Cicero, pre-eminent Roman statesman and orator of the first century bc and a prolific writer, composed the first substantial body of philosophical work in Latin. Rising from small-town obscurity to the pinnacle of Rome’s staunchly conservative aristocracy, he devoted most of his life to public affairs. But he was deeply interested in philosophy throughout his life, and during two intervals of forced withdrawal from politics wrote two series of dialogues, first elaborating his political ideals and later examining central issues in epistemology, ethics and theology. Designed to establish philosophical study as an integral part of Roman culture, these works are heavily indebted to Greek philosophy, and some of the later dialogues are largely summaries of Hellenistic debates. But Cicero reworked his sources substantially, and his methodical expositions are thoughtful, judicious and, on questions of politics and morals, often creative. An adherent of the sceptical New Academy, he was opposed to dogmatism but ready to accept the most cogent arguments on topics important to him. His vigorously argued and eloquent critical discussions of perennial problems greatly enriched the intellectual and moral heritage of Rome and shaped Western traditions of liberal education, republican government and rationalism in religion and ethics. These works also afford invaluable insight into the course of philosophy during the three centuries after Aristotle.

1 Life and writings

Marcus Tullius Cicero, elder son of a locally influential family in the town of Arpinum, moved to Rome in his youth to pursue a career in law and government. There he studied with several Greek philosophers, including the Academic Philo of Larissa, and after a brilliant legal debut he spent two years in Greece studying philosophy and rhetoric with Antiochus and the Stoic Posidonius. Upon his return he won election to a major office that brought lifelong membership of the Senate (Rome’s supreme governing body) and soon established himself as the foremost advocate of the age. Elected consul (Rome’s chief executive office) in 63 bc, he suppressed an insurrection and was hailed his country’s saviour. But opponents contrived his exile in 58 bc, and when he returned the following year, he found his influence severely diminished. Turning to writing, he formulated and defended his political ideals in three pioneering dialogues (see §2). When Julius Caesar precipitated civil war in 49 bc, Cicero sided reluctantly with the opposition as the lesser threat to Roman institutions. Caesar’s swift victory brought dictatorship, and Cicero, although granted clemency, was excluded from politics. Returning to writing, he championed free political discussion in a series of rhetorical works, then composed in twenty months a dozen works (nine survive whole or in large part) discussing central problems in Hellenistic philosophy (see §§3–4) (see Hellenistic philosophy ). The political turmoil that followed Caesar’s assassination in March 44 bc slowed, then halted this astonishing pace as he rallied resistance to Mark Antony’s despotic designs. His campaign might well have succeeded had Antony not colluded with Caesar’s adoptive heir, the future Augustus: Cicero was assassinated and his head impaled in the Forum where he had spoken so often and so eloquently.

Cicero’s extant works, although only part of his enormous output, comprise over fifty speeches, nearly a thousand letters to friends and associates, several works on rhetorical theory and practice, and twelve on philosophical topics. This vast corpus, besides displaying great intellectual range and stylistic virtuosity, embodies Cicero’s conviction that philosophy and rhetoric are interdependent and both essential for the improvement of human life and society. His oratory bears the stamp of his theoretical studies, and his treatises and dialogues are richly oratorical. The philosophical works in particular unite the rhetorical techniques and ample style of Roman oratory with the analytical methods and conceptual apparatus of Greek philosophy in a unique fusion of eloquence and insight. All but one of these works are fictional dialogues. Some portray Cicero or eminent Romans of the previous century discoursing at length among friends; others, employing a format that reflects Roman political and legal practice but also the critical spirit Cicero admired in Plato and his sceptical heirs in the New Academy (see §3), present paired speeches for and against Epicurean and Stoic theories. Composed for audiences unused to abstract theory and systematic analysis, the discussions lapse at times into earnest declamation, and the close questioning found in Plato’s dialogues is rare. However, they are methodically organized and often incisive, and by presenting opposing views and arguments in clear and engaging terms, they dramatize the significance of fundamental problems and encourage critical reflection.

2 Classical republicanism

Cicero’s first philosophical works are three long dialogues that analyse and evaluate the political institutions and practices of contemporary Rome in the light of Greek theory. Although largely conservative, they provided the first political theory in Latin and remain the most systematic ancient account of Roman government; while others described events, only Cicero advanced a structural analysis. Written when Rome’s republican traditions were collapsing under unprecedented concentrations of economic and military power, these dialogues champion political liberty, rational debate and rule by law. Articulating the principles behind his lifelong goal of harmonizing Rome’s competing interests in a just and stable ‘concord of the orders’, they propound a comprehensive vision of civil society directed by an elected elite schooled in rhetoric and philosophy, devoted to constitutional government and able to shape public opinion through effective oratory.

The first of these works, On the Orator , explores the role of rhetoric and philosophy in public life. Oratory had long been a potent tool in Roman politics, and Cicero aims to reinforce its prestige and legitimize its influence by showing that its success requires wide learning and sound reasoning. Much of the discussion focuses on education, as he weighs the merits of the traditional Roman emphasis on history, poetry and practical experience against the Greek disciplines of formal rhetoric and philosophy. His model orator, who clearly reflects Cicero’s own proficiencies, unites thorough knowledge of history and law with complete command of logical method, philosophical theory and rhetorical techniques in a Romanized version of Plato’s philosopher-rulers. Both expect philosophical education to produce the best statesmen; but whereas Plato’s ideal hinges on mathematical training and transcendental metaphysics, Cicero proposes a thoroughly pragmatic programme of instruction designed to foster eloquence and informed civic debate.

Cicero delineates the institutional framework behind his conception of leadership in On the Republic, which was almost entirely lost until most of the first third was recovered in 1820. Challenging the utopian bent of Greek political theory exemplified by Plato’s Republic , Cicero argues that the best constitution, far from being unattainable, was largely realized in Rome, where a unique blend of monarchy, aristocracy and democracy formed a ‘mixed constitution’ that provided a system of government ostensibly stable and just. This account, which recasts Rome’s narrow oligarchy as a paradigm of aristocratic paternalism, rests on an incisive and apparently original analysis of political legitimacy. Defining a republic (res publica) as ‘a people’s affair’ (res populi) and a people as a community ‘united by consensus about right and by mutual interest’, Cicero criticizes all other constitutions for contravening the people’s rights and interests, then argues that no political system is legitimate unless it distributes legal rights equally to all, but electoral, legislative and judicial authority proportionally according to merit and wealth. Other extant sections of the dialogue classify types of constitution in classical Greek fashion and survey the development of Roman institutions; lost or poorly preserved sections summarized Hellenistic debates about the nature and rewards of justice, discussed Roman education and measured past Roman statesman against Cicero’s ideal. Ending the work is the famous ‘dream of Scipio’, which interweaves astronomy and eschatology to sketch a theodicy that rewrites Plato’s myth of Er in the light of Stoic cosmology and exalts public service by reserving the finest posthumous rewards for outstanding statesmen.

On Laws, a sequel probably left incomplete and published only after Cicero’s death, fills in his constitutional model by outlining a comprehensive legal system. Vying with Plato’s Laws this time, he continues his argument that Rome already embodied much of the ideal. His treatment of religion and political administration, which is all that survives, is deeply conservative, largely an explanation and defence of existing statutes and institutions by appeal to Greek theory and Roman history, with proposals for change limited to streamlining and archaizing reforms rather than extensive revision or thorough codification. The work was extremely influential, especially on Christian and early modern thought, because it contains the fullest surviving ancient account of natural law (see Natural law ). Drawing heavily on Stoic ideas, Cicero argues that the natural world exhibits a divinely ordained and rationally intelligible order that can be codified in legislation and provides the ultimate tribunal for all positive law.

Civil war interrupted Cicero’s writing and he never returned to constitutional or legal theory. But his very last book rounds out his political thinking by examining the role of personal morality in public life. In the guise of an extended epistle to his son, On Duties maps out a code of conduct for the Roman nobility that emphasizes justice, benefaction and public service. The focus throughout is on men of high station and the problems of integrating personal ambitions and social obligations. Borrowing heavily from the Stoic Panaetius (§2) , Cicero argues that virtuous conduct is always expedient as well as morally required, and that apparent conflicts between morality and personal advantage are illusory because virtuous action is always the best option.

3 Scepticism and natural theology

Cicero embraced the sceptical stance of the New Academy in his youth when he studied under Philo of Larissa. Modern scholars debate whether he shifted soon afterward to positions he learned from Antiochus, or remained an Academic sceptic throughout life. But a likely explanation for this debate is the moderate tenor of Cicero’s scepticism, which eschews dogmatic certainty but accepts ‘convincing’ (probabile) arguments. Inclined by profession to adversarial argument, Cicero employed the Academic strategy of arguing on both sides of questions in his earlier as well as his later dialogues. But his aim was never suspension of judgment, as in the early New Academy under Arcesilaus (§2). Rather, adopting a pragmatic fallibilism that derives from Carneades (§4) but probably owes most to Philo, he considered the careful weighing of opposing arguments the most reliable route to truth. On this basis, he endorsed an eclectic synthesis of ideas derived mainly from Antiochus and Stoics but also from Aristotle and Plato (whom he styles the prince of philosophers and quotes often in original translations); and while reserving judgment on many theoretical issues, he held firm views about many practical matters of politics and morality.

Cicero discusses his sceptical stance most fully in Academics, which inaugurates his second series of dialogues by recounting a pivotal debate within the later New Academy about the nature and possibility of knowledge. Only parts of the work survive. But one fully preserved book (entitled Lucullus) summarizes Antiochus’ heavily Stoicizing brand of empiricism, then rehearses a battery of sceptical objections drawn mainly from Carneades and Philo. Debate centres on Antiochus’ defence of the Stoic ‘criterion of truth’ (see Stoicism §§12–13 ), a class of allegedly self-certifying perceptions that would be the source and foundation of any knowledge. Cicero himself delivers the sceptical rebuttal, arguing forcefully that knowledge requires certainty, but that certainty is neither attainable nor necessary for the rational conduct of life. Both sides are powerfully argued, and in line with Philo’s fallibilism, readers are obliged to assess which seems more ‘convincing’. Though frequently elliptical, the discussion maintains a level of argument rarely matched in Latin before Augustine.

