This is a difficult and rewarding book. Its four principal threads each deserve attention. First, philosophy in Charlton's view is an art rather than a science: its goal is not so much to discover hitherto unknown facts as to help us realize what we already, inchoately, know. Second, we have as much (or probably much more) to learn from classic texts both ancient and early modern as from the latest REF-approved essay. Third, philosophical puzzles may be unravelled by attention to the grammar of our ordinary speech (not only English). Fourth, it is by following this route that we can recover from the numbed incomprehension to which Socrates, it is said, reduced his victims: ‘before enlightenment; chop wood, carry water; after enlightenment; chop wood, carry water’ – not that Charlton cites that particular Zen proverb!

Only philosophers, so Charlton tells us, ask what it is to be true, to exist, to happen or to have happened, to think, or to be right or wrong. Others may ask ‘what time is it?’, ‘what are you thinking?’ or ‘what caused the crash?’, but only philosophers are puzzled to understand what time or thought or causation really are. These are the questions Charlton identifies as ‘metaphysical’, and reckons that they are not to be answered simply by ‘research’ (or not, at least, by merely ‘empirical’ research of the sort we now identify as ‘scientific’). Philosophy – like pure mathematics – is an ‘arts subject’, and its goal is ‘to put into words’ what we already, practically, understand, or at any rate can recognize, rather as a great literary or art critic can help us to notice what, in a way, we had seen already.

So how are philosophers to address these abstract, metaphysical concepts? Charlton proposes that we need to attend to the ordinary constructions of our human languages. The question (for example), ‘what is it to exist’ is not to be answered by investigating the first three seconds after the Big Bang, or laying out a table of elementary particles. At best such investigations answer only questions about what exists, for how long or with what properties. What matters is how we employ such terms as ‘existence’ in our common speech – a speech which is itself the context within which more arcane or specialized enquiries can take place. This is not to forget that common speech may be misleading: surface features of a particular language may lead us, notoriously, to suppose that ‘existing’ as well as ‘purring’ is something that tigers do and unicorns do not (since unicorns neither exist nor purr). By attending to the proper grammar of our sentences, both indicative (‘the tiger is waiting to pounce’) and non-indicative (‘run!’), we may understand our own sentences more clearly. Charlton employs similar tactics to explore what we mean in speaking of goodness, rightness, causality, agency, pastness or futurity. Philosophy, on these terms, is a continual process of avoiding familiar errors, not uncovering radically new truths, and its value – this is an implication only lightly emphasised by Charlton – lies not in the number of peer-reviewed monographs judged to have added to the ‘sum of human knowledge’ but in assisting us all to see more clearly what we already ‘know’. This task is not well served by professional philosophers who employ technical jargon with the implicit claim that their new ways of speaking embody unfamiliar truths intelligible only to the highly-educated, narrowly-specialized elite. ‘Is it really credible’, Charlton enquires, ‘that philosophers today have a better knowledge of what truth is than ordinary people 50 or even 2,000 years ago?’ (17).

At first it seems that Charlton is committed to answering that it isn't. But on the other hand, Charlton himself acknowledges (even insists), that a good many University applicants – like the Athenians of Plato's day – have great difficulty identifying invalid patterns of inference – which at least opens the possibility that ordinary people even now may be similarly blind to factors that have only become evident after many centuries of exploration. Even University graduates – even highly acclaimed experts in particular fields – are liable (to use a different example than Charlton's) to gross errors in assessing probabilities, errors that are routinely reinforced by popular presentations of ‘average’ attainments or tests reckoned to be ‘99% accurate’ (without troubling to distinguish modes, medians or averages, or detailing the percentage of false positives and false negatives). ‘Early Greek enquirers into nature who lived before the advent of grammar seem not to have distinguished objects like elephants from qualities like heat; they were prepared to think that the world consists of hot and cold or dense and rare’ (30). Maybe our successors will be similarly amused by our own fallacies, identified as such only by tedious exploration of a kind that no normal person would ever bother to read.

