Harari’s study is the latest book-length investigation of the Aristotle’s Posterior Analytics ( APo). But where many recent predecessors — Goldin’s Explaining an Eclipse (Michigan, 1996), Ferejohn’s Origins of Aristotelian Science (Yale, 1991), and Kal’s On Intuition and Discursive Reasoning in Aristotle (Brill, 1988) — focus on a more or less limited set of issues (indicated by their titles), Harari (H) offers a comprehensive interpretation of the theory of demonstration, resembling in its ambition McKirahan’s Principles and Proofs (Princeton, 1992).
Unfortunately, the comparison with McKirahan points to the deficiency of H’s work. H claims to explore the theoretical relationship between Aristotle’s theory of demonstration and his syllogistic theory, proposing to do so from what she terms a metaphysical perspective based on the theory of substance. What is more, she promises not to detach Aristotle’s theory from its historical context, which she claims other studies do by assessing the theory from the standpoint of modern philosophy of science. In both these respects, the stated aims remind the reader of McKirahan, who gives extended analysis of predication in APo I 4-6 and of possible parallels between the APo and proof in Euclid. But where McKirahan provides a meticulously argued analysis of the APo organized in a logically compelling progression, H races through a dizzying number of topics and passages from throughout the APo and elsewhere, jumping from one to another with insufficient argument or explanation and with only glancing and incomplete references to secondary literature, of which she is frequently critical. The result is an uneven and ultimately unpersuasive work of limited originality.
The book, which began as the author’s doctoral dissertation, is divided into main five chapters, plus an introduction and conclusion. “Intellect as a First Principle” explores what is claimed to be two different senses of arche in the APo. “The Immediate Premiss” turns to the formal structure and modal status of demonstrative premises. “The Logic of Demonstration” turns to the logic behind syllogistic inference. The fourth chapter, “Syllogism and the Object of Knowledge,” claims to explain the discrepancy between the theory of demonstration and Aristotle’s scientific practice. The final chapter, “Knowledge and Demonstration,” explores the difference between knowledge of the fact and knowledge of the reason why, arguing that demonstration effects a transition from a perceptual understanding of the fact to a conceptual understanding of the reason explaining the fact.
It should be evident from the chapter titles and descriptions that H takes on several major and controversial areas of Aristotelian demonstration. Within each chapter, many subsidiary matters are also included. In the first chapter, for instance, one encounters (in rapid succession) discussion of hypotheses vs. definitions, induction in the Prior Analytics, induction in the APo, nous, knowledge of the universal and knowledge of the particular, the Meno problem and Plato’s theory of recollection, and the relation between APo I 1 and Metaphysics I 1 (but not APo II 19, which is part of the second chapter). Along the way, some claims of real interest are made. H argues, for instance, that induction is a grasp of the form of a thing. But the argument for this claim, as for most claims in the book, is far too brief. Key passages ( APo I 1 and APr II 21) are referred to and occasionally quoted in brief along with others from the APo and elsewhere, but interpretive cruces and alternate interpretations are hardly discussed. One feels that the author is making a string of assertions that might hang together well enough on their own terms but which lack anything like a sufficient engagement with the extensive secondary literature on the passages. In fact, though H cites a fairly long list of other scholars, only the briefest indication of their arguments is given.
H’s treatment of the epistemological implications of APo II 19 can serve as an example. Is the process that leads to a grasp of first principles rationalistic or empiricist? H charges Barnes with being a proponent of the latter and Kahn of the former. But no attempt is made to render these highly ambiguous labels precise, either in H’s terms or in those of the scholars H criticizes. Nor is there any laying out of the reasons behind each scholar’s position. The whole issue is settled in a single paragraph. So too with the next question, whether the process in II 19 leads to a grasp of concepts or propositions. This is certainly one of most contested areas of the whole APo. But again, H gives only the barest indication of the state of critical opinion, sweeping it all aside by claiming that it ignores the historical context of Aristotle’s issue.
This context, it turns out, derives from Plato’s Meno. I must say that I had expected more in the way of contexualizing, and certainly do not regard H’s brief mentions of this dialogue and a few others as anything original. A more extensive look at historical practices comes in comparisons with Euclidean proof. But here, of course, the historical context is reversed — it is Aristotle who provides the context for Euclid, not the other way around. And again, the approach is hardly original. One can mention again McKirahan as just one scholar who has provided precisely this sort of comparison.
Many of these deficiencies could have been avoided, I think, if the book had been filled out beyond its compressed 130 pages or if it had focused more assiduously on its stated theme. As it stands, the steps of argument and progression of topics can be very hard to follow, with the reader often confused by seemingly discontinuous leaps. The relationship between demonstration and syllogistic is a rich and complex topic, with an extensive secondary literature that requires careful consideration. H’s central argument, that demonstration produces a change in the knower from perceptual to conceptual understanding, is interesting. But as it stands, Knowledge and Demonstration cannot be recommended.
[[For a response to this review by Orna Harari, please see BMCR 2005.10.03.]]