Twelfth century (The)

The twelfth century John Marenbon INTRODUCTION The twelfth century began and ended with events which mark it off, at least symbolically, as a discrete period in the history of Western philosophy. It was in about 1100 that Abelard—the most wide-ranging and profound philosopher of the period—arrived in Paris to study, and very soon to teach, logic. The competing, quarrelling, disorganized schools of Paris, whose growth Abelard did so much to stimulate, would be the setting for much of what was liveliest and most sophisticated in twelfth-century philosophy. It was in the year 1200 that Philip Augustus issued the privilege to the schools of Paris which, symbolically at least, marks the beginning of Paris University. The schools would henceforth become a more homogeneous and tightly-regulated organization, imposing a rigid framework on thirteenth- and fourteenth-century scholastic thought. Works newly translated from the Greek and Arabic gradually entered the curriculum and the work of almost all the twelfthcentury philosophers was rapidly forgotten. Modern historians of philosophy have set out to repair this neglect. But (at least until very recently) they have characterized the period in two main ways, each of which leaves in question whether twelfthcentury philosophy itself contains much worth studying. The first way has been to see the time as one of beginnings. Between 1100 and 1200, it is said, the ground was prepared for the great flourishing of scholasticism in the mid- and late thirteenth century. Such a description might well suggest that the twelfth century, fascinating as it may be for the intellectual historian who wishes to see how, and against what background, ideas develop, produced little of independent philosophical interest. The second way of characterizing the twelfth century has been in terms of its ‘humanism’ (and, closely linked to this, as a time of ‘renaissance’). The period is presented as one of revived activity in all branches of learning, closely connected with a respect for the classical past and a wish to rediscover its literature in all its various branches, poetic, scientific, legal and philosophical. The achievements of the thirteenth century are presented as being narrower: its sophistication in logic, philosophy and theology must be balanced against the aridity of the scholastics’ style, their rejection of the variety and complexity in form found in twelfth-century writing, their apparent contempt for poetry and fine latinity. Such a contrast can easily be turned against the thinkers of the twelfth century by the modern reader of philosophy. They are suspected of being dilettantes; their writings, full of interest to the literary historian, are thought to lack the precision and singlemindedness necessary for good philosophy.¹ There is some truth in the rationale behind each approach. Twelfthcentury scholars did, indeed, elaborate the logical and theological techniques which served the philosophers of the thirteenth- and fourteenth-century universities. Many were attentive to the literary form of their writings and enthusiasts of ancient literature as well as ancient philosophy. And even the sharpest-minded of them, such as Abelard, can be clumsy or imprecise in their technical vocabulary and sometimes inattentive to the complexity of the issues they are treating; whilst some well-known thinkers of the time, especially those most influenced by Platonism, are more inclined to system building than to detailed argument and analysis. Yet there is a substantial body of twelfth-century thought sufficiently rigorous to require careful philosophical analysis and certainly interesting and unusual enough to deserve attention in its own right, rather than just as the forerunner of something else. Much of it is linked to the most striking feature of intellectual life in the period: the importance of the ‘trivium’: the three language-based disciplines of grammar, rhetoric and, most prominently, logic. (Indeed, the humanistic interest in ancient literature and in rhetoric was part of a general enthusiasm for the verbal arts, among which logic was dominant.) This chapter must be selective. More than half of it is devoted to the outstanding philosophers of the time: Peter Abelard and his nearcontemporary, Gilbert of Poitiers. The section following this one looks more briefly at four important masters working at the turn of the century. A later section sketches the Platonic current in twelfth-century thought, looking especially at the work of William of Conches and Thierry of Chartres. The concluding sections provide a quick introduction to the logical schools and theological methods of the period from 1150 to 1200, a time still far less well investigated than the previous half century. FOUR MASTERS AT THE BEGINNING OF THE CENTURY Four masters, already established by 1100 or shortly afterwards, indicate the most