Cicero continued to use this Academic method of presenting opposing arguments in most of his subsequent dialogues. On the Nature of the Gods presents two pairs of speeches expounding, then criticizing, Epicurean and Stoic theology. The central issue is not whether gods exist, but rather their role in nature and human affairs. Epicureanism envisages immortals residing far beyond this world and oblivious to human affairs; Stoicism maintains that the world itself is rational and divine, operating in accord with providence to the benefit of humanity; and the Academic critiques, without denying the existence of gods, argue that both schools rely on highly dubious assumptions and wholly fail to prove their cases. The debate covers questions of cosmology at length, including both an impassioned defence and sustained criticism of Stoic arguments from design; and the beliefs and rituals of Greek and Roman polytheism are criticized and reinterpreted throughout. The whole discussion, in its reliance on canons of rational inquiry and analysis, is a model of natural theology. Refusing to rely on revelation or authority, all sides agree that religious issues must be decided by evidence and argument; Stoic appeals to divination and widespread belief in gods, for example, are advanced as inferences to the best explanation. In the end, Cicero endorses the Stoic conception of the world as governed benevolently by divine reason, which he clearly found more compelling both intellectually and morally than either the stark amorality of Epicurean atomism or the official religion of Rome which his public station obliged him to uphold.

Two subsequent works examine related problems of knowledge and causation. On Divination presents a richly illustrated defence followed by a scathing critique of various religious claims to foreknowledge. Highlighting epistemological issues, Cicero constructs remarkably balanced cases. The defence, appealing to Stoic arguments that divination is a natural consequence of divine providence and often based on systematic observation, assembles a vast gallery of empirical evidence; his reply, in keeping with the rigorous rationalism of Academic scepticism, challenges the validity of this evidence and argues that divination is inexplicable and inconsistent with the evident regularity of nature. But Cicero, despite ridiculing the defence and pronouncing much of Roman religion sheer superstition, offers no final verdict and endorses traditional rites on grounds of social utility. On Fate analyses the closely connected topic of determinism. Cicero defends a theory of free will which he attributes to Carneades, while criticizing the Epicureans for postulating uncaused atomic ‘swerves’ (see Epicureanism §12 ), and arguing that the Stoic conception of fate as a universal and eternal causal web precludes voluntary action and moral responsibility. Detailed analysis of Stoic theories of causation and logical consequence makes this perhaps Cicero’s most exacting work.

4 Humanist ethics

Informing all of Cicero’s work is a profound faith in the natural goodness of humanity and the power of reason to direct and improve human life. The assumptions behind this humanist outlook, which is probably Cicero’s most constructive synthesis of Greek ideas, are systematically examined in On the Ends of Good and Evil (usually known as the De finibus). Tackling the central question of ancient ethical theory, the dialogue inquires into the ultimate end of human action and how happiness is attained (see Eudaimonia ). Paired speeches expound and criticize Epicurean and Stoic accounts of human nature and the status of moral virtue; ending the work and receiving only brief criticism is a neo-Aristotelian account derived from Antiochus. Cicero speaks throughout as an Academic sceptic, arguing that Epicurean hedonism is incoherent and morally subversive, and challenging the Stoic doctrine that moral virtue is the sole good and hence sufficient for happiness. But he commends the moral austerity and theoretical rigour of Stoic ethics, and while he finds the Aristotelian position intuitively attractive and most conducive to public service, he questions whether Antiochus’ view that non-moral interests are also intrinsically good undermines the supremacy of moral virtue.

Setting aside sceptical worries to address practical questions, Cicero explores some applications of his ethical rationalism in two substantial works and a pair of earnest but elegant moralizing essays. On Duties employs a Stoic framework to spell out systematic rules of conduct (see §2). Tusculan Disputations similarly uses Stoic theory to analyse problems in moral psychology. Adopting a format used by Carneades, Cicero presents five lengthy disquisitions refuting common beliefs about the emotions; but since each speech upholds a Stoic paradox, the result is a sustained defence of Stoic doctrines. Invoking a legion of philosophical and literary authorities, Cicero argues vigorously that philosophy is the medicine of the soul, and that it alone enables us to scorn death, endure pain, overcome grief and other passions, and lead good lives. Much of the argument rests on an acute Stoic analysis of emotions as governed by beliefs. But Cicero’s ideal of rational restraint and self-control as the source of mental tranquillity and happiness distills ideas central to Greek and Roman culture alike. Two shorter dialogues portray eminent Romans from the previous century as sage advisors on more personal topics. On Old Age enumerates the lasting rewards of honourable character and education, including a glorious afterlife. On Friendship extols a paradigm of aristocratic male companionship based on mutual benefit but also integrity and loyalty. Both works, while distinctly less systematic than Cicero’s other dialogues, exemplify his ability to illuminate vital human concerns with philosophical insight and graceful eloquence.

5 Influence

Cicero was the most influential writer and intellect of his time, and his impact on Western culture has been lasting and profound. His philosophical writings, by forging expressions essential for theoretical discussion, inaugurated over sixteen centuries of philosophy in Latin. They also fuelled the rise of Christianity in the West, as the Latin Fathers mined his dialogues in their campaigns against pagan religion and philosophy (see Patristic philosophy). Augustine, whose life was transformed by the exhortation to philosophy in Cicero’s lost Hortensius, drew on his writings extensively, especially in Against the Academics and City of God . Through these and other writers, most notably Ambrose, Jerome and the pagan Macrobius (whose Neoplatonic commentary on ‘Scipio’s dream’ was widely studied), Cicero’s ideas shaped medieval thought, especially ethics and theories of natural law. His influence reached its zenith in the Renaissance, when Erasmus and other humanists emulated his critical spirit and reaffirmed his secular outlook and ecumenical ideals (see Renaissance philosophy). By making his writings the foundation of a liberal education, they also increased his moral authority; and in the following centuries his ethical and political works (above all On Duties, dubbed ‘Tully’s Offices’ in English) fostered the revival of republicanism and the development of liberalism, while his dialogues on religion were inspirational to deism. Voltaire proclaimed him the model of enlightened reason; and Hume was deeply indebted to Cicero, especially in his critique of religious dogmatism and his conception of ‘mitigated’ scepticism. The rise of idealism in the nineteenth century lowered Cicero’s philosophical reputation considerably. But renewed interest in scepticism and virtue ethics, along with improved understanding of Hellenistic philosophy, has in recent decades stimulated intense discussion of his work yet again.

Aristotle of Stagira is one of the two most important philosophers of the ancient world, and one of the four or five most important of any time or place. He was not an Athenian, but he spent most of his life as a student and teacher of philosophy in Athens. For twenty years he was a member of Plato’s Academy; later he set up his own philosophical school, the Lyceum. During his lifetime he published philosophical dialogues, of which only fragments now survive. The ‘Aristotelian corpus’ (1462 pages of Greek text, including some spurious works) is probably derived from the lectures that he gave in the Lyceum.

Aristotle is the founder not only of philosophy as a discipline with distinct areas or branches, but, still more generally, of the conception of intellectual inquiry as falling into distinct disciplines. He insists, for instance, that the standards of proof and evidence for deductive logic and mathematics should not be applied to the study of nature, and that neither of these disciplines should be taken as a proper model for moral and political inquiry. He distinguishes philosophical reflection on a discipline from the practice of the discipline itself. The corpus contains contributions to many different disciplines, not only to philosophy.

Some areas of inquiry in which Aristotle makes a fundamental contribution are these:

(1) Logic. Aristotle’s Prior Analytics constitutes the first attempt to formulate a system of deductive formal logic, based on the theory of the ‘syllogism’. The Posterior Analytics uses this system to formulate an account of rigorous scientific knowledge. ‘Logic’, as Aristotle conceives it, also includes the study of language, meaning and their relation to non- linguistic reality; hence it includes many topics that might now be assigned to philosophy of language or philosophical logic (Categories, De Interpretatione, Topics ).

(2) The study of nature. About a quarter of the corpus (see especially the History of Animals, Parts of Animals, and Generation of Animals; also Movement of Animals, Progression of Animals) consists of works concerned with biology. Some of these contain collections of detailed observations. (The Meteorology contains a similar collection on inanimate nature.) Others try to explain these observations in the light of the explanatory scheme that Aristotle defends in his more theoretical reflections on the study of nature. These reflections (especially in the Physics and in Generation and Corruption) develop an account of nature, form, matter, cause and change that expresses Aristotle’s views about the understanding and explanation of natural organisms and their behaviour. Natural philosophy and cosmology are combined in On the Heavens .

(3) Metaphysics. In his reflections on the foundations and presuppositions of other disciplines, Aristotle describes a universal ‘science of being qua being’, the concern of the Metaphysics . Part of this universal science examines the foundations of inquiry into nature. Aristotle formulates his doctrine of substance, which he explains through the connected contrasts between form and matter, and between potentiality and actuality. One of his aims is to describe the distinctive and irreducible character of living organisms. Another aim of the universal science is to use his examination of substance to give an account of divine substance, the ultimate principle of the cosmic order.

(4) Philosophy of mind. The doctrine of form and matter is used to explain the relation of soul and body, and the different types of soul found in different types of living creatures. In Aristotle’s view, the soul is the form of a living body. He examines the different aspects of this form in plants, non-rational animals and human beings, by describing nutrition, perception, thought and desire. His discussion (in On the Soul, and also in the Parva Naturalia ) ranges over topics in philosophy of mind, psychology, physiology, epistemology and theory of action.

