Fine on 'Knowledge and Belief in <i>Republic</i> 5-7': the argument about Book 5

Gail Fine on Knowledge and Belief in Republic 5-7: Fine's interpretation of the end of Book 5

Summary of "Knowledge and Belief in Republic 5-7" as found in Gail Fine, ed. Plato I as well as in S. Everson, ed., Cambridge Companions to Ancient Thought I: Epistemology (CUP 1990), 85-115.

The Meno says that knowledge is true belief tied down by an aitias logismos, a 'causal explanation.' It also says that true belief can become knowledge, and it seems the same thing can be an object of knowledge and belief.

In the Republic, however, most people think that Plato has changed his mind. They hold what is called the 'TWO WORLDS' hypothesis:

there is no knowledge of sensibles, only of forms, and
no belief about forms, only about sensibles. The two worlds hypothesis will be abbreviated TW hereafter.

Gail rejects TW.

Some consequences of TW:
1. objects of belief and knowledge are disjoint.
2. This amounts to a rejection of the Meno's position. True beliefs cannot become knowledge, for they are mutually exclusive.
3. Scepticism about knowledge of the sensible world: it is impossible.

Those consequences are surprising, since philosophers, who know Forms, are supposed to rule in the sensible world. What good would knowledge of Forms do them if TW is true?
What is more, TW is contradicted by passages in the R.: 506c has beliefs about a Form. At 520c the philosopher returning to the cave KNOWS things there in the cave. To be fair, Plato could be speaking loosely in those passages, but they give us pause.

Gail thinks that:

All Knowledge requires knowledge of Forms, but it is not restricted to Forms.
Those who are restricted to sensibles will never have Knowledge.
Plato is a coherentist, not a foundationalist about justification. All beliefs, to be known, must be justified by other beliefs: none are self-evident or self-justified.
Knowledge is always articulate: it does not consist of a vision or acquaintance, but always involves the ability to explain the object of knowledge.

In R. 5, 476c and following, Plato is trying to justify to the sight-lovers why THEY should believe in Forms. The term 'sight-lovers' refers to people who hold that the sensible world is the source of Knowledge. Plato wants to argue against that view. The "Dialectical Requirement" (DR) requires that he use only terms which sight-lovers understand and claims to which they agree in his argument.

Arguments:

I. the objects of knowledge and belief

1. Knowledge is set over what is (477a9-10).
2. Belief is set over what is and is not.

A key point in how we interpret I.1 and I.2: 'is' can mean:

a) exists
b) is F (where 'F' is some predicate to be determined by context: e.g. is true, is good, is bad, is dastardly, etc.)
c) is true

1 and 2 thus have three possible versions, or some combination if Plato has a different conceptual apparatus. Fine wants to determine which version is the right version for the rest of the argument.

If we use a) exists, we find that 1) and 2) identify sets of objects of knowledge and belief: knowledge is set over existent things and belief is set over both existent and non-existent things. If we use b) is F, 1) and 2) identify the objects of knowledge as things that have some predicate F, and the objects of belief as things that do and don't have some predicate F. If we use c) is true, we find that the objects of knowledge are things that are true and objects of belief are things that both are true and aren't true.

Now, on c, the objects are propositions (usually represented by 'P,' 'Q,' 'R,' etc.), for propositions are what can be true or not true. That should be acceptable to the sight-lovers, because it seems right to say that one can know only true propositions, but one can believe propositions that are both true and false (albeit without being aware that the false ones are false). That seems like an everyday way of speaking about belief and knowledge.

But if we use a or b, we violate DR, for the sight-lovers are not likely to want to separate the objects of knowledge and belief without some further argument. a and b assume something that cannot be assumed if one is following DR:

a) includes a violation of DR: it would amount to the claim that we cannot know someone who is dead, or anything that does not exist, like Santa Claus. A sight-lover might not want to accept that.
b) includes a violation of DR as well: it introduces the notion of both being and not being F: is that half-existence? Why can't I know that this pencil is both equal to other pencils and unequal to the Washington Monument?What sense of being F and not being F applies?

c satisfies DR! and Plato says it elsewhere: Meno 98a and Grg. 454d6-7. Therefore, argues Fine, we should accept c as the right version.