important directions philosophy would take in the century which followed. Three of them, Garlandus of Besançon, Roscelin of Compiègne and William of Champeaux, were logicians, although Roscelin also put forward controversial views on the Trinity and William would write on moral and theological topics, such as natural law, sin and free will. Bernard of Chartres, however, was a grammarian, interested both in grammatical theory and in careful reading of ancient pagan philosophical texts, in particular Plato’s Timaeus. The tradition of medieval logic was well established by the late eleventh century. Like other branches of study in the Middle Ages, it was based on ancient texts. Six were in common use by 1100: Aristotle’s Categories and On Interpretation, Porphyry’s Isagoge (all in Boethius’ translations, and approached using Boethius’ commentaries) and Boethius’ own On Division, treatises on categorical and hypothetical syllogisms and On Topical ‘differentiae’. Taken together, these works provided an introduction to constructing and analysing arguments. The Isagoge and the Categories could be read as guides to the various sorts of term which can appear as subject or predicate in a statement. On Interpretation explained how terms are combined to make statements, and how statements are related to each other as, for instance, contraries (‘All men are bald’—‘No man is bald’) or contradictories (‘All men are bald’— ‘Some man is not bald’). Students could then learn from Boethius’ own textbooks how to construct syllogistic arguments, either using atomic statements as premisses so as to form categorical syllogisms or molecular statements as premisses to form hypothetical syllogisms, and they could study arguments based on ‘topics’, commonly accepted maxims of reasoning.² These texts, especially the two by Aristotle, also contain far more than such an introduction. The Categories can be read as the concise statement of an ontology, whilst the On Interpretation raises problems about the nature of truth and meaning, about perception and knowledge, and about modality and free will. Sporadic evidence—occasional glosses, and passages in Peter Damian and Anselm of Canterbury—suggests that eleventh-century scholars were already aware of some wider implications of the logical texts. But the main emphasis at this stage seems to have been on mastering the basic skills of logic through careful study of the texts. If more digressive discussion was wanted, the earliest commentators were happy to turn to Boethius and copy passages of his commentaries verbatim.³ Garlandus of Besançon is known for his Dialectica, a comprehensive textbook on logic which was probably written at the turn of the twelfth century. Some scholars have described Garlandus as an early nominalist: an exponent of the view that nothing exists which is not a particular. But it is more accurate to see him as following a particular interpretative method in his approach to the Isagoge and the Categories. In common with a number of other scholars of the time (including the young Abelard), Garlandus read these texts in voce rather than in re: as talking not about things but about words.⁴ For instance, when Porphyry writes about genera or about accidents, Garlandus takes his remarks as concerning words such as ‘animal’ and ‘whiteness’. Roscelin’s views can be surmised only from allusions (usually hostile) by other writers. The most famous of these is Anselm of Canterbury’s comment that, according to Roscelin, universals are merely the puffs of air made when we speak. This might be just a jibe which draws out the consequences of in voce exegesis. But some scholars—especially Jean Jolivet—have given a more ambitious reconstruction of Roscelin’s thinking, using other evidence too. Roscelin, they say, focused on the reference a word has to an individual, whole object in the world. In the case of the words ‘genus’ and ‘species’, he would have argued that there is no individual object in the world to which they refer, so their reference is just to other words (such as ‘animal’, ‘man’)—and, considered as things, words are just puffs of air. Whether Roscelin ever propounded this view coherently, and if so when (he was still alive in the early 1120s), is uncertain.