(5) Ethics and politics (Nicomachean Ethics, Eudemian Ethics, Magna Moralia). In Aristotle’s view, the understanding of the natural and essential aims of human agents is the right basis for a grasp of principles guiding moral and political practice. These principles are expressed in his account of human wellbeing, and of the different virtues that constitute a good person and promote wellbeing. The description of a society that embodies these virtues in individual and social life is a task for the Politics , which also examines the virtues and vices of actual states and societies, measuring them against the principles derived from ethical theory.

(6) Literary criticism and rhetorical theory (Poetics, Rhetoric ). These works are closely connected both to Aristotle’s logic and to his ethical and political theory.

1 Life

Aristotle was born in 384 bc, in the Macedonian city of Stagira, now part of northern Greece. In his lifetime the kingdom of Macedon, first under Philip and then under Philip’s son Alexander (‘the Great’), conquered both the Greek cities of Europe and Asia and the Persian Empire. Although Aristotle spent much of his adult life in Athens, he was not an Athenian citizen. He was closely linked to the kings of Macedon, whom many Greeks regarded as foreign invaders; hence, he was affected by the volatile relations between Macedon and the Greek cities, especially Athens.

Aristotle was the son of Nicomachus, a doctor attached to the Macedonian court. In 367 bc Aristotle came to Athens. He belonged to Plato’s Academy until the death of Plato in 347; during these years Plato wrote his important later dialogues (including the Sophist, Timaeus, Philebus, Statesman, and Laws ), which reconsider many of the doctrines of his earlier dialogues and pursue new lines of thought. Since there was no dogmatic system of ‘Platonism’, Aristotle was neither a disciple of such a system nor a rebel against it. The exploratory and critical outlook of the Academy probably encouraged Aristotle’s own philosophical growth.

In 347 bc Aristotle left Athens, for Assos in Asia Minor. Later he moved to Lesbos, in the eastern Aegean, and then to Macedon, where he was a tutor of Alexander. In 334 he returned to Athens and founded his own school, the Lyceum. In 323 Alexander died; in the resulting outbreak of anti-Macedonian feeling in Athens Aristotle left for Chalcis, on the island of Euboea, where he died in 322.

Aristotle married Pythias, a niece of Hermeias, the ruler of Assos. They had a daughter, also called Pythias. After the death of his wife, Aristotle formed an attachment to Herpyllis, and they had a son Nicomachus.

2 Order of Aristotle’s works

By the end of Aristotle’s life the Lyceum must have become a well-established school. It lasted after Aristotle’s death; his successor as head of the school was his pupil Theophrastus. Many of the works in the Aristotelian corpus appear to be closely related to Aristotle’s lectures in the Lyceum. The polished character of some passages suggests preparation for publication (for example, Parts of Animals I 5), but many passages contain incomplete sentences and compressed allusions, suggesting notes that a lecturer might expand (for example, Metaphysics VII 13). We cannot tell how many of his treatises Aristotle regarded as ‘finished’ (see §11 on the Metaphysics and §21 on the Ethics ).

It may be wrong, therefore, to ask about the ‘date’ of a particular treatise. If Aristotle neither published nor intended to publish the treatises, a given treatise may easily contain contributions from different dates. For similar reasons, we cannot plausibly take cross-references from one work to another as evidence of the order of the works. External, biographical considerations are unhelpful, since we lack the evidence to support any detailed intellectual biography of Aristotle.

A few points, however, may suggest a partial chronology.

(1) Some of Aristotle’s frequent critical discussions of Plato and other Academics may have been written (in some version) during Aristotle’s years in the Academy. The Topics may reflect the character of dialectical debates in the Academy.

(2) It is easier to understand the relation of the doctrine of substance in the Categories and Physics I–II to the doctrine and argument of Metaphysics VII if we suppose that Metaphysics VII is later.

(3) The Organon (see §4) does not mention matter, perhaps because (a) Aristotle had not yet thought of it, or because (b) he regarded it as irrelevant to the topics considered in the Organon. If (a) is correct, the Organon precedes the works on natural philosophy.

(4) Some of the observations used in Aristotle’s biological works probably came from the eastern Aegean. Hence, Aristotle probably pursued his biological research during his years away from Athens. We might trace his biological interests to the Academy (see Plato’s Timaeus ); he may also have acquired them from his father Nicomachus, who was a doctor. Probably, then, at least some of the biological works (or versions of them) are not the latest works in the corpus.

(5) The Magna Moralia (if it is genuine) and the Eudemian Ethics probably precede the Nicomachean Ethics (see §21).

The order in which Aristotle’s works appear in the Greek manuscripts goes back to early editors and commentators (from the first century bc to the sixth century ad); it reflects their view not about the order in which the works were written, but about the order in which they should be studied. This entry generally follows the order of the corpus, except that it discusses On the Soul after the Metaphysics (see §17), not among the works on natural philosophy (where it appears in the manuscripts).

3 Appearances

The general aim of rational inquiry, according to Aristotle, is to advance from what is ‘better known to us’ to what is ‘better known by nature’ (see Physics I 1; Posterior Analytics 71b33; Metaphysics 1029b3). We achieve this aim if: (1) we replace propositions that we thought we knew with propositions that we really know because they are true and we understand them; (2) we find general principles that explain and justify the more specific truths that we began from; (3) we find those aspects of reality that explain the aspects that are more familiar to us.

The things better known to us in a particular area are the relevant ‘appearances’ (phainomena). Aristotle presents them through detailed collections of empirical data, reached as a result of ‘inquiry’ (historia; for example, Parts of Animals 646a8). Empirical inquiry proceeds from particular observations, by means of generalizations through induction (epagōgē) from these particular cases, until we reach experience (empeiria). Experience leads us to principles that are better known by nature ( Prior Analytics 46a17); we also rely on it to test principles we have found ( Generation of Animals 760b28).

Philosophical inquiry also relies on ‘appearances’. However, the appearances that concern it are not empirical observations, but common beliefs, assumptions widely shared by ‘the many and the wise’. The critical and constructive study of these common beliefs is ‘dialectic’. Aristotle’s method is basically Socratic. He raises puzzles in the common beliefs, looking for an account that will do them justice as a whole. Among common beliefs Aristotle considers the views of his predecessors (for example, Metaphysics I; On the Soul I; Politics II), because the puzzles raised by their views help us to find better solutions than they found.

Inquiry leads us to causes and to universals. Aristotle has a realist conception of inquiry and knowledge; beliefs and theories are true in so far as they grasp the reality that we inquire into (see Realism and antirealism §2 ). Universals and causes are ‘prior by nature’; they are not created by, or dependent on, any theory, but a true theory must fit them.

If we attended only to Aristotle’s remarks on what is better known to us and on the process of inquiry, we might regard his position as a form of empiricism (see Empiricism ). But in his remarks on what is better known by nature, he insists on the reality of universals and on the importance of non-sensory forms of knowledge (see §15 on universals, §19 on thought).

4 Thought and language

One means of access to appearances, and especially to common beliefs, is the study of what words and sentences ‘signify’ (sēmainein). This is part of ‘logic’ (logikē, derived from logos, which may be translated ‘word’, ‘speech’, ‘statement’, ‘argument’ or ‘reason’: see Logos), which is discussed in the first section of Aristotle’s works (Categories, De Interpretatione, Prior Analytics, Posterior Analytics, Topics ). This section of the corpus came to be called the ‘Organon’ (‘instrument’), because logic, as Aristotle conceives it, concerns statements and arguments in general, without restriction to any specific subject matter; it is therefore an instrument of philosophical inquiry in general, rather than a branch of philosophy coordinate with natural philosophy or ethics. The Organon includes some elements of philosophy of language, as well as formal logic (syllogistic; see §5) and epistemology (see §6).

According to Aristotle’s account of signification (see especially De Interpretatione 1–4), as commonly understood, the word ‘horse’ signifies horse by signifying the thought of horse; in using the word, we communicate thoughts about horses. When the thoughts about horses we communicate are true, we communicate truths about the universal horse; even when our thoughts are not completely true, we may signify the same universal horse.

To understand the signification of a name ‘F’, we look for the corresponding definition (logos, horismos) of F. Aristotle distinguishes nominal definitions, stating the beliefs associated with the name, from real definitions, giving a true account of the universal that underlies the beliefs embodied in the nominal definition (see Posterior Analytics II 8–10. Aristotle himself does not use the labels ‘nominal definition’ and ‘real definition’.).

Not every name corresponds to one nominal and one real definition. Some names correspond to no genuine universal; ‘goatstag’ signifies (in one way) animals that are both goats and stags, but it does not signify a genuine universal, since there is no natural kind of goatstag. Other names correspond to more than one universal, as ‘chest’ signifies both a container and a part of an animal. Chests are ‘homonymous’ (homōnyma) or ‘multivocal’ (pollachōs legomena; ‘spoken of in many ways’); more than one definition is needed to capture the signification of the name. By contrast, since only one definition corresponds to the name ‘horse’, horses are ‘synonymous’ ( Categories 1).

Other philosophers make serious errors, Aristotle believes, because they suppose they can give a single account of things or properties that are really multivocal. Once we see that different Fs are F in different ways, we see that different, although (in many cases) connected, accounts of what it is to be F must be given. Some philosophically important cases of multivocity are cause (Aristotle’s doctrine of the four causes; see §9), being (the doctrine of the categories; see §7) and good (the criticism of Plato’s belief in a Form of the Good; Nicomachean Ethics I 6).

