Nally, E. G. - Is Plato A Coherentist The Theory of Knowledge in Republic V-VII

apeiron 2015; 48(2): 149–175
Edith Gwendolyn Nally

Is Plato a Coherentist?
The Theory of Knowledge in Republic V–VII
Abstract: Although many scholars take a foundationalist approach to the theo-
ry of knowledge in Republic V–VII, few if any have responded to a growing
number of coherentist interpretations, which hold that, for Plato, justification
rests not on a first principle but rather on the coherence of a sufficiently large
number of beliefs. This paper argues that the structure of knowledge in the Re-
public analogies is incompatible with the coherentist reading. Plato’s analogies
provide ample evidence that he holds something, at the very least, approximat-
ing a foundationalist first principle. In addition, this paper puts forward a Pla-
tonic solution to the regress problem. It argues that the form of the good is
plausibly self-justifying. Because the form of the good serves as the “measure”
by which the philosopher is able to determine the goodness, fittingness, and
truth of all that he encounters (517b), it seems reasonable to conclude that the
philosopher’s ultimate source of justification – what it is that puts an end to
the search for explanations and serves as an explanation of his claim to know
the form of the good – is nothing other than the form of the good itself.
Keywords: Plato, foundationalism, form of the good, coherentism, Republic
DOI 10.1515/apeiron-2014-0055
In what follows, I argue that Plato is a not a coherentist about justification in

the Republic. My argument proceeds in four parts. First, I introduce the regress
problem as it arises in Book VII. Second, I examine the analogies of the Sun,
Divided Line, and Cave in order to show that the form of the good, at the very
least, approximates a foundationalist first principle. Third, I argue that Plato is
very likely an implicit foundationalist, in that he holds a first principle or basic
belief, although he does not clearly recognize the regress problem. Fourth, I
present a solution to the regress problem on Plato’s behalf. The regress can be
put to rest, I argue, insofar as the philosopher has direct apprehension of the
form of the good.1

Edith Gwendolyn Nally: University of Richmond – Philosophy, Richmond, Virginia,
United States, E-Mail: egn9b@virginia.edu
Brought to you by | Yale University

Authenticated
Download Date | 7/11/15 12:53 AM
150  Edith Gwendolyn Nally
I The Regress Problem

Gail Fine has argued that, in Republic VII, Plato sets out two necessary condi-
tions for knowledge.2 The first condition (KL) is that one must be able to give a
logos for what one knows (533c1–2).3 I assume for the purposes of this paper
that a logos is an explanatory account which serves to justify a knowledge
claim.4 In order to count as knowing some proposition p, Plato believes that
one must be able to give an explanatory account of p.5 The second condition is
that one can know p by way of deduction from a hypothesis only insofar as one
also has knowledge of the hypothesis from which p follows (533c3–4). Fine
characterizes this provision as the view that (KBK) “knowledge must be based
on knowledge.”6 She reads Plato as claiming that one can know p only if one
has an account of p, which one also knows.
These two provisions – (KL) knowledge requires a logos or explanatory ac-
count and (KBK) knowledge must be based on knowledge – give rise to a parti-
cularly vicious version of what contemporary philosophers know as the “the
regress problem.”7 The regress problem can be put roughly as follows. (KL)
Knowing some proposition p requires a logos or explanation q. Yet explanation
q, (from KBK) must itself be a piece of knowledge. It follows (from KL) that
explanation q must have some further explanation r. The same will be true of

1 Many thanks to Dominic Scott, Daniel Devereux, and Harold Langsam for their insightful
comments on this paper. I am also grateful to Benjamin Jasnow, Evan Keeling, Courtney Evans,
and David Hewett. I have benefited from our many discussions of the Republic.
2 Gail Fine (1999) 238–41.
3 We find a similar condition stated also at 402a, 531e, Thea. 202c, 208d, and Phae. 78c–d 95d.
4 Myles Burnyeat (1980) argues that Plato is not concerned with justification but rather with
explanation. For Plato knowledge is not a justified true belief that x, but rather a true belief
that x plus some explanation as to why x is the case. For the purposes of this paper, I follow
Gail Fine in her response to Burnyeat’s view: “I agree that Plato takes knowledge to require
explanation. But we shouldn’t infer that he doesn’t take knowledge to be justified true belief.
We should infer instead that the sort of justification he takes to be necessary for knowledge
requires explaining why things are so.” (2003; 4–5)
5 As to whether Plato is concerned with knowledge of objects or propositions, scholars are of
two minds. Gail Fine (1999) takes it that, for Plato, knowledge is of propositions. Francisco
Gonzales (1996) 245–275 argues for the alternative. I am inclined to allow for both; however, I
am concerned primarily with propositions in this paper.
6 Gail Fine (1999) 238. See also (1979) 367. Fine cites Theaetetus and Meno 75c8–d7 for evi-
dence of this view. Later in the paper, I shall amend Fine’s summary of this point.
7 This problem has gone by many names throughout history. Aristotle appears to be the first
to formulate it explicitly it in the Post. Anal. (72b5). Sextus Empiricus later named the problem
“the diallelus” (Outlines of Pyrrhonism, II. Chapter IV. 20). For a contemporary formulation of
the problem see Anthony Quinton (1973) 119 and Laurence BonJour (1985) 18–19.

Authenticated
Is Plato a Coherentist?  151
each further explanation: For every r, one must appeal to some further s in
order to explain it. And the same will be true of s; because (KBK) each explana-
tion must itself be a piece of knowledge, (KL) each s will require some further
explanation t. The demand for explanation continues this way seemingly ad
infinitum. Substantiating a single claim seems to require an infinite string of
explanations.
There are three general strategies for contending with the regress problem.
The first strategy, “infinitism” it is sometimes called, is to embrace the idea of a
regress and give up on establishing a finite explanatory chain.8 Of course, Plato
never seems to embrace the idea of an infinite chain of logoi, and so I devote
little attention in what follows to the infinitist solution.9 The second strategy,
often called the “coherentist” solution, is to embrace the view that a chain of
explanations might be infinitely long but ultimately circular.10 11 According to
this view, a claim qualifies as knowledge if it coheres with other claims which
are “mutually supporting and explanatory, and form a sufficiently large
group.”12 The coherentist believes that an explanation of p might eventually ap-
peal to p itself, and that this circularity is not vicious so long as the chain of
explanations is sufficiently long. The third and final strategy for dealing with
the regress problem is the “foundationalist” solution. Foundationalism, at least
the version I consider here, is the view that something like a basic belief or first

8 It is worth noting that this strategy is not a new one. In the Post. Anal., Aristotle sets out the
infinitist solution as one school of thought common at the time (72b8–13). Ancient thinkers
may, therefore, have held some version of infinitism. For more on the contemporary argument
for infinitism see Peter Klein (1999) or Scott Aikin (2005).
9 Plato does consider an infinite regress of forms in the third man argument in the Parme-
nides. This argument is at least tangentially related to the present problem. As far as I know,
however, Plato never considers the infinitist account with respect to the question of justifica-
tion. If someone were to defend an infinitist reading of the Republic analogies, I imagine it
would suffer from many of the same problems as the coherentist interpretation which I address
below.
10 There are two different views commonly called “coherentist”. “Coherentism about truth” is
the view that a belief is true insofar as it “coheres” with another set of beliefs. “Coherentism
about explanation” or “coherentism about justification” is the view that a belief is explained/
justified insofar as it coheres with a thinker’s other beliefs. Although there may be some con-
nection between these views, only the latter will be relevant here. For more on contemporary
formulations of coherentism about justification see Laurence BonJour (1998).
11 Again, Aristotle anticipates this solution; he discusses a school of thought that halts the
regress by an appeal to circular reasoning (Post. Anal., 72b15–20).
12 Gail Fine (1999) 240–41. Fine takes her views about coherentism largely from Laurence
BonJour (1985) Part 2, chapters 5 and 7.