⁵ The earliest definite signs of a serious interest in the semantic and metaphysical problems about universals comes from those logicians who adopted a realist view. William of Champeaux, who taught at the school of Notre Dame in Paris, was one of their leaders. For modern philosophers, the problem of universals concerns properties and relations. But William and his contemporaries approached the question mainly in the context of a remark in Porphyry’s Isagoge about species and genera, that is to say, universal substances. They had then to consider primarily the semantics not of sentences such as ‘Socrates is white’, but of those such as ‘Socrates is a man’. A simple view, derived from Boethius, held that there is a universal essence shared by all men, who are then individuated by their accidental attributes (being six foot tall, sitting just here at six o’clock). William was forced to abandon this theory (‘material/essential essence realism’) by the attacks of his former pupil, Abelard, and then espoused an ‘indifference theory’, according to which the many particulars of the same species are at the same time one in that they are ‘not different’ from each other in respect of their nature. William’s interests as a logician were not, however, confined to speculations about universals. He was probably the author of a general Introduction to logical method, and it is clear from Abelard’s Dialectica that he discussed the problems raised by the different meanings of the verb ‘to be’—as the copula and as implying present existence. William was a theologian as well as a logician. He studied under Anselm of Laon, a leading scriptural exegete. Like Anselm’s, William’s theological teaching survives in the form of ‘Sentences’ (sententiae), ranging in length from a couple of lines to several hundred words, which may originally have been stimulated by dispute over the interpretation of a passage from the Bible, but take the form of freestanding discussions of a problem. William is more speculative and more analytical than Anselm, ranging over topics such as intentions and acts, the ontological status of evil, and implicit faith. He approaches the question of divine prescience ([7.28] 195–6) and human free will in the manner of a logician, trying (though not very successfully) to show that the statement, ‘It is possible for things to happen other than as they will happen’ does not imply ‘It is possible for God [who foresees all things] to be mistaken.’ Bernard, master at the cathedral school of Chartres in the first two decades of the century, represents a different tradition of early medieval teaching. He concentrated, not on logic, but on grammar. Part of his work as a grammarian was connected with the theory of grammar, as expounded by Priscian in his elaborate Institutiones grammaticae. Eleventh-century scholars had already composed a running commentary (the Glosule) to the Institutiones, which dealt with philosophical questions about semantics far more thoroughly than Priscian himself had done. Unfortunately, Bernard’s work in this area is known only through a few remarks made, long after his death, by John of Salisbury.⁶ John also records how Bernard commented on the Latin classics, drawing out their moral teaching and he commemorates him as the leading Platonist of his time. Bernard’s Platonism seems to have owed a good deal to Boethius’ Theological Treatises, especially in its introduction of secondary forms, enmattered images of the immaterial primary forms. But there was one work by Plato himself that was available to him and other twelfth-century scholars: the Timaeus in Calcidius’ partial Latin translation. Recently, a strong argument has been made for attributing to Bernard a commentary on the Timaeus. Although these Glosae in Platonem are for the most part straightforwardly literal and heavily reliant on the commentary Calcidius had written, they contain Bernard’s characteristic views about secondary forms and they exercised an influence on later twelfth-century exegesis of the work.⁷ PETER ABELARD Peter Abelard is probably better known than any medieval philosopher, not as a thinker but as the husband of Heloise and participant in a remarkable exchange of love-letters which have held their appeal from the time of Petrarch to the present. It was in fact the castration plotted by Heloise’s relatives and its consequences which gave Abelard’s career its distinctive shape, splitting it roughly into two halves. From about 1102 until his castration in 1117, Abelard was a brilliant teacher of logic in Paris and in Melun and Corbeil, small towns both connected with the royal court. He had studied under both Roscelin and William of Champeaux—and quarrelled with both. Although he did teach Christian doctrine, it was a relatively unimportant part of his work. The Dialectica, a textbook of logic, independent in form but closely linked to exegesis of the six standard ancient texts, probably dates from the end of this period. The Logica (‘Ingredientibui’)—logical commentaries, of which those on the Isagoge, Categories, De interpretatione survive in full—was probably written up a little later, but it too reflects his teaching at this time.