5 Deduction

Part of logic, as Aristotle conceives it, is the study of good and bad arguments. In the Topics Aristotle treats dialectical arguments in general. In the Prior Analytics he examines one type of argument, a ‘deduction’ (syllogismos; literally, ‘reasoning’, hence the standard term ‘syllogism’). This is an argument in which, if propositions p and q are assumed, something else r, different from p and q, follows necessarily because of the truth of p and q (Prior Analytics 24b18–20, paraphrased). Aristotle insists that it is not possible for the premises of a deduction to be true and the conclusion false (‘follows necessarily’); that a deduction must have more than one premise (‘if p and q are assumed’); that the conclusion cannot be identical to any premise (‘different from p and q’); and that no redundant premises are allowed (‘because of the truth of p and q’). He takes deductions to express affirmative or negative relations between universals, taken either universally (‘Animal belongs to every (no) man’) or not universally (‘Animal belongs (does not belong) to some man’). He takes the affirmative and negative claims to imply existence (so that ‘Biped belongs to some dodo’ follows from ‘Biped belongs to every dodo’; the latter affirmation is not equivalent, therefore, to ‘If anything is a dodo, it is biped’).

These different features of an Aristotelian deduction differentiate Aristotle’s account of a deduction from a more familiar account of deductively valid arguments. An argument may be valid even if it is redundant, or a premise is identical to the conclusion, or it has only one premise, or it is about particulars, or it contains neither ‘some’ nor ‘every’ nor ‘belongs’; but no such argument is an Aristotelian deduction. Aristotle’s theory of the different forms of deduction (often called ‘the moods of the syllogism’) examines the various forms of argument that necessarily preserve the truth of their premises. He begins from ‘complete’ (or ‘perfect’) deductions whose validity is evident, and classifies the different types of arguments that can be derived from (shown to be equivalent to) the complete deductions. He also explores the logical relations between propositions involving modalities (‘Necessarily (possibly) animal belongs to every man’ and so on). Since Aristotle accepts this relatively narrow account of a deduction, his exploration of the different forms of deduction is not a theory of valid arguments in general; the Stoics come much closer to offering such a theory (see Stoicism §11; Logic, ancient ).

Aristotle’s theory of deduction is developed for its own sake, but it also has two main philosophical applications. (1) Deduction is one type of argument appropriate to dialectic (and, with modifications, to rhetoric; see §29). Aristotle contrasts it with inductive argument (also used in dialectic), in which the conclusion does not follow necessarily from the premises, but is made plausible by them. (2) It is essential for demonstration (apodeixis), which Aristotle takes to be the appropriate form for exhibiting scientific knowledge.

6 Knowledge, science and demonstration

The progress from what is known to us to what is known by nature aims at epistēmē, the scientific knowledge whose structure is exhibited in the demonstrative pattern described in the Posterior Analytics. A demonstration is a deduction in which the premises are necessarily true, prior to and better known than the conclusions, and explanatory of the conclusions derived from them. Aristotle assumes that if I know that p, then I can cite some justification q, to justify my belief that p, and I also know why q justifies p ( Posterior Analytics I 2). The right sort of justification relies on things better known by nature – the general laws and principles that explain the truth of p. Since these are embodied in demonstrations, grasp of a demonstration of p expresses knowledge of p. Aristotle’s theory of demonstration, then, is not intended to describe a procedure of scientific inquiry that begins from appearances; it is an account of the knowledge that is achieved by successful inquiry.

To show that a deduction is a demonstration, we must show that its premises are better known than the conclusion. Sometimes we can show this by demonstrating them from higher premises that are even better known. This process of justification, Aristotle claims, must be linear and finite. A circular ‘justification’ must eventually ‘justify’ a given belief by appeal to itself, and an infinite regress imposes on us a task that we can never complete. Since, therefore, neither a circle nor an infinite regress can really justify, a proper justification must ultimately appeal to primary principles of a science.

These primary principles are ‘assumptions’ (hypotheseis); we must see that they are better known and prior to other truths of a science, without being derived from any further principles. Since they are the basis of all demonstration, they cannot themselves be demonstrated; Aristotle claims that we have non-demonstrative understanding (nous: Posterior Analytics II 19) of the ultimate principles of each science (see Nous ).

How are we entitled to claim understanding of an ultimate principle? Aristotle believes that the principles of a science are reached from appearances (perceptual or dialectical or both), which are the starting points known to us. He may believe that this relation of the principles to appearances justifies us in accepting them as first principles and in claiming to have understanding of them. This explanation, however, does not easily fit Aristotle’s demand for linear and finite chains of justification. That demand suggests that the assumptions of a science must be self- evident (seen to be true without any inferential justification), so that his conception of knowledge expresses a foundationalist position (see Foundationalism §3). (On difficulties in foundationalism see Agrippa .)

Although Aristotle’s aim of reaching a demonstrative science reveals some of his epistemological doctrines and assumptions, it does not evidently influence most of the structure or content of most of the surviving treatises. In his main philosophical works, the influence of dialectical methods and aims is more apparent.

7 Categories and beings

Part of the task of logic is to explain the nature of predication (‘A is B’, analysed by Aristotle as ‘B is predicated of A’ or ‘B belongs to A’, as in ‘Animal belongs to every man’), which is presupposed by complex logoi (statements and arguments). In the Categories (katēgoriai; predications), Aristotle introduces ten ‘categories’ (usually called schēmata tēs katēgorias, ‘figures (that is, types) of predication’). The categories correspond to different sorts of words (for example, count-nouns, adjectives, verbs) and to different grammatical functions (for example, subject, predicate), but they primarily classify the different non-linguistic items introduced in predications. The sentences ‘Socrates is a man’ and ‘Socrates is a musician’ are grammatically similar, but they introduce different sorts of things; the first predicates a second substance of a first substance, whereas the second predicates a non-substance of a first substance.

The first category is called ousia (literally, ‘being’), which is translated into Latin as ‘substantia’, and hence usually called ‘substance’ (see Substance §1). The nine non- substance categories include quality, quantity and relative (the only ones that Aristotle refers to often; the categories are listed in Categories 4, Topics I 9). Each category contains both particulars and universals. The statement that this individual man is an animal predicates a second substance (that is, a universal in the category of substance) of a first substance (that is, a particular in the category of substance). ‘White is a colour’ predicates one universal quality of another.

The categories display the multivocity of beings (see §4). Whereas animals constitute an ordinary univocal genus with a single definition, beings do not constitute an ordinary genus; hence there is no single account of what it is for something to be a being. Aristotle believes Plato mistakenly pursued a single account of beings; the theory of categories is meant to avoid Platonic errors.

In marking categorial divisions, Aristotle is influenced by grammar and syntax, but also by his ontology – his classification of beings. This classification rests on his view of nature and change, which clarifies his analysis of predication.

8 Change and substance

Aristotle’s Physics discusses nature, physis. The nature of x is a principle (or ‘source’; archē), internal to x, of change and stability in x; hence the inquiry into nature leads to a discussion of change in natural substances (the elements, plants and animals). Aristotle proceeds dialectically, raising and solving puzzles involved in the understanding of natural change. In solving the puzzles, he introduces the different types of beings that are presupposed by a coherent account of natural change.

In Physics I 7–8, Aristotle analyses a simple example of change – Socrates changing from being pale to being tanned. This change involves a subject (or ‘underlying thing’; hypokeimenon), Socrates, who loses one contrary (his pale colour) and acquires another contrary (his tan). Neither of the contraries persists, but the subject persists (otherwise there would not be a change in Socrates). This particular subject that persists through change is what the Categories calls a first substance. First substances differ both from second substances and from non-substances by being capable of undergoing change; they persist while receiving opposites (as Socrates is first pale and then tanned). They cannot, however, remain in existence irrespective of any properties gained or lost; Socrates’ ceasing to be a man is not a change in Socrates, but the perishing of Socrates.

The properties that a first substance cannot lose without perishing constitute (approximately) the essence of that first substance (see Essentialism ). These essential properties define a kind to which the first substance belongs. A kind may be a species (eidos), for example, man or horse, or a genus (genos), for example, animal. In predicating a second substance of a first substance (as in ‘Socrates is a man’), we place the first substance in the kind it belongs to. If we predicate one of the contraries that the first substance can lose without perishing, we introduce an item (Socrates’ pale colour, his particular height, his ignorance, his being the husband of Xanthippe) in one of the non-substance categories (quality, quantity, relative, and so on). The kinds to which these non-substantial items belong are non- substantial universals.

Aristotle also examines the coming to be and perishing of a first substance. Here again, he distinguishes a persisting subject and two contraries. If we make a statue from bronze, the lump of bronze (the subject) acquires the shape of the statue, and loses the shapelessness it had, and so changes between contraries. But although the lump remains in existence, a new subject, the statue, has come into being. In this case, the subject of the change is the matter (hylē), and what it acquires is the form (eidos, also rendered ‘species’).

This analysis of change suggests an argument ( Physics II 1) to show that the genuine subject, and hence the genuine substance, is the matter, whereas the apparent substance (for example, the statue) is simply matter with a certain shape. Socrates does not become another subject if he changes shape; hence (we may argue) the lump of bronze does not become another subject simply by acquiring the shape of a statue. Similarly, then, a natural organism might be understood as a piece of matter shaped in a certain way so as to embody Socrates. Natural organic ‘substances’, such as Socrates and this tree, turn out to be not genuine subjects, but mere configurations of the matter that is the real substance.

Aristotle does not endorse this eliminative attitude to natural organic substances. He uses the argument to raise a puzzle about whether matter or form is substance. He discusses this puzzle in Metaphysics VII (see §12–14). This discussion relies on his account of causation and explanation.

