‘Anything Peter published will undoubtedly be worth republishing even if he did not collect it himself’. Anthony Kenny's remark is amply justified by the 22 papers in this volume, all but two of which were published between 1949 and 2003. The collection begins with a deservedly influential paper with which Strawson himself was dissatisfied, ‘Ethical Intuitionism’. The source of his dissatisfaction is unclear. Perhaps it lies in his incautious claim that ‘when we say “Welshmen tend to be good singers” we mean that most Welshmen sing well’ and that, analogously, if we say ‘All acts of promise keeping tend as such to be right (or have prima facie rightness)’ we can only mean that most acts of promise-keeping are right, while brusquely dismissing the suggestion that ‘it is not the bare presence of the promise-keeping feature that entails the rightness of the act, but the presence of this feature, coupled with the absence of any features which would entail its wrongness’ (8–9). The paper ends on a characteristically Strawsonian note: ethical sentences are related to, but not reducible to, sentences about feelings nor do they describe intuitable ethical properties, rather as material object-sentences are related, but not reducible, to sentences about sense-experiences nor dependent on the ‘metaphysics of substance’ (11). There follow two renowned sallies against Quine (and Carnap), ‘In Defence of a Dogma’ and ‘Construction and Analysis’, which express, among other things, Strawson's preference for the description of natural languages over the construction of artificial languages. The fourth paper, ‘Proper Names’, elucidates the relationship of the grammatical (or logical) distinction between subject (or reference) and predicate, to the ontological or categorial distinction between particular and universal. Like Strawson's later work on this theme, it expresses his ‘demythologized Platonism’, 244), a genial tolerance of the abstract, intensional entities, such as universals and propositions, that Quine and others reject: ‘It is objected that there is no clear general criterion of identity for such items. Never mind: we get on well enough, and communicate well enough, without one’ (256). ‘The Post-Linguistic Thaw’ defies its title with a spirited defence of the iconoclastic linguistic movement, while welcoming signs of a return to systematic generality and to greater respect for philosophers of the past. After a skirmish with Carnap, ‘Analysis, Science and Metaphysics’ introduces the concept of ‘descriptive’ (in contrast to ‘reforming’) metaphysics, the exhibition of the ‘general structure’ of our ‘conceptual apparatus’ (88). ‘Bennett on Kant's Analytic’ introduces Kant, who helped to inspire Strawson's project of descriptive metaphysics and whom Strawson in turn did much to bring into the centre of contemporary philosophical discussion. In ‘Does Knowledge have Foundations?’ Strawson dissects and dismantles the doctrine that it does, but finds a truth lurking at the bottom of the muddle: the ‘intimate and necessary relation of belief, concept, and experience’ (108). ‘Knowledge and Truth’ continues this theme, arguing that while the truth of some types of proposition amounts to correspondence with the world, whereas the truth of e.g. mathematical and ethical propositions does not, we should resist the temptation to regard the former as foundations of the latter – in part because he agrees with Quine that no type of statement is immune from revision in the light of experience and that no statement can be confirmed or disconfirmed in isolation (though he famously rejects Quine's conclusion that this requires the abandonment of such intensional notions as analyticity and synonymy). Observation propositions are ‘checkpoints’ of knowledge, not foundations (115). In ‘Scruton and Wright on Anti-Realism’ Strawson insists that whatever may be the case in mathematics, statements about feelings and about past events must have the same meaning in the mouths of those who can directly verify them and in the mouths of those who cannot: the ‘conception of verification-transcendent truth-conditions …, and its link with that of meaning, is an essential part of a general view of the world which is in no way contrary to reason and to which we are in any case inescapably committed’ (123). ‘Perception and its Objects’ argues that common-sense realism, our immediate awareness of physical things, allows for a causal connexion between objects and our perceptual experience. Science is no threat, as long as we acknowledge ‘relativity of description to standpoint’: ‘Science is not only the offspring of common sense; it remains its dependant’ (145) ‘Liberty and Necessity' argues that since such knowledge as we have of the causes of our behaviour has not undermined our attachment to freedom and responsibility, any defensible thesis of determinism would equally leave it unscathed. ‘Sensibility, Understanding, and the Doctrine of Synthesis’ challenges Kant's belief that the functions of our judgements and our spatio-temporal forms of intuition are inexplicable facts. Such logical operations as predication and generalization are ‘analytically implicit in the very notion of judgement’ (159). On space and time Strawson's view echoes Trendelenburg's: objects, including ourselves, are in space and time, and space and time enable us to individuate the numerically distinguishable objects falling under any general concept. As conditions of empirical knowledge, they are nevertheless a priori forms of intuition. Strawson emphasizes the intertwining of concepts and intuitions. Synthesis does not retrospectively reproduce conceptually uncontaminated impressions and combine them with a present impression by imposing a unifying objective concept, of an enduring, causally efficacious object; each impression is already concept-laden. Nevertheless Strawson suggests that Kant might also be read as anticipating the unconscious ‘information processing’ postulated by cognitive psychology. ‘Two Conceptions of Philosophy’ returns to the differences between Strawson and Quine. Properties are rehabilitated, to be linked to a singular term by a ‘copulative device’ (176). ‘Review of Paul Grice, Studies in the Way of Words’ assesses Grice's accounts of meaning and of conversational implicature, and concludes with some reflections on ‘conceptual analysis’. Analysis of a concept should not contravene the ordinary usage of competent speakers, but must go beyond it, exploring ‘structural features of our thought- connections, analogies, and distinctions – which do not readily display themselves on the linguistic surface’ (184). Fundamental concepts cannot be reduced to simpler concepts, but form a network of interconnected concepts. (Strawson's ‘connective model’ echoes Plato's later dialogues as well as Kant.) In ‘Knowing from Words’, Strawson argues that humans are social animals, who would be ‘pitifully short-sighted’ without the testimony of spoken and written words (189). Perception, memory and testimony are equally essential to the formation of our belief-systems. In ‘What Have We Learned from Philosophy in the Twentieth Century?’ Strawson answers ‘general human conceptual self-understanding’ (193). He counts Wittgenstein as an ally, but reproaches his disdain for systematic theories, for the ‘reality of subjective experience’, and for the ‘inescapable presence in our experience of abstract intensional objects’, which are, among other things, the bearers of truth and falsity (193f.) ‘A Category of Particulars’ considers non-substantial particulars, ‘property-instances’, or what are sometimes called ‘tropes’, though this ‘expression either has no relevant suggestiveness at all or, if it has any, the suggestion it carries is entirely inappropriate’ (201). In ‘Paul Grice’ Strawson returns to the concerns of his former tutor: meaning, conversational implicature and the correspondence theory of truth. He follows Grice in stressing the ‘latitudinal’ unity of philosophy, the inter-connexions between its various specialisms, and its ‘longitudinal’ unity, the continuing relevance of the great dead philosophers. ‘Why Philosophy?’ repeats Strawson's commitment to conceptual self-understanding and aims a volley at those who hanker for a ‘new and revealing vision of human life, the world, and experience’ and thinkers who provide it, such as ‘Nietzsche (with his notion of the “Übermensch”), Sartre (with “mauvaise foi” and “existential choices”), and Heidegger (with something altogether more mysterious and probably more sinister)’ (224–5). In ‘Intellectual Biography’, Strawson writes: ‘I have no religious beliefs. When asked whether I believe in God, I am obliged to answer “No”; I have difficulty with the concept. But I am sometimes tempted to add that I believe in grace – a quality which eludes precise description, but is sometimes manifested in the words and actions of human beings’ (246). Finally, ‘A Bit of Intellectual Biography’ records Strawson's debt to Kant and Kant's debt to Strawson, for his ‘somewhat ahistorical attempt to recruit Kant to the ranks of the analytical metaphysicians, while discarding those metaphysical elements that refused any such absorption’ (250). He criticises Rae Langton's attempt to resurrect unknowable things-in-themselves (‘spatial relations are ideal, but they make experience of real dynamical relations possible’) for its neglect of time and the contents of inner sense (252).

Strawson's relentless defence of common sense evokes Johnson's response to Berkeley: ‘The seas of argument may wash for ever around these rocks of truth, but the rocks are not worn away’ (122). His common-sense realism perhaps accounts for his aversion to some of Kant's successors, despite his general respect for his own philosophical forebears and despite his own significant contribution to the revival of interest in the ‘post-Kantian’ thinkers. Hegel, it is true, figures in a list of philosophers – ‘Plato, Aristotle, Aquinas, Hume, Kant, Hegel, Frege, Wittgenstein’ – whose greatness does not entitle them to a ‘valid and exclusive claim to have got things finally right on some substantial matter’ (223). Strawson was undoubtedly less mysterious and sinister than Heidegger, but he noted his own affinity to him: ‘In common with Hume and Wittgenstein (and even Heidegger) I argued that the attempt to combat such doubts [about e.g.the external world and induction] by rational argument was misguided: for we are dealing here with the presuppositions, the framework, of all human thought and enquiry’ (242). But Nietzsche and Sartre are summarily dismissed – as if there were no more to Nietzsche than the Übermensch, and as if bad faith were not a manifest failing, to which philosophers as well as waiters are especially prone. Kant ignored animals, but Schopenhauer did not. He revised Kant's account of perceptual experience so as to accommodate the dog that chases a cat and the cat that tracks a mouse. Strawson ignores both Schopenhauer and animals, so we miss his answer to the question whether the application of concepts is really so essential to perception of enduring objects as he and Kant suppose (163). Along with animals, Strawson also neglects God and does not extend to God his tolerance of abstract entities. The concept of God no doubt presents difficulties and his existence is questionable. But confrontation with the difficulties has inspired much great philosophy and the concept surely has some claim to inclusion among the pervasive fundamental concepts whose interrelationships Strawson explores. The concept and the belief are not, of course, inescapable, since serious atheists are more abundant than serious doubters in the external world or other minds. But the concept and the belief have invariably been socially available, especially in childhood training, and it is not clear that they could be eliminated entirely without excessive strain on the rest of our cognitive and practical equipment. Kant and Nietzsche believed that they could not, and Heidegger's obscure but suggestive remarks about our transcendence to ‘world’ or to ‘beings as a whole’, carrying us beyond the local environment to which animals are confined, has a similar implication. The dog is immediately aware of the cat it is chasing, but it does not transcend to beings as a whole and it does not believe in God. Strawson, in a his aspiration to human self-understanding, did transcend to beings as a whole and, allowing for a certain flexibility in the concept of religion, we might say that he was a profoundly religious man. However that may be, Peter Strawson, both as a man and as a thinker, displayed the grace in which he half-believed. He has earned a secure place among the great dead fallible philosophers.