Kenny Anthony. Medieval Philosophy: A New History of Western Philosophy. Volume 2

Подождите немного. Документ загружается.

but it seems inadequate in two ways. First of all, we think with assent

whenever we call to mind a belief on any topic, whether religious or not.

Secondly, as Augustine himself often points out, at any moment there are

many things we believe even though we are not thinking about them at all.

A thought, that is to say a thinking (cogitatio), is a dateable event in our

mental life; belief (including the special kind of belief that is faith) is

something diVerent, a disposition rather than an episode.

When Augustine talks of faith, he is less concerned to expound its

epistemic status than to emphasize its nature as a gratuitous virtue, one

of the Pauline triad of faith, hope, and charity, infused in us by God. And

when he is most eloquent in expounding its role, his language once again

uses the metaphor of light, but in a manner that goes contrary to his

explanation of our knowledge of eternal truths. Thus, we read in The City of

God, ‘The human mind, the natural seat of reason and understanding, is

enfeebled by the darkening eVect of inveterate vice. It is too weak to bear,

let alone to embrace and enjoy, the changeless light. To be capable of such

bliss it needs daily medication and renovation. It must submit to be

cleansed by faith’ (DCD IX. 2).

Bonaventure on Illumin ation

The relation of faith to reason occupied a principal place in the epistemol-

ogy of Augustine’s successors in the high Middle Ages. St Bonaventure, like

Augustine, preferred Plato’s philosophy to that of Aristotle, but he believed

that even Plato’s greatest successors, Cicero and Plotinus, were grievously

in error about the true nature of human happiness. Without faith, no one

can learn the mystery of the Trinity or the supernatural fate that awaits

humans after death (I Sent. 3. 4). But, for Bonaventure, the philosopher,

however gifted, is in a position worse than that of mere ignorance: he is in

positive error about the most important things there are to know. ‘Philo-

sophical science is the way to other sciences; but he who wishes to stop

there, falls into darkness’ (De Donis, 3. 12).

A Christian philosopher, enlightened by the grace of faith, can make

good use of the arguments of philosophers to broaden his understanding

of saving truth. This Bonaventure himself does, oVering various proofs of

the existence of God: defective being implies perfect being, he argues,

KNOWLEDGE

162

dependent being implies independent being, mobile being implies immo-

bile being, and so on. These proofs he interprets, in Platonic manner, as

being mere stimuli to bring to full consciousness a knowledge of God’s

existence that is implanted by nature in the human mind (Itin., c. 1). He

oVers his own version of Anselm’s ontological argument to show that

nothing more than reXection on what is already in our minds is needed to

produce an explicit awareness of God’s existence.2 ReXection on the desire

for happiness, which every human being has, will show that it is a desire

that cannot be satisWed without possession of the supreme Good, which is

God (De Myst. Trin. 1. 17, conclusio).

For Bonaventure, the inborn notion of God was a special case. He did

not believe, in general, that our ideas were innate; he agreed with Aristotle

that the mind was initially a tabula rasa, and that even the most general

intellectual principles were only acquired subsequent to sense-experience

(II Sent. 24. 1. 2. 4). The notion of God was, uniquely, innate because the

mind itself was an image of God, a mirror in which God’s features could be

dimly seen. (De Myst. Trin. 1. 1). Somewhere between the inborn knowledge

of God and the acquired knowledge of intellectual principles stands our

knowledge of virtue: not an innate idea nor an abstraction from the senses,

but a natural capacity to tell right from wrong (I Sent. 17. 1).

The knowledge acquired from the changeable and perishable objects of

sense-perception is itself subject to doubt and error. If we are to acquire

stable certainties, we need assistance from the unchangeable truth which is

God. The Ideas in God’s mind, the ‘eternal reasons’, are not, in this life,

visible to us; but they exercise an invisible, causal, inXuence on our

thought. This is the divine illumination that enables us to grasp the stable

essences that underlie the X eeting phenomena of the world (Itin.2.9).

