Reviewed by Joachim Aufderheide, University of St Andrews (firstname.lastname@example.org)
One of the major difficulties posed by Plato's dialogue Philebus is its apparent lack of coherence, nicely summed up by Rodier: 'one could maintain. . .that the Philebus was nothing but an assemblage of pieces originally distinct and devoid of organic unity.'[] Another major problem is pinpointing the elusive aim of the dialogue, as can be seen in Damascius' introduction to his commentary on the Philebus: the dialogue could be about pleasure, intelligence (nous), the final cause of the universe, or the good which is present in all animals, from the most godlike to the lowest.[]
In his Le Philèbe de Platon: introduction à l'agathologie platonicienne Sylvain Delcomminette provides solutions to both problems by first identifying the specific goal of the Philebus and then showing that the dialogue coherently moves towards this goal. The Philebus offers something at which other dialogues, most prominently the Republic, only hint. The Republic offers a glimpse of the Form of the Good based on a description of its function, but the reader is not told what the Good is (see esp. Rep. 506d8-e7). Only the Philebus, or so Delcomminette claims, enables the reader to find out what the Good essentially is: its essence (logos) is revealed (13). He calls this project 'agathologie', i.e. the science of the good (12), and claims that this science can only be dialectics, the only real science according to the Republic (Rep. 533c7-e2). The Philebus' particular structure is explained by the requirements of dialectics: the text provides the platform for the dialectical movement requisite for uncovering the Form of the Good.
These claims receive strong and well argued support in Delcomminette's 613 pages of commentary-cum-interpretation of the Philebus. Prefaced by a helpful five-page table of contents and a nineteen-page introduction, the book is divided into three parts, following the apparently tripartite composition of the Philebus. The first part (289 pp.) interprets Phlb. 11a-31b where the problem is formulated and the proper method of solving it is developed. This method, dialectics, is put to work in the second part of the book (250 pp.), which is concerned with the general account of pleasure and true and false pleasures as well as a classification of sciences (Phlb. 31b-59d). This enables the reader to see in the third part (100 pp.) what the Good consists in, with a commentary focusing on Phlb. 59d-67b. The book is further subdivided into 17 chapters which correspond mostly to what Delcomminette takes to be sections of the dialogue.
Despite its name Delcomminette's 'introduction to agathology' is not aimed at beginners. First, the Philebus remains a difficult dialogue despite all the insights contained in Delcomminette's book. Secondly, at times those are themselves hard to understand. This is due to both the high demands Delcomminette makes on his readers, and to the complexity of the book. The reader is asked to suspend judgement about the validity of the interpretation until the end of the book. This is not a modest request, for the reader is expected not only to read the whole book (637pp.), but also to engage actively in the reconstruction of all points and their internal connection (17-8; I will say more about this in the next paragraph). This requires time, energy and, I am afraid, a high level of tolerance for frustration. Two reasons for this stand out most: i) the book contains a host of philosophically substantial arguments which can stretch effectively over some 200 pages - the most interesting of which is perhaps the argument against eudaimonism, culminating in Chapter X (see esp. 504-5); ii) Delcomminette's argument is sometimes marred by reticence, paradoxically for a book of this length, rather than loquacity in explaining the preferred terminology. (This might be due to its being a revised and abridged (!) doctoral thesis which has been prepared for publication in less than three years.)
I will now turn to some details, beginning with the methodological requirement that the reader should participate in the philosophical development of the discussion between Socrates and his interlocutors. This requirement follows neatly from the first part of the book which develops a method for finding out what the Good is, namely the method of dialectics. The frequently novel and always illuminating interpretations of notoriously difficult passages such as 'the one and the many' (Phlb. 15a ff.), 'the god-given method' (Phlb. 16c-18e), and 'the fourfold division' (Phlb. 23b-27c) result in a perhaps non-standard conception of dialectics. Very broadly, Delcomminette understands dialectics as the (intellectual) movement of determining what is previously undetermined (cf. 536). The goal of dialectics is later summed up (528-30) as grasping the content of a Form.[] If this is the only method of grasping Forms (as is argued in the introduction, 1-19), and if the goal of the Philebus is to reveal the essence of the Good, then, of course, the reader cannot understand the dialogue fully unless he or she engages in dialectics.
This straightforward point is, unfortunately, not backed up by a straightforward method for interpreting the dialogue. Delcomminette's appeal to 'the method of an internal commentary which avoids approaching the text from an external perspective' (17) is ineffective because the contrast here between internal and external perspective is unclear.