It is probably inevitable therefore that ‘philosophy’, and the academic life in general, will always be a specialist, potentially arcane, pursuit, dependent on ‘the threefold task of distinguishing different kinds of question, introducing rigour into language and reasoning, and setting up institutions of higher education’ (23). That task was accomplished, or at least begun, ‘by three individuals, Socrates, Plato and Aristotle’. Not only did they establish such institutions, they also established a distinctively ‘Western’ political ambience: ‘we want political leaders with brains’ (32), and expect people to be governed by their ‘beliefs’, as these are expressed in their moral, political and religious practices and customs (33). We are also outraged when our rulers turn out to be personally corrupt (gross institutional corruption apparently bothers us much less). Other peoples have other ways of conceiving what matters most in individuals and societies – though Charlton is mostly concerned with what he thinks is shared across the human community. In insisting – not implausibly – that Plato ‘was the most rationalistic and down to earth of thinkers’ (34) he likewise neglects the actual development of Platonic theory beyond its founder's conscious wish, in case that development runs counter to his own implicit metaphysics.

Charlton draws on grammar – the rules that govern the way that words are put together to make intelligible sentences – to distinguish predication and quantification, and so also ‘things we predicate’ (like being red, or running) and ‘things we quantify’ (like poppies or children). In English these uses may be confused, encouraging the innocent to suppose that ‘exists’ may signify something we predicate, or ‘red’ stand for a quantifiable thing. Alongside all such words are words that can be explained ‘only by giving the constructions to which they contribute’: ‘the’, ‘not, ‘than’ and the like. And the same can be said of words like ‘good’: ‘that wine is red’ and ‘that wine is good’, he says (43), may seem to predicate different things of the wine. But really ‘red’ and ‘good’ do not signify different qualities: ‘red’ signifies a certain colour, but ‘good’ merely indicates that the wine is being recommended for some unspecified reason. He applies a similar analysis to many other familiar philosophical terms (quality, relation, mode of action): they neither ‘stand for’ nor ‘signify’ anything, but their meaning is revealed in the way that they are used, the constructions to which they contribute. Philosophy, so Wittgenstein declared (Blue and Brown Books 27: cited by Charlton, 47), ‘is a fight against the fascination which forms of expression [aka grammatical forms] exert upon us’ – a notion which some have elevated into the dream of a perfectly transparent, unambiguous language that will no longer mislead us. Alternatively, he was rejecting the very notion of such a transparent language, one that unveiled an extra-linguistic reality, and saying only that we speak in many different ways, for many different purposes, and should not build a metaphysical, metalinguistic fantasy from those differences.

At each stage of his presentation Charlton offers succinct accounts of what other philosophers, from Plato to Locke to Wittgenstein to Timothy Williamson, have thought or said, and then worries at these accounts to modify or update them. ‘True’, for example, he interprets not as a property of statements or opinions, but of acts of saying (71). There are no actual entities, ‘thoughts’ or ‘propositions’, to be true or false, nor are there eternal untruths (‘that Caesar died in his bed’ or ‘that 7+ 5 = 11’): better suppose that we sometimes speak falsely as well as truly (pseudesthai or aletheuein). ‘That Caesar died in his bed is false, if saying that Caesar died in his bed is speaking falsely’ (73). Here and occasionally elsewhere Charlton almost suggests that (ancient) Greek offers a less misleading mode of speech: the speech act of comparing or linking one thing to another (its being pros ti, in Aristotle's categories) would not trigger, as the Latinate term ‘relation’ does, any suggestion that there are such countable things as ‘relations’ which must then be connected (related) at both ends to the original subjects (76, 87). Nor is there any such thing as ‘being’ to link a subject with its predicate: that Mars is spherical does not require the conjunction of three somewhats, but only that we may truly say ‘spherical’ of Mars (the planet), but not of Phobos or Deimos (its unrounded moons). Nor, merely because things may ‘look different’ from one angle or another, are there indefinitely many quantifiable ‘looks’ between the thing that is seen and the sentient creature that sees (71, 184).