BUT, 2c can mean either of the following:

2ci. Every proposition that can be believed is both true and false.
2cii. The set of propositions that can be believed includes some truths and some falsehoods.

2ci is controversial: that a proposition be both true AND false is not readily acceptable. So it violates DR.

2cii is not controversial.

Our best candidates are 1c and 2cii.

1c. Knowledge is set over what is true (477a9-10)
2cii. The set of propositions that can be believed includes some truths and some falsehoods.

If P is known, it is true.
If P is believed, it is true or false.

II. Capacities

477b-478b: knowledge and belief are different capacities. Capacities are different because

a) they have different scopes, they are 'set over' different things.
b) they do different work.

So all capacities that have the same scope and do the same work are one single capacity.
Those which have different scopes and/or do different work are different capacities.Remember the principle in the Republic used to detect the existence of parts in the soul: this capacity-definition-move is a similar argumentative move.

xx) knowledge and belief satisfy b differently: they do different work.
xy) thus they are different capacities.

The different work they do is that knowledge does infallible work, while belief does fallible work. Knowledge's being infallible means that "if you know, you cannot be wrong."

Plato may say that because of xx and xy, knowledge and belief also have different scopes.
BUT that would not be obvious: butchery and husbandry are different capacities but set over the same scope (domestic animals).

Fine suggests that Plato can be right if he is saying that knowledge has true propositions as its scope: it is "set over" true propositions. Belief, however, is set over true AND false propositions. Thus the scope of knowledge and belief is different, but not disjoint: it overlaps. The scope of knowledge is a subset of the scope of belief.

On Fine's interpretation, Plato is warranted in concluding from the fact that knowledge and belief do different work that they have different scopes.

SO FAR, Fine has managed to come up with a noncontroversial reading that satisfies DR and does not commit Plato to TW.

III. Known v. believed

"What is known is not the same as what is believed."

That claim may be a problem for Fine: it looks like TW.

It may mean that the objects of knowledge and belief are different or, as Fine thinks it does, merely that the quality "known" is different from the quality "believed." Fine is clearly working here to suggest an unexpected but not implausible meaning of that sentence.

Sometimes, one gets the feeling that scholars are extending a theory along the lines the author might have extended it, but maybe the author did not think that far. Fine operates by figuring out how to make various claims consistent and how to "close" Plato's theory. But Fine is scrupulously honest: she herself identifies two points at which Plato seems to have violated DR (the one-over-many assumption below and the priority of definition claim below).

IV. 479a and following:

3) Each of the many Sensibles both is (F) and is not (F).

Use "beautiful" as an example of F. Because the sight-lovers accept 3, they deny that beauty is one thing. They think there are many beauties.

Plato, however, accepts the One-over-Many assumption: that is to say, he thinks beauty is one thing that somehow applies to or causes or explains or is the model for many things: the relation between the one and the many is a problematic interface. What is more, Fine wants to take the One-over-Many assumption and "build it into" the argument to make it work. (QUESTION: is that a violation of DR? It sure looks like it, as Fine footnote 13 admits) The One-over-Many assumption is that there is only one property F, in virtue of which all and only F things are F.

4) The sight-lovers' beliefs (nomima) about the many Fs are and are not (479d3-5).
5) Therefore, the sight-lovers have belief, not knowledge, about the many Fs (479e1-5)

Fine claims that in 3, Plato is using the predicative sense of 'is' on which 'is' links a subject with a predicate (e.g. 'this rose is beautiful'), but in 4, Plato has shifted back to the veridical meaning of 'is' on which 'is' means 'is true.' She makes that claim because 4 makes best sense on the veridical reading: the sight-lovers beliefs are both true and not true, because that is what they have agreed that beliefs are, both true and not true.

Plato thinks that knowledge entails truth. Think of the following situation: The sight-lovers claim that beauty in a particular painting is defined as 'bright color.' Plato, however, points out that bright color by itself does not necessarily entail beauty: some brightly colored things are ugly.

Thus the sight-lovers' attempt to have knowledge of beauty (or any sensible property) is doomed to failure, because any property-type-W they choose to define a property-X will have some tokens that are not-property-X.

Plato believes that one can know something about x only if one knows what x is. One must have a definition before one can have knowledge. That too seems to violate DR, as Fine footnote 15 admits. There are nuances here that Fine does not fully explore: there is a lot to be said about "know" and "definition" and "having" that might make things better or worse for Fine or Plato.