'External perspective' could be understood here as 'the point of view of another philosophy' or (what probably means the same) 'would-be "objectivity"' (17). None of this helps. Leaving aside the pejorative force of 'would-be', why should the "objective" standpoint be such a hindrance for interpreting a Platonic dialogue? After all, Delcomminette understands that interpreting a Platonic dialogue requires doing philosophy. Yet one crucial aspect of doing philosophy is, as Plato would be the first to admit, a concern for truth. But how can one aim at truth without trying to be objective? Coherence alone does not entail truth. So it is not enough to consider merely the coherence of the Philebus. Rather, the interpreter should step back from the dialogue and consider whether the points in the dialogue are true, and if not, why not. Note that 'stepping back' is both incomplete and a metaphor. Yet once the metaphor is complete, it can reveal something about Delcomminette's approach. The relevant logical form of 'stepping back' here is 'stepping back from X to Y' where X ranges over the Philebus or passages taken from it. It might seem as if Delcomminette wants to restrict Y only to other passages of the Philebus: this, at least, is the most natural understanding of 'internal' commentary. According to another understanding of 'internal' one could restrict Y such that it ranges only over Platonic dialogues. There is reason to doubt, however, that these restricted ranges of Y leave room for doing philosophy. This doubt is fueled by Rodier's remarks on the Philebus that an internal commentary (where Y ranges over the Platonic corpus) does not entail or presuppose a concern for truth: 'we are not concerned with knowing whether the doctrine at hand is true, but whether it is Platonic.'[]
Here is not the place to discuss in detail the question of how to interpret a Platonic dialogue. The relevant question here is how Delcomminette deals with the tension between the requirements for doing philosophy on the one hand, and giving a purely internal commentary on the other. Much to the book's advantage, it is not restricted to a purely internal commentary in either of the two senses above. Delcomminette interrupts the flow of the argument frequently to pause over the plausibility of the theses he discovers in the Philebus. This is obviously not a novel approach, but Delcomminette does well in avoiding irrelevant points. His reflections are frequently not introduced by means of the concerns of modern or contemporary philosophers, but by means of objections Plato's contemporaries could have leveled. Examples are discussions of Aristotle's objection in De Memoria that Plato's theory of memory is too static (322-4) or a Calliclean objection that living a happy life in Socrates' sense amounts to being as dead as a stone (501-3). This seems to be a good way of trying to discern what Plato meant or how a particular argument works. Incidentally, this way of assessing the philosophical merit of Plato's argument provides a sense of the domain over which Y could range so that stepping back from the original text allows for doing philosophy without importing misleading concerns. Whether the interpreter can arrive at truth through this kind of 'objectivity' is a question for another occasion.
Delcomminette is not, however, always faithful to the requirements of a 'philosophical' "philosophical" interpretation of the Philebus. Rodier's approach seems to be unexpectedly close when Delcomminette confesses that he has 'tried to be Platonic' in his interpretation of the Philebus (19). By this he means that one's attempt to understand the Philebus is best helped by reading more Plato and drawing connections between the dialogues, a feat remarkably well done by Delcomminette. Consequently, the book contains a plethora of illuminating points on dialogues other than the Philebus, so that this book is worth studying for those who are interested in Plato's philosophy in general. There are, however, some disadvantages to this approach which will become clear through the discussion of a problem which does not primarily concern methodology.
The problem stems from one of the main themes in the Philebus: determining the relationship between pleasure and the good life. The hedonist interlocutor seems to hold (at some point) that pleasure is sufficient for a good life, whereas Socrates seems to contend that pleasure is not even necessary. To find out who is on the right track, one must become clear about pleasure. What is pleasure? It is not difficult to see that the place of pleasure in the good life depends on the theory of pleasure. If pleasure were only a remedial good, e.g. the alleviation of pain, then Socrates' thesis seems more convincing that the hedonist's. Yet if pleasure were something that is good not only because it brings about another good, Socrates' claim loses support. A major task, then, for the interpreter of the Philebus is to clarify the account or rather accounts of pleasure. For there is good reason to believe that the Philebus distinguishes different kinds of pleasure and that they play different roles in the good life accordingly. So, before the interpreter can give the account of pleasure, he or she must argue that Socrates seeks to unify pleasures under this account. That this is not an easy task becomes clearer from considering these four propositions:
A) Pleasure is a coming-to-be (Phlb. 54c6);
B) If X is a coming-to-be, then X is not in the class of the good (54c6-11);
C) Pleasure is not in the class of the good (54d1-2, from A and B)
D) Some pleasures are good (indicated by the appearance of certain pleasures on the list of things which possess features of the Form of the Good; see 65a1-5 for the features, 65d4-6 for the introduction of the list, and 66c4-6 for the occurrence of pleasure).
