Dawson's main concern is with book 3 of John Locke's Essay Concerning Human Understanding, which is devoted to spoken language, how it functions, and how it can be abused. The basis of Locke's theory is that words, principally ordinary nouns, signify the ideas of the speaker primarily and immediately, that the link between words and thought is purely conventional, and that our general terms do not and cannot correspond to concepts that are natural representations of Aristotelian essences, of genera and species, for no such concepts exist. Instead, we construct our own complex ideas of natural kinds by making a choice, to some extent arbitrary, from the clusters of properties that we experience in a regular way, and we use our general terms in accordance with the nominal essences thus formed. If we stir into this mixture, which is already dangerously individualistic, a strong distrust of two human tendencies—the careless tendency to use words without clarifying the ideas (if any) that they represent, and the morally culpable tendency to use words to impress or mystify—we find ourselves in a parlous situation. We have access to the ideas of others only through their words, but we can never be sure that their words represent ideas similar to our own. Insofar as our political life depends on communication and consensus, the very possibility of a civil society is thrown into doubt.
In her rich, detailed study of Locke's theory of language and its context, Hannah Dawson fulfils two aims. First, she shows, through a survey of the full range of early-modern discussions of language, not only by thinkers ranging from Montaigne to Boyle, but also by the conventional logicians and grammarians read in schools and universities, that unease about problems arising from the very nature of language was already a feature of early-modern thought. Locke's achievement was to provide a full account of reasons for this unease. Second, in her lengthy treatment of Locke's own writings, she makes clear the difficulties to which Locke's skepticism about our ability to communicate successfully gave rise, notably with respect to his political theory.
The strength of the book lies in its deft handling of a mass of material. On a rough count, the bibliography of printed primary sources contains about 200 items from the sixteenth and seventeenth centuries, and there is a very full bibliography of secondary sources. Dawson covers all this material with the aid of numerous quotations, and she could easily have lost the reader in a morass of detail. Fortunately, the thread of her argument is always clear, and her engaging presentation holds the reader's attention from beginning to end.
Viewed simply as a work of reference for those wanting a guide to early-modern literature on theories of language, Dawson's book will be invaluable. Yet it is more than a work of reference. Dawson's thesis about the general recognition of problems caused by language is thoroughly convincing, and her thesis about the consequences of Locke's skepticism will be of great interest to those working on his political theory. If I do not describe this second thesis as convincing, it is because of the well-known tension, emphasized by Dawson herself, between Locke's professed common-sense approach, and the philosophical implications of his doctrines. Just as he combines an epistemology that seems to cut us off from the external world by interposing a veil of ideas with an apparently robust confidence that we are acquainted with a public world of continuous physical objects, so he combines a theory of language that seems to cut us off from the possibility of problem-free communication with an apparently robust confidence that his writings can change minds, and that “candid and intelligent Readers” will be able to grasp “the true meaning” of a discourse (Essay 3.xi.27).
Given the amount of material handled, there is little room for the detailed philosophical analysis that some of the topics demand. To take one example, a closer investigation of the difference between Aristotelian-scholastic simple concepts and Lockean simple concepts would have been relevant (cf. 30–33). One can argue that the Aristotelian-scholastic simple concept was one that represented a complex reality but did not involve composition. Epistemologically speaking, what we grasp when we look around us is medium-sized physical objects such as cats and apples, and while our concepts are subject to later analysis, they are not initially formed by the combination of more primitive items. Given this approach, along with the belief that living things are, indeed, divided into natural kinds, it is not silly to classify terms such as “cat” as simple, lexical items, to be contrasted with phrasal items such as “the cat on the mat.” Moreover, once a distinction has been made between simple but complex concepts and concepts that are simple in the sense of being primitive and not complex, one can more easily understand the difference between the Aristotelian-scholastic and the Lockean account of such concepts as “cat.” For Locke, such a concept is not merely complex; it arises out of more primitive concepts that are epistemologically prior, and it is formed by composition in accordance with rules.
Sometimes the absence of useful philosophical distinctions seems to be related to a misunderstanding of the texts. For instance, in her discussion of what words signify (16–17, 93), Dawson castigates logicians for carelessly switching between two accounts of the relation between words and things, a binary account and a standard account that interposes ideas between words and things. However, the sources she cites for the so-called binary account seem to be concerned only with the logical distinction between use and mention. We can talk about the word “man,” saying that it is a one-syllable English noun, or we can use the word “man,” saying that man is an animal. The standard account, whereby words signify ideas, which in turn represent things, presupposes that we are considering words as used rather than mentioned.
The book is very nicely produced, with footnotes (sometimes amounting to mini-articles) rather than endnotes, and it has a well-constructed index. I noted only a few minor typographical errors. The one major irritant was Dawson's failure to use the standard method of referring to Aristotle's works by Bekker numbers. Instead, the reader is given the impossible task of checking page references in a number of different translations. While the reader should be cautious about some of the details, Dawson's book is, nonetheless, an exciting and important contribution to the literature on early-modern philosophy in general and on Locke in particular.