So sightlovers (who are confined to sensible things) have beliefs, but not knowledge. Some of their beliefs are true, but they cannot be knowledge since they necessarily lack a correct aitias logismos (causal explanation).

So all beliefs, which are confined to the predicative-is world, will be separated from the veridical-is world, but not because there are Two Worlds, but rather because any account phrased in terms of things that are (predicative) cannot be knowledge.

V. 479e7-480a5 knowledge

6. Knowledge is possible.
7. There must, then, be objects of knowledge (sensible objects cannot be known, as the argument in section IV above shows, so the object of knowledge must be non-sensible).
8. Therefore, there are Forms ("Form" is the label for any non-sensible object of knowledge here: it is nothing more than whatever the non-sensible objects of knowledge are).
9. Those who know the Forms have knowledge: those who are restricted to the many Fs have belief (at best).
10 (to be refined below). Therefore knowledge is set over Forms, and belief is set over the many Fs.

Number 6 violates DR. Sight-lovers who accept the argument so far might well say that it means that knowledge is impossible and we are stuck with beliefs-only, but many might not, so let's let that slide.

Note that nothing here requires that there be Two Worlds. In other words, nothing requires us to say that the sensible world does not exist, while the Forms do.

There is, Fine thinks, a tacit claim that knowledge requires that nonsensibles exist, but it is not claimed that they are all that exists. Nor is it claimed that knowledge is restricted to what exists.

The kind of Forms that 8 is warranted in positing are unitary, non-sensible, and explanatory. Other qualities of Forms are not warranted by this argument.

Fine says that 10 is worrisome because it appears to say that knowledge and belief have disjoint objects. But that is what Fine took so long to reject earlier. She can reject it now for similar reasons and claim that 10 is really shorthand for a much longer claim:

10 (refined) all knowledge requires knowledge of Forms, and if one is completely restricted to sensibles, the most one can achieve is belief. Knowledge of sensibles may be possible, but it has as a prerequisite knowledge of Forms.

Advantages of Fine's interpretation:

It starts out satisfying DR.
It is valid.
It involves no equivocative 'is.'

Important results of Fine's interpretation:

Two Worlds are not necessary.
It is possible to know sensibles, although Plato nowhere says that explicitly.

Disadvantages:

DR is violated at 2, possibly 3, important points in the argument.

Republic 6-7

Gail Fine argues that Plato believes in a coherentist theory of knowledge. Coherentism and foundationalism are two ways to respond to the problem raised by the claim that knowledge is based upon knowledge. In other words, if I know P, I know it because of Q (which I know), and I know Q because of R (which I know), and I know R because of . . . . You can see the regress going on infinitely.

Foundationalism says that the regress stops at a foundation: "things that are self-evident" is one candidate for the foundation. "Self-evidence" is a problematic notion, however.

Coherentism, on the other hand, says that somewhere in the chain of "I know x because of y, and I know y because of z, and ...," you might know R because of something earlier in the chain! And coherentism claims that that sort of a "circle" is not necessarily vicious. The circle, or web, that is sufficiently broad and explanatory counts as knowledge. There must be mutual support among the consistent beliefs. In other words, it is not simply a circle, but many small and large circles that form a web or even a tangle that is sufficient to support itself.

What role does Truth play in both? Neither actually assures that Truth will be present. Gail Fine does not seem to do much with this point, except perhaps at the very end where she suggests that modern epistemology has moved away from certainty and truth.

Perhaps what is being discussed is something we might call 'understanding' rather than 'knowledge' (but don't worry too much about the name: concentrate on the distinction). The distinction is that 'Understanding' something does not involve a claim to access some quality that is called 'truth.' 'Knowing' does, however, seem to involve access whatever 'truth' is.

The claim that knowledge is based on knowledge, a cornerstone of Fine's coherentist version of Platonic epistemology, will be referred to as KBK.

Plato also thinks that part of knowing is being able to give a 'reasoned account,' an (aitias) logismos (Phaedo 76b, Republic 531e, Meno 98a). So knowledge requires a logismos, a 'reasoned account.' We'll call that requirement KL.

The two upper stages of the divided line (509D+) (the stages will be referred to as L1, L2, L3, and L4, with L4 being the most advanced stage) involve knowledge.