9 Causes

When we correctly answer questions such as ‘Why does this event happen?’ or ‘Why is this object as it is?’, we state the cause (or explanation; aition) of the event or object. Aristotle believes that causes are multivocal (see Physics II 3; Metaphysics I 3). Different accounts of a cause correspond to different answers to why-questions about (for example) a statue. (1) ‘It is made of bronze’ states the material cause. (2) ‘It is a statue representing Pericles’ states the formal cause, by stating the definition that says what the thing is. (3) ‘A sculptor made it’ states the ‘source of change’, by mentioning the source of the process that brought the statue into being; later writers call this the ‘moving cause’ or ‘efficient cause’. (4) ‘It is made to represent Pericles’ states ‘that for the sake of which’, since it mentions the goal or end for the sake of which the statue was made; this is often called the ‘final’ (Latin finis; ‘end’) cause.

Each of the four causes answers a why-question. Sometimes (as in our example) a complete answer requires all four causes. Not all four, however, are always appropriate; the (universal) triangle, for example, has a formal cause, stating its definition, but no efficient cause, since it does not come into being, and no final cause, since it is not made to promote any goal or end.

Some have claimed that Aristotle’s ‘four causes’ are not really causes at all, pointing out that he takes an aition to be available even in cases where the why- question (for example, ‘Why do the interior angles of this figure add up to two right angles?’) does not seek what we would call a cause (in Aristotle’s division, an efficient cause). When explanations of changes are being sought, however, Aristotle seems to provide recognizably causal explanations. Even the aitia (material, formal, final) that do not initially seem to be causes turn out to play an important role in causal explanation; for this reason, the label ‘four causes’ gives a reasonably accurate impression of Aristotle’s doctrine.

His comparison between artefacts and natural organisms clarifies his claims about formal and final causes. The definition of an artefact requires reference to the goal and the intended function. A hammer’s form and essence is a capacity to hammer nails into wood. The hammer was designed to have this capacity for performing this function; and if this had not been its function, it would not have been made in the way it was, to have the properties it has. The form includes the final cause, by specifying the functions that explain why the hammer is made as it is.

Similarly, Aristotle claims, a natural organism has a formal cause specifying the function that is the final cause of the organism. The parts of an organism seem to perform functions that benefit the whole (the heart pumps blood, the senses convey useful information). Aristotle claims that organs have final causes; they exist in order to carry out the beneficial functions they actually carry out. The form of an organism is determined by the pattern of activity that contains the final causes of its different vital processes. Hence Aristotle believes that form as well as matter plays a causal role in natural organisms.

To claim that a heart is for pumping blood to benefit the organism is to claim that there is some causal connection between the benefit to the organism and the processes that constitute the heart’s pumping blood. Aristotle makes this causal claim without saying why it is true. He does not say, for instance, either (1) that organisms are the products of intelligent design (as Plato and the Stoics believe), or (2) that they are the outcome of a process of evolution.

Aristotle’s account of causation and explanation is expressed in the content and argument of many of his biological works (including those connected with psychology). In the Parts of Animals and Generation of Animals for instance, he examines the behaviour and structure of organisms and their parts both to find the final causes and to describe the material and efficient basis of the goal-direction that he finds in nature ( Parts of Animals I 1). He often argues that different physiological processes in different animals have the same final cause.

Some ascribe to Aristotle an ‘incompatibilist’ view of the relation between final causes and the underlying material and efficient causes. Incompatibilists concede that every goal-directed process (state, event) requires some material process (as nutrition, for example, requires the various processes involved in digesting food), but they argue that the goal-directed process cannot be wholly constituted by any material process or processes; any process wholly constituted by material processes is (according to the incompatibilist) fully explicable in material-efficient terms, and therefore has no final cause.

Probably, however, Aristotle takes a ‘compatibilist’ view. He seems to believe that even if every goal-directed process were wholly constituted by material processes, each of which can be explained in material-efficient terms, the final-causal explanation would still be the only adequate explanation of the process as a whole. According to this view, final causes are irreducible to material-efficient causes, because the explanations given by final causes cannot be replaced by equally good explanations referring only to these other causes. This irreducibility, however, does not require the denial of material constitution.

10 Change

Aristotle studies nature as an internal principle of change and stability; and so he examines the different types of change (or ‘motion’; kinēsis) that are found in the natural elements and in the natural organisms composed of them. In Physics III 1 he defines change as ‘the actuality of the potential qua potential’. His definition marks the importance of his views on potentiality (or ‘capacity’; dynamis) and actuality (or ‘realization’; energeia or entelecheia) (see Metaphysics IX 1–9).

The primary type of potentiality is a principle (archē) of change and stability. If x has the potentiality F for G, then (1) G is the actuality of F, and (2) x has F because G is the actuality of F. Marathon runners, for instance, have the potentiality to run 26 miles because they have been trained to run this distance; hearts have the capacity to pump blood because this is the function that explains the character of hearts. In these cases, potentialities correspond to final causes.

Potentiality and possibility do not, therefore, imply each other. (1) Not everything that is possible for x realizes a potentiality of x. Perhaps it is possible for us to speak words of Italian (because we recall them from an opera) without having a potentiality to speak Italian (if we have not learnt Italian). (2) Not everything that x is capable of is possible for x; some creatures would still have a potentiality to swim even if their environment lost all its water.

These points about potentiality help to clarify Aristotle’s definition of change. The building of a house is a change because it is the actuality of what is potentially built in so far as it is potentially built. ‘What is potentially built’ refers to the bricks (and so on). The completed house is their complete actuality, and when it is reached, their potentiality to be built is lost. The process of building is their actuality in so far as they are potentially built. ‘In so far… ’ picks out the incomplete actuality that is present only as long as the potentiality to be built (lost in the completed house) is still present. Aristotle’s definition picks out the kind of actuality that is to be identified with change, by appealing to some prior understanding of potentiality and actuality, which in turn rests on an understanding of final causation.

In the rest of the Physics , Aristotle explores different properties of change in relation to place and time. He discusses infinity and continuity at length, arguing that both change and time are infinitely divisible. He tries to show that the relevant type of infinity can be defined by reference to potentiality, so as to avoid self- contradiction, paradox or metaphysical extravagance. In his view, infinite divisibility requires a series that can always be continued, but does not require the actual existence of an infinitely long series. Once again, the reference to potentiality (in ‘can always… ’) has a crucial explanatory role.

11 Metaphysics

Some of the basic concepts of the Categories and Physics – including substance, particular, universal, form, matter, cause and potentiality – are discussed more fully in the Metaphysics . This is a collection of fourteen books, some of them loosely connected. Aristotle probably did not deliver a course of lectures in the order of the present treatise. Parts of book I are almost repeated in book XIII. Book V is a ‘philosophical dictionary’ that seems to interrupt the argument of books IV and VI. Book XI summarizes parts of book IV. Books II and XI were probably not written entirely by Aristotle.

Still, whatever their literary origins, all these books have a common subject matter, since they all contribute to the universal science that studies the common presuppositions of the other sciences. This universal science has four names. (1) ‘First philosophy’: it studies the ‘first principles’ and ‘highest causes’ (including the four causes of the Physics ) presupposed by the other sciences. (2) ‘The science of being’: every science presupposes that it studies some sort of being, and the science of being examines and defends this presupposition. (3) ‘Theology’: first philosophy is not only first in so far it is most universal, but also in so far as it deals with the primary sort of being, the sort on which all other beings depend. The primary sort of being is substance, and the primary sort of substance is divine substance; hence the science of being must study divine substance. (4) ‘Metaphysics’ (ta meta ta physika; ‘the things after the natural things’): it is ‘after’ or ‘beyond’ the study of nature because (a) as theology, it studies entities outside the natural order, and (b) as first philosophy, it starts from the study of nature (which is prior and better known ‘to us’) and goes beyond it to its foundations and presuppositions (which are prior and better known ‘by nature’; see §3).

The first three of these names are used by Aristotle himself ( Metaphysics IV 1–3, VI 1). The fourth was given to the treatise in antiquity (at an uncertain date); its use of ‘after’ captures Aristotle’s different claims about the relation of the universal science to other sciences.

The universal science is the science of being qua being – that is, being in so far as it is being – just as mathematics is the science of some beings qua mathematical objects (see §16) and physics is the science of some beings qua changeable. The science of being studies the beings that are also studied by other sciences, but it isolates the relevant properties of beings by a different level of abstraction; it does not rely on the fact that they have the properties of mathematical or natural objects, but simply on the fact that they are beings studied by a science ( Metaphysics IV 1–2).

A special science assumes that it begins with a subject that has properties. The universal science is the science of being because it studies the sort of subject that is presupposed by the other sciences; and it is primarily the science of substance because substance is the primary sort of being. Aristotle’s analysis of change in Physics I introduces substances as subjects; the Metaphysics asks what sorts of subjects and substances must be recognized by special sciences.

Aristotle argues that if we are to signify a subject, it is impossible for each of its properties both to belong and not to belong to it. This principle is often called the ‘Principle of Non-Contradiction’ ( Metaphysics IV 3–4). To defend the principle, Aristotle considers an opponent who is willing to assert that a single subject, man, is both a bipedal animal and not a biped animal. If the opponent really says this about a single subject, then, when he uses ‘man’, he must signify one and the same subject, man. If he agrees that in using ‘man’ he signifies a biped animal, then he cannot also deny that man is a biped animal; for if he denies this, he can no longer say what ‘man’ signifies, and hence he cannot say what subject it is that he takes to be both a biped animal and not a biped animal. This property (which one cannot also deny of a subject) is an essential property. Hence, the attempt to reject subjects with essential properties is self-undermining.

Subjects of change must also, according to Aristotle, have objective properties (that is, properties that they have whether or not they appear to have them). An argument against Protagoras seeks to show that any attempt to reject objective properties undermines itself ( Metaphysics IV 5). Protagoras denies that there are any objective properties, because he claims that how things appear to someone is how they are. If he is to maintain the infallibility of appearances against any possibility of correction, then, Aristotle argues, he must claim that it is possible for the same subject to change in every respect at every time (to match different appearances). This is possible, however, only if the same subject can remain in being, but change in all respects. Aristotle replies that if the same subject persists, it must keep the same essential property (the ‘form’); hence it cannot change in every respect (IV 5).