⁸ After the castration, Abelard became a monk of St Denis. Although he continued to teach logic, theological questions came more and more to occupy him. The first fruit of this new interest was the Theologia summi boni, a treatise on the Trinity, rich in philosophical discussion, which was promptly condemned at the Council of Soissons in 1121. Undeterred, Abelard greatly extended the work, developing his logical analysis of Trinitarian relations and adding a long eulogistic account of the ancient philosophers and their virtues, to form the Theologia Christiana. One logical work dates from the same time (the Glossulae on Porphyry, often called the Logica Nostrorum petitioni sociorum), but Abelard’s main energies were given to theology and, increasingly, to ethical questions within theology. His Collationes (Dialogue between a Christian, a Philosopher and Jew), probably written c. 1130, discuss the virtues, evil and the highest good. By this time, Abelard—who had left St Denis and, for a time, taught students at his own monastic/ eremitic foundation, the Paraclete—was abbot of St Gildas, a monastery on a remote peninsula in Brittany. His attempts to reform the Breton monks proved disastrous and, from about 1133 to (probably) 1140, Abelard was teaching again in Paris. Although he gave some lectures on logic, he devoted most of his energy to developing his ethicallybased theological system. The final version of his Theologia, the Theologia scholarium, a commentary on St Paul’s letter to the Romans, Sententie recording his theology lectures, and (from c. 1138 to 1139) Scito teipsum (or, as he also called it, his Ethics) are the most important works from this highly productive period. At the same time, Abelard wrote extensively at the request of Heloise, who had taken over the Paraclete as abbess of a group of nuns, providing her with sermons, letters, scriptural exegesis, answers to theological queries and poetry. Even as a young logician in Paris, Abelard had been a controversial figure, competing with William of Champeaux for students and reputation and patronized by William’s enemies in the Church and at court. From the time of the Council of Soissons onwards he became a target for the hostility of the reforming party in the Church and, by the late 1130s if not earlier, for that of its leader, Bernard of Clairvaux. His campaign culminated in the Council of Sens of 1140, where Abelard was accused of nineteen heresies listed by Bernard. Abelard denied all charges of heresy, but the charges were upheld by the Pope. Abelard, now sick, spent the last two years of his life at the great abbey of Cluny and one of its dependencies. There the abbot, Peter the Venerable, ensured that the sentence of excommunication was lifted and engineered a reconciliation with Bernard. Perhaps because of the controversies which accompanied and ended his career, Abelard has gone down in the history of philosophy as a brilliant, daring but unconstructive thinker: powerful as a logician but, otherwise, to be blamed or praised for merely applying the tools of logic to theology. This judgement is unjust, but it does reflect an important difference between Abelard, the most wide-ranging and inventive Western philosopher of the twelfth century, and the great thinkers of the thirteenth and fourteenth centuries. Unlike Aquinas, Duns Scotus and Ockham, Abelard did not combine his developments in different areas of philosophy into a single, coherent and distinctive pattern. Rather, his work falls into two separate parts, corresponding roughly to the division in his career. In his logical works he not only makes startling discoveries of a technical nature; he also reconsiders the metaphysical questions raised by Porphyry and Aristotle in the light of his nominalism, trying to arrive at an account of the basic structure of things and tackling the various aspects, semantic and metaphysical, of the problem of universals with great sophistication. In his theological writing he concentrates on developing a philosophical ethics which, where necessary, shapes the understanding of Christian doctrine to suit its requirements. The following pages can give only an impression of some of the most philosophically interesting aspects of a thinker whose originality and breadth of vision would entitle him to much more space in a History of Philosophy had it not been his misfortune to live in the Middle Ages. First, his ideas on two areas connected with the more formal side of logic will be sketched: his treatment of conditionals, and his analysis of modal statements. (Many other of his more formal developments are also of great interest: for instance, his treatment of ‘impersonal statements’, such as ‘It is good that you are here’, and his discussion of die copula.)⁹ Second, after a glance at his basic metaphysics, Abelard’s approach to the problem of universals will be examined. Third, his account of a central area in ethics, the ethical act, will be discussed and set in context. Abelard developed his ideas about conditionals (‘if…then…’ statements) mainly in considering the theory of ‘topical arguments’ put forward by Boethius in On Topical ‘differentiae’.