Aquinas on Concept-Formation

So, following a long line of predecessors, Bonaventure appeals to the

supernatural to explain how the human mind works. His contemporary,

Aquinas, rejects this approach. Aquinas does use the metaphor of light to

explain the working of the intellect: the agent intellect provides light,

2 See Ch. 9 below.

KNOWLEDGE

163

which turns potentially thinkable individual objects in the world into

actually thinkable objects in the mind. But Aquinas insists that the agent

intellect is a natural faculty within the individual huma n being, not—as in

the tradition of Avicenna and Averroes—a supernatural entity operating

on the mind from outside.3

In the Summa Theologiae la 79. 3–4 Aquinas states with great emphasis that

the agent intellect is something in the human soul. To be sure, there is an

intellect superior to the human intellect, namely the divine intellect; but

for human thought there needs to be a human power derived from that

superior intellect. God enlightens every man coming into the world, as

St John says, but only as the universal cause who gives the human soul its

characteristic powers (4 ad 1).

Aquinas sets out his attitude to theories such as Bonaventure’s in

question 84 of the First Part, where he asks whether the intellectual soul

knows material things ‘in their eternal natures’ (in rationibus aeternis). In the

Sed contra we are told:

Augustine says: If we both see that what you say is true, and we both see that what

I say is true, then where do we see that? Not I in you, nor you in me, but both of us

in that unalterable truth that is above our minds (Conf. XIII. 25. 35). But the

unalterable truth is in the eternal natures. Therefore the intellectual soul knows

all things in their eternal natures.

In his usual courteous style, Aquinas in the sequel rejects the doctrine of

divine illumination, but phrases his rejection in such a way as not to

criticize St Augustine more than is absolutely necessary.4

There is no doubt that Aquinas is not an empiricist: that is to say, he

denies that sensory experience is suYcient by itself for intellectual thought

(ST 1a 84. 6c). In addition to sense-experience, there is needed the action of

the agent intellect. But if Aquinas is not an empiricist, he is not an

illuminist either. The agent intellect by itself is insuYcient for the acquisi-

tion of intellectual knowledge. ‘Beside the intellectual light within us,

there is a need for thinkable species taken from outward things, if we are

to have knowledge of material things’ (ST 1a 84. 6c). The human intellect,

in this life, is a faculty for the understanding of material objects. Without

3 See Ch. 7 below.

4 I am here taking issue with the account in R. Pasnau, Thomas Aquinas on Human Nature

(Cambridge: Cambridge University Press, 2001), from which I have learnt much.

KNOWLEDGE

164

the senses no object would be given to us; without the agent intellect no

object would be thinkable. Thoughts without phantasms are empty;

phantasms without species are darkness to the mind.

The agent intellect is not, for Aquin as, something supernatural: it is part

of human nature. When he discusses the nature of teaching (ST 1a 111. 1),

Aquinas says: ‘There is within each human being a principle of knowledge,

namely the light of the agent intellect, by means of which from the

beginning there are known certain uni versal principles of all sciences.’

Aquinas compares the role of the agent intellect in teaching to the role of

our bodily nature in medicine. The doctor’s art imitates nature, which

heals a patient by temperature control, by digestion, and by the expulsion

of noxious matter. When a pupil is learning, the teacher is assisting him to

make use of his intellect’s natural light in order to progress to new

knowledge. The analogy is telling: the action of the agent intellect is no

more supernatural than the action of the digestive system. Both of them,

equally, are products of the creator God; but if being a creature of God

makes something supernatural, then the whole world is supernatural, and

the distinction between nature and supernature loses its point.

But does not God, as creator of the agent intellect, infuse a special

insight in a way in which he does not in creating other things? In the Summa

contra Gentiles 3. 47 Aquinas distinguishes between the likeness of God that is

present in every creature and the special likeness in the intellect because of

its capacity for the knowledge of truth. There are some truths on which all

human beings agree, the Wrst principles of speculative and practical

reasoning. It is the presence of these truths in the mind that makes the

mind an image of God. These truths are not inborn, nor are they acquired

from experience or induction. What is inborn is the faculty for recognizing

them when experience presents us with their instances.

The agent intellect is essentially a concept-forming capacity, which

operates upon phantasms. It turns the potentially thinkable data of sense-

experience into the actually thinkable species. The formation of concepts

involves the application of principles such as that of non-contradiction:

possession of the concept of X involves the ability to distinguish what is X

from what is not X. In that sense the agent intellect can be said to be aware

of such principles: but of course, by itself without any sensory input, such

an awareness contributes nothing to the knowledge of the essence of

material objects which is the intellect’s proper task in our present life.