The context of A) to C) is a general criticism of pleasure (Phlb. 53-55), whereas D) is taken from the final ranking of what makes life good. There seems to be a tension between C) and D) which becomes more apparent by substituting 'good' in C) and D) to highlight the fact that the referents are similar on both occasions. So, 'class of the good' could be rendered as 'good in itself', and there is reason to believe that things which possess features of the Good are also good in themselves (at least Delcomminette does not deny this). There is thus a tension between
C*) Pleasure is not good in itself, and
D*) Some pleasures are good in themselves.
A ready solution to this problem is to argue that there are two accounts of pleasure in the Philebus: C*) does not mean that all pleasures are not good in themselves. Rather the scope is limited to those pleasures which are cases of coming-to-be, but not all pleasures are cases of coming-to-be. There is a distinction, one could argue, between the pleasures of replenishing, e.g. eating and drinking, and other pleasures such as seeing, smelling or thinking (Phlb. 50e-53c). The former are clearly comings-to-be, whereas the latter are not. Since the text says that only the latter pleasures are good (Phlb. 66c4-6), there is not inconsistency between D*) and C*).[]
Delcomminette accepts the challenge, but rejects the proposed solution. He contends that A) means that all pleasures are cases of coming-to-be. With B), this would imply that no pleasure is good in itself. If this is the correct interpretation, Delcomminette must explain:
1. how the pleasure of seeing or thinking can be a coming-to-be;
2. the fact that the passage Phlb. 53-55 is concerned with all pleasures;
3. why some pleasures are good.
In the remainder of this review I will outline and assess how Delcomminette's deals with these tasks in order to indicate how Delcomminette tackles difficult questions in interpreting the Philebus. As regards 1, it seems to be an advantage if one is not pressed to subsume the pleasures of seeing and thinking, the so-called 'pure pleasures', under the heading of 'coming-to-be'. The pleasures of eating and drinking, so-called 'impure pleasures', are clearly comings-to-be because they are, at root, restorative processes. It is not obvious (and sounds false) that e.g. seeing should be a restorative process too. Yet Delcomminette has no qualms here: all pleasures are explained as felt restorations, the difference between the two kinds being only that the lack is felt in impure pleasures, but not in pure pleasures. Consequently, no pleasure is good in itself (452-3). This claim is, in my view, not well supported. A first warning prefaces the discussion of pure pleasures (Phlb. 50e-53c), 'to achieve a complete and coherent interpretation ... comparison with other dialogues is sometimes necessary' (455). Given the brevity of the passage in question this method seems appropriate. It is inappropriate, however, to use points from other dialogues as premises to reach the desired conclusions - which is what Delcomminette does. Accordingly we learn (452-3 n.3; 456) that the pleasures of smelling are to be understood as replenishment because a similar picture is given in the Timaeus (64b-65d). This seems to be a weak counter to the argument which aims to rule out the possibility that pleasures of smelling can be subsumed under replenishment-type pleasures in the Philebus. Of course one could argue that in this case importing a point from another dialogue is warranted -- but Delcomminette does not take the opportunity to do so. More material for understanding the pleasures of smelling is provided in Delcomminette's discussion of the second kind of true pleasures, dubbed 'aesthetic pleasures' (457). Yet again, as some points from the Symposium are imported here, the initial worry remains that some problems are too readily explained by reference to an (alleged) overarching Platonic doctrine. This is the disadvantage of the attempt to be Platonic to which I alluded earlier.
Even if it is not clear how all kinds of pleasures could be at issue in Phlb. 53-55, Delcomminette provides an interesting argument which turns on considerations about the structure of the dialogue to support the claim that all pleasures are meant (cf. 2.). Unlike e.g. D. Frede,[] Delcomminette argues that the method of collection and division is applied to pleasures. It is characteristic for this method to bring the divisions made in a certain domain under one heading, whether we begin with a) a unit and then understand how it is many, or begin with b) multiplicity and unify later on (Phlb. 16c-d and 17c-18b). If the method of collection and division applies to pleasure and if b) is indeed the right way of thinking about pleasures, then Delcomminette seems right in holding that 'it is completely natural that Socrates should come back to pleasure in general in the course of his analysis, for we have repeatedly pointed out that the inverse way of the god-given method [i.e. b)] implies that the unity of a kind cannot be fully determined before the division into species is complete.' (493). This would support the claim that at the end of the discussion of pleasure, at Phlb. 53-55, all pleasures should be meant, and hence that all pleasures share the feature of being only cases of coming-to-be. This, Delcomminette claims, is the generic account of pleasure (498). Unfortunately, this argument is undermined by Delcomminette's consideration in support of 3.
