Table of Contents


Series page

Title page

Copyright page


Introduction: The Unity of Locke’s Thought

A Word about Empiricism

1 Life, Contexts and Concerns

1. Locke’s Life

2. The Contexts for the Major Works

3. Locke’s Correspondence

4. Religion, Politics and Epistemology: Questions of Authority and Allegiance

2 The Theory of Ideas

1. The ‘New Way of Ideas’

2. Three Key Ambiguities

3. The Attack on Innate Principles

4. The Empiricist Project

3 Human Knowledge and Its Limits

1. Locke’s Definition of Knowledge

2. Intuitive, Demonstrative and Sensitive Knowledge

3. Human Knowledge and Ignorance

4. Knowledge, Probability and Faith

4 The Material World

1. Locke’s Common-sense Realism

2. Locke’s Scientific Realism: The Corpuscular Philosophy

3. Primary and Secondary Qualities

4. Real and Nominal Essences

5. Angelic and Human Chemistry

5 God and Religion

1. Faith and Reason in Religion

2. Locke on Faith and Reason

3. Locke’s Cosmological Proof

4. The Reasonableness of Christianity

5. God and the Law of Nature

6 The Soul and the Afterlife

1. The Immortality of the Soul

2. Can we Prove the Immateriality of the Soul?

3. Personal Identity

4. Immortality Revisited

7 The Two Treatises of Government

1. The Context: Locke’s Fears of Absolutism

2. Bad Principles Exposed

3. Better Principles Expounded

4. The Case for Private Property

5. The Conservative Revolutionary

6. Equality, Women and Slaves

8 Problems of Church and State

1. The Problem of Divided Authority

2. Argument 1: Belief is Not Subject to Coercion

3. Argument Two: The State has no Authority to Enforce Religious Conformity

4. Argument Three: Toleration Does and Coercion Does Not Serve the Cause of Truth

5. The Limits of Toleration

6. Toleration After the ‘Glorious Revolution’

Select Bibliography

References to Works of Locke Used in This Volume

Electronic Resources


Other titles in the Clinic Thinkers series:

Bernard Gert, Hobbes

Andrew Ward, Kant

Title page


CThe Correspondence of John Locke
CUOf the Conduct of the Understanding
ECTAn Essay Concerning Toleration, and Other Writings on Law and Politics, 1667–1683
EHUAn Essay Concerning Human Understanding
ELNEssays on the Law of Nature and Other Associated Writings
LTLocke on Toleration
PWPosthumous Works of Mr John Locke
RCThe Reasonableness of Christianity, as Delivered in the Scriptures
STCESome Thoughts Concerning Education
TTGTwo Treatises of Government
WJLThe Works of John Locke


The Unity of Locke’s Thought

John Locke (1632–1704) has been described both as the greatest English philosopher (Hume, of course, was Scottish) and as the father of modern empiricism. Both judgements might be challenged: Ockham, Hobbes, Mill and Russell might vie for the former title, while Bacon and Gassendi clearly played major roles in the development and articulation of empiricism long before Locke. But Locke’s place in philosophy’s premier league is not in doubt; his influence on subsequent epistemology, metaphysics, religion and politics was enormous; and his philosophical reputation has if anything grown over the last thirty years. No scholar today would endorse Gilbert Ryle’s celebrated verdict, delivered in a lecture in 1965, that ‘nearly every youthful student of philosophy both can and does in about his second essay refute Locke’s entire theory of knowledge’.1 The scholarship of the past generation has helped to strip away centuries of misunderstanding and misrepresentation, and the real Locke turns out to have been a much more powerful and profound thinker than the caricature still sometimes found – regrettably – in textbook accounts of ‘the British empiricists’.