KNOWLEDGE

165

It is the agent intellect itself that is the reXection, the mirroring, of the

uncreated light of the divine intell ect. When the agent intellect employs its

principles in forming concepts out of sense-experience, it needs no further

divine illumination, as Thomas emphasizes.

In all awareness of the truth, the human mind needs the divine

operation. But in the case of things known naturally it does not need

any new light, but only divine movement and direction (IBT 1. 1c).

St Thomas did, of course, believe that there was a supervenient, super-

natural divine illumination of the human mind: this was the grace that

produced faith in those fortunate enough to possess it. But he carefully

distinguishes this from the innate, natural light that is the agent intellect.

‘Whatever we understand and judge, we understand and judge in the light

of the primary truth, in so far as any light of our intellect—whether it be

the product of nature or of grace—is an impression of the primary truth’

(ST 1a 88. 3 ad 1).

Aquinas on Faith, Knowledge, and Science

A sharp distinction between truths knowable by natural light, and those

accessible only by the supernatural light of faith, is indeed one of St

Thomas’ principal contributions to medieval epistemology. Natural reason,

he believed, was capable of reaching a limited number of truths about God:

that he existed, was omniscient, omnipotent, benevolent, and so on.

Doctrines such as the Trinity and the Incarnation were known only by

revelation and unprovable by unaided reason. Faith, in the theological

sense, is belief in something on the word of God. Faith is diVerent from the

kind of belief in the existence of God which a successful philosophical proof

would produce. The faithful believer takes God’s word for many things,

but one cannot take God’s word for it that he exists. Belief in God, in this

sense, is not part of faith, but is presupposed by it. Thomas calls it a

‘preamble’ of faith.

Truths about God that are reached by natural reason are the province of

natural theology; the mysteries of faith are the subject of revealed the-

ology. But there is an ambiguity in the expression ‘unaided reason’. It may

mean that in arguing for its conclusions, natural theology rests only on

premisses derived from experience or reXection, and that it has no need to

KNOWLEDGE

166

call in aid any premisses derived from sacred texts or special revelation.

In anoth er sense it may mean that the natural theologian reaches his

conclusions without the aid of divine grace. When we talk about ‘unaided

reason’ in the Wrst sense, we are talking about the premisses from which

reason reaches its conclusion, and we are talking about logical relation-

ships. On the other hand, when we contrast unaided reason with the aid of

grace, we have moved from the realm of reasons to that of causes: we are

talking about the causal, not the logical, anteceden ts of the reasoning

process.

Even those truths that are in principle open to reason, such as the

existence of God and the immortality of the soul, must, according to

Aquinas, in practice be accepted by many people on authority. To establish

them by philosophical argument demands more intelligence, leisure, and

energy than can be expected from the majority of humankind. In setting

out the structure of natural theol ogy, St Thomas makes a distinction

between the beliefs of the learned and the belief s of the simple. The simple

believer need not be capable of following proofs such as the Five Ways

which, in the philosopher, produce (if successful) knowledge that God

exists. The simple believer only believes that there is a God. This belief is not

faith, for the reason given; it is a belief on human, not divine, authority.

But it is perfectly reasonable, provided that arguments for the existence of

God are available to the believing community, even if intelligible only to

the learned members of it (ScG 1. 3–6).

Aquinas’ distinction between faith and reason and between natural and

revealed theology marked a turning point in medieval epistemology.

Epistemology is the philosophical discipline that studies knowledge and

belief: what kinds of things we can know, and how we can know them;

what kinds of things we should believe, and why we should believe them.

Aquinas’ work sharpened the distinction between knowledge and belief;

more than any of his predecessors he emphasized that a Christian’s grasp of

the mystery of the Trinity was a matter not of knowledge or understand-

ing, but of faith. Within the realm of belief he made a distinction between

faith and opinion on the basis of degrees of certitude: faith, but not

opinion, involves a commitment to the truth of the proposition believed

parallel to that of knowledge. Corresponding to this diVerence of certitude,

there is a diVerence in the type of justiWcation: faith depends on supernat-

ural testimony, opinion rests on everyday evidence.