¹⁰ He did not, however, believe that what Boethius said there about inferences could be applied directly to conditionals, partly because many of Boethius’ maxims were concerned to provide probable, rhetorically convincing arguments rather than irrefragable ones, partly because—unusually for a medieval thinker—Abelard clearly distinguished between the validity of an argument and the truth of a conditional. For the truth of a conditional, his requirement was more stringent even than the modern notion of strict implication (it is impossible for the antecedent to be true and the consequent false): he also insisted on a strict criterion of relevance. For Abelard, ‘if p then q’ is true if and only if p ‘of itself requires’ q, by which Abelard means that the sense of q must be contained in that of p (Abelard [7.19] 284:l–4). One of Abelard’s reasons for imposing this criterion was that, from conditionals which fail this criterion (for instance, ‘If it’s a man, it’s not a stone’, based on the topic ‘from opposites’), Abelard was able to infer a conclusion of the form ‘if p, then not-p’, and he looked on this as a reductio (although most modern logicians would certainly not).¹¹ Unfortunately for Abelard, his great rival and critic in the 1130s, Alberic, was able to show that, even observing Abelard’s criterion, arguments could be constructed which led to ‘if p, then not-p’. Abelard appears to have had no answer to this problem, which would exercise the next generation of logicians (see below, pp. 175–6). Abelard also thought deeply about the semantics of conditionals.¹² On what does the truth of ‘if p, then q’ depend? He rules out two apparently promising answers: that the truth of a conditional is based on a relation between thoughts, or that it is based on the things to which the conditional refers. Thoughts cannot provide the basis, Abelard considers, since one can think of a true statement without thinking of all the numberless statements it entails. Nor can things provide the basis, because (for instance) the conditional ‘If it is a rose, it is a flower’ would remain true even if there were no roses or flowers of any kind. Abelard concludes that the truth of conditionals is based on dicta: on ‘what is said’ by statements. ‘It is a rose’ says of something that it is a rose. ‘It is a flower’ says of something that it is a flower. ‘If it’s a rose, it’s a flower’ is true because that something is a rose not only cannot be true unless it is also true that it is a flower, but also it requires that it is a flower. What, then, are dicta? Often, Abelard treats them rather as some modern philosophers treat propositions. They are non-linguistic bearers of truth and falsity. At other times Abelard seems to regard them more as states of affairs, truth-makers rather than truthbearers. At all times, however, he insists that dicta are not things. This position, too, is far from clear, since it suggests that statements about dicta must be analysed into statements using some other terms, but it is hard to see what these terms could be. Abelard discussed modal logic in the Dialectica and, in greater detail, in his Logica commentary on the On Interpretation.¹³ He was the first medieval logician clearly to distinguish between the de dicto (or ‘de sensu’) and de re readings of modal statements such as ‘It is possible that the sitting man stands’. De dicto this is read as a false statement: ‘Possibly the man is sitting and standing.’ De re it is read as a statement which (provided the man actually is sitting) is true: ‘The man is sitting and possibly he is standing.’ But the exact interpretation of the de re reading gave Abelard difficulties. To a considerable extent, he shared an ancient view of modality which did not allow for synchronous alternative possible states of affairs. According to this view, 1 The man is sitting at t and possibly he is standing at t’ (where t’ is any time other than t) is, under the right circumstances, true, but 2 The man is sitting at t and possibly he is standing at t must be false. The de re modal statement ‘The man is sitting and possibly he is standing’ must therefore be interpreted as (1). Yet Abelard also believes that anything, according to its nature (the sort of thing it is), always has various potencies: a man, for instance, can sit or stand at any time. This view, it might seem, should have led him to acknowledge synchronous possible states of affairs. Instead, Abelard prefers to think about possibility just in terms of what is possible for some thing, according to its nature, without thinking about the possibility or impossibility of states of affairs involving the thing. For instance, being able to walk is part of human nature. Therefore, Abelard believes, it is possible for a man who has had his legs amputated to walk; but he does not think that this commits him to holding that the man might actually walk at some time in the future, nor does he explicitly recognize any possible state of affairs in which the man has not lost his legs, synchronous with