Authenticated
principle justifies all other beliefs. For the foundationalist, the regress problem
halts at the point where one comprehends this first principle.13
In the Republic, it has long been thought that the form of the good serves as
a foundational first principle – it is, after all, the highest point on the Divided
Line (511b). There are nevertheless a number of commentators who have defend
coherentist interpretations of the Republic.14 For example, Jyl Gentzler examines
Socrates’ account of justice at 442d–443b and concludes that he presents a co-
herentist argument. What it is that “confirms his hypothetical account of jus-
tice” is that it “explains why the people that we regard as just behave as they
do.”15 Gentzler concludes that the interlocutors accept Socrates’ view “because
of its coherence with his and his interlocutors’ other views, its simplicity, and
its explanatory force.” The interlocutors accept Socrates’ view because it co-
heres with their wider system of beliefs. Similarly, Kenneth Sayre, in his reading
of the Divided Line gives a coherentist interpretation of the form of the good.
He rejects the notion that the form of the good is some kind of “super proposi-
tion”, or a first principle, which serves to justify all other entities.16 He argues,
instead, that dialectic results in “synoptic vision that displays [the] kinships
and interconnections in the nature of things.” Gail Fine has given one of the
more extreme readings of the Sun, Divided Line, and Cave, suggesting that, in-
stead of being the highest principle and the last thing the philosopher must
understand, the form of the good is but one “link” in a circular explanatory
chain.17 According to her view, the philosopher does not possess knowledge of
the good itself until he descends from the world of the forms and considers their

13 The only version of foundationalism I consider in this paper is the view that there is some-
thing like first principle, or basic belief, which serves to justify all other beliefs. But not all
foundationalist positions depend on something like a first principle. For more other possibili-
ties see William Alston (1976) 287–305.
14 Others who have held this view include: Jyl Gentzler (2005) 469–96, Kenneth Sayre (1995)
177–81, and D. T. J. Bailey (2005) 102–103. See also Gail Fine (1979), Myles Burnyeat (1980)
173–91 and (1990) 216–18, and Anton and Preus (1989) 278 for a similar reading of the Theaete-
tus. Burnyeat describes of Plato’s theory of knowledge in the Theaetetus as follows: “Knowing
o is having a true judgement concerning o and an explanation based on the relationships of
elements and complexes within the domain of o.” This is not to be confused with the more
common position that Terence Irwin (1977) 133–159 and Gregory Vlastos (1982) 711–714 defend
in their reading of the Socratic dialogues, where Socrates does at times appear to give a coher-
ence view of truth.
15 Gentzler (2005) 480–81.
16 Sayre (1995) 177.
17 Fine (1999) 240–41. Kenneth Sayre (1995) 177–81 appears to hold some form of this view as
well.

Authenticated
relationship to sensible particulars. It is at this point that he realizes that the

form of the good is what furnishes him with an account of his entire system of
beliefs. In addition to this novel reading, Fine offers a powerful objection to the
foundationalist interpretation.18 She argues that the form of the good cannot be
what halts the regress of explanations. Plato is explicit that the the form of the
good must have a logos of its own; the philosopher must be able to say what is
good in all cases and defend his view against all arguments (534b). But the
logos of the good must be something that is itself known (from KBK) and there-
fore seems to require some further explanation (from KL). The only way to break
free of the need for some further explanation is to deny either (KL) that the form
of the good requires some logos or else (KBK) that its logos is something that is
itself known. Because Fine believes that Plato subscribes to both of these condi-
tions without exception, she takes the resulting dilemma as evidence that Plato
must not be a foundationalist at the time of the Republic.19 She concludes that
he must be a coherentist.
Few have responded to these coherentist interpretations of the Republic.20
My arguments in the next section are intended to do just this. I establish that
the analogies of Books V–VII, especially the Divided Line, are ill-fitted to the
coherentist reading. They suggest instead that Plato holds something at the very
least approximating a foundationalist first principle.

18 Ibid. 243.
19 See also Fine (2008) 181–182. Here her objection to the foundationalist reading is more
explicit: “If understanding what goodness is requires that one explain what it is in terms of
other concepts, those concepts are now more basic than goodness. We have here an instance
of the classic problem of the hierarchical structure of knowledge. If knowledge of X is founded
on knowledge of Y, and that, in turn, on knowledge of Z, then either we have an infinite regress
of knowledge or we have some foundations of knowledge, knowledge of which is grounded on
nothing but themselves [in other words, they have no logos]. Plato’s insistence that we must be
able to give an account of what we know [a logos] seems to rule out self-evident foundations of
knowledge; he asserts the necessity of a logos in many passages, notably Rep. 534b–c, where
the philosopher’s task is that of giving the logos of each of the forms and his ultimate aim that
of differentiating the form of the good from the other forms.”
20 As far as I know John Uglietta (2006) 331-40 is the only other scholar to have tackled this
issue. Uglietta argues in passing that the analogy of the Cave weighs against Fine’s coherentist
interpretation. I am very much in agreement with Uglietta; aim to show that it is an improbable
reading and that it is inconsistent with much of the Divided Line. Additionally, whereas Ugliet-
ta does not address the plausibility of the foundationalist interpretation, I shall argue in favor
of it by dismissing the threat of an infinite regress.

Authenticated
II The Republic Analogies

The Sun, Divided Line, and Cave are among the most widely discussed passages
in all the dialogues. Even giving a brief outline of them can feel like venturing
out onto shifting ground. That being said, the following descriptions are in-
tended to provide us with a more secure footing. My interpretation of these ana-
logies, especially the “hypothetical-unhypothetical” distinction in the Divided
Line, plays a crucial role in my later analysis of the regress problem.
The Sun
In Book VI, Plato argues that there must be a cause of all intelligible things and
draws an analogy with the sun. Just as the light of the sun illuminates the visi-
ble world, Plato surmises that there must be something, which “illuminates”
the intelligible world (508b–c). What it is that “lights up” intelligible things is
the form of the good (508e). Just as the sun shines on sensible particulars, mak-
ing them visible, so too does the form of the good render all that is intelligible
knowable to the thinker. It is also notable that, as a part of explaining this me-
taphor, Plato remarks that the form of the good is what “reigns” over the intelli-
gible world (509d1–3).21
The Divided Line
As a matter of further explaining the division between the sensible and intelligi-
ble worlds, Plato then introduces the analogy of the Divided Line. He tells us to
imagine a line broken into four parts; the bottom half concerns sensible particu-
lars and the top concerns that which is purely intelligible (509d). The levels are
divided according to the different objects; the first and second levels represent
thought about sensibles, while the third and fourth represent thought about in-
telligible objects. Although a great deal might be said of levels one and two of
the Line, they are of little interest in the present discussion insofar as a person
who remains at the bottommost levels appears to be concerned with “images”
or “likenesses” of the true nature of reality (509e). It is not until the third level

21 He says: νόησον τοίνυν, ἦν δ᾽ ἐγώ, ὥσπερ λέγομεν, δύο αὐτὼ εἶναι, καὶ βασιλεύειν τὸ μὲν
νοητοῦ γένους τε καὶ τόπου, τὸ δ᾽ αὖ ὁρατοῦ, ἵνα μὴ οὐρανοῦ εἰπὼν δόξω σοι σοφίζεσθαι περὶ
τὸ ὄνομα. ἀλλ᾽ οὖν ἔχεις ταῦτα διττὰ εἴδη, ὁρατόν, νοητόν (509d1–3);

Authenticated
that a thinker abandons his preoccupation with the sensible particulars and be-
gins to engage with intelligible objects (510b). The third and fourth levels of the
Line are outlined in more detail below.
Level 3
A thinker who remains at level three of the Divided Line has yet to grasp the
true nature of reality. He has given up his preoccupations with sensibles but is
nevertheless limited to reasoning by the process of “hypothesis” (ἐξ ὑποθέσεων)
(510b4–6).22, 23 In his explanation of level three, Plato gives the example of the
geometer, who spends his time considering figures and constructing mathemati-
cal proofs. Presumably, every proof that the geometer undertakes will begin
with an assumption of some sort, i. e. ‘a triangle is a three-sided enclosed figu-
re’ or ‘a line is the shortest distance between two points’. It is from these as-
sumptions that the geometer draws his conclusions. Yet the geometer’s work is
not concerned with the truth of his assumptions themselves (511a).24 Instead, he
bases his claims on premises, whose truth he cannot readily explain – not at
least while he remains at level three.
Level 4
By beginning to recognize that his assumptions are not firmly established prin-
ciples, and by engaging in dialectic, a thinker might eventually reach what
Plato calls the “unhypothetical” (ἀνυπόθετος) principle of everything (511b4–

22 ἧι τὸ μὲν αὐτοῦ τοῖς τότε μιμηθεῖσιν ὡς εἰκόσιν χρωμένη ψυχὴ ζητεῖν ἀναγκάζεται ἐξ ὑπ-
οθέσεων, οὐκ ἐπ᾽ ἀρχὴν πορευομένη ἀλλ᾽ ἐπὶ τελευτήν, τὸ δ᾽ αὖ ἕτερον—τὸ ἐπ᾽ ἀρχὴν ἀνυ-
πόθετον—ἐξ ὑποθέσεως ἰοῦσα καὶ ἄνευ τῶν περὶ ἐκεῖνο εἰκόνων, αὐτοῖς εἴδεσι δι᾽ αὐτῶν τὴν
μέθοδον ποιουμένη (501b4–6).
23 Plato thinks those at level three make claims that might apply to the forms themselves;
their conclusions are drawn from hypotheses, but these conclusion might nevertheless be true.
In other words, although they have yet to realize it, the level-three thinkers are concerned with
claims which, it might so happen, map onto reality (510d). Later on, Plato also remarks that
thinkers at the third level are better off than those with mere opinion (533b–c).
24 Santas (1999) 272 observes that the geometer does not leave his assumptions completely
undefended. What Plato means, Santas argues, is just that the geometer does not give an epis-
temological account of his assumptions. The geometer is not concerned with determining what
must be the true objects of mathematics. As such, he fails to see that the conclusion that he
derives are not just true of the geometrical figures he uses in his derivations but also true of
the forms themselves.

