Bryn Mawr Classical Review

Arguably one of the most famous Aristotelian statements is made at the beginning of Metaphysics Z: "Being is said in many ways" -- to on legetai pollachôs. Not only being is said in many ways;¹ nature and medical are also such. In each of these cases we are confronted with a problem: what is that which is said in many ways and how are these ways related to one another? The what-it-is question requires an analysis of the ways in which something is, directing us to the problem of the univocity or non-univocity of Aristotelian concepts and so to the notion of homonymy. It is here that Julie K. Ward's fine book Aristotle on Homonymy: Dialectic and Science makes its contribution.

Ward's book comes in the so-called realist tradition of Aristotelian scholarship. Her treatment of homonymy is based on the statement that "the referents of words are extra-linguistic entities, not other words" (13). The question of the connection between homonyms becomes then a question about the ways in which the things designated by the same word are connected. This allows Ward to construct a theory about what she calls, borrowing a phrase coined by Christopher Shields ², "core-dependent homonymy," in which homonyms are causally connected in one of the four Aristotelian causal ways. For example, in the case of the medical, we understand why both the scalpel and the medical art are called medical by looking at a causal connection between the two entities. The bulk of Ward's book is concerned with explaining and demonstrating how this kind of homonymy works.

Aristotle on Homonymy: Dialectic and Science has six chapters, which analyze four main subjects: (1) placing Aristotelian homonymy in its historical context and showing its Platonic influence; (2) explaining systematic homonymy as it appears in the Topics and the Metaphysics; (3) applying core-dependent homonymy to the Aristotelian concepts being, nature, and friendship; (4) showing the role of homonymy in the process of scientific inquiry.

Ward begins by pointing out different aspects of homonymy. Taken simply, as it appears in the Categories, homonymy refers to things which share the name but have different definitions. This can be interpreted in two ways. On the one hand, homonymy might refer to things which only have a name in common and nothing more (Accidental Homonymy, or Discrete Homonymy in Christopher Shields' book mentioned above). On the other hand, homonymy might refer to things with the same name which share some characteristics (Systematic Homonymy, or Comprehensive Homonymy for Shields). Ward argues for systematic homonymy. The main argument she proposes is historical, argung that Aristotle's view of homonymy was not influenced by Speusippus, as some suppose, but by Plato. Ward says that Plato tries to make sense of how Forms and particulars have the same name, while having different natures. When he employs the term homônymos, Plato, in Ward's view, distinguishes between the different characters that a Form and the particulars have while having the same name. Nevertheless, while the Form and the particulars have different natures, they also share similar characteristics and are causally connected. It is not accidental, then, that a Form and a particular share the same name. On the contrary, it is precisely because of a certain relation of participation in the Beautiful that, for example, a beautiful flower is called such. If we think back to the distinction between accidental and systematic homonymy, Ward claims that the latter is similar to Plato's use of homonymy. Aristotle uses this concept of homonymy when he deals with unicategorial predicates -- things like being, one, nature, life, and friendship. Ward claims that they "belong to a category of terms referring to things having an overlapping set of natures but lacking a single nature. Things related in this way fail to possess the conditions for being specifically the same, or synonymous, and yet possess more in common than simply being called by the same name" (39).

In the second chapter, Ward turns to an analysis of homonymy as it appears in the Topics. Accidental homonymy and systematic homonymy, which both have their place in Aristotle's theory according to Ward, are described here as having two different functions. That of accidental homonymy is a negative one, which is to clarify terms by pointing out things which have only the name in common and nothing else. This clarification avoids equivocation. The function of systematic homonymy (which Ward also calls "core-dependent homonymy" and also "core-related homonymy") goes beyond that. The process of finding connections among the relata having the same name -- the core-related homonyms -- contributes to the understanding of what a thing is, as Ward explains in the following chapters.

Ward's book is then dedicated to this systematic or core-related homonymy --philosophically, the more interesting type. First, Ward engages in an analysis of two cases, the Healthy and the Medical, which are supposed to provide clarification as to how core-related homonymy works. Ward points out that in these cases we have primary and secondary instances of a character which are connected causally. Take, for instance, the Medical. If the primary instance is medicine itself, i.e., the medical art (and it is), all the other things which are called medical are such because of a causal connection to medicine itself. The scalpel, then, is called medical because it is useful to medicine. An operation is called medical because it is a function of medicine. In each case, "medical" signifies a different character or quality, but these characteristics are somehow connected with the medical art itself. The account of "medical" according to which the scalpel is medical is clearly different from the account of "medical" according to which the art is medical, but the definitions have overlapping points; this fact provides the explanation for their being named by a common predicate, medical. Moreover, the scalpel is also causally connected to medicine because it is an efficient cause of it, Ward claims, for the scalpel serves as a means to medical knowledge. So, in Ward's words, core-related homonymous things "named by a common predicate 'F' must be related such that one thing's being called 'F' is causally related to another thing's being called 'F'" (101).