At L3, the stage of 'Thought,' we can satisfy KL by giving accounts of what we know. But our knowledge is based on hypotheses, and so may not satisfy KBK. Gail Fine claims, however, that at L3, we can give accounts of the hypotheses in terms of the conclusions, and accounts of the conclusions in terms of the hypotheses, and that is sufficient to satisfy KBK.

In other words, within any given discipline, the basic hypotheses are justified by their explanatory power relative to conclusions, and the conclusions are justified by the hypotheses. As long as we remain within one discipline, we are at L3, and to the uncharitable, we will appear to be in a vicious circle.

At L4, however, we are able to see the entire structure of all knowledge, and all hypotheses are justified in terms of how they fit into the whole structure of knowledge. All conclusions are justified in terms of hypotheses. Thus L4 is, in a way, just a larger version of L3. It's coherence and breadth is what makes it knowledge rather than a vicious circle.

The Sun
The Sun is an image Plato uses at 509B+. In the image, there are 2 sides. On one side are people using their sight who are either in the dark or out in the sunlight. On the other side are people using the capacities of their soul to become aware of things: they either use Forms or do not. As the sun is to the sense of sight, so the Form of the Good is to the soul's capacities to know and understand.

Sight elements (small 's')

s1 Sight looks at visible objects in the dark, unilluminated by the Form of the good or other forms, and so has belief.
s2 Sight looks at visible objects illuminated by the sun.

Soul elements (capital 'S' label)

S1 The soul is aware only of sensibles, unilluminated by the Form of the good or by other Forms, and so has belief.

S2 The soul considers knowable objects illuminated by the Form of the good, and so has (the best sort of) knowledge.

Note that Plato does not explicitly claim and discuss the idea that there is knowledge of sensibles or belief about Forms. It is crucial to Gail Fine's interpretation that these not be excluded, however, for remember that she is rejecting the Two Worlds hypothesis (TW). In support of her interpretation, she cites 506c, where Socrates says he has no knowledge of, only BELIEF about, the Form of the good. She cites 520c to claim that the philosopher has KNOWLEDGE about the sensibles. The Sun image adds to what is found at the end of Book 5 the claim that the best sort of knowledge requires knowing the Form of the good (505a, 508a5).

The Form of the good is greater than other knowables (504c9-e3, 509b6-10), although it is not a being (509b9-10).

Gail Fine says the Form of the good is the teleological structure of things (that is, the structure of things that leads to the goal of all the things): individual Forms are its parts, and particular sensible objects instantiate it. Think of a house: the form of the house is not any additional part of the house: it is the ordering of all the parts to make a house.

Fine calls her interpretation of the Form of the good the "best explanation" of what Plato says: in other words, she is acknowledging that Plato nowhere explicitly endorses this as his account of the Form of the good. Rather, it is simply the best way to explain what he does say about the Form of the good. She does not tie this interpretation of the Form of the good to Plato's text, except in that she cites the Timaeus 27dff, where the demiurge, being good, wants the world to be good, and tries to instantiate the Form of the good as much as possible.

The Line briefly outlined:

L1 is 'imagination,' and is explained in terms of images of physical objects, but there is nothing about L1 that confines it exclusively to images:
L2 is confidence, and is explained in terms of physical objects.
L3 is thought, and is explained in terms of using hypotheses to arrive at conclusions using images of the Forms.
L4 is knowledge/understanding and uses dialectic (511b, 533c) to remove the hypotheses (533c8) by explaining them in terms of an unhypothetical first principle (the Form of the good). At L4, reasoning is performed with the Forms themselves, not images of them.

The Cave briefly outlined:

C1: shadows of artificial objects are illuminated by a fire internal to the cave: one cannot tell the difference between shadows and objects.
C2: shadows of artificial objects are illuminated by a fire internal to the cave: shadows and objects can be discriminated systematically.
C3: outside of the cave: one sees the shadows of objects, then the objects themselves.
C4: outside of the cave: one sees the sun (and everything in its light).

Note that the Line and the Sun describe static conditions: cognitive states are described but there is no explanation of movement from one to another. In the Cave, however, the situation is dynamic: one can move from one state to another.