Authenticated
6).25 This principle, which I assume is the form of the good, occupies the high-
est position on the Divided Line (511b). After his encounter with this principle,
a thinker is able to turn his attention back to everything that he had once only
assumed and to draw conclusions, which are based on his knowledge of the
forms themselves (511b8–9).
What process is Plato describing in levels three and four of the Divided
Line? To begin with, it will be helpful to mark out two distinct phases in the
philosopher’s education: his ascent, roughly an “upward” movement, ending in
his apprehension of the form of the good (510a2–511b4), and his descent, where
he turns his attention back “downward” and begins to reason about other forms
and particulars using his knowledge of the good (511b4–c2).26 In examining the
philosopher’s ascent, let us consider the case of the geometer once more. Sup-
pose that the geometer, a level-three thinker, has completed a host of geometri-
cal proofs, based on a number of assumptions about geometric figures. Suppose
now that he has begun to wonder about the nature and truth of the assumptions
on which he has been basing his conclusions. Are the fundamental principles
of geometry true? And what does geometrical analysis actually show? Do math-
ematical claims apply only to the physical world or do they hold more gener-
ally?27 Grappling with these and other related questions, the geometer rises to
the fourth level of the Line. It is here that he will eventually realize that the
study of mathematics concerns the forms themselves, and that his proofs are
true in virtue of their appealing to these perfect and enduring entities. Having
realized this, the thinker will presumably turn his attention to other subjects as
well, engaging in similar investigations into music, astronomy, ethics and poli-
tics. After a long period of study, having worked through each discipline and its
relationship to the forms, he will eventually inquire about the nature of the

25 τὸ τοίνυν ἕτερον μάνθανε τμῆμα τοῦ νοητοῦ λέγοντά με τοῦτο οὗ αὐτὸς ὁ λόγος ἅπτεται
τῇ τοῦ διαλέγεσθαι δυνάμει, τὰς ὑποθέσεις ποιούμενος οὐκ ἀρχὰς ἀλλὰ τῷ ὄντι ὑποθέσεις, οἷον
ἐπιβάσεις τε καὶ ὁρμάς, ἵνα μέχρι τοῦ ἀνυποθέτου ἐπὶ τὴν τοῦ παντὸς ἀρχὴν ἰών, ἁψάμενος
αὐτῆς, πάλιν αὖ ἐχόμενος τῶν ἐκείνης ἐχομένων, οὕτως ἐπὶ τελευτὴν καταβαίνῃ, αἰσθητῷ παν-
τάπασιν οὐδενὶ προσχρώμενος, ἀλλ᾽ εἴδεσιν αὐτοῖς δι᾽ αὐτῶν εἰς αὐτά, καὶ τελευτᾷ εἰς εἴδη
(511b2–6).
26 This division is supported in 511b2–6: τὸ τοίνυν ἕτερον μάνθανε τμῆμα τοῦ νοητοῦ λέγοντά
με τοῦτο οὗ αὐτὸς ὁ λόγος ἅπτεται τῇ τοῦ διαλέγεσθαι δυνάμει, τὰς ὑποθέσεις ποιούμενος οὐκ
ἀρχὰς ἀλλὰ τῷ ὄντι ὑποθέσεις, οἷον ἐπιβάσεις τε καὶ ὁρμάς, ἵνα μέχρι τοῦ ἀνυποθέτου ἐπὶ τὴν
τοῦ παντὸς ἀρχὴν ἰών, ἁψάμενος αὐτῆς, πάλιν αὖ ἐχόμενος τῶν ἐκείνης ἐχομένων, οὕτως ἐπὶ
τελευτὴν καταβαίνῃ, αἰσθητῷ παντάπασιν οὐδενὶ προσχρώμενος, ἀλλ᾽ εἴδεσιν αὐτοῖς δι᾽ αὐτῶν
εἰς αὐτά, καὶ τελευτᾷ εἰς εἴδη.
27 This point follows Santas (1999) 270–3 in that the shift from level three to four constitutes
a shift in what a thinker considers the proper objects of his inquiry.

Authenticated
forms themselves. What it is that makes the forms “forms”? In virtue of what
are these entities perfect and enduring? This, I take it, is the course by which a
thinker begins to search for the form of the good – this is the course by which
the philosopher eventually discovers the “unhypothetical” principle which
stands at the top of the Line (511b4).
Plato is not clear about what it means to be an “unhypothetical” (ἀνυ-
πόθετος) principle. He does, however, employ the term in a telling way, in order
to draw a distinction between the method by which the geometer reasons, rest-
ing all of his claims on assumptions, and the way that the philosopher comes to
know the form of the good itself (511b7). At level three of the Divided Line, with
respect to geometry, a “hypothetical” claim is one that the geometer entertains
without any proof or explanation of its truth (510b). He holds the axioms of
geometry to be true, conducting proofs within the system of geometry, without
ever having examined the axioms themselves.
In what way, then, does the philosopher’s grasp of the form of the good,
the “unhypothetical” principle of everything, differ from the geometer’s purely
“hypothetical” investigations? Some commentators take the relevant difference
between the geometer’s purely “hypothetical” investigations and the philoso-
pher’s grasp of the form of the good to be a difference in proof or demonstra-
tion.28 This is the view that the axioms of mathematics are “hypothetical” inso-
far as the geometer can offer no proof of what he believes to be the case. By
contrast, the form of the good is “unhypothetical” insofar as the philosopher
will have a proof or demonstration of its truth.
In opposition to this view, Gerasimos Santas has argued that the form of
the good cannot be deduced from anything else; it can neither be proven nor
demonstrated.29 If Santas is correct, if the form of the good cannot be deduced
from anything else, then Plato must have some other reason for labeling it “un-
hypothetical”.30 Santas suggests that the relevant difference, what sets the ge-
ometer’s reasoning about principles of geometry apart from the philosopher’s
encounter with the form of the good, must instead be a difference in the degree
of epistemological certainty. The philosopher has certain knowledge of the
good, while the geometer, never having questioned the axioms of mathematics,
cannot be said to know them for sure. This view is well supported by evidence
from Book VII, where Plato describes dialectical reasoning as a process that
does away with hypothetical claims and draws the philosopher up to a first

28 See Cornford (1965) 65–6 and Ross (1951) 54–5.
29 Francisco Gonzalez (1996) 223 also makes the argument that we ought not to think the
difference as a matter of proof.
30 Santas (1999) 269–70.

Authenticated
principle in such a way that it will be “secure” or “confirmed” (ἵνα βεβαιώσηται)

(533c5–d1).31, 32 In Book VII, the relevant distinction between hypothetical
claims and the form of the good appears to be just that the good is something
that the philosopher grasps as a certain truth, while a hypothetical claim is one
whose truth has not yet been firmly established.
The hypothetical-unhypothetical distinction is not, however, fully ex-
hausted by the notion of epistemological certainty. The distinction in question
might also be thought of as a difference in “epistemological priority”. The no-
tion of epistemological priority can be clarified as follows: for two claims, the
first can be said to be “epistemologically prior” to the second, if knowing the
second requires an appeal to one’s knowledge of the first. In this way, knowing
the truth of some proposition p is epistemologically prior to knowing the truth
the conjunction p & q.33 This distinction gives us a tidy way of carving up the
hypothetical-unhypothetical distinction. To say that some claim h is “hypotheti-
cal” would mean that there is something else, some “higher” principle, which is
epistemologically prior to h. In order to count as knowing h, one must first in-
vestigate the truth of those claims epistemologically prior to h, on which h de-
pends. This distinction maps neatly onto the text of the Divided Line insofar as
the geometer cannot be said to know his subject matter until he rises to the
fourth level to investigate the truth of his assumed axioms (511d1–3). His knowl-
edge of mathematical claims seems to depend in some way on his knowledge of
the forms. By contrast, a claim which is “unhypothetical” is one for which there
is no principle that is epistemologically prior to it. It is a claim for which there
is nothing “higher” to which one must appeal in order to count as having
knowledge. I take it that this is what Plato has in mind in placing the form of
the good atop the Divided Line: one need not appeal to anything “higher” or
more epistemologically basic in order to count as knowing it.34

31 Where Socrates describes the dialectical method in Book VII, he says: οὐκοῦν, ἦν δ᾽ ἐγώ, ἡ
διαλεκτικὴ μέθοδος μόνη ταύτῃ πορεύεται, τὰς ὑποθέσεις ἀναιροῦσα, ἐπ᾽ αὐτὴν τὴν ἀρχὴν ἵνα
βεβαιώσηται… (533c5–d1).
32 Bailey (2005).
33 Jim Pryor (2000) defines epistemic priority as follows: “Your justification for believing p1 is
antecedent to your justification for believing p2 just in case your reasons for believing p1 do
not presuppose or rest on your reasons for believing p2. Your reasons for believing p1 can not
beg the question whether p2.” (524–25)
34 Bailey (2005), in his helpful discussion of what it means to be “unhypothetical”, claims
that Aristotle uses the label “unhypothetical” to refer to principles which anyone who knows
the ways things are must know. He calls this the “priority” condition; this is the view that an
“unhypothetical” principle is the sort of thing which anyone who has knowledge must grasp.
Bailey goes on to argue that Plato and Aristotle share roughly the same conception of the “un-