Charlton takes a similar line with respect to being right and being good (or their opposites). Saying, for example, that ‘white is a good colour for the woodwork’ is as much as to say ‘if only the woodwork were white’: it is a recommendation which may be wise or foolish, but not ‘true’ or ‘false’ (103), unless we meant only that white appears in an authorized list of recommended colours (without ourselves endorsing that decree). Such recommendations need not be merely arbitrary: at some point they will turn on what serves some real needs, what best pleases or best benefits some entity. Sorting out what it is best to do is a matter, as Aristotle said, of working out our priorities, in the light of long experience of likely outcomes and more dispassionate judgement. ‘To be good is to be an objective’ (106) – chiefly or entirely, in Charlton's exposition, an objective of human beings like us, motivated by personal health and pleasure, by the laws of our community, and at last a disinterested concern for the wellbeing of others (even including the non-human). To ask for further reasons for these motives (identified in Plato's tripartite soul) when they themselves constitute the fundamental principles of all practical reasoning (107) is as inept as requiring further defence of modus ponendo ponens: neither that nor the wider principles of reasoning are premises to be defended, but the very mode of rational deliberation. It is possible, perhaps, to abandon them, but this is only to say that it is possible to act and think at random, not that we can do so and still claim to be thinking clearly. ‘Acquisition of these metaphysical concepts [truth, existence, goodness] must go in step with mastery of elementary grammar’ – and so be inaccessible to animals without language (114).

Turning from these most general of metaphysical topics (truth, existence, goodness) Charlton next examines change, time, causation, space, and how these concepts are expressed through ordinary language. As with the earlier cases, he proposes that the terms do not identify extra-linguistic realities, either by ‘standing for’ or ‘signifying’ them. There are changes and there are states: ‘becoming red’ is not the same as ‘being red’ – but perhaps such becomings can be analysed away into successive states? Something that wasn't red is now red: or even (dispensing with any implication that, somehow, time passes) it is not-red at T1, red at T2. Are time and change really real at all (a question, Charlton says (120), that only philosophers – and mystics – ever ask)? His own insistence that some things have happened ‘in the past’ and others will happen ‘in the future’, and that it would be ‘ungrammatical’ to replace these constructions by descriptions of what ‘timelessly is’ at distinct moments of a temporal dimension, is too rapid, resting as it does on a blank assertion that phrases indicating the date or time must be ‘adverbial’: ‘in “the poker was red on Monday” “on Monday” must be construed neither with “the poker” [as David Lewis] nor with “red” [as J.E. McTaggart] but with “was”’ (125). His argument against Bertrand Russell's prescription (simply that the poker has different properties at different moments) is not grounded in grammar but, oddly perhaps, in fantasy or thought-experiment: if the poker merely has different properties at different moments, then each momentary instantiation is logically independent of every other, and there need be no continuity at all. To which it is not unreasonable to respond that there is indeed no logical reason why there is: the poker might be successively red, green and colourless (or even briefly non-existent or made of chocolate), and the fact (if it is one) that its four-dimensional extension instead is visibly red at one end and black at another, with all appropriate gradations in between is a matter of physical and not logical law.

Rejecting Russell's account is an answer to the mere chance that reasonable prediction or retrodiction might be derailed. Conversely, Charlton's motive in insisting that remarks, and even accurate predictions, about the future only ‘become true’ in that future is to evade the dead hand of fatalism (that we can no more determine here and now what will happen than what did), but there are other ways of avoiding that moral – rather than metaphysical – disadvantage. Even his own account of ‘presentness’ as merely indicating that something is ‘contemporaneous’ with a particular utterance (‘it is no longer raining here in Bristol’, as I write), leaves it open that this ‘present now’ is no more universal or unique than ‘this present here’. The typing of these words is an event located in Bristol, in October 2014: neither place nor date are uniquely here and now (since there are many observers situated elsewhere and elsewhen). Conversely, if only the shifting moment is ever really experienced, then ‘the past’ exists no longer, ‘the future’ does not yet exist, and all our present claims about either have no accessible referent (any more than we can have access to the many merely ‘possible worlds’ of metaphysical speculation): we can no more say or assess what is true of past or future worlds than of possible worlds arising from any simple change (witness the large number of plausible ‘alternate histories’ stemming, say, from the Spanish Armada's success). Charlton is inclined simply to reiterate that past events ‘really did occur’ and future events ‘will’, and that the reality of these events would only be in question if nothing had ever happened, or that nothing further would, but still insists that ‘the present’ is somehow unique. To which the obvious reply is that every moment is contemporary with itself, and cannot cease to be so. The other problems posed by modern physicists (that ‘contemporaneity’ is not as unambiguously identified as we might have thought, and that there may even be closed time-like loops in the near vicinity of black holes) are not ones that Charlton reckons relevant to our first understanding of the terms we use (131). But we did not need those problems for time and change to be problematic! Nor is the thought that all times (whether we reckon them ‘earlier’ or ‘later’) are as real as October 2014 confined to metaphysicians and mystics.