the actual state of affairs in which the man is without them ([7.20] 229:34–6, 273:39–274:18). Such an approach may be rather unsatisfactory, but it had its advantages when Abelard came to the theological problem of predestination. Is it possible for God to save a man who is predestined to damnation? Abelard thought not. God would predestine to damnation only those fitting to be damned, and it is not possible for God not to damn someone fit for damnation. Yet, Abelard insisted, it is possible for the man to be saved, since this is a possibility open to any man ([7.18] 521:669–79). Abelard approached the question of what things there are with the presumption of nominalism already firmly in mind. Everything, he believed, is a particular. He thought he had strong arguments for rejecting any of the positions according to which his contemporaries held that there are some things which are not particulars but universals.¹⁴ As a consequence, Abelard had to make a radical adaptation of what might be called the ‘traditional’ metaphysics of his time, taken over from Aristotle’s Categories and Porphyry’s Isagoge. Here is a sketch of this traditional metaphysics. It is not intended to give an accurate account of Aristotle’s or Porphyry’s intentions, but rather an impression of how their textbooks tended to be read by early twelfth-century scholars. According to the traditional ontology, things are of four basic sorts (see Figure 1). There are particular substances: the particular members of natural kinds (such as this man, or Socrates, to take the standard twelfth-century example of a particular substance). Natural kinds like water which do not obviously divide into particular members tend to be ignored. There are universal substances, the natural kinds themselves such as Man and Animal. Then there are what were called ‘accidents’: non-essential properties, and relations, of substances. Like substances, accidents were considered to be universal or particular; so, for instance, Socrates would be white by his own particular whiteness. In practice, however, particular accidents were rarely mentioned. These, then, are the four basic sorts of things: particular substances, universal substances, particular accidents and universal accidents. Man-made objects (houses, ships and so on) were considered to be composites of natural substances. This picture derives mainly from the Categories. The Isagoge added to it Porphyry’s famous ‘tree’ (see Figure 2). Universal substances are arranged into a hierarchy of genera and species. ‘Genus’ and ‘species’ Figure 1 SUBSTANCE ACCIDENT UNIVERSAL Animal, Man colour, whiteness PARTICULAR Socrates, this man this whiteness are relative terms: Man is a species of Animal, and so Animal is the genus of Man, but Animal is itself a species of Living Thing. Each species is distinguished from its genus by a specifying characteristic or, as it was called, differentia: having-sense-perception is, for instance, what differentiates Animal from Living Thing. As a nominalist, Abelard had to make some drastic changes to this traditional scheme.¹⁵ Holding that every thing is a particular, he simply cancels out the first line in Figure 1. For him, things are of just two sorts: particular substances and particular accidents (and differentiae). And Abelard stresses that although particular accidents cannot exist except in dependence on a particular substance, they are each separate things, which might have been attached to different substances from those to which in fact they are attached.¹⁶ Since there are no universal substances, there cannot be a hierarchy of genera and species. But Abelard translates Porphyry’s tree into the structure of particular things. He regards differentiae as particular, non-substance things, exactly like accidents (he had a convenient word which meant either an accident or a differentia: a ‘form’), except that each substance of a given kind must have attaching to it certain given sorts of differentiae, as indicated by Porphyry’s tree: for instance, a man cannot be without rationality, mortality, having-sense-perceptions and so on. This scheme is not without problems. It might seem to imply that a particular substance of a given kind is not really one thing at all, but rather a bundle of differentiae. Some of Abelard’s discussions do appear to favour a bundle theory, in which these bundles would be attached to body, which would be regarded as fundamental rather than as just one type of substance. But elsewhere Abelard explicitly recognizes that particular substances exist in a way which, in theory, is independent from the differentiae which must attach to them (see esp. [7.19] 420:30– 421:8, and cf. [7.68] 128–30). Another difficulty concerns accidents. Aristotle had given nine classes of accidents, which included, for Figure 2 instance, relations, time and posture. Abelard initially accepted that even accidents in these categories are particular things. In the mid- 1120s—after he had done his most important work as a logician—he came to think it implausible that a relation such as fatherhood is a thing of any sort. He therefore revised his treatment of accidents and accepted the existence of particular accidental forms only in some categories; but he was left (at least to judge by surviving texts) without an account of accidents in the remaining categories.