L1

To recap: Fine is rejecting TW, according to which the way of distinguishing L1 from L2 from L3 from L4 is by the OBJECTS that each state involves (TW holds that the objects are sensible images, sensible objects, Forms, Form of the good; and that L1 is confined to images, L2 to physical objects, etc.) Gail Fine rejects that objects analysis in favor of what she calls a "contents" analysis, but might better be called a "process" analysis because it identifies processes characteristic of the parts of the line. L1, L2, L3, and L4 are distinguished by different reasoning in them. She needs to show that Plato does not confine each stage to particular objects.

Most of us are at L1, as are most people of the ideal city (515a5, 517d4-e2, 520c1-d1): thus it cannot be confined to images, because we are not plausibly confined to images. What is more important, the released person is confused by the objects she sees: in other words, she is still at L1, even though she is looking at an object, not an image. So it's not necessary to be looking at images to be at L1: at L1, one may also see other things, even Forms, but one is seeing them as if they were images: one is not distinguishing them. The philosopher returning KNOWS the images (520c): he is not at L1, however. So it's not sufficient to be looking at an image to be at L1. People at other stages can also look at images.

What IS characteristic of L1 is an inability to discriminate between objects and images. We are at L1 morally (520c7-d1, 517d-e).

L2

L2 is the first stage of the elenchus, when we learn that we are ignorant and to distinguish between images and objects. Beliefs become more reliable: the slave boy in the Meno moves from guessing to confidence (i.e. from L1 to L2), but is not at L3, because he cannot give a reasoned account (aitias logismos). Euthyphro was confidently at L1, but over the course of the dialogue realizes that there is a difference between what he thought he knew and what piety really is, so he can discriminate an image from an object. He has begun L2.

To be at L2, one must be able to discriminate between two sorts of things: images and objects. One still lacks the ability to explain the difference. So what is the difference? One version is physical objects and their images (me and my mirror image). Another one might be morality: Socrates knows that what most people think are virtues are not virtues, he can make distinctions they cannot, and yet he cannot fully explain why.

L3

One turns from sensibles towards non-sensibles, which are what are called 'Forms.' 510 c2-d3 has a mathematical illustration of what that means. One

a) uses sensibles as images of Forms and
b) one proceeds from a hypothesis to conclusions.

At 511d, Plato makes it clear that one can know the square in both L3 and L4, which means that the objects distinction won't work here either: L3 is not a matter of what sort of objects one sees.

The Republic itself is at L3: Socrates uses images of the Sun and the Cave to illustrate things about Forms (reality: health illustrates justice), etc. So it satisfies a above by using sensibles as images of Forms. It also satisfies b (proceeding from hypothesis to conclusion), because Plato claimed that a longer way would be better (504c9-e2, 435d), which would relate virtue to the Form of the good.

Mathematics is at L3.

L3 is knowledge, but it is inferior. It is coherentist, but within strict confines.

L4

L4 involves relating all the L3 knowledge to the Form of the good to achieve a broader holistic knowledge.

One no longer needs images of the forms, just as one need not discuss particular animals in order to discuss taxonomy, and one can connect different systems of taxonomies of different things without reference to particular individual members of the taxonomies.

Many have held that L4 is a sort of vision, and so does not fulfill KBK or KL at all. Gail Fine argues that is not so. The route to L4 is dialectic (511b-c, 533a-d: cf. 534b3-6).

Rep. 533a1-5 says:

Socrates: You won't be able to follow me any longer, Glaucon, even though there is no lack of eagerness on my part to lead you, for you would no longer be seeing an image of what we're describing, but the truth itself. At any rate, that's how it seems to me. That it is really so is not worth insisting on any further. But that there is some such thing to be seen, that is something we must insist on. Isn't that so?
Glaucon: Of course.
Socrates: And mustn't we also insist that the power of dialectic could reveal it only to someone experienced in the subjects we've described and that it cannot reveal it in any other way?
Glaucon: That too is worth insisting on.
Socrates: At any rate, no one will dispute it when we say that there is no other inquiry that systematically attempts to grasp with respect to each thing itself what the being of it is, for all the other crafts are concerned with human opinions and desires, with growing or construction, or with the care of growing or constructed things.

Conclusion: there is L4 type knowledge of all of the following sensibles, objects, and Forms. Plato did not believe in TW. Knowledge requires Forms, but it is not confined to them.