Bryn Mawr Classical Review

Starting with Aristotle, there is a long tradition of investigating the origins/first principles (Greek ἀρχαί) of Plato's philosophy. Fritz-Gregor Herrmann innovates in this tradition by investigating the linguistic histories of key terms employed in Plato's philosophy. In this sense, then, the 'roots' of Herrmann's title refer both to the origins of Plato's philosophy, and to the etyma of the words that Plato used. It is a fundamental premise of Herrmann's study that Plato's language and, consequently, his thought grew out of a fertile field of semantic and lexical possibilities ranging from Indo-European roots in pre-Homeric oral traditions to the linguistic practices of his competitors. This premise functions as a gentle exhortation that scholars of ancient philosophy can produce more comprehensive and nuanced understandings of Plato's philosophy when they apply careful philological investigation in the process of reconstructing it. Herrmann's study of the key terms employed in Plato's metaphysics promises to show that Plato did not operate in a vacuum, and, moreover, it highlights Plato's activity of modifying pre-philosophical language to develop a linguistic system that approximates (at worst) or harmonizes with (at best) the concepts with which he was experimenting from the early period through the middle-late dialogue Parmenides, the terminus ante quem for this study. Little further investigation into the post-Parmenides dialogues occurs, although there is a promise of future work on the metaphysics of the Sophist and Statesman,¹ which would be a welcome sequel to this work.

The basic argument of Herrmann's study is that key terms in the Phaedo -- those associated with the "Theory of the Forms"² by those who attribute such a theory to Plato, including "form", "idea", "being", and "sharing" -- cannot be extricated from traditions of Greek history and culture as demonstrated in pre-Platonic texts of both philosophical and non-philosophical character. In constructing a semantic range for these terms, Herrmann overlooks few ancient authors, with recurrent appeal to not only the requisite Presocratic philosophers and Sophists (e.g. Anaxagoras, Empedocles, Democritus, Gorgias, even Diogenes of Apollonia) but also religious, literary, historical, and medical sources (e.g. Homer, Aristophanes, Herodotus, Thucydides, the Hippocratic Corpus). An examination of the range for these usages in pre-Platonic texts reveals the "semantic development"³ of the terms before Plato began to select and shift these semantic fields in order to fashion a language that would represent his philosophical propositions. This concept of "semantic development," as it is deployed alongside historical linguistics, is problematic: one illustrative example of this problem is the teleology involved in the examination of the history of the term ἰδέα, for which Herrmann's preferred translation is "appearance." Basing his study of this term on the Indo-European root *weid- (the same root for the congener εἶδος), Herrmann claims that "[o]n etymological grounds, one would expect ἰδέα...to refer to the totality of a visual impression as given or perceived in an instant."⁴ His reasoning is, from a philological point-of-view, apparently sound enough: it is possible (although by no means necessary) that the presence of the collectivizing/generic -a ending in a zero grade of the root may suggest these semantic characteristics.⁵ The problem with this approach is that, by relying on the Indo-European root and historical linguistics, Herrmann seeks to approximate historical occurrences of the term ἰδέα throughout pre-Platonic Greek literature to an unhistorical Ur-meaning that might or might not be built into the employment of the term. That is to say, Herrmann wants to argue that the term ἰδέα develops historically over time, but also that it has always had (or sought to have) these ideal semantic characteristics; these two claims are in tension with one another. As an issue of methodology, this tension is problematic: the notion that a "totality" is both visualized and impressed "in an instant" precludes the possibility that visualization and cognitive memory themselves function in a dialogical relationship, correcting one another in a negotiation that mirrors the process of meanings changing over time in the discursive fields of speech and writing. What this issue suggests to us, then, is that, if we grant that individual words may have a meaning that is stable in an historical present, we are still required to imagine language in its "root" sense -- one that is, for better or worse, a fantasy and empirically unattainable -- in order to establish an ideal to which that historical usage can approximate. The occurrence of this problem in Herrmann's book, however, provides scholars with an opportunity to consider further the possible ways in which we can integrate linguistics, history, and philology in the study of ancient philosophy; Herrmann has provided a promising, if not fully developed, model that deserves (and requires) further discussion.