¹⁷ Abelard’s basic metaphysics set the problem which his treatment of universals had to answer.¹⁸ As a nominalist, his explicit answer to the question which his contemporaries usually posed—‘Are there universal things or just universal words?’—was unequivocal: only words could be universals. How then, he had to explain, could universal words be used meaningfully? His theory of the semantics of universals is designed to answer this question, and does so with remarkable success. But there still remained a metaphysical question for Abelard to tackle: if species and genera are not things, what is the real basis for the system of natural kinds, which Abelard recognized as a feature of reality, not a mind or language-imposed construct? Abelard’s answer to this question, less satisfactory than his treatment of the semantic one, ends by taking an unexpected turn. To begin, however, with the semantic problem. It is not, as a modern philosopher might expect, a problem about deciding the reference of predicates. For Abelard (along with most of his contemporaries), in a statement ‘S is P’, the reference of ‘S’ and ‘P’ is the same. Latin grammar makes this position plausible: there are no articles, and an adjective ‘_’ always includes the meaning of _-man/woman/thing, according to its gender. So in ‘Socrates est homo’, ‘Socrates’ and ‘homo’ (‘a/the man’) are thought to refer to the same thing: Socrates; and similarly in ‘Socrates est albus’, ‘albus’ (‘a/the white man’) is taken to refer to Socrates. There is no difficulty about any of this for a nominalist, since Socrates is a particular thing, a perfectly acceptable referent for words. The nominalist’s problem concerns, rather, the signification of universal words. To signify x to someone is to cause there to be a thought of x in his mind. Twelfth-century logicians were primarily concerned with signification in their semantic analyses. It is through the signification of the predicate, they held, that the speaker conveys his meaning: that Socrates is a man (not a donkey), that he is white (not turquoise or indigo). And, whilst Socrates is a man on account of particular differentiae of rationality and mortality, and white by a particular accident of whiteness, in the statement ‘Socrates is white’ the signification of ‘white’ is universal: it produces a thought of whiteness in general, not of the particular whiteness by which Socrates happens to be white. But, according to Abelard, there is nothing which is whiteness in general (or man-ness in general): there are just particular accidents of whiteness, just particular men. So there seems to be no x of which universals can produce a thought when they signify. Abelard’s earlier way of tackling this problem, in the Logica ([7.20] 20:15–22:24), is to posit an x which is not a thing. When universal words are heard, they produce a thought (which is a thing—a particular accident—Abelard holds). They also cause a mental image which, Abelard says, is not a thing at all, but a figment. (Abelard’s explanation of why it is not a thing shows that he thinks of the image solely in terms of its content: my mental image of a castle cannot be a thing because it is not really made of stone, and so on.) In the case of universal words, the mental image is a common, undifferentiated one, of man rather than of Socrates. It is these common mental images or conceptions which, he says, are the objects of the thoughts produced by universal words. Abelard can therefore claim both that universal words signify— there is an x of which they produce a thought—and that there is no thing which they signify. In the Glossulae ([7.20] 530:24–531:29) Abelard simplifies this picture. There he argues that a word signifies so long as it produces in its hearers thoughts with content. The content of the thoughts produced by universal words is (or, at least, can be) universal, derived from particulars by a process of abstraction. And so universal words signify, but it does not follow from this that there are any universal things which they signify. Although any account of signification must involve the mind, in both Abelard’s earlier and his later theories universal words have their signification as a result of how things really are. We form a common conception of man—or, in the later theory, abstract a universal thought content for man—because men are really alike. But, although the signification of universal words is based on how things really are, it is not always based