12 From being to substance

In Metaphysics IV 2 and VII 1 Aristotle argues that, since substance is the primary type of being and other beings are in some way dependent on substances, the science of being must primarily be concerned with substance. The arguments of IV 4–5 describe some features of substances; they must be subjects with stable, objective, essential properties. Books VII–IX describe these subjects more fully, by re-examining the conception of substance that is presented in the Categories and Physics (see §§7–8).

Aristotle observes that we regard substance both as ‘a this’ and as ‘essence’ (or ‘what it is’). We might assume that these two descriptions pick out two sorts of substances – a particular subject (‘this’) and a universal (‘what it is’), corresponding to the first and second substances of the Categories . Aristotle, however, insists that his question ‘What is substance?’ will be satisfactorily answered only when we have found the one thing that best satisfies the conditions for being both a subject (a ‘this’) and an essence (‘what it is’). Whatever best satisfies these conditions is primary substance.

The different candidates that Aristotle considers for this role are matter, form and the compound of the two. He argues against the first and third candidates, and defends the second. He regards matter and compound as types of substance, but argues that they are secondary to form because they do not meet the relevant conditions to the same degree. To show that form is primary substance, he argues that a form is both a subject and an essence of the right sort. In books VIII–IX he clarifies his answer by identifying form with the actuality for which the matter is the potentiality.

13 Why is form substance?

In claiming that form is substance, Aristotle relies on the connections between form, cause, essence and identity. He rejects the eliminative view (§8) that the so-called ‘coming-to-be’ or ‘perishing’ of an artefact or organism is simply an alteration of the matter. According to the eliminative view, this alteration does not involve the existence or non-existence of a distinct substance, any more than Socrates’ coming to be musical involves the existence of a distinct substance, musical Socrates. Aristotle replies that the production of an artefact and the generation of an organism introduce a new subject, a substance that is neither identical to nor wholly dependent on the matter that constitutes it at a time (see Identity §2). Although this statue of Pericles has come into being from a particular piece of bronze, we may repair the statue by replacing damaged bits; we preserve the same statue but we cause a different bit of bronze to constitute it. Similarly, an organism remains in existence as long as it replaces its matter with new matter: it persists as long as its form persists ( Generation and Corruption I 5).

When Aristotle speaks of the relation of form to matter, he may refer to either of two kinds of matter: (1) the proximate, organic matter (for example, the organs and limbs making up the organic body); and (2) the remote, non-organic matter (for example, blood, earth, water) of which the organic body is made. Remote matter can exist without the form of the organism, but the organism can persist without any particular piece of remote matter. Proximate matter cannot exist without the form (since it is the function of an arm or heart that makes it the limb or organ it is); the form is the actuality of which the proximate matter is the potentiality ( On the Soul 412a10; Metaphysics 1038b6, 1042b10).

The role of the form in determining the persistence of an organism results from its role as the source of unity. The form, including the organism’s vital functions, makes a heap of material constituents into a single organism ( Metaphysics VII 16). A collection of flesh and bones constitutes a single living organism in so far as it has the form of a man or a horse; the vital functions of the single organism are the final cause of the movements of the different parts. The organism remains in being through changes of matter, as long as it retains its formal, functional properties. Since the structure, behaviour and persistence of the organism must be understood by reference to its form, the form is irreducible to matter (see §9); the organism, defined by its form, must be treated as a subject in its own right, not simply as a heap of matter.

These facts about organisms explain why Aristotle sees a close connection between primary substance and form. Organisms are substances primarily because of their formal properties, not because of their material composition; hence we cannot identify all the basic subjects there are unless we recognize the reality of formal properties and of subjects that are essentially formal.

14 What are substantial forms?

The conclusion that primary substance and form are closely connected, however, explains only why some substances are essentially formal; it does not explain why form itself is substance. To explain this further claim, we need to decide whether Aristotle regards a substantial form as (1) a species form (shared by all members of a given species, for example, the form of man or horse), normally taken to be a universal, or as (2) a particular form, proprietary to (for example) Socrates. (See Metaphysics VII 10–16, XII 5, XIII 10, Generation of Animals IV 3 for important evidence.)

Some points favouring the ‘universal solution’ are the following. (1) Aristotle often contrasts the form with the compound of form and matter, and describes particulars as compounds; hence he apparently does not regard particulars as forms. (2) Similarly, he says that a particular differs from a universal in having both form and matter; hence no particular seems to be simply a form. (3) He says the form is what is specified in a definition, but there is no definition of a particular; hence a particular apparently cannot be a form. (4) He says that substance is prior in knowledge to non-substance, but scientific knowledge of particulars is impossible; hence they apparently cannot be substances, and only a universal can be a substance.

In favour of the ‘particular solution’ it may be argued: (1) a substance must be a subject, whereas all universals are said of subjects; (2) a substance must be a ‘this’, as opposed to a ‘such’, and hence, apparently, some sort of particular; (3) Aristotle argues at length that no universal can be a substance.

We might be tempted to conclude that Aristotle’s position is inconsistent. His conviction that substance as ‘this’ and substance as ‘what is it’ must be the same thing leads him to insist that the successful candidate for substance must satisfy the criteria for being both a this (a subject, and hence a particular) and an essence (a property, and hence a universal). If one and the same thing cannot satisfy both criteria, then no one thing can satisfy all Aristotle’s conditions for being a substance.

We need not draw this conclusion, however. We can maintain that Aristotle consistently favours the universal solution, if we can show: (1) a ‘this’ need not be a particular; (2) some universals are subjects; (3) a species form is not the sort of universal that cannot be a substance.

We can maintain that he consistently favours the particular solution, if we can show the following. (1) The contrast between form and matter does not imply that they are always mutually exclusive; some forms may be constituted by, or embodied in, particular bits of matter. Sometimes, indeed, Aristotle speaks as though a form is a subject that can persist and perish and can exchange its matter. (2) The sense in which particulars do not allow definition and scientific knowledge does not prevent them from also being, in an appropriate sense, prior in definition and knowledge to universals ( Metaphysics XIII 10 may attribute the relevant priority to particular substances).

These two solutions are different ways of expressing Aristotle’s belief that substances are basic. Both his metaphysics and his natural philosophy express and defend the conviction that natural organisms and their kinds are substances because they are fundamental; they are fundamental because they are irreducible to their constituent matter. It is more difficult to decide whether the individuals or their kinds are more fundamental. Perhaps, indeed, we ought not to decide; different things may be fundamental or irreducible in different ways.

15 Universals, Platonic Forms, mathematics

These disputes partly concern Aristotle’s attitude to the reality of universals. One- sided concentration on some of his remarks may encourage a nominalist or conceptualist interpretation (see Nominalism §§1, 2 ). (1) He rejects Plato’s belief (as he understands it) in separated universal Forms (see Plato §§10, 12–16; Forms, Platonic), claiming that only particulars are separable. (2) In Metaphysics VII 13–16 he appears to argue that no universal can be a substance. (3) He claims that the universal as object of knowledge is – in a way – identical to the knowledge of it ( On the Soul 417b23).

Other remarks, however, suggest realism about universals. (4) He claims they are better known by nature; this status seems to belong only to things that really exist. (5) He believes that if there is knowledge, then there must be universals to be objects of it; for our knowledge is about external nature, not about the contents of our own minds.

Aristotle’s position is consistent if (1)–(3) are consistent with the realist tendency of (4)–(5). The denial of separation in (1) allows the reality of universals. Similarly, (2) may simply say that no universals are primary substances (which are his main concern in Metaphysics VII). And (3) may simply mean (depending on how we take ‘in a way’) that the mind’s conception of the extra-mental universal has some of the features of the universal (as a map has some of the features of the area that it maps). While Aristotle denies that universals can exist without sensible particulars to embody them, he believes they are real properties of these sensible particulars.

He offers a rather similar defence of the reality, without separability, of mathematical objects ( Physics II 2; Metaphysics XIII 3). While agreeing with the Platonist view that there are truths about, for example, numbers or triangles that do not describe the sensible properties of sensible objects, he denies that these truths have to be about independently-existing mathematical objects. He claims that they are truths about certain properties of sensible objects, which we can grasp when we ‘take away’ (or ‘abstract’) the irrelevant properties (for example, the fact that this triangular object is made of bronze). Even though there are no separate objects that have simply mathematical properties, there are real mathematical properties of sensible objects.

16 Metaphysics: God

When Aristotle claims that first philosophy is also theology (see §11), he implies that the general discussion of being and substance is the basis for the special discussion of divine substance. (Hence later writers distinguish ‘special metaphysics’, dealing with God, from ‘general metaphysics’, dealing with being in general.) The different features of substance explained in Metaphysics VII–IX are included in the divine substance of XII. (1) Primary substance is to be identified in some way with form rather than with matter or with the compound of form and matter; divine substance is pure form without matter. (2) Primary substance is in some way numerically one, a ‘this’ rather than a ‘such’; divine substance is completely one and indivisible. (3) Primary substance is in some way actuality rather than potentiality; divine substance is pure actuality with no potentiality. (4) Primary substance is soul rather than body (see §17); divine substance is pure intellect without sense or body.

In each case the properties of primary substance are found in a sensible substance (an animal or a plant) only in so far as they belong to an object that also has other properties; hence primary substance in sensible reality is the form and actuality of an object (a horse, for example) that also has matter and potentiality. In divine substance, however, each feature is found in separation from these other properties; that is why a divine substance lacks matter, multiplicity, parts or potentiality. Aristotle argues that a substance with these pure substantial properties must exist if any sensible substances are to exist; for the existence of potentialities that can be actualized presupposes the existence of an actuality that does not itself include any potentiality (to avoid an infinite regress).