KNOWLEDGE

167

Having distinguished it from faith, Aquinas gives an account of know-

ledge (scientia) that is heavily inXuenced by the ideal of a deductive science

that Aristotle set out in his Posterior Analytics. Every truth that is capable of

being strictly known, he maintained, is a conclusion that can be reached by

syllogistic reasoning from self-evident premisses. There are some propos-

itions that have only to be understood in order to command assent: such

are the law of non-contradiction and other similar primary principles. The

ability to grasp and exercise these principles is the fundamental endow-

ment of the intellect: it is called intellectus in the strictest sense. The human

intellect also has the power to deduce conclusions from these self-evident

principles by syllogistic processes: this is called the ratio, or reasoning

faculty. First principles are related to the conclusions of reason as axioms

to theorems. The grasp of Wrst principles is called the habitus principiorum; the

knowledge of theorems deduced from them is the habitus scientiae (ST 1a 2ae.

57. 2).

St Thomas nowhere gives a list of the self-evident principles that are the

premisses of all scientiWc knowledge, nor does he try, like Spinoza, to

exhibit his own philosophical theses as conclusions from self-evident

axioms. But he tells us that the Wndings of any scientiWc discipline consti-

tute an ordered set of theorems in a deductive system whose axioms are

either theorems of a higher science or the self-evident principles them-

selves. A theor em may be provable in more than one system: that the earth

is round, for instance, can be shown both by the astronomer and by the

physicist. Sciences diVer from each other if they have diVerent formal

objects: the astronomer and the geometer, we might say, know about a

single material object, the sun, under two diVerent formal descriptions: qua

heavenly body or qua spherical solid. Conclusions derivable from diVerent

sciences will be deduced from syllogisms with diVerent middle terms. More

than one chain of reasoning may lead from the Wrst principles to a

particular theorem; but from any theorem at least one chain must lead

back to the axioms. The ideal of science thus set out seemed most obviously

realized by Euclid’s formalization of geometry.

Such a theory of scientia is clearly inadequate a s a general epistemology.

In the Wrst place, many of the things that we are common ly, and rightly,

said to know are not propositions of any deductive system. It may be

claimed that this point is simply an issue of translation: the Latin verb

‘scire’ and the noun ‘scientia’ are concerned not with knowledge but with

KNOWLEDGE

168

science. In fact, Aquinas often uses the verb as equivalent simply to ‘know’;

but it is true that he has a pair of terms, the verb ‘cognoscere’ and the noun

‘cognitio’ which have a much broader and less technical scope. These

words are used in a variety of contexts to refer to very diVerent things:

sense-perception as well as intellectual understanding; knowledge by de-

scription as well as knowledge by acquaintance; acquiring concepts as well

as making use of them. Careful attention to context is needed to Wnd the

appropriate translation in diVerent contexts. Sadly, some medievalists in

recent years have abandoned translation for transliterat ion, which not only

produces ugly English but leads to intellectual confusion. The pseudo-verb

‘cognize’ looks like an episode verb; and so all kinds of diVerent cognitive

states, activities, and acts are made to look as if they referred to a moment-

ary event of which there could be a mental snapshot. But it remains true

that if we are to look for a rewarding epistemology in Aquinas we should

examine his practice with ‘cognitio’ rather than his theory of scientia.

However, let us look for a moment at Aquinas’ theory as an account of

science, rather than as a general epistemology. It is important to realize

that it is not intended as an account of scientiWc method: we are not meant

to understand that the scientist starts with self-evide nt prin ciples and

proceeds to conclusions about the world by rolling out a priori deductions.

The procedure goes in the opposite direction: the scientist starts with a

phenomenon—an eclipse of the moon, say—and looks for the cause of it.

Finding the cause is the same thing as Wnding the middle term in a

syllogism which will have as its conclusion the occurrence of the eclipse.

The task of science is only completed when this syllogism, in turn, is traced

back, through other syllogisms, to arrive at Wrst principles. But the Wrst

principle thus arrived at forms the conclusion, not the starting point, of

the scientiWc inquiry.5 The chain of deduction is not the vehicle, but the

output, of the venture.