Authenticated
We are now in a better position to examine the philosopher’s ascent and

descent. How does the philosopher finally come to know the form of the good?
And, by what process does he eventually turn his focus downward? In answer-
ing this question, it is helpful to think of the philosopher’s ascent like setting
up a course of dominoes. In the beginning, he builds a structure of beliefs. He
deduces conclusions from assumed premises and develops a complex system of
interrelated opinions. At this stage, the philosopher cannot be said to “know”
the claims within this system (533c2–3). Only once he grasps the unhypothetical
first principle – something that he knows for certain and without appealing to
any prior principle – like knocking down a domino at the head of a chain, is he
able to descend, working back through the structure of his beliefs, re-deducing
his former claims, from a principle that he holds to be a certain truth. Just as
the structure of dominoes is in place before knocking down the first domino, so
too has the philosopher constructed a system of interrelated beliefs and depen-
dencies before he comes to understand the form of the good. It is not until he
knocks down the head domino, not until he encounters the form of the good
itself, that the chain reaction is set off. When he does encounter the form of the
good, the philosopher is finally able to see that all the claims he had once only
assumed to be the case, and all the dependencies and logical entailments that
he discovered at some earlier point in his education, do in fact follow from this
first principle. In this way, his encounter with the form of the good is what con-
verts his prior assumptions into knowledge.
After encountering the good, the philosopher might, for example, take up
one of his former geometrical proofs. When he first constructed the proof, he
would have begun with an assumed principle and worked his way down to a
conclusion. Although the conclusion might have followed logically, it would
nevertheless have been entailed by a principle he had only assumed, a mere
“working hypothesis”. Now, however, after having investigated the forms and
encountered their ultimate explanation, the philosopher has in his possession a
piece of knowledge. From this piece of knowledge he can reason his way to the
first principles of each discipline.35 This is how the philosopher comes to have

hypothetical”. Although this is a somewhat different formulation of “priority”, and one that
comes from Aristotle, Bailey’s view might be taken as further support for the reading that the
form of the good is epistemically prior to other hypothetical principles.
35 It should be noted that I am using “reason” as a kind of shorthand for whatever process
the philosopher uses to justify the first principles of each the disciplines. Santas (1999) ques-
tions whether deduction is the proper way of thinking about this process. How, after all, might
we deduce the principles of mathematics from knowledge of what is good? I am aware of this
worry but, for the sake of space, remain neutral on this question. All that matters in the present
context is that the philosopher’s encounter with the form of the good allows him to “discharge”

Authenticated
knowledge of the first principles of geometry, establishing them by some logical

process (like deduction)36 so that they are no longer assumed. This is the pro-
cess of his descent; beginning with his most basic assumptions, the philosopher
is now able to reason through his former arguments, discharging his hypotheses
and replacing them with certain truths.
The Cave
Although I shall not focus on the analogy of the cave in nearly as much detail,
the later stages of this metaphor also play an important role in my arguments
against a coherentist interpretation of the Republic. In Book VII, Plato tells us to
imagine prisoners living in chains deep in a cave. While they are imprisoned,
these people concern themselves with shadows of artifacts projected onto the
walls (514a–515c). Moreover, they take the shadows to constitute what is real
(515c). Plato then asks us to imagine a prisoner who breaks free from his chains
and leaves the cave, where he encounters daylight for the first time (515c–516a).
Once he recovers from a temporary period of confusion and blindness (516a),
the freed man looks out upon real entities illuminated by the light of the sun.
After he has explored this world for some time, the free man realizes that the
sun is what provides light to everything else (516b–c). I presume that this final
stage of the journey represents the point at with the philosopher encounters the
form of the good, recognizing that it provides justification to his beliefs about
all other intelligible and sensible entities (517b).
Having established a basic reading of the Republic analogies, let us turn
our attention to the main arguments against the coherentist interpretation. As I
mentioned above, Gail Fine puts forward an interpretation of the Divided Line,
according to which, even at the highest point of his ascent, the philosopher is
in need of some further explanation.37 Her view is that the philosopher encoun-
ters the form of the good but does not yet possess an explanation of its essence.
He does not yet possess a logos of the good – and so, the need for explanation
still exists. Fine speculates that the logos requirement (KL) can be met only once
the philosopher sees that the form of the good has some explanatory power over
other forms and sensible particulars. She argues that the form of the good offers
the philosopher a kind of teleological map of the nature of reality, explaining

his former hypotheses in some manner—regardless of whether it occurs through deduction or
through some other rational process.
36 See n. 20.
37 Fine (1999) 243.

Authenticated
how all of his beliefs are interconnected, but that he can know the good itself
only once he sees “how well it allows us to understand the natures of, and
interconnections between, other forms and sensibles.”38 To put her view an-
other way, Fine believes that the philosopher can be said to “know” the form of
the good only once he travels back down the Divided Line and sees that it fur-
nishes him with an explanation of everything that he encounters.39
The analogies of Books V–VII suggest, however, that the form of the good
occupies a more privileged epistemological position. In the analogy of the Sun,
for instance, Plato even goes so far as to label the good that which “reigns”
(βασιλεύειν) over the physical and intelligible world (509d).40 Like a ruler, the
good is what governs the intelligibility of other entities. The Divided Line pro-
vides further support for the epistemological supremacy of the good. Here, Plato
labels the form of the good the “unhypothetical” first principle (510b). We have
seen that this description stands in contrast to the principles of geometry, which
are merely “hypothetical” in virtue of the fact that the geometer has yet to
search out their explanations. I suggested above that we might think of the hy-
pothetical-unhypothetical distinction in terms of epistemological priority; the
form of the good is unlike the axioms of geometry in that that it can be known
without an appeal to anything prior or more basic. Each of these analogies
places the form of the good in a more privileged position than the rest of the
philosopher’s beliefs. Given such hierarchical descriptions, the coherentist inter-
pretation – in that it takes the form of the good to be one of many links in a
chain of explanations – is a somewhat unnatural reading.
The analogies of the Republic also suggest that the philosopher need not
make his descent in order to have knowledge of the good. In the Divided Line
passage, at the height of his ascent, the philosopher’s reason “fastens onto” or
“lays hold of” (ἅπτω) the first principle (511b2–6).41 This suggests that the philo-
sopher has a secure grasp of the essence of goodness at the point where he
encounters the form, and not at some point thereafter. The analogy of the cave
provides further support for this conclusion. John Uglietta, in his discussion of
Fine’s view, notices that “Plato’s account suggests that the search for knowl-

38 Ibid. 244–6.
39 Ibid. 245.
40 For the full passage, see note 19.
41 In fact, ἅπτω appears twice in the passage: τὸ τοίνυν ἕτερον μάνθανε τμῆμα τοῦ νοητοῦ
λέγοντά με τοῦτο οὗ αὐτὸς ὁ λόγος ἅπτεται τῇ τοῦ διαλέγεσθαι δυνάμει, τὰς ὑποθέσεις ποιού-
μενος οὐκ ἀρχὰς ἀλλὰ τῷ ὄντι ὑποθέσεις, οἷον ἐπιβάσεις τε καὶ ὁρμάς, ἵνα μέχρι τοῦ ἀνυ-
ποθέτου ἐπὶ τὴν τοῦ παντὸς ἀρχὴν ἰών, ἁψάμενος αὐτῆς, πάλιν αὖ ἐχόμενος τῶν ἐκείνης
ἐχομένων, οὕτως ἐπὶ τελευτὴν καταβαίνῃ, αἰσθητῷ παντάπασιν οὐδενὶ προσχρώμενος, ἀλλ᾽ εἴ-
δεσιν αὐτοῖς δι᾽ αὐτῶν εἰς αὐτά, καὶ τελευτᾷ εἰς εἴδη (511b2–6).