Since this primary type of substance is divine, it is what traditional belief in the Olympian gods was about, what the Presocratics were talking about when they spoke of ‘the divine’, and what Plato was talking about in speaking of a supreme god. Aristotle mentions the traditional Olympian gods without committing himself to acceptance of the traditional conception of them. He rejects anthropomorphic views of the gods, but he speaks of the divine nature as a kind of mind. He believes that there is something divine about the order and workings of nature, and still more divine in the heavenly substances ( Parts of Animals I 5). Although he continues to speak of gods in the plural, he also speaks of one divine mind as the ultimate cause of the whole universe; these remarks help to justify the later interpreters who take him to speak of the one God who is the subject of (for example) Aquinas’ ‘Five Ways’ (Summa Theologiae 1a q.2 a.3) (see Aquinas, T. §11 ).

Aristotle’s God is the ultimate cause of the physical universe, but not its creator (as Plato’s demiurge is), since Aristotle believes the universe is eternal. Nor does Aristotle suggest that God has providence or foreknowledge concerned with future contingent events. But he believes that the physical universe is dependent on God. In Physics VIII he argues that the explanation of motion requires recognition of a first cause of motion, and in Metaphysics XII this first cause is identified with divine, immaterial, substance. This first mover is itself unmoved; it initiates motion only as an object of love initiates motion by attraction. It is the ultimate final cause of the various movements in the universe.

In treating the divine substance as a god, and hence as a being with a soul and an intellect, Aristotle attributes some mental life to it. But since it would be imperfect if it thought of objects outside itself (because it would not be self-sufficient), it thinks only of its own thinking. This restriction, however, is not as severe as it may seem, since Aristotle believes that the various objects of thought are in some way identical to the mind that thinks them (see §15). In so far as God thinks of his own mind, he thereby also contemplates the order of the universe as a whole; this is the order that the different movements in the universe seek to embody.

Sometimes (as in Physics VIII) Aristotle argues for a single first mover. In Metaphysics XII, however, he argues that an unmoved mover must be postulated for each of the distinct movements of the heavenly bodies. This astronomical interpretation of his theological doctrine is difficult to reconcile with his belief, reaffirmed in Metaphysics XII 10, that in some way the universe is unified by a single first unmoved mover.

17 Soul and body

Aristotle’s treatise On the Soul is placed among the works on natural philosophy, but should be read with Metaphysics VII–IX. In Aristotle’s view, disputes about soul and body are simply a special case of the more general disputes about form and matter. He rejects both the Presocratic materialist assumption that the soul is simply non- organic matter, and the Platonic dualist claim that it must be something entirely non-bodily. He argues that soul is substance because it is the form of a natural body, and that the body is the matter informed by the soul. Although the soul is a substance distinct from the non-organic body (the collection of non-organic matter belonging to a living organism; see §13), it is not immaterial (if being immaterial excludes being composed of matter), nor is it independent of some non-organic body or other.

Aristotle assumes that the soul is the primary principle of life, and hence that it distinguishes the living from the non-living. A living organism is nourished, grows and diminishes, through itself – from a causal origin within itself rather than from the action of external agents. A living organism must, therefore, be teleologically ordered, since (for Aristotle) nutrition and growth cannot be understood without appeal to final causation (see Teleology ).

If life must be conceived teleologically, and the soul is the primary principle of life, then the soul is form rather than matter. For the primary principle is whatever explains our vital activities; since these are goal-directed activities, their explanation must refer to the goal-directed features of the subject, and so to the form rather than the matter. If the soul is what we live by primarily, it must be the final cause of the body, and so a formal, not a material, aspect of the subject. Soul must, therefore, be substance as form.

Aristotle attributes to the soul the features of substantial form (see §13). (1) It is a substance that is irreducible to a material non-organic body (remote matter); to that extent the soul is incorporeal, and not just some ordinary material stuff. (2) It is the source of unity that makes a heap of material constituents into a single organism. For a collection of flesh and bones constitutes a single living organism in so far as it is teleologically organized; the activities of the single organism are the final cause of the movements of the different parts. Since a single organism has a single final cause, it has a single soul and a single body. (3) The identity and persistence of the soul determine the identity and persistence of the creature that has it. If something has a soul in so far as it has life, then Socrates perishes if and only if his soul does. The truth of this Platonic claim (Phaedo 115c–e) does not imply Platonic dualism. (4) The definition of a soul must mention the proximate material subject (the organic body and its parts) whose capacities are actualized in the functions of the organism ( Metaphysics 1036b28–30). A soul must be non-coincidentally connected to a specific sort of organic body ( On the Soul 407b20–4).

Some of the puzzles in Aristotle’s doctrine of substantial form arise in his doctrine of soul and body. If, for instance, he recognizes particular substantial forms, then he also recognizes (as the previous paragraph assumes) the individual souls of Socrates and Callias; if, however, he recognizes only one substantial form for each species, then he recognizes only one soul for human beings, another for horses, and so on.

Since the soul is the form of the living body, an account of the different ‘parts’ or ‘capacities’ (or ‘faculties’; dynameis) of the soul does not describe the different physiological processes underlying the different activities of a living organism, but describes their formal and goal-directed aspects. Aristotle describes the capacities that distinguish the different types of souls: nutrition (characteristic of plants), perception and appearance (characteristic of animals) and rational thought (characteristic of rational animals) (see Psychē). He describes some of the physiological basis of these psychic capacities in the shorter treatises on natural philosophy, including the Parva Naturalia, the Movement of Animals, and the Progression of Animals .

18 Perception

To define perception, Aristotle returns to his contrast between form and matter. Perception happens in so far as (1) the perceiver becomes like the object ( On the Soul 417a18); (2) the perceiver that was potentially F (for example, white) becomes actually F when it perceives the actually F object (418a3); (3) the perceiver acquires the form, but not the matter, of the object (424a18–24). These descriptions express a realist view of perception and its objects; Aristotle assumes in (2) that an object is actually white, square, and so on in its own right, before we perceive it.

He is sometimes taken to imply in (1) that perception requires physical similarity; but (3) counts against this interpretation. A sense receives the form without the matter in the way in which a house without matter is in the soul of the architect before the house is built. In the latter case, nothing that looks like a house is in the builder, but features of the house correspond to features of the builder’s design. Similarly, when we hear a tune, our ears do not necessarily sound like the tune, but a state of us systematically corresponds to the tune (as features of a map correspond to features of the area it maps).

A ‘common sense’ perceives common properties of sensible objects, such as size, shape and number, which are all perceived through the perception of motion ( On the Soul 425a14–20). This is not a sixth sense independent of the other five, but the result of the cooperation of the five senses. Aristotle argues that we can explain our grasp of these common properties without supposing that they are objects of intellect rather than sense (contrast Plato, Theaetetus 184–6).

19 Appearance and thought

Appearance (or ‘imagination’; phantasia) links perception to goal-directed movement. A lion sees or smells a deer; it takes pleasure in the prospect of eating the deer, and so wants to catch the deer. To connect perception with pleasure and desire, we need to say how the deer appears to the lion (as prey); this is what Aristotle calls the lion’s appearance of the deer ( On the Soul III 3, 7).

Aristotle denies that this appearance constitutes a belief (doxa). He argues that belief requires reason and inference, which non-human animals lack; in his view, they lack any grasp of a universal, and have only appearances and memory of particulars ( Nicomachean Ethics 1147b4–5). The operations of sense, memory and experience are necessary, but not sufficient, for the grasp of a universal that is expressed in concepts and beliefs ( Posterior Analytics II 19; Metaphysics I 1).

Concepts and beliefs require intellect (nous) actualized in ‘understanding’ or ‘thinking’ (noein; On the Soul III 4)(see Nous ). Thought differs from perception in so far as it grasps universal essences – for example, what flesh is, as opposed to flesh. Perception does not include grasp of the universal as such; in grasping the universal, we recognize some feature of our experience as a ground for attributing the universal to a particular that we experience.

To explain how the mind is capable of grasping universals when we interact causally with particular perceptible objects, Aristotle distinguishes two aspects of intellect – passive and ‘productive’ (or ‘active’ or ‘agent’) – claiming that these two aspects must combine to produce thought of universals ( On the Soul III 5). He does not say how productive intellect contributes to our grasp of universals. Later interpreters suggest that productive intellect abstracts the aspects relevant to the universal from the other features of particulars that are combined with them in perception (Aquinas, Summa Theologiae 1a q.79 a.3).

Aristotle takes the presence of this productive intellect to be necessary for any thinking at all. Moreover, he believes that productive intellect is capable of existing without a body. He still maintains his belief in the inseparability of soul from body; for since productive intellect is not a type of soul, its separate existence is not the separate existence of a soul.

20 Desire and voluntary action

Perception, appearance and thought are connected to goal-directed movement by means of desire. The appearance of something as desirable is the source of an animal’s tendency to pursue one sort of thing rather than another. External objects, however, appear desirable to different agents in different ways. Aristotle distinguishes the appetite (epithymia) that animals have from the wish (rational desire; boulēsis) that only rational agents have; appetite is for the pleasant and wish is for the good ( On the Soul 414b2–6, 432b5–7, Politics 1253a15–18).

A rational agent’s wish differs from appetite in so far as it is guided by deliberation resting on one’s conception of one’s good. Such a conception extends beyond one’s present inclinations both at a particular time and over time. Rational agents are aware of themselves as extending into past and future. Deliberation that is guided by reference to these broader aspects of one’s aims and nature results in the rational choice that Aristotle calls ‘decision’ (prohairesis; Nicomachean Ethics III 3).

Agents who act on desire and appearance also act voluntarily (hekousiōs), in so far as they act on some internal principle (archē). While voluntary action is not confined to rational agents, their voluntary action has special significance, because it is an appropriate basis for praise and blame. Since it has an internal principle, it is in our control as rational agents, and therefore we are justly praised and blamed for it. We are held responsible for our actions in so far as they reflect our character and decisions ( Nicomachean Ethics III 1–5).