The range of Locke’s thought is as impressive as its depth. Here is a man who thought long and hard about natural and revealed theology, metaphysics, epistemology, the natural sciences, medicine, psychology, physical and human geography, semantics, ethics, politics, education and economics, and made significant original contributions to several of these disciplines. This presents an obvious problem for commentators and authors of textbooks. Does one attempt an overview of Locke’s thought as a whole, or does one choose to focus on the Essay Concerning Human Understanding, or the Two Treatises of Government, or the Reasonableness of Christianity? Does one sacrifice depth for breadth or vice versa? In accordance with the stated aims of this series, this book aims to be fairly comprehensive in its scope, covering at least Locke’s metaphysics, religion, epistemology and politics, while glancing briefly at his views on other subjects. But the danger then looms that the book will simply fall apart into a series of entirely distinct chapters, one on each key text or each aspect of Locke’s multifaceted thought. To avoid this peril, a leading idea or guiding thread is clearly needed.

At this point, some recent scholars might ask whether it is right even to attempt to find such a leading idea. Each of Locke’s main works, it has been argued, is written as a response to some definite problem of his day, and cannot be understood except in that historically fixed context. There is no reason to assume that Locke ever attempted to create a philosophical system of his own or to found a school of philosophy with a distinctive set of characteristic doctrines. There is considerable truth to both of these assertions. If we look at the genesis of Locke’s major works, they are clearly prompted by issues highly specific to his time and place, and such questions about their origins are of obvious relevance to their comprehension. To take them entirely out of context is to run evident risks of misunderstanding. And it is undeniably true that Locke never took himself to be establishing a ‘Lockean’ school or teaching a special body of distinct ‘Lockean’ doctrines. So there is a challenge here that must be met by the author of any attempt to see Locke’s work as a coherent whole.

The first point that needs to be made in response to this challenge is to insist that setting a philosophical classic in its historical context is in no sense incompatible with treating it as a work of philosophy. Historians of philosophy sometimes oppose a ‘contextualist’ approach to an ‘analytical’ one, but this is an entirely false antithesis. Both in general and in the case of Locke in particular, the ‘contextualist’ and the ‘analytical’ approaches are complementary rather than antagonistic. If, for example, we treat Locke in the context provided by contemporaries such as Boyle and Newton, rather than that of successors such as Berkeley and Hume, we will find that we arrive at a picture of Locke that is simultaneously more historically accurate and more philosophically impressive.2 We can thus grant the contextualist’s point about the genesis of Locke’s major works without abandoning the search for a guiding thread. Even if Locke was not a system-builder, he was still a philosopher. His correspondence shows clearly that he was often asked to comment on the consistency of the views expressed in his various works. Can he, for example, deny that morality is an innate law written by God on each and every human heart (Essay, Book 1), and still claim that we have natural rights and duties discernible by Reason (Second Treatise of Government)? His opponents claimed that his empiricism undermines the moral law; friends such as William Molyneux and James Tyrrell urged him repeatedly to provide an account of the foundations of morality consistent with his rejection of innate principles. And is the tabula rasa thesis of the Essay consistent with the admission, in Some Thoughts Concerning Education, of seemingly inborn differences of temperament among children? Locke is not a system-builder, but he cared about the truth, and a man who cares about truth cannot be indifferent to consistency. There is no Lockean system of philosophy, but there is a distinctively Lockean point of view, grounded in his opinions concerning human knowledge and its limits.

My suggestion is that we focus on the epistemology, and start with Locke’s often-stated claim that we humans have been given enough knowledge for our needs. We can know enough of the natural world for human arts such as agriculture, manufacturing, navigation and medicine, even though we remain hopelessly ignorant of what might be called ‘the nature of Nature’, that is, the underlying metaphysics of the supposed real essences of things. And we can know enough of God’s existence and of the God-given moral law to grasp our duties towards one another as human beings and as citizens, despite remaining massively ignorant of the nature of God and even of our own souls. This combination of a modest agnosticism about metaphysics with confidence in ordinary empirical and moral knowledge is Locke’s most distinctive feature among early modern philosophers. Many of his contemporaries would have denied that we can know our moral duties without an intensive training in their particular religion, or that we can gain an adequate grasp of the natural world without settling deep issues of matter theory such as the truth or falsity of atomism. Here Locke’s thinking begins to sound strikingly modern: there are many living philosophers who ask precisely Locke’s question, ‘How can I build the widest possible consensus?’, bringing in people from different backgrounds and traditions and getting them to sign up to some proposition of great practical importance despite their very different (and still unreconciled) starting points. This central Lockean distinction, between what we need to know for our practical concerns on the one hand, and what we don’t need to know (and should learn to dispute peacefully and amicably) on the other hand, will inform the whole book.