Words and Ideas is organized into three parts: Parts I and II, each of which is subdivided into three chapters, deal with the histories of the key terms in Plato's metaphysics throughout Greek literature before the Phaedo; Part III, which is subdivided into five chapters, applies this information to a reading of Phaedo 95e-107b. In keeping with the encyclopedic project that Herrmann proposes -- at times, the book reads like a heavily glossed concordance -- each chapter deals with concepts as they are delegated to semantic fields, e.g. Chapter Two, which investigates the occurrences of the words παρουσία, παρεῖναι, and παραγίγνεσθαι in pre-Platonic literature and the early dialogues of Plato. The organization of the book is clear and allows the reader to approach the key terms in Plato's metaphysics and ontology by seeking out the word under investigation in the comprehensive and well-tabulated Table of Contents and turning right to the chapter that deals with that term.⁶ Herrmann treats as many significant terms as possible, including illuminating studies of terms not often examined, such as μορφή and κοινωνία.⁷ The painstaking work that Herrmann has done to contextualize these terms leads to several conclusions that complement those of recent studies published on Plato's early and middle Theories of the Form, most notably monographs by Dancy and Rickless.⁸ Herrmann is careful to distinguish the employment of words that were semantically related to Plato's metaphysics from dialogue to dialogue, focusing on the Meno, Euthyphro, Charmides, Gorgias and Phaedo. One would have appreciated a more extensive treatment of the Republic and Symposium, which would have required a dedicated response to Francesco Fronterotta's magisterial treatment of Plato's early Form Theory in his monograph ΜΕΘΕΞΙΣ: La Teoria Platonica delle Idee e la Participazione delle Cose Empiriche (Scuola Normale Superiore, 2001): Herrmann cites it only once and without serious comment or response.⁹ This oversight does not detract from the value of Herrmann's contribution as a "preliminary step towards a study of [Plato's] philosophy,"¹⁰ although the absence of reaction to recent (i.e. within the past 10 years) critical scholarship on Plato's Formal Theories does not help to make this book relevant to those scholarly traditions more firmly based in so-called "anglophone" or "analytical" criticism.

The payoff for Herrmann's study is the demonstration that Plato's metaphysical and ontological terminology was adapted from a mixture of pre-Platonic philosophical paradigms. While this conclusion is predictable, the specific details are novel and remarkable. Words and Ideas closes by suggesting that the most relevant influences on the development of Plato's language of the Forms in the Phaedo were the Pythagorean Philolaus of Croton, the atomist Democritus of Abdera, and Anaxagoras of Clazomenae. The significance of Democritus is perhaps most doubtful here, in part because the passages of the Phaedo that Herrmann adduces for comparison -- as he himself acknowledges -- are "Presocratic in a general sense"¹¹ and also share characteristics with both Anaxagoras and Philolaus, not to speak of Empedocles. For Philolaus and Anaxagoras, on the other hand, the arguments are highly persuasive and represent significant advances in our understanding of how Plato manipulated concepts that were already present in texts that were certainly known to Plato and available for his perusal. The Plato that results is a remarkable combination of Philolaic metaphysician and Anaxagorean physicist, with some influence of Democritean [or Archytan?] materialist.¹² In this sense, Herrmann's Plato is a bricoleur, a daring and selective linguistic artist whose adaptation of antecedent semantic paradigms problematizes the relationship between words and ideas and brings them into reconciliation in the process of (re)creating Greek philosophy. As Herrmann recognizes, Plato is not entirely consistent in his terminological usage, especially regarding words that we readily associate with "Form," but Herrmann also notes the semantic consistency of οὐσία ("Being" or, as Herrmann translates, "what something is") qua philosophical term through the early and middle dialogues.¹³ Further investigation into the significance of Being in the extant writings and testimonia of the Pythagoreans Philolaus and Archytas, whose concepts Plato adapted to suit his own philosophy, would be welcome.