Authenticated
edge comes to a final end when one understands or discovers the Form of the
good.” Although Uglietta does not elaborate, I take it that he has in mind the
point at which the philosopher exits the cave, and after some time and toil fi-
nally looks upon the sun itself. In his unpacking of this metaphor, Plato writes
(517b–c):
τὰ δ᾽ οὖν ἐμοὶ φαινόμενα οὕτω φαίνεται, ἐν τῷ γνωστῷ τελευταία ἡ τοῦ ἀγαθοῦ ἰδέα καὶ
μόγις ὁρᾶσθαι, ὀφθεῖσα δὲ συλλογιστέα εἶναι ὡς ἄρα πᾶσι πάντων αὕτη ὀρθῶν τε καὶ
καλῶν αἰτία, ἔν τε ὁρατῷ φῶς καὶ τὸν τούτου κύριον τεκοῦσα, ἔν τε νοητῷ αὐτὴ κυρία
ἀλήθειαν καὶ νοῦν παρασχομένη, καὶ ὅτι δεῖ ταύτην ἰδεῖν τὸν μέλλοντα ἐμφρόνως πράξειν
ἢ ἰδίᾳ ἢ δημοσίᾳ.
In the knowable realm, the form of the good is the last thing to be seen and it is reached
only with difficulty. Once one has seen it, however, one must conclude that it is the
cause of all that is correct and beautiful in anything, that it produces both light and its
source in the visible realm, and that in the intelligible realm it controls and provides
truth and understanding, so that anyone who is to act sensibly in private or public must
see it.42
It is highly relevant that Plato describes a final point in the philosopher’s as-
cent. Plato calls the form of the good the “last thing” (τελευταία) which the
philosopher must encounter on his journey from the cave. This suggests that,
once the philosopher encounters the form of the good, the search for explana-
tions ceases.
It may be objected, however, that the cave just shows that the philosopher
has seen all that he needs to see. He may, nevertheless, still need to work out
what exactly goodness is, even after his encounter with the form. According to
Fine, the philosopher only understands the nature of goodness fully once he
returns to thinking about sensible particulars. Yet, this reading is inconsistent
with similar descriptions from later in the dialogue. When Plato describes the
dialectical process in Book VII, the philosopher’s encounter with the good is
“the limit of the intelligible” and “the end of journeying for the one who
reaches it” (532d). Shortly thereafter, we find the characterization of dialectic,
as a method of inquiry, which leads a thinker up to the first principle in a way
so as to be “secured” or “confirmed” (ἵνα βεβαιώσηται) (533c5–d1).43 The place-
ment of the good as the “highest” or “final” principle, which is at this point
“secure” or “confirmed”, supports the view that the philosopher has secure

42 Unless stated otherwise, translations of the Republic follow Grube and Reeve, in Cooper
(1997).
43 Where Socrates describes the dialectical method in Book VII, he says: οὐκοῦν, ἦν δ᾽ ἐγώ, ἡ
διαλεκτικὴ μέθοδος μόνη ταύτῃ πορεύεται, τὰς ὑποθέσεις ἀναιροῦσα, ἐπ᾽ αὐτὴν τὴν ἀρχὴν ἵνα
βεβαιώσηται… (533c5–d1).

Authenticated
knowledge of the good at the pinnacle of his ascent, and not, as Fine would
have it, only once he turns his focus back to other forms and sensible particu-
lars.
Although Fine is right to think that the form of the good serves as an ex-
planatory account of everything else, i. e. other forms and sensible particulars
(511b, 516c–d, 517c), it is not this function that furnishes the philosopher with
an account of its essence. For Fine, the philosopher can be said to “know” the
good itself only once he sees that it furnishes him with an account of everything
else that exists. It would seem, however, that the endpoint of the philosopher’s
search for explanations, the point at which he has a substantiated grasp of the
form of good, is the point at which he apprehends goodness itself.
In sum, the epistemological supremacy of the good and its placement as
the last or final thing encountered by the philosopher weighs heavily against
the coherentist interpretation. At the end of this section, it is interesting to note,
however, that this argument does not preclude the general spirit of Fine’s view.
She summarizes her reading of Plato’s theory of knowledge as follows:
One knows more to the extent that one can explain more; knowledge requires not a vision,
and not some special sort of certainty or infallibility, but sufficiently rich, mutually sup-
porting, explanatory accounts. Knowledge for Plato does not proceed piecemeal; to know
one must master a whole field, by interrelating and explaining its diverse elements.44
This point is still very much true of Plato’s theory of knowledge. The philoso-
pher is at each stage of his education attempting to master a field of study,
creating a rich and mutually supporting system of beliefs. Plato is quite clear
about this; the only way to reach the top of the line is through rigorous dialec-
tical reasoning. The philosopher must take up many disciplines and weave to-
gether a complex system of beliefs. What the above argument is intended to
show is just that the philosopher need not continue the search for explanations
after his encounter with the form of the good. Instead, at the point where the
philosopher encounters the good, he has a secure grasp of its truth.
III Implicit Foundationalism

Based on the above evidence, it is rather tempting to conclude that Plato must
be a foundationalist at the time of the Republic. But this conclusion does not
follow quite so readily; Fine has argued that the foundationalist interpretation

44 Fine (1999) 245.

Authenticated
invites a vicious form of the regress problem. This problem arises insofar as
Plato stipulates both that the form of the good is something that is known and
that it has its own logos (508e, 534b). If the form of the good has its own logos,
that logos must (from KBK) be something that is itself known and therefore (KL)
seems to require some further explanation. The same will be true of each further
explanation ad infinitum. The only way to break free of the regress seems to be
to deny either (KL) that the form of the good has its own distinct logos or else
(KBK) that its logos is something that is itself known.
Despite the initial force of this critique, it is not clear that the foundational-
ist reading is doomed to failure. Fine’s argument against the foundationalist
interpretation attributes a certain awareness on the part of Plato that the regress
problem is a threat. Because the regress threatens foundationalism, she argues,
Plato must have opted instead for a coherence view. It seems much more likely,
however, given the evidence from the analogies of Books V–VII, that Plato is
simply an “implicit foundationalist”. He might accept some rough brand of
foundationalism at the time of the Republic, despite being either unaware or
unperturbed by the threat of the regress problem. According to this view, he
might think that the form of the good serves to explain or justify all other pieces
of knowledge, without ever worrying about what it is that serves to explain or
justify the good itself.
This view is supported by the fact that Plato would likely not have felt the
need to address the regress problem in the context of Books V–VII. The regress
problem only arises if KL and KBK are taken to be universal claims, applying to
all pieces of knowledge, including the form of the good. In what follows, I argue
that the provisions set out in 533c1–4, where many have found evidence of
KBK, concern only hypothetical claims and therefore do not extend to the form
of the good. If I am correct, if Plato does not endorse a version of KBK that
extends to the form of the good, then the regress problem is not a natural worry
at this point in the dialogue.
I began this paper with the two provisions at 533c1–4. In this passage Plato
is in the process of criticizing the geometer; he cannot be said to “know” insofar
as he makes use of hypotheses without any account of their truth. The geometer
fails to have knowledge, Plato thinks, because he fails to satisfy the first neces-
sary condition (KL), that all knowledge have an explanatory account (533b).
The criticism is just that the geometer, if he is reasoning about purely hypothe-
tical claims, has no logos for what he claims to know. Plato then sets out the
second condition as follows (533c1–4):
ᾧ γὰρ ἀρχὴ μὲν ὃ μὴ οἶδε, τελευτὴ δὲ καὶ τὰ μεταξὺ ἐξ οὗ μὴ οἶδεν συμπέπλεκται, τίς
μηχανὴ τὴν τοιαύτην ὁμολογίαν ποτὲ ἐπιστήμην γενέσθαι;

Authenticated
For where the starting-point is something that the reasoner does not know, and the con-
clusion and all that intervenes is a tissue of things not really known, what possibility is
there that assent in such cases can ever be converted into true knowledge or science?
The reply to this question is “None”. Something which the geometer deduces
from a mere hypothesis cannot be known with any certainty.
Fine and others have taken this passage as evidence for KBK, the condition
that “all knowledge must be based on knowledge”. It is my view, however, that
Plato sets out a decidedly different and weaker provision in this part of the dia-
logue. His point in this passage might be better summarized as the view that:
(KBK1) a proposition can be known by way of deduction from some principle only if the
principle is itself known with certainty.
Whereas KBK claims that all knowledges is based on some other piece of
knowledge, KBK1 claims that one cannot claim to know a proposition based
solely on the fact it follows logically from some assumed premise; to know
such a proposition one must also know the assumed premise. The mistake
against which Plato warns in KBK1 might be thought of as the tendency to con-
fuse an argument from hypothesis for one that guarantees the truth of its con-
clusion.
We find a sentiment similar to KBK1 in Plato’s critique of the geometer in
the Divided Line passage. Unlike the philosopher, the geometer bases his beliefs
on unconfirmed assumptions, whose truth he cannot explain (511b). He mis-
takes these claims for knowledge, believing his conclusions to be true in virtue
of the fact that they follow from certain axioms, although he has failed to inves-
tigate the truth of the axioms themselves. It appears that in KBK1 Plato means
only to revisit this sentiment, warning his interlocutors against making the ge-
ometer’s mistake, believing themselves to be wise, when their conclusions rest
on untested assumptions.
If this reading is correct, then, Plato is not stipulating that all knowledge
must follow from, or be based on, some other piece of knowledge. Where many
have read this passage as setting out KBK, a universal condition, Plato is in-
stead introducing KBK1, a view that highlights a common weakness in hypothe-
tical argumentation – hypothetical arguments base their conclusions on axioms
or principles which are not known for certain. Unlike KBK, KBK1 singles out the
class of claims that follow from some “higher” or more epistemologically basic
principle. It is nevertheless possible that there could be a class of claims, i. e.
the philosopher’s knowledge of the form of the good, which, in virtue of not
following from anything more epistemologically basic, are exempt from the pro-
vision in question (KBK1). I argued above that the form of the good is “unhy-
pothetical” (ἀνυπόθετος) in the sense that there are no principles that are epis-