But perhaps there is still scope for a roughly Aristotelian approach to all these issues: time is only what we measure against assorted cycles, of the moon or the sun or the hands of a clock, and even space is only what we measure in checking distances or container volumes. There is no empty time, nor any absolutely empty void within which there are no landmarks nor measurable metres. Charlton repeatedly brings his readers back to the obvious context of our lives together, attempting answers to the familiar questions: What time is it? How far to the next station? Are we there yet?

In addressing causation he pays more attention than usual to physical theories of motion, and on that basis presents an account of causal action as simply the continuation of a process. When a human agent brings about an event, the ongoing event and the bringing it about are, in a way, the very same: ‘the wheelbarrow moved because I pushed it’ (154); my pushing and its moving require no further connection, and the cause of the effect does not precede the effect. Which explains in turn why Charlton finds Russell's analysis of temporal difference unsatisfactory (as above). Without such deep continuities as underlie our apparatus of causal explanation every moment, every place, must be conceived as different, and there can be no rational answer to the Humean challenge. It is not merely that in a merely discontinuous world we could make no predictions: we could know nothing of what was going on; indeed, it is hard to see how we could even exist. Similarly in dealing with ‘materiality’ Charlton acknowledges both ancient and Enlightenment discussions of matter, suggesting that the effort to identify what causal powers were possessed by such mere stuff (solidity, perhaps, as well as mutability) was to miss the point. The causal powers displayed by any identifiable object to affect or change another just are its materiality: its having a particular location, speed, resilience and the like, from which a plausibly continuous story can be told (for example about the motion of billiard balls). ‘Matter’ does not identify an homogenous, barely characterizable, stuff from which all actual things are constituted, and into which they dissolve: just as time is only what we measure changes by, matter is what we weigh. Or rather, the matter of any identifiable object is its power to affect other such objects in certain definable ways: it can push them aside, dissolve, ignite or consume them. The alternative – a currently easy orthodoxy – is that the surface continuities, the seeming causes of our ordinary experience and action are more like moving pictures, in which it may seem that it is the swordsman who kills his enemy or the potter moulds his pot, but the reality is that all these seemings are like Russell's discontinuous states, a sequence of stills explained by who knows what (165).

In conclusion Charlton asks what it is to think, and follows a familiar thread to suggest that thought – though not all sentient experience – requires language. In attributing beliefs and desires to someone we are not speaking of distinguishable entities, let alone specific neural states, but offering teleological explanations for behaviour, normatively assessing that behaviour, whether or not it is also ‘causally determined’. An animal without the appropriate language cannot attribute beliefs or desires to others, nor even to herself, since she cannot be assessing the terms under which their (her) behaviour would be sensible (pace modern ethologists' suggestion that at least some non-human animals have ‘a theory of mind’, and so can recognize what other creatures do or do not know). Charlton comes close (188) to Richard Rorty's suggestion that even to attribute feelings to such an animal is only to express our determination to treat that animal in a certain way, for no very obvious reason: ‘the moral prohibitions [against hurting babies ‘and the better looking sorts of animals’] are expressions of a sense of community based on the imagined possibility of conversation, and the attribution of feelings is little more than a reminder of these prohibitions' (Philosophy and the Mirror of Nature (Princeton University Press 1979), 190). This is probably not what Charlton himself intends.

In conclusion, ‘our world is one of change, time and causation, in which we act on purpose and for reasons’ (209), and (too) ‘much philosophy in modern times has been devoted to depriving our world of these features, and trying to reconcile us to this impoverishment’ (209). Charlton by contrast insists that ‘nothing is more important than that the world contains conscious beings, and that we ourselves are not just conscious but intelligent’ (211). His own intelligence is clear, and even those who might be inclined to disagree with him in detail must find their own intelligence challenged – and perhaps enhanced – by his bracing attention to past philosophies, to grammar, and to the purported puzzles of metaphysics.