Given this overall approach, we can expect problems of two kinds to arise for Locke. From the sceptical side, we can expect challenges to the positive knowledge-claims that he does make. He thinks, for example, that the existence of God can be demonstrated by Reason. But since he also says that the moral law is God-given, his demonstration of the existence of God had better be sound! If morality is independent of religion, sceptical doubts about the so-called ‘proofs’ of the existence of God are harmless intellectual games; if morality depends on religion for its foundations, a great deal more is clearly at stake. From the dogmatists’ side, we can expect challenges to Locke’s repeated claims that we don’t need to know the metaphysics. He tells us, for example, in his famous account of personal identity, that my assurance of my own personal survival (and hence of the possibility of rewards and punishments after death) doesn’t depend on any particular philosophical theory about the soul. But if he is wrong about this, it might turn out that the knowledge he tells us we don’t possess (knowledge of the nature of the soul) is essential to the ordinary believer’s assurance of an afterlife. We can thus expect Locke’s position to be vulnerable to attacks from both sides – and this is exactly what we find.

A Word about Empiricism

Historians of early modern philosophy of a previous generation used to tell a story of two opposed schools or traditions, pitting ‘The Continental Rationalists’ (Descartes, Spinoza and Leibniz) against ‘The British Empiricists’ (Locke, Berkeley and Hume), with Kant’s critical philosophy providing a final synthesis of the two schools at the end of the eighteenth century. The rationalists, we were told, thought that reason was supreme, and neglected experience; the empiricists trusted experience and had their doubts and reservations about reason. The story had a pleasing symmetry, and of course appealed to the prejudices of Anglo-Saxon audiences, who liked to think of themselves as plain no-nonsense folk, modestly following the teachings of experience and justifiably suspicious of too much high-flown speculation. This story still provides the narrative structure for many an undergraduate textbook or introductory course on early modern philosophy.

The problem with the story is that it is completely unhistorical, in two distinct but equally important senses.3 In the first place (1), there is no reason to think that the historical figures in question thought of themselves as belonging to these two opposed camps. Locke might think of Descartes as rather too bold in his claims to grasp the respective essences of material and spiritual substances, but he nevertheless regards Descartes as a fellow ‘modern’ and an ally against the Aristotelianism of the schools. Berkeley constructs a metaphysical system that can be shown to have significant debts to Descartes and his disciple Nicolas Malebranche. When Hume writes to Michael Ramsay,4 listing the books he should read before attempting Hume’s own Treatise, he cites Descartes and Malebranche, Berkeley and Bayle, but not Locke! The ‘two schools’ story is an artefact of later historiography, not something that would have been recognized by the early modern philosophers themselves. And (2), any suggestion that ‘the rationalists’ appealed exclusively to reason and never to experience, while ‘the empiricists’ appealed exclusively to experience and never to reason is promptly refuted by the most cursory reading of the relevant texts. It would be very easy to document any number of appeals to experience in Descartes, Spinoza and Leibniz, and some crucial argumentative appeals to supposedly self-evident rational principles in Locke and Berkeley.