Authenticated
temologically prior to it. This is just to say that, in order to count as knowing
the good, one need not appeal to any more basic principles for explanation or
justification. Because the form of the good does not follow from any prior prin-
ciple, KBK1 insofar as it concerns the class of claims that follow from prior or
more basic principles simply does not hold.
If I am correct, if this passage is not setting out KBK, then it is easy to see
how Plato might have missed or set aside the argument which Fine raises in
opposition to the foundationalist interpretation; it simply does not arise in this
context. Because KBK1 applies to claims that follow from some more basic prin-
ciple, a warning that does not apply to the philosopher’s knowledge of the form
of the good, the infinite regress of explanations is not an immediate or natural
worry at this point in the dialogue.
It may be helpful to think of the above solution in terms of the following
image: picture the philosopher’s system of beliefs as a hanging mobile, with
each claim suspended from some principle “higher up” on the mobile. The
theorems of geometry would hang on threads below the axioms from which
they follow. (The individual forms would presumably hang below the form of
the good.) Given this imagery, we might think of the passage at 533c1–4 as a
warning about those knowledge claims which are “suspended from” some
higher principle on the mobile. For all those claims that depend on a higher
principle (this is just what it means to be “suspended” in the mobile meta-
phor), Plato thinks that such a claim can be known for certain only if the prin-
ciple from which it hangs is also known for certain. It remains an open possi-
bility, however, that there may be a claim or a group of claims, like the form
of the good, presumably at the top of the mobile, which are not “suspended
from” any higher principles. With respect to a claim of this sort, it makes no
sense to demand (KBK1) that one can only know it only in virtue of knowing
the principles from which it follows – precisely because there no principles
from which it follows.
Fine takes the threat of the regress problem as evidence that Plato must not
be a foundationalist in Republic V–VII. Because he is firmly committed to KL
and KBK, Fine argues that Plato must see the threat of a regress and embrace it.
If I am correct, however, Plato is not concerned at 533c1–4 with KBK. It is un-
wise, therefore, to conclude on the basis of the regress problem that Plato must
be a coherentist. Furthermore, it is fully consistent with what Plato does say in
this passage, he gives a warning against unconsidered mathematical assump-
tions, that the philosopher’s encounter with the form of the good might never-
theless be what serves as the ultimate source of justification in his system of
knowledge.

Authenticated
IV A Platonic Solution: Direct Apprehension

If I am correct, then Plato’s remarks at 533c1–4 are intended only to apply to
the class of claims that are deduced from some prior principle. The form of the
good, because it is not deduced from some prior principle, is therefore exempt
from the provision at 533c1–4. Nevertheless, it might be objected that this solu-
tion does little to meet the philosophical challenge at hand. Worse yet, it might
be objected that the regress problem does arise naturally in the context of Books
V–VII, once we consider that the form of the good has a logos (534b) and that
this logos must itself be a piece of knowledge. Even if the form of the good does
not follow from anything prior, it must nevertheless have a logos of its own.
Plato states this explicitly shortly after setting out the two claims that have been
of central interest to us in this paper. He says (534b5–c3):
οὐκοῦν καὶ περὶ τοῦ ἀγαθοῦ ὡσαύτως: ὃς ἂν μὴ ἔχῃ διορίσασθαι τῷ λόγῳ ἀπὸ τῶν ἄλλων
πάντων ἀφελὼν τὴν τοῦ ἀγαθοῦ ἰδέαν, καὶ ὥσπερ ἐν μάχῃ διὰ πάντων ἐλέγχων διεξιών,
μὴ κατὰ δόξαν ἀλλὰ κατ᾽ οὐσίαν προθυμούμενος ἐλέγχειν, ἐν πᾶσι τούτοις ἀπτῶτι τῷ
aλόγῳ διαπορεύηται, οὔτε αὐτὸ τὸ ἀγαθὸν φήσεις εἰδέναι τὸν οὕτως ἔχοντα οὔτε ἄλλο
ἀγαθὸν οὐδέν...
The same applies to the good. Unless someone can distinguish in an account the form of
the good from everything else, can survive all refutation, as if in battle, striving to judge
things in accordance with being, and can come through all this with his account still
intact, you’ll say that he doesn’t know the good itself or any other good...
The philosopher must be able to give an account, which distinguishes what is

good from everything else, and to maintain this position through extensive at-
tempts by his interlocutors to refute him. In this context, having a logos seems
to require possessing something like a definitional account of goodness. Yet, if
the philosopher must have a definitional account of goodness, and this account
is something that he knows, then (from KL) this account would seem to require
a logos of its own.45 Presumably the same will also be true of this logos, and
every further logos, ad infinitum.
In the last section, I argued that 533c1–4 is little more than a warning about
hypothetical claims (KBK1), and that the regress problem might not arise natu-
rally in this context. The worry here is that the regress problem might follow
readily from certain other features of Plato’s theory of knowledge. Cast in this

45 One possible solution to the regress problem that I do not entertain is the view that the
philosopher does not know his own definition of the form of goodness. I do not entertain this
view in that it seems to be an implausible way of thinking about explanations. Something that
is not itself known does not seem to be an adequate explanation or justification.

Authenticated
light, intuitive foundationalism, the view that Plato would not have felt the
need to consider the regress problem at this point in the Republic, appears to be
an uncharitable interpretation, one that attributes to Plato a substantial lack of
insight.46 It might also be objected that in order to set aside the coherentist
reading once and for all, the burden of proof lies with the foundationalist inter-
pretation to provide a more thorough explanation of the way in which the phi-
losopher comes to have secure knowledge of the form of the good. If the foun-
dationalist interpretation fails to offer a convincing solution to the regress
problem, leaving implicit foundationalism to rest on its textual merits, then the
coherentist interpretation might be preferred due to its explanatory power and
“completeness”.
In response to these objections, I would first like to note that the implicit
foundationalist reading is loyal to the text of this part of the Republic. In other
words, although it may not be a satisfying philosophical result, it is very likely
that Plato was not concerned to give a solution to the regress problem anywhere
in Books V–VII. I recognize, however, that attributing an incomplete view to
Plato might be taken as a weakness, both in the context of arguing against the
coherentist interpretation and given the widespread commitment that, as inter-
preters, we ought to give complete philosophical interpretations. It is in this
spirit that I offer the following solution to the regress problem, not because the
text of the Republic demands it, but rather as a way of making the foundational-
ist reading a more convincing and complete view.
The Regress Problem Revisited
Traditionally, scholars who have strayed into a discussion of the regress pro-
blem in the Republic have left us with a murky picture of the role of explanation
in the philosopher’s final ascent. Cornford resorts to an enigmatic comparison
to religious revelation. For him, the philosopher has something like “an upward
movement of intuition” or an “immediate act of vision” by which he suddenly
grasps or sees the form of the good.47 As a matter of defending this view, Corn-
ford quotes the Seventh Letter: “After long dwelling on a common life of philo-

46 This objection might arise from the widespread assumption that, as interpreters of Plato,
where we encounter an incomplete argument on the part of Socrates, we ought to, as Crombie
(1962) puts it, “find an interpretation of it which renders it valid, or at least to reconstruct the
valid train of thought the presence of which in Plato’s mind allowed the fallacy to pass unde-
tected.”
47 Cornford (1932) 48.

Authenticated
sophic converse, suddenly, as if from a leaping spark, a light is kindled, which,

once it has arisen in the soul, thenceforth feeds itself” (Ep. VII, 341). The soul
simply leaps, grasping the essence of the good in a single unmediated act of
insight. Others following in this tradition have proposed that the philosopher
has a kind of immediate intuition or “direct acquaintance” with the form of the
good.48 This view, a version of which I defend below, holds that the philosopher
encounters the form of the good in an unmediated way; he encounters the form
in such a way that he comes to know it immediately and with certainty. Cross
and Woozley, sensing the conjectural nature of this discussion, remark that
“there is nothing in this part of the Republic to help us… what [Plato] does say
is so brief and so obscure that most comment tends to become speculation, and
to take us too far from the text of the Republic, which is our immediate inter-
est.”49 Before moving on, I am inclined to give a similar disclaimer. Although
the following analysis is rooted in the theory of knowledge set out in Books V–
VII of the Republic, it enters into somewhat speculative territory.
The philosopher may have something roughly like “direct acquaintance” of
the form of the the good, and this encounter is what halts his search for expla-
nations. Others who have defended this view argue that the philosopher en-
counters the form of the good in such a way that he is immediately justified in
accepting it as true.50 Yet these scholars have done little to explain the relation-
ship between the form of the good and its logos. In order to dismiss the regress
once and for all – in order to show that form of the good might have a logos
without giving rise to the regress anew – I argue that having a logos of the good
must not itself be what justifies the philosopher’s knowledge of goodness. Hav-
ing a logos, or explanation, of what is good must be in the philosopher’s power,
yet it is not the mechanism by which he comes to know goodness. The philoso-
pher comes to have knowledge of goodness through a kind of direct acquain-
tance with the form. I conclude with a brief description of what it means to have
a logos of goodness. The philosopher’s logos of the good, his ability to explain
what is good in every context, is itself justified by his direct acquaintance with

48 Robinson (1953) 172–3 describes a kind of “intuition” by which the philosopher comes to
have knowledge. Others who consider this view include Cross and Woozley (1964) 252–53; Hare
(1965) 35; Ross (1951) 67; Vlastos (1985) 13–15; and White (1976) 99–100. Guthrie (1978) 133 in
his analysis of the Sophist describes just how widespread this view is: “...to grasp the one Form
above the many is not simply the last stage in a process of thought but an achievement of
direct acquaintance with the divine world in an act analogous to vision. This for many is the
core of Platonism…”
49 Cross and Woozley (1964) 261.
50 See note 48.