The serious problem with Aquinas’ theory is that it leaves quite unclear

what is the role of experience and experiment in science. True, ‘scientia’ is

broad enough to include mathematics and metaphysics; but it is clear from

Aquinas’ examples that his account is meant to cover disciplines such as

5 Aquinas clearly distinguishes the two procedures in ST 1a 79. 8, but rather confusingly he

calls the deductive process ‘inquiry’ and the process of inquiry ‘judgement’. But in his commen-

tary on the Posterior Analytics he makes clear that that work is concerned with ‘judgement’. See

Eleonore Stump, Aquinas (London: Routledge, 2003), 525, to which I am much indebted.

KNOWLEDGE

169

astronomy and medicine. Scientia, he tells us, concerns universal and

necessary truths: but how can the Xuctuating world we encounter in

sense-experience provi de any such truths? How can it be that—as Aquinas

himself says (ST 1a 101. 1)—human beings depend on the senses for the

acquisition of scientia?

The role that Aquinas assigns to the senses in the scientiWc enterprise

concerns the acquisition of concepts and the understanding of principles,

rather than the establishment of any contingent laws of nature. He

describes how the deliverances of the senses are necessary for the abstrac-

tion of universal concepts, and he shows how we grasp universal principles

by reXecting on particular instances of them. In each case he used the word

‘inductio’ to describe this process (CPA 1. 30, 2. 30). But the word, like so

many of Aquinas’ Latin technical terms, is a false friend. In inductio individ-

ual instances provide an illustration of, not an argument for, a proposition

which, once clearly understood, is self-evidently true. This is something

quite diVerent from induction as understood since the time of Bacon, in

which instances provide statistical support for a scientiWc generalization.

Since early modern times, epistemology has often taken the form of a

response to scepticism: what reasons do we have for relying on the

evidence of our senses, for accepting the existence of an external world,

for believing in the existence of other minds? Aquinas shows very little

interest in epistemology as thus understood. He accepts the general reli-

ability of our senses, regards the nature of material objects as the proper

object of the human intellect as we know it, a nd argues about the nature

Duns Scotus, as imagined by a

Wfteenth-century illuminator.

KNOWLEDGE

170

and number, rather than the existence, of human and superhuman minds.

In the intellectual climate of his time there was not a clear distinction to be

drawn between psychology and epistemology, that is to say between the

description and the vindication of the activities of our mental faculties.

Aquinas himself did not seek to develop such a distinction, in a manner

parallel to the way in which he sharpened the dichotomy between faith

and reason. A reader, therefore, who wishes to follow further his discussion

of the operation of the senses and the intellect should turn to the chapter

on philosophy of mind (Chapter 7).

The Epistemology of Duns Scotus

It is arguable that epistemology, as understood in modern times, makes its

Wrst appearance in the writings of Duns Scotus. This may seem a surprising

claim. At Wrst sight, Scotus is much further removed than Aquinas is from

any concern with scepticism. Whereas Aquinas thought that the proper

object of the intellect, in this life, was the nature of material objects, Scotus

believed the intellect was powerful enough to include all things in heaven

and earth, ranging over the full scale of being, inWnite as well as Wnite.

Moreover, while Aquinas believed that material individuals were the

subject of sensory rather than intellectual knowledge, Scotus was willing

to attribute to the intellect a direct knowledge of individuals in themselves

(Quodl. 13 p. 32). But while Scotus thus extended the scope of the intellect,

he diminished the degree of certainty it could attain.

A particular individual, Scotus argues in his commentary on the De

Anima (22. 3), is something capable of being grasped by the human intellect,

even in the present life when its faculties are dimmed by sin. If it were not,

we would never be able to attain knowledge of universals by induction,

and we would not be able to have a rational love for a human individual.

But our knowledge of individuals is obscure and incomplete. If two

individuals did not diVer at all in their sensory properties, the intellect

would not be able to tell one from the other, even though they would have

two diVerent haecceities and thus be two diVerent individuals. This obscur-

ity in our knowledge of individuals must carry with it also a clouding of

our knowledge of universals; for ‘it is impossible to abstract universals from

the singular without previous knowledge of the singular; for in this case

KNOWLEDGE

171