The aim of N.-O.'s book is to provide a re-evaluation of the first chapter of Aristotle's De interpretatione that reconstructs its theory of signification and the psychological foundations thereof. It also challenges the traditional interpretation, according to which the chapter puts forth a semantic theory in which linguistic expressions relate first to thoughts and second to things, where thoughts are understood as pictorial representations of things. N.-O.'s alternative interpretation denies that thoughts are here pictorial representations and proposes that they are intellectual items that grasp the primordial aspect of things, but not in a pictorial manner. Linguistic expressions represent thus the primordial aspect that thoughts share with the things they are representations of. N.-O. proposes that his reading is not only more faithful but also philosophically stronger, not least because it prevents Aristotle's theory of signification from falling under ‘semantic representationalism’.

The book is divided into five chapters: the first contains a discussion of the term ϕωνή; the second contains a discussion of the terms σύμβολον and σημεῖον; the third and fourth are discussions of παθήματα τῆς ψυχῆς, ὁμοιώματα and νοήματα; and the fifth concludes with a discussion of the conventional relation (κατὰ συνθήκην) between linguistic expressions and their significates.

The discussion of the first chapter is based on a careful analysis of the relevant passages in De an., Hist. an. and Part. an. Its main conclusions are (i) that vocalised sounds in Aristotle are not to be dissociated from communication, for they imply a link to the psychological content that they intend to communicate; (ii) that it is the faculty of phantasia and not of sense-perception that is responsible for the production of vocalised sound; and (iii) that human beings are the only animal species that can engage in deliberation, discussion and debate through the linguistic expression of their thoughts.Footnote ¹

The second chapter presents a philological analysis of the terms σύμβολον, σημεῖον, σημαίνειν and σημαντικόν that leads N.-O. to the conclusion that Aristotle uses the first two terms interchangeably. We are also told that a linguistic expression signifies a psychological item that is a unity. This unity, however, does not need to be an ontological unity, nor does it need to correspond to something that exists in the external world; it only needs to be a logical unity – it needs to be susceptible of account or nominal definition (e.g. the goat-stag which, although does not exist in reality, is susceptible of nominal definition). I could not agree more with this second claim. As to the first one, it is the qualification of σημεῖον as σύμβολον in Int. that makes them synonymous in this particular context. They are not generally synonymous, however, and the qualification of signs as symbols in this context is not superfluous, but has the very important implication of narrowing the signs with which Aristotle is here concerned to only conventional signs.

The chapter about thoughts (νοήματα) aims at highlighting that in Int. 1 Aristotle is using νοήματα in a general sense that embraces opinions, calculations, knowledge, universals and principles. N.-O. also shows that this general use of the term is coherent with its use in De an., which also justifies the introduction of the account of νοήματα in De an. into the interpretation of the first chapter of Int. This general sense of νοήματα embraces a wide range of logico-psychological unities and explains Aristotle's consideration of a wide range of linguistic expressions that includes simple terms, compound expressions, assertions and syllogisms.

Chapter 4 is the cornerstone of the book's main argument. The chapter starts with the important and well-argued point that the qualification of thoughts as likenesses (ὁμοιώματα) does not amount to their qualification as pictorial images – thoughts are basically intellectual items that grasp the form(s) of things, but not in a pictorial manner. More problematic, though, is the reading of ὣν … πρώτων in Int. 1.16a6 and its justification. I agree that here πρώτων should not be read as a qualification of a relation between linguistic expressions and thoughts that implies a second one between linguistic expressions and things. Nevertheless, N.-O.'s crucial claim that πρώτων should be read, not in an adverbial manner, but as an adjective that qualifies thoughts, so that it indicates primordial thoughts (i.e. a particular kind of thought that grasps the primordial aspect of things) seems far-fetched; for it is weakly justified with other passages in the Aristotelian corpus where there is mention of such primordial thoughts. We are not given conclusive evidence that Aristotle is here exclusively referring to some primordial thoughts that contain the primordial aspect of things. I would say that, rather than something that is primordial, this πρώτων aims at indicating that what is primarily targeted by the intention of communication of a speaker are her thoughts. Hence, it should be read adverbially. The primordiality of the signification of thoughts is more simply (and also more correctly) explained by the fact that this passage presents us with the claim that the main purpose of the production of human language is the accurate representation of thoughts or beliefs about things (accurate in the sense that linguistic expressions preserve fundamental logical features of the thoughts that they represent, such as their compositional form and their epistemic modality).Footnote ² Although I agree that the realm of language in Int. is the human intellect, I also think that the justification of the claim that linguistic expressions signify the primordial formal aspect of things through the signification of their thoughts is in need of stronger support. Where we disagree, then, is in the weak emphasis that N.-O. puts on the effectiveness and accuracy of linguistic communication as the notion that leads the development of Chapters 1–5 of Int.

The book ends with a chapter about κατὰ συνθήκην and concludes that the signification at stake in Aristotle's account is conventional, in opposition to some ancient accounts that consider linguistic signification as a natural phenomenon.

The volume presents a thorough analysis of the key terms of Aristotle's Int. 1, particularly of the controversial passage 16a3–9. Most of the book's conclusions are accurate and well argued. This book will thus become a useful companion to readers of this elliptic and elusive chapter. It also presents a pleasant combination of concision, philological analysis and philosophical insight. However, an analysis of this chapter that takes into consideration the rest of De interpretatione and other treatises of the Organon (e.g. the Topics and the Sophistical Refutations) is regrettably absent. The book has more typos than one would reasonably expect.