Authenticated
the good. The form of the good serves as a paradigm against which the philoso-
pher can “measure” any action or object under his consideration.
Before moving on, it will be useful to be clear about what it means to stand
in a relationship of “direct acquaintance” to something. Bertrand Russell de-
scribes this relationship as follows:
I say that I am acquainted with an object when I have a direct cognitive relation to that
object, i. e. when I am directly aware of the object itself. When I speak of a cognitive
relation here, I do not mean the sort of relation which constitutes judgment, but the sort
which constitutes presentation… When we ask what are the kinds of objects with which
we are acquainted, the first and most obvious example is sense-data. When I see a co-
lour or hear a noise, I have direct acquaintance with the colour or the noise.”51
In the case of sense perception, Russell claims, there is something like a direct
relation between the subject and the object, that is, between the viewer and the
sense-data. He goes on to argue that this relation also exists in cases in which
the object is a concept. He gives the following example: “Not only are we aware
of particular yellows, but if we have seen a sufficient number of yellows and
have sufficient intelligence, we are aware of the universal yellow”. After seeing
enough instances of a particular phenomenon, like the color yellow, Russell
thinks that we develop an abstract concept, which, like the original sense data,
also stands before the mind in a way that it direct or immediate. The concept of
yellow, when called to mind, has a similar immediateness to the particular in-
stances of yellow from which it was derived. Russell makes a number of claims
about the proper objects of this relation. He thinks, for example that we can be
directly acquainted with sense data and with concepts, but not physical objects.
What is of particular interest to us in the present context is not so much the
nature of the objects of direct acquaintance, but rather the relation itself – be-
tween the viewer and the object of acquaintance.
If the philosopher encounters the form of the good through direct acquain-
tance, it makes sense that Plato should often write about the philosopher’s en-
counter with the form of the good as a kind of “gripping” experience. On the
Divided Line there is a moment at which the philosopher’s reason “fastens
onto” or “grasps” (ἅπτω) the good itself (511b2).52 Similarly, in the later discus-
sion of the philosophical method, Socrates reports that the form of the good is
ultimately “revealed” (φήνειεν) through the power of dialectic (533a5–6).53

51 Russell (1917) 154.
52 See note 41.
53 Just after describing the philosopher’s encounter with the good, Socrates asks οὐκοῦν καὶ
ὅτι ἡ τοῦ διαλέγεσθαι δύναμις μόνη ἂν φήνειεν ἐμπείρῳ ὄντι ὧν νυνδὴ διήλθομεν, ἄλλῃ δὲ
οὐδαμῇ δυνατόν;

Authenticated
Although I am generally reluctant to put much stock in these metaphors, it is

worth noting that Plato describes the result of the process of dialectic, the en-
counter with the form of the good itself, as a kind of direct or unmediated ex-
perience. It seems plausible, therefore, to read the philosopher’s encounter with
the form of the good as a kind of direct acquaintance.
This view is extremely helpful in the present discussion, in that this sort of
direct acquaintance might provide the philosopher with a prima facie reason for
thinking that what he encounters is, in fact, the true nature of goodness. It may
be that the form of the good appears to the philosopher in such a way that he is
immediately justified in accepting it as true. It stands before his mind, much in
the way of sense-data, so immediately that he has every reason to believe his
apprehension is accurate. To return to the metaphor I adopted at the end of the
last section, it would appear that standing in the relationship of direct acquain-
tance to the form of the good might serve as the “thread” from which the philo-
sopher’s system of beliefs, the mobile, is suspended from the ceiling. His direct
acquaintance with goodness itself, the fact that it is present before his mind in
a direct or unmediated way, justifies the philosopher in accepting his apprehen-
sion of the form as true and accurate.
If, however, the philosopher’s direct acquaintance with the form of the good
is what justifies his knowing it, then we might wonder why he must also have a
logos. Plato is clear that in order to count as knowing the good, the philosopher
must have an explanation, a definitional account, of what it means to be good
in every context. The question is, what function does this logos serve on the
direct acquaintance view? I argued above that, in the case of the philosopher’s
beliefs about mathematics (and other disciplines), having a logos, or explana-
tion, is what justifies a thinker in holding a belief as true. But, if the direct ap-
prehension view is correct, the logos or explanation of goodness is importantly
different. Although the philosopher must have a logos of goodness – he must be
able to explain what is good in every context – this is not sufficient for justifica-
tion. Knowledge of goodness also requires his direct apprehension of the form
of the good. According to this view, the philosopher comes to have knowledge
of the form of the good through direct acquaintance. He is justified insofar as
he has a kind of unmediated apprehension of the form, and this unmediated
apprehension furnishes him with the ability to discern what is good in every
context.
If this is correct, if the philosopher is directly acquainted with goodness,
then he need not appeal to some additional logos or explanation in order to
account for what is good in every context. Instead, the philosopher’s direct ac-
quaintance with the form of the good will provide him with a means of discern-
ing what is good in particular situations. His direct apprehension of goodness

Authenticated
enables him to test whether any object or action, whether it be a mathematical

proof or a proposed course of action, is in fact good, appropriate, or fitting. The
philosopher will be able to determine whether there is a kind of “fit” between
the objects under his consideration and the form of the good, as it stands before
his mind. This view is made more plausible by thinking about the form of the
good as an entity, which embodies all that is correct and beautiful (ὀρθῶν καὶ
καλῶν) (517c). If we think of the good as an entity that can stand before the
mind, a concept that conveys correctness and beauty in a manner that the phi-
losopher can entertain (much in the way that Russell entertains the universal
concept of yellow), then rather than appeal to some further explanation in order
to justify how he knows what is good in every context, the philosopher might
simply appeal to the form of goodness itself, measuring, as it were, the extent
to which an object or proposed course of action compares to this entity, the
perfect conception of goodness, correctness, and fittingness.
This proposal is strengthened by the fact that we find evidence of the philo-
sopher’s ability to “measure” goodness throughout the middle books of the Re-
public. Where Plato sets out the qualities of the philosopher, he remarks that he
would make an ideal ruler in that he is naturally endowed with the capacity of
“measurement” and “grace” (ἔμμετρον καὶ εὔχαριν) (486d). In his exposition of
this claim, Plato reports that the philosopher will be naturally attuned to the
extent to which all things express ideal qualities (486e). I take it that what Plato
has in mind here is something similar to the view I have expressed above: the
philosopher will make an especially apt ruler in that he is able to measure all
that he encounters against the ideal nature of the forms, in order that he might
determine whether and to what extent the objects of his consideration are good,
true, fitting, just, beautiful, etc. In Book X, we find a similar view: Socrates
argues that the logical part of the soul is superior to the appetite and spirit in-
sofar as it is the part which “trusts in measurement and calculation” (μέτρῳ καὶ
λογισμῷ) (603a). The logical part of the soul is superior in that it can detect the
extent to which certain objects, propositions, or actions resemble the ideal
forms.
This view is also supported by the fact that, in Books V–VII, Plato repeat-
edly stresses the explanatory nature of the form of the good. We are told several
times that the good is what “provides truth and understanding” to the intelligi-
ble world (508e, 517b, 532b). At 517b, it is this feature – the fact that form of the
good is what “produces both light and its source in the visible realm” and “in
the intelligible realm controls and provides truth and understanding” – which
enables the philosopher to act sensibly both in public and private matters. I
take it that the form of the good provides the philosopher with a way of deter-
mining the correctness or fittingness of anything and everything that he comes