I have found very few errors in Greek translation and none of transcription,¹⁴ and Herrmann has done well to provide the reader with as much Greek as is necessary to make clear the semantic value of the terms under investigation. It is a testament to the fine editing and proofreading that Words and Ideas, a work that features extensive Greek text and translation on every single page, possesses only a couple of minor typographical oversights.

Words and Ideas will be most useful to professors and graduate students who have an interest in the history of philosophy and in the language and concepts that underlie the structure of ancient metaphysics. What is more, anyone with a strong investment in the history of ideas will benefit greatly from Herrmann's documentation of the traditions of philosophical concepts as traced throughout non- or pre-philosophical Greek literature. Scholars of various disciplines other than Classics whose Greek is not strong will profit greatly from the methodical philological work that Herrmann has done. Given the linguistic knowledge that is required to navigate through Herrmann's arguments, only the most advanced undergraduates who have been trained in Greek, Latin, and (preferably) some Indo-European linguistics will find the book accessible. Readers without knowledge of these subjects will find this work to be a stimulus for further language study.

1.   P. 309 endnote 270.
2.   Herrmann is careful to avoid using "Form" throughout his book, preferring instead to leave those terms that refer to what we call "Forms" untranslated or to emphasize the alternative semantics such as "type" or "looks" (pp. 148-9), although the description of the book on the dust jacket does refer to the "Theory of the Forms."
3.   P. 148.
4.   See pp. 96, 147-8, 151, 163, 310 endnote 281.
5.   For "in an instant," Herrmann claims (p. 310 endnote 281): "if the function of the zero-grade of graded roots is just to state the verbal action, as witnessed in the aorist." It is perhaps the case that, at least in this endnote, Herrmann himself wishes to back away from so decisive a lexicalization: "We cannot be ultimately certain about the word's formation, and connotations can only be gleaned from the actual contexts in which the word occurs."
6.   Unfortunately, this design results in a significant oversight, namely the underdeveloped Index Rerum at the back of the book, which is frustratingly vague (to the point of being practically useless).
7.   Pp. 179-85, 254-62.
8.   R.M. Dancy, Plato's Introduction of Forms (Cambridge, 2004). Samuel C. Rickless, Plato's Forms in Transition: A Reading of the Parmenides (Cambridge, 2007). Neither work has been cited, although surely Rickless' book was still in press when Words and Ideas was completed.
9.   P. 334 endnote 472.
10.   P. 1.
11.   P. 240.
12.   The term ἰδέαι ("figures" or "types") occurs in the treatise ascribed to Archytas called On Law and Justice (Thesleff 1965: 34.10-11) and is there contrasted against εἰκόνες ("representations"), on which see C. Huffman, Archytas of Tarentum (Cambridge, 2005, p. 601). The significance of Archytas to the semantic range of ἰδέα or λογισμός ('calculation') in the Phaedo (e.g. 79a3), and more generally to the development of Plato's metaphysics, has not been fully examined by Herrmann (pp. 312-3 endnote 304, pp. 326-7 endnote 433).
13.   P. 273.
14.   Translations of Parmenides 127e1-4 are quite different on pp. 11 and 21; τῶν μεγίστων on p. 35 is untranslated; from time to time, Herrmann will translate certain terms that possess semantic ambiguities as "x and y" or "x or y" which leads to confusion, e.g. ἐξ ἐπιμελείας as "from exercising and practising" on p. 66 or ἐγγιγνομένῳ as "that which comes into, or comes into being in the soul" on p. 87; "Demokritos" for "Democritus" on p. 315 endnote 329; "Huffman 2006" for "Huffman 2005" on p. 327 endnote 433.