Delcomminette accepts that only something which has being (ousia) can earn a rank on the list of the goods (cf. 500). He also accepts the challenge to explain the appearance of pleasure on this list despite the fact that pleasure is a coming-to-be. In outline, his solution is that certain pleasures, the pure pleasures, have a 'derived being' which in turn allows these pleasures to appear on the list of goods (500). Suppose the obvious questions could be answered (What does 'derivative being' mean? Why do these pleasures have derivative being? If it is constitutive of pleasures to belong to the category of coming-to-be, how can they have being? ); s uppose the argument would work. In this case it would undermine the reasoning for 2. by undermining its presupposition that prior to Phlb. 53-55 no generic account of pleasure is given.
If at Phlb. 53-55 a generic account of pleasure were given, then all pleasures would have to be subsumed under this account. In particular, those pleasures of 'derived being' would not prove exceptions from the generic account. Rather, that they have only 'derived' being seems to suggest that they do not belong to the category of being, but still to that of coming-to-be. If this point about 'derived' being carries over, Delcomminette's interpretation faces problems. Think of the other potential generic account of pleasure, that pleasure belongs to the ontological category of the unlimited (cf. Phlb. 31a6-10). It seems that this can be seen, with equal right, as a generic account of pleasure. It is true that the text suggests otherwise since unlimited pleasures are equated with unmeasured pleasures, measured pleasures with limited pleasures(Phlb. 52c1-d1), so that there would seem to be two kinds of pleasures, limited and unlimited. Yet this suggestion can be refuted by material which Delcomminette uses in his argument for 2. One could argue that the limit some pleasures have is only derivative. For at Phlb. 28a1-3, Socrates points out that we should look to something other than the character of being unlimited if there is anything good about pleasure. Together with the reminder at Phlb. 31a7-10 that pleasure in itself is unlimited, it seems that only the addition of something else can make pleasure both measured and good (measure or measuredness is one feature of the Good: see Phlb. 65a1-5). If, however, the limit pleasures can have is only derivative, then it is fair to say that a generic account of pleasure as unlimited is given already in Phlb. 24-31: there pleasure is unlimited. Analogous to the use of 'derived being' in Delcomminette's argument, here too pleasure's having 'derived limit' does not rule out pleasure's belonging to the category of the unlimited. If so, it is not clear why Socrates should provide another generic account of pleasure at Phlb. 53-55. Hence, the presupposition for Delcomminette's reasoning to the conclusion that all pleasures are at issue at Phlb. 53-55 seems to be undermined.
The reader is likely to disagree with Delcomminette on particular points of the interpretation -- for a book of this scope this should not come as a surprise. Yet, in my opinion, this disagreement does nothing to diminish the book's value. Delcomminette's Introduction to agathology is a thought-provoking contribution to the study of Plato's philosophy -- even if the reader eventually fails the course in agathology, i.e. fails to grasp the Form of the Good (as was my case). Apart from the felicitous way of illuminating the philosophical merit of Plato's arguments, the most impressive feature of the book is perhaps its outstanding scholarship: adherence to two research maxims 'Read as much as possible!' and 'Don't be stuck in a tradition!' (cf. 19) result in a sixteen-page bibliography of books and articles cited in which almost every searching reader will find previously unknown gems. Anyone with a scholarly interest in Plato's thought can benefit from working through Delcomminette's book. But in particular those who are interested in the Philebus are given a fantastic working tool, containing both a first-class discussion of the secondary literature on any controversial issue in the Philebus and Delcomminette's often novel solutions to intricate problems.
[] Rodier, G. (1900), 'Remarques sur le Philèbe, in (id.) Études sur la philosophie grecque (2nd ed. 1957), Paris: J. Vrin, 47.
[] Westerink, L. G. (1959), Damascius: Lectures on the Philebus wrongly attributed to Olympiodorus, with text, translation, notes and indices, Amsterdam: North-Holland Publishing Company, paragraphs 1-6.
[] The content of a Form seems to be what Nehamas calls 'essential properties of Forms'. See Nehamas, A. (1975), 'Plato on the imperfection of the sensible world', American Philosophical Quarterly (12):105-17.
[] Rodier 1900:97. It should be clear that the remark cannot alone suffice as a guideline for interpreting Plato. One would have to establish beforehand what counts as Platonic, a task which involves all sorts of problems not only about the transparency of the dialogues, but also about Plato's possible philosophical development or the possibility of his discussing different philosophical solutions to certain problems without necessarily endorsing any of them.
[] Cf. Carone, G. (2000), 'Hedonism and the pleasureless life', Phronesis (45): 257-83. See 262-70 for a similar argument.
[] Frede, D. (1997), Platon: Philebos, Übersetzung und Kommentar, Göttingen: Vandenhoeck & Ruprecht, 227.