Authenticated
across, whether it be a mathematical proof, a poem, or a course of action. The

form of the good, insofar as it is the perfect embodiment of correctness or fitt-
ingness allows the philosopher to measure any and every definition, every ac-
tion, every object against it in order to determine the extent to which the thing
in question is a good or fitting one.
Because the form of the good is the source of all intelligibility, because it is
what enables the philosopher to determine what is good in all pursuits, it seems
reasonable to conclude that it might also serve as the ultimate account or expla-
nation as to whether the philosopher’s individual accounts of goodness, in par-
ticular situations, are in fact good or fitting. By looking to the form of the good
in all contexts, the philosopher can confirm that he is on the right track. This is
the case even if the question is about goodness itself. If I am correct, if the form
of the good serves as the ultimate explanation of the way in which the philoso-
pher knows what is good or fitting in every situation, then the regress problem
can be put to rest once and for all. In order to justify his definitional account of
goodness, the philosopher must have direct apprehension of the form of the
good. He must have an account (logos) of what it means to be good in all con-
texts, from geometry to public affairs, but this is not problematic in that in order
to justify his account of goodness, the philosopher need not appeal to anything
other than the form of the good itself. Looking to the form of the good, through
a kind of measurement or comparison to its ideal nature, the philosopher is able
to determine whether any object or proposed action is a good or fitting one.
In Books V–VII the philosopher’s search for explanations ends with his ap-
prehension of a first principle, yet the structure of his justification is never made
explicit. Implicit foundationalism, even if it is not a satisfying philosophical ac-
count, is nevertheless exceedingly loyal to the text of the Republic. The above
solution to the regress problem is, therefore, intended only as an appendix to
Plato’s theory of knowledge, an attempt to make the structure of justification in
the philosopher’s final ascent a more convincing and complete brand of founda-
tionalism. To this end, I have suggested that the philosopher has something
roughly analogous to what others have called “direct acquaintance” of the form
of the good. If I am correct, then the philosopher might appeal to the form of
the good itself in order to determine whether and to what extent his individual
accounts of goodness are fitting. This type of measurement is what allows him
to determine the truth, accuracy, fittingness, beauty and goodness of all things.
Although Plato may not have held this solution to the regress problem with any
deliberateness, it is not without textual virtues. Thinking about the form of the
good as an entity of sorts, against which the philosopher might measure the
objects of his consideration provides us with a tidy solution to the regress pro-
blem, but it also helps us to make some sense of a number of highly rarefied

Authenticated
passages in the dialogue. For example, if we think of the form of the good as a
kind of measure against which the philosopher might set any object under his
consideration, it makes perfect sense that it should be what “controls and pro-
vides truth and understanding” to both the material and immaterial world. The
form of the good is what allows the philosopher to say whether and to what
extent something under his consideration is good, fitting, or appropriate,
whether it be a proof, a work of art, or a course of action for his fellow man.
Works Cited
Aikin, Scott. ‘Who is Afraid of Epistemology’s Regress Problem?’ Philosophical Studies. Vol.
126, no. 2, (2005) 191–217.
Alston, William P. ‘Has Foundationalism been Refuted?’ Philosophical Studies. Vol. 29, no. 5,
(1976) 287–305.
Aristotle. ‘Posterior Analytics’ in Barnes, Jonathan ed. The Complete Works of Aristotle: The
Revised Oxford Translation, Princeton, NJ, Princeton University Press, (1984).
Bailey, D. T. J. ‘Plato and Aristotle on the Unhypothetical’ Oxford Studies in Ancient Philoso-
phy. Vol. 30, (2006) 101–104.
BonJour, Laurence. ‘The Structure of Empirical Knowledge’ Cambridge, Harvard UP, (1985)
18–19.
BonJour, Laurence. ‘In Defense of Pure Reason: A Rationalist Account of A Priori Justification’
Cambridge, Cambridge UP, (1998).
Burnyeat, M. F. and Barnes, Jonathan. ‘Socrates and the Jury: Paradoxes in Plato’s Distinction
between Knowledge and True Belief’ Proceedings of the Aristotelian Society, Supplemen-
tary Volumes. Vol. 54, (1980) 173–206.
Burnyeat, M. F. and Levett, M. J. ‘The Theaetetus of Plato’ Indianapolis, IN, Hackett, (1990).
Cornford, F. M. ‘Mathematics and Dialectic in Republic V-VII’ in Allen, Reginald ed., Studies in
Plato’s Metaphysics, London; New York, Routledge, (1965) [1932] 61–95.
Crombie, I. M. , ‘An Examination of Plato’s Doctrines’ New York, Humanities Press, (1962).
Cross, R. C. and Woozley, A. D. ‘Plato’s Republic; a Philosophical Commentary’ New York, St.
Martin’s Press, (1964) 230–61.
Fine, Gail. ‘Knowledge and Belief in Republic V–VII’ in Fine, Gail ed., Plato 1: Metaphysics
and Epistemology, Oxford, Oxford UP, (1999) 215–75.
Fine, Gail. ‘Knowledge and Logos in the Theaetetus’ The Philosophical Review. Vol. 88, no. 3,
(1979), 366–397.
Fine, Gail. ‘The Oxford Handbook of Plato’ Oxford, Oxford UP, (2008) 181–182.
Gentzler, Jyl. ‘How to Know the good: The Moral Epistemology of Plato’s Republic’ The Phi-
losophical Review. Vol. 114, no. 4, (2005) 469–96.
Gonzalez, Francisco J. ‘Propositions Or Objects? A Critique of Gail Fine on Knowledge and
Belief in Republic V’ Phronesis. Vol. 41, no. 3, (1996) 245–275.
Gonzalez, Francisco J. ‘Dialectic and Dialogue’ Northwestern UP, (1998) 209–44.
Guthrie, W.C.K. ‘A History of Greek Philosophy, Vol.5: The later Plato and the Academy’ Cam-
bridge, Cambridge UP, (1978), 122–60.

Authenticated
Irwin, Terence. ‘The Middle Dialogues: Criticisms of Socratic Ethics; 10.5’ in Plato’s Moral
Theory; the Early and Middle Dialogues, Oxford, Clarendon Press, (1977), 133–159.
Klein, Peter. ‘Human Knowledge and the Infinite Regress of Reasons’ Nous. Vol. 33, Supple-
ment: Philosophical Perspectives, Vol. 13, Epistemology, (1999) 297–325.
Mackie, J. L. ‘Ethics: Inventing Right and Wrong’ Harmondsworth, NY, Penguin, (1977) 38–9.
Moore, G. E. ‘Principia Ethica’ Cambridge, Cambridge UP, (1903 [2000]).
Plato. Republic: trans. Grube, G. M. A. and Reeve C. D. C. in Cooper, John and D. S. Hutchin-
son eds., Complete Works, Indianapolis, Hackett Publishing, (1997).
Pryor, James. ‘The Skeptic and the Dogmatist’ Nous. Vol. 34, no. 4, (2000) 517–49.
Quinton, Anthony. ‘The Nature of Things’ London, Routledge, (1973) 119.
Robinson, Richard. ‘Plato’s Earlier Dialectic’ 2nd Ed., Oxford, Clarendon Press, (1953) 172–3.
Ross, W. D. ‘Plato’s Theory of Ideas’ Oxford, Clarendon Press, (1951) 54–5.
Russell, Bertrand. ‘Knowledge by Description and Knowledge by Acquaintance’ Mysticism and
Logic, Barnes & Noble Books, (1917, Reprinted 1951) 152–167.
Ryle, Gilbert. ‘The Concept of Mind’ London, Hutchinson’s UL, (1949) 107–10.
Santas, Gerasimos. ‘The form of the good in Plato’s Republic’ in Fine, Gail ed., Plato 1: Meta-
physics and Epistemology, New York, Oxford UP, (1999) 247–74.
Sayre, Kenneth M. ‘Plato’s Literary Garden: How to Read a Platonic Dialogue’ Notre Dame,
University of Notre Dame Press, (1995) 177–81.
Sextus Empiricus. Outlines of Pyrrhonism, Cambridge, Harvard UP, (1976).
Uglietta, J. ‘Knowledge and Coherence in Plato’s Republic’ Apeiron: A Journal for Ancient Phi-
losophy and Science. Vol. 39, no. 4, (2006), 331–340.
Vlastos, Gregory. ‘The Socratic Elenchus’ The Journal of Philosophy. Vol. 79, no. 11, 79th An-
nual Meeting of the APA, Eastern Division, (1982), 711–714.

Authenticated

Nally, E. G. - Is Plato A Coherentist The Theory of Knowledge in Republic V-VII

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Nally, E. G. - Is Plato A Coherentist The Theory of Knowledge in Republic V-VII

Enviado por

Direitos autorais:

Formatos disponíveis

apeiron 2015; 48(2): 149–175

Edith Gwendolyn Nally

Keywords: Plato, foundationalism, form of the good, coherentism, Republic

In what follows, I argue that Plato is a not a coherentist about justification in

Brought to you by | Yale University

I The Regress Problem

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

relationship to sensible particulars. It is at this point that he realizes that the

Brought to you by | Yale University

II The Republic Analogies

The Divided Line

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

principle in such a way that it will be “secure” or “confirmed” (ἵνα βεβαιώσηται)

Brought to you by | Yale University

We are now in a better position to examine the philosopher’s ascent and

Brought to you by | Yale University

knowledge of the first principles of geometry, establishing them by some logical

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

III Implicit Foundationalism

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

IV A Platonic Solution: Direct Apprehension

The philosopher must be able to give an account, which distinguishes what is

Brought to you by | Yale University

The Regress Problem Revisited

Brought to you by | Yale University

sophic converse, suddenly, as if from a leaping spark, a light is kindled, which,

Brought to you by | Yale University

Brought to you by | Yale University

Although I am generally reluctant to put much stock in these metaphors, it is

Brought to you by | Yale University

enables him to test whether any object or action, whether it be a mathematical

Brought to you by | Yale University

across, whether it be a mathematical proof, a poem, or a course of action. The

Brought to you by | Yale University

Brought to you by | Yale University

Brought to you by | Yale University

Você também pode gostar