and all intellectual learning arises from pre-existing knowledge. This is evident if we look at all the examples. For the mathematical sciences are acquired in this way, as is each of the other arts. The same goes for arguments – both syllogistic and inductive, for both produce instruction by means of what we are already aware of …
We consider we have scientific knowledge or understanding of something … when- ever we consider we know that the cause of the item in question is its cause, and that it is not possible for it to be otherwise. So it is clear that having scientific knowledge is something of this sort. For both those who do and those who do not have knowledge think that they are in this situation, the latter merely believing it, while the former are actually in it. Hence scientific knowledge relates to that which cannot be otherwise.
We shall discuss later whether there is also another way of knowing. But we can state now that there is knowledge through demonstration. By demonstration I mean a scientific syllogism, and by this I mean one whose possession constitutes scientific knowledge.
If knowledge is indeed what we have just proposed, demonstrative knowledge must necessarily depend on premises which are true, primary, immediate, and better known than, and prior to and causes of, the conclusion … Without such conditions there can be syllogisms, but not be a demonstration, since it will not produce scientific knowledge …
Since it is impossible for that of which there is knowledge to be otherwise, that which is known through demonstrative knowledge must be necessary. Demonstrative knowledge is the knowledge we have in virtue of having a demonstration. A demonstration therefore is a syllogistic deduction from necessary premises …
It is evident that if the premises of a syllogism are universal, then the conclusion of such a demonstration – demonstration in the strict sense – must also be an eternal truth. So there can be no demonstration with respect to perishable things, nor any scientific knowledge of them strictly speaking but only in the accidental sense; for in such cases the attribute does not belong to the subject universally, but only at a particular time and in some respect … Demonstrations and knowledge of things that occur often, such as eclipses of the moon, do hold good permanently in so far as they relate to events of a certain kind; but in so far as they do not hold good permanently, they are [not universal but merely] particular. And so in other cases …
We have said that it is not possible to achieve scientific knowledge through demonstration unless we know the premises that are primary and immediate … With regard to these starting-points, it would be strange if we possessed them all along, since then we would possess knowledge superior to demonstration without being aware of it. But if, by contrast, we acquire them, and did not possess them earlier, how would we come to know them and learn them in the absence of any pre-existing knowledge? That is impossible, as we said earlier with regard to demonstration.1 Thus it is clearly impossible either for us to possess them all along, or for us to acquire them if we are ignorant and have no predisposition for knowledge. So we must already have some capacity …
This capacity evidently belongs to all animals, since they have an innate power of discernment – what we call sense-perception. Though it is innate, there are some animals in which the sense-impression persists, while in others it does not. For the latter group … there is no knowledge outside the act of perceiving; but others can retain something in the mind after perceiving it. And when this happens frequently, we get a difference arising as a result of the retention, some come to develop a logos2 and others do not.
Thus from sense-perception there arises memory; and when there is repeated memory of the same thing, there arises experience (for though there are many memories, they make up a single experience). And from experience – the whole universal now established in the mind (the one distinct from the many, whatever is one and the same in all the many instances) – there arises the starting-point of a skill, or of scientific knowledge (skill if it concerns what merely comes to be, scientific knowledge if it concerns what is).
Thus these dispositions are neither innate in a determinate form, nor on the other hand do they arise from other higher states of knowledge, but they come about from sense-perception. It happens just as in battle when there is a rout: if one man stands fast, another does, and then another, until a position of strength is reached. The mind is so constituted as to be capable of this.
Let us now restate the account we have just given, which was not very clear. When one of the undifferentiated particular things ‘stands fast’, a primitive universal is