Aristotle’s defence of his belief that we are appropriately responsible agents does not confront the questions later raised by Epicurus’ claim that responsibility is incompatible with the complete causal determination of our actions (see Epicureanism §12). An incompatibilist position is ascribed to Aristotle by Alexander in On Fate (see Alexander of Aphrodisias §4 .) Aristotle neither explicitly presents an incompatibilist position nor explicitly endorses a compatibilist position of the sort later defended by the Stoics.

A discussion of time, truth and necessity (the ‘Sea Battle’; De Interpretatione 9) has suggested to some interpreters that Aristotle is an indeterminist. His opponent is a fatalist, who assumes that (1) future-tensed statements about human actions (for example, ‘There will be a sea battle tomorrow’) were true in the past, and infers that (2) the future is necessarily determined, independently of what we choose. Aristotle certainly rejects (2). If he accepts the validity of the fatalist’s argument, and rejects (1), then he accepts indeterminism.

An alternative reply to the fatalist would be to accept (1) and to deny the validity of the argument. We might argue that the past truth of statements about my actions does not imply that my actions are determined independently of my choices. If on Friday Socrates decides to walk, and he acts on his decision on Friday, then it was true on Thursday that Socrates would walk on Friday, and also true that on Friday he would act on his decision to walk, but it was not true on Thursday that he would walk whether or not he decided to (see Stoicism §21 ). Probably Aristotle accepts this alternative reply to the fatalist, and hence does not endorse indeterminism.

21 The human good

Aristotle’s account of rational agents, choice, deliberation and action is an appropriate starting point for his ethical theory. Ethics is concerned with the praiseworthy and blameworthy actions and states of character of rational agents; that is why it concerns virtues (praiseworthy states) and vices (blameworthy states) (see Aretē ).

Aristotle’s ethical theory is mostly contained in three treatises: the Magna Moralia, the Eudemian Ethics and the Nicomachean Ethics. The titles of the last two works may reflect a tradition that Eudemus (a member of the Lyceum) and Nicomachus (the son of Aristotle and Herpyllis) edited Aristotle’s lectures. The Magna Moralia is widely agreed not to have been written by Aristotle; some believe, with good reason, that it contains a student’s notes on an early course of lectures by Aristotle. The Eudemian Ethics is now widely agreed to be authentic, and generally (not universally) and reasonably taken to be earlier than the Nicomachean Ethics. Three books (Nicomachean Ethics V–VII = Eudemian Ethics IV–VI) are assigned by the manuscripts to both the Eudemian Ethics and the Nicomachean Ethics .

Aristotle conceives ‘ethics’ ( Magna Moralia 1181a24) as a part of political science; he treats the Nicomachean Ethics and the Politics as parts of a single inquiry ( Nicomachean Ethics X 9). Ethics seeks to discover the good for an individual and a community (Nicomachean Ethics I 2), and so it begins with an examination of happiness, (eudaimonia). (‘Wellbeing’ and ‘welfare’ are alternative renderings of eudaimonia that may avoid some of the misleading associations carried by ‘happiness’; see Eudaimonia.) Happiness is the right starting point for an ethical theory because, in Aristotle’s view, rational agents necessarily choose and deliberate with a view to their ultimate good, which is happiness; it is the end that we want for its own sake, and for the sake of which we want other things (so that it is the ultimate non- instrumental good). If it is to be an ultimate end, happiness must be complete (or ‘final’; teleion) and self-sufficient ( Nicomachean Ethics I 1–5, 7).

To find a more definite account of the nature of this ultimate and complete end, Aristotle argues from the human function (ergon), the characteristic activity that is essential to a human being in the same way that a purely nutritive life is essential to a plant and a life guided by sense perception and desire is essential to an animal ( Nicomachean Ethics I 7). Since a human being is essentially a rational agent, the essential activity of a human being is a life guided by practical reason. The good life for a human being must be good for a being with the essential activity of a human being; hence it must be a good life guided by practical reason, and hence it must be a life in accordance with the virtue (aretē) that is needed for achieving one’s good. The human good, therefore, is an actualization of the soul in accordance with complete virtue in a complete life. This ‘complete virtue’ appears to include the various virtues described in the following books of the Nicomachean Ethics; this appearance, however, may be challenged by Nicomachean Ethics X (see §26).

22 Virtue of character

From the general conception of happiness Aristotle infers the general features of a virtue of character (ēthikēaretē; Nicomachean Ethics I 13). He agrees with Plato in recognizing both rational and non-rational desires (see Plato §14 ). One’s soul is in a virtuous condition in so far as the non-rational elements cooperate with reason; in this condition human beings fulfil their function well. The argument from the human function does not make it clear what states of a rational agent count as fulfilling the human function. Aristotle seeks to make this clearer, first through his general account of virtue of character, and then through his sketches of the individual virtues.

A virtue of character must be a ‘mean’ or ‘intermediate’ state, since it must achieve the appropriate cooperation between rational and non-rational desires; such a state is intermediate between complete indulgence of non-rational desires and complete suppression of them. (Aristotle is not recommending ‘moderation’ – for example, a moderate degree of anger or pleasure – in all circumstances.) The demand for cooperation between desires implies that virtue is more than simply control over desires; mere control is ‘continence’ (enkrateia) rather than genuine virtue.

The task of moral education, therefore, is to harmonize non-rational desires with practical reason. Virtuous people allow reasonable satisfaction to their appetites; they do not suppress all their fears; they do not disregard all their feelings of pride or shame or resentment ( Nicomachean Ethics 1126a3–8), or their desire for other people’s good opinion. Aristotle’s sketches of the different virtues show how different non-rational desires can cooperate with practical reason.

23 Virtue, practical reason and incontinence

A virtuous person makes a decision (prohairesis) to do the virtuous action for its own sake. The correct decision requires deliberation; the virtue of intellect that ensures good deliberation is prudence (or ‘wisdom’, phronēsis; Nicomachean Ethics VI 4–5); hence the mean in which a virtue lies must be determined by the sort of reason by which the prudent person would determine it (1107a1–2). Virtue of character is, therefore, inseparable from prudence. Each virtue is subject to the direction of prudence because each virtue aims at what is best, as identified by prudence.

In claiming that prudence involves deliberation, Aristotle also emphasizes the importance of its grasping the relevant features of a particular situation; we need to grasp the right particulars if deliberation is to result in a correct decision about what to do here and now. The right moral choice requires experience of particular situations, since general rules cannot be applied mechanically. Aristotle describes the relevant aspect of prudence as a sort of perception or intuitive understanding of the right aspects of particular situations ( Nicomachean Ethics VI 8, 11).

These aspects of prudence distinguish the virtuous person from ‘continent’ and ‘incontinent’ people ( Nicomachean Ethics VII 1–10). Aristotle accepts the reality of incontinent action (akrasia), rejecting Socrates’ view that only ignorance of what is better and worse underlies apparent incontinence (see Socrates §6; Akrasia §1 ). He argues that incontinents make the right decision, but act contrary to it. Their failure to stick to their decision is the result of strong non-rational desires, not simply of cognitive error. Still, Aristotle agrees with Socrates in believing that ignorance is an important component of a correct explanation of incontinence, because no one can act contrary to a correct decision fully accepted at the very moment of incontinent action.

The error of incontinents lies in their failure to harmonize the demands of their appetites with the requirements of virtue; their strong appetites cause them to lose part of the reasoning that formed their decision. When they act, they fail to see clearly how their general principles apply to their present situation. If their failure results from an error in deliberation, it is clear why Aristotle insists that incontinent people lack prudence.

24 Choice, virtue, and pleasure

It is initially puzzling that virtuous people decide to act virtuously for its own sake as a result of deliberation. If they decide on virtuous action for its own sake, then their deliberation causes them to choose it as an end in itself, not simply as a means. Decision and deliberation, however, are not about ends but about ‘the things promoting ends’ (ta pros ta telē, often rendered ‘means to ends’). Aristotle’s description of the virtuous person, then, seems to attribute to decision a role that is excluded by his explicit account of decision.

This puzzle is less severe once we recognize that Aristotle regards different sorts of things as ‘promoting’ an end. Sometimes he means (1) that the action is external and purely instrumental to the end; in this way buying food ‘promotes’ eating dinner. Sometimes, however, he means (2) that the action is a part or component of the end, or that performing the action partly constitutes the achieving of the end; in this way eating the main course ‘promotes’ eating dinner. Deliberation about this second sort of ‘promotion’ shows that an action is worth choosing for its own sake, in so far as it partly constitutes our end.

This role for deliberation explains how virtuous people can decide, as a result of deliberation, on virtuous action for its own sake; they choose it as a part of happiness, not as a merely instrumental means. Prudence finds the actions that promote happiness in so far as they are parts of the happy life. Such actions are to be chosen for their own sake, as being their own end; they are not simply instrumental means to some further end. The virtuous person’s decision results from deliberation about the composition of happiness; virtuous people decide on the actions that, by being non-instrumentally good, are components of happiness in their own right.

Aristotle’s demand for the virtuous person to decide on the virtuous action for its own sake is connected with two further claims: (1) the virtuous person must take pleasure in virtuous action as such; (2) in doing so, the virtuous person has the pleasantest life. In these claims Aristotle relies on his views about the nature of pleasure and its role in happiness ( Nicomachean Ethics VII 11–14, X 1–5).

He denies that pleasure is some uniform sensation to which different kinds of pleasant action are connected only causally (in the way that the reading of many boring books on different subjects might induce the same feeling of boredom). Instead he argues that the specific pleasure taken in x rather than y is internally related to doing x rather than y, and essentially depends