Should we, then, simply abandon the traditional terms ‘Rationalist’ and Empiricist’ altogether? That might be an over-reaction. I intend to follow John Cottingham in treating ‘rationalism’ and ‘empiricism’ as cluster-concepts.5 The obvious analogy here is with medicine. Doctors generally find it difficult to come up with neat lists of necessary and sufficient conditions for the presence of diseases, and often resort to formulae such as ‘patients with disease D will tend to manifest some of the following symptoms S1. … n’. In similar manner, says Cottingham, historians of early modern philosophy should treat the terms ‘rationalist’ and ‘empiricist’. A philosopher might be properly called a rationalist, we are told, if he exhibits a sufficient number of the following ‘symptoms’:

1. Distrust of the senses as a source of knowledge.

2. Reliance on reason as a source of informative knowledge about the world (not just about our concepts or our language).

3. Belief in innate ideas and innate principles.

4. Belief in an intelligible order of nature that the human mind can in principle grasp and somehow ‘mirror’.6

5. Belief that causal relations are necessary (causal rationalism).

6. Appeals to and reliance on a mathematical or demonstrative model for human knowledge.

If we apply this test to Locke, we find that he shows only very weak symptoms of rationalism. By at least four of Cottingham’s criteria, Locke emerges as a definite negative. He thinks that the senses do furnish us with knowledge appropriate to our nature and needs; he flatly denies the existence of innate ideas and innate principles; he rejects the presumptuous thought that we can or should try to think God’s thoughts or see our world from a God’s eye point of view; and he tells us in a number of places that natural sciences such as physics and chemistry cannot be done in a demonstrative manner. With regard to his reliance on reason, the evidence is more equivocal, but it is clear that he places significantly less weight on reason and more on experience than, for example, Descartes and Leibniz. And with regard to causal rationalism, the evidence is again equivocal, but Locke is clear that even if there are ‘necessary connections’ in nature, as he suspects, they will not generally be recognized and known as necessary by us. So if we look for these six symptoms characteristic of rationalism, we get four clear ‘nos’ and two equivocal ‘maybes’. If we applied the same test to Spinoza or Leibniz, we would get a string of ‘yeses’.

Many contemporary historians of early modern philosophy find themselves in broad agreement with Cottingham in regarding the old terms ‘rationalism’ and ‘empiricism’ as still capable of doing useful explanatory work, despite the serious reservations we have noted. Instead of artificially dividing up the philosophers of the seventeenth and eighteenth centuries into two opposed schools, they now ask, of any given philosopher, how much – or how little – he thinks reason can achieve a priori, working independently of experience from supposedly self-evident first principles. A philosopher who confidently appeals to innate ideas and innate principles, and who thinks that, working from such materials, reason can provide us with a wealth of a priori knowledge, will properly be characterized as a rationalist. A philosopher who rejects innate ideas and innate principles, and who appeals to experience far more than to reason in his account of human knowledge, will properly be characterized as an empiricist. This is how I propose to use the term ‘empiricist’ throughout the rest of this book. I shall use the traditional terms ‘empiricism’ and ‘empiricist’ in discussing Locke’s philosophy, but the terms must be understood with two clearly stated provisos: (a) that we are not seeking to situate Locke in a distinct school or tradition, and (b) that an empiricist philosopher can and does at times appeal to reason and to supposedly self-evident rational principles. The difference here may be one of degree rather than of kind.


1 Gilbert Ryle, ‘John Locke’, in his Collected Papers, volume 1, 147.

2 See Peter Alexander (1985) Introduction, 1–12, for the importance of setting Locke’s philosophy in the right context – that provided by Boyle and Newton rather than by Berkeley and Hume.

3 For a sharp and incisive statement of the recent attack on the rationalist–empiricist distinction, see Loeb (1981).

4 David Hume to Michael Ramsay, 26 August 1737. Quoted from Mossner, 104.

5 Cottingham (1988), 10–11.

6 For an illuminating discussion of this theme, see Edward Craig on what he calls ‘the insight ideal’: Craig (1987), chapter 1, ‘The Mind of God’, 13–68.