With this in mind, Ward proceeds to an analysis of the homonymy of three Aristotelian concepts: being, nature, and friendship. In the case of being, the thesis of core-related homonymy is based on the idea that the primary instance of being is that of substance and that all the other secondary instances are related to this one. Ward correctly points out that for Aristotle a quality and a quantity (which qualify in this account as secondary instances of being) are always of some thing, a primary substance, and the quality and the quantity are only insofar as there is a substance that manifests them. So Socrates' paleness is only insofar as Socrates is. Both Socrates' paleness and Socrates are said to be, but each in a different way. Socrates' paleness is a secondary instance of being related to a primary instance of being, that of the substance Socrates.

The other two cases which Ward analyses, nature and friendship, produce further evidence for core-related homonymy. The first one, nature, does not pose problems. Aristotle points out, Ward says, that there is a primary use of physis, signifying that which has in itself a cause of motion and rest. He also says that physis is non-univocal. The secondary instances of physis are all connected to this primary instance. The second case, friendship, has the difficulty of apparently lacking a primary instance which is causally connected to the other two. The utility and pleasure friendships are not called friendships because of a causal connection they have with complete friendship. Ward attempts to resolve this problem by considering the basis of each of the three types of friendships. They all appear to have as their purposes a good. The difference is that complete friendship is based on the real good, as Ward calls it, while the other two on what appears to be good, namely utility and pleasure. In order to provide a causal analysis of friendship, then, one needs to have an analysis of the good. The definitions of everything that appears to be good depends on the account of what is good indeed, giving thus to the good a priority in logos. For Ward, then, the good itself is the primary instance, while utility and pleasure and everything else that appears to be good are secondary instances. Returning to the types of friendship, their connection would be now seen by considering the kind of good that they are based on. Since the complete one is based on the real good, it is the primary instance of friendship, while the other two, based on what appears to be good, are secondary instances.

At the end of the book, Ward returns to the contribution of homonymy in scientific inquiry. She explains that homonymy provides a useful tool for the classification by division. Used as a method, homonymy can reveal accidental homonyms so as to avoid equivocation, but it can also show the "inter-connection among the uses of the terms that possess the shared feature to point to the primary instance" (198).

Julie Ward's book is well conceived and an important contribution on a difficult topic. Her treatment of homonymy is partly indebted to Shields (1999), and she does a good job pointing out both their agreements and their differences.

The study moves from a clear explanation of what Ward takes to be homonymy in Aristotle, to showing it at work, and to deciphering its contribution in constructing definitions. This progressive form helps the reader go through a difficult topic but might also be the source of some repetitive claims. Without a doubt, the book will generate fruitful discussions.

One of the main questions that this volume raises is the connection between Aristotle's theory of homonymy and Plato's treatment of the subject. The connection between Plato and Aristotle is important for Ward, since she begins the book by pointing it out and ends it with an afterword in which she brings the two philosophers together again. But one may reasonably wonder whether she brings them too close. Consider, for instance, the notion of primary instance. When analyzing the non-univocity of the good in order to make sense of the core-dependence of friendship, Ward says that the real good is the primary instance while those things which appear to be good are secondary instances. But one might wonder what this real good as a primary instance really is. Certainly, the good is not univocal. "Good" picks out different characters when applied to different things. A good human is not good in the same way a good dog is. In the case of the human, the primary instance is the human's real good, while the secondary instances are what appear to be good. But the same analysis can be done in the dog's case. There is a real good for a dog, but also things which appear to be good but are not. Should we then say that the good has several primary instances, and if so are they connected to a super-primary instance, which might resemble a Platonic Form? To be sure, Ward does not claim such a thing, but it is not obvious that her theory would not allow it.

What certainly will cause much discussion is the causal connection between primary and secondary instances of a homonym. Ward explicitly rejects Shields' theory that the secondary instances are not causally connected to the primary instance. Instead, she believes that the causal connections should be explained by making reference to one of the four Aristotelian causes. But this needs further discussion. One might desire to have a better explanation of how the secondary instances which bring about the primary instance (as, for example, a certain diet is healthy because it efficiently causes health) are different than those which are only indicative of the primary instance (as a healthy body is receptive of health). Further, even if we accept that a term "F" applies differently to primary and secondary instances, it is not obvious that there must be a causal connection among them.

To conclude, Julie Ward's volume is a valuable study, on a subject that needs further inquiries. One cannot but be engaged with Ward's fine and thorough analysis of Aristotelian homonymy.

1.  Aristotle says also in Metaphysics Gamma that there are many senses in which a thing may be said to be, and there he also mentions that these senses are not homonymous (1003a33-4). As Ward points out, Aristotle uses homonymy here with the meaning of accidental homonymy. The different senses of being are still, Ward will argue, core-related homonyms.
2.   Christopher Shields. Order in Multiplicity: Homonymy in the Philosophy of Aristotle. Oxford: OUP, 1999.