on a complete understanding of how they really are. Abelard takes it for granted that we unproblematically group things correctly according to their natural kinds, but he considers that only in some cases do we know the structure of differentiae characteristic of a given natural kind. There is another semantic relationship, however, which does link universal words to this very structure in all cases: Abelard calls it ‘imposition’. He envisages the first user of a word ‘imposing’ a certain group of sounds on a particular substance and every substance of the same kind. The impositor may well not know what is the structure of differentiae that characterizes the substance in question, but when he imposes the word he does so according to the structure the substance really has, whatever that may be. He thus creates a link—though an open, unspecified one—between the universal word and the real structure of the objects to which it can be used to refer and on which its signification is based.¹⁹ The metaphysical side of the problem of universals for Abelard is to explain in what the real resemblance between members of a kind consists, given that there are no real universals. He tackles it with his notion of status ([7.20] 19:21–20:14). Men, for instance, are alike in sharing the status of man; and this, he explains, means just that they are alike in being men or in that they are men. And the status of man (being a man), he insists, is not a thing of any sort. There is nothing wrong with this explanation, but it raises another question in its turn. On what in reality is the status of man founded? What is involved in being a man? Abelard has all along made it clear that what characterizes men is a certain structure of particular differentiae. No one is a man who does not have (his own particular) differentiae of mortality, rationality and so on. Abelard might, at this stage, have proposed a variety of resemblance nominalism, which would hold that particular differentiae of a given sort are exactly similar and this similarity is unanalysable. Each status would be defined as having a certain structure of such particular differentiae; x would share the status of y if and only if his particular differentiae were exactly similar to y’s. Surprisingly, in the only discussion ([7.20] 569:32–573:5) where he directly answers the question of what makes different particular differentiae of the same sort similar, Abelard opts for a different solution. He suggests that, whilst a man may have been rational by any one of infinitely many actual or hypothetical particular rationalities, there is a universal differentia of rationality, which no man can lack. He goes on to say that, in a sense, this universal differentia is the same as a particular differentia, since it differs from it not as one thing from another, but merely by ‘definition’—a type of difference Abelard had introduced in discussing the Trinity and never entirely clarified.²⁰ Although this discussion therefore remains obscure, it suggests that, pressed to give the metaphysical basis of his theory of universals, Abelard has sacrificed much of the nominalist ground he so strenuously defends at every other stage. Abelard developed his ethical theory on three different levels. He attempted (especially in the Collationes) to answer the most general questions about the nature of good and evil and their relation to God (see [7.65] and [7.68] 233–50). Whereas Abelard’s contemporaries and medieval and patristic predecessors tended to argue, in Neoplatonic fashion, that evil is a privation not a thing, Abelard was ready to admit that there are evil things—particular accidents of, for instance, pain or sorrow—although not evil substances. He reconciled this position with God’s goodness and omnipotence by explaining that, when we assert the goodness of God’s providence, we are predicating ‘good’ not of things but of dicta: ‘it is good that there are evil things’ (‘good’ is predicated of the dictum that-there-are-evil-things) and does not entail ‘evil things are good’. In some of his sermons and poetry, the Rule he wrote for Heloise and her nuns and a long poem of advice to his son (the technocratically-named Astralabius), Abelard developed the practical implications of his ethics and examined the tensions between moral standards and the thoughts and feelings of individuals in difficult, ethically problematic circumstances.²¹ At the centre of Abelard’s moral theory, however, is his discussion of the ethical act; and, although his treatment of virtues and merit is also interesting and innovative, it is Abelard’s treatment of sin which, rightly, has attracted the interest of historians of philosophy.²² Yet it has often been misconstrued. Historians have frequently described it as ‘intentionalist’ and contrasted it with a crude, externalist approach to ethical judgement, where a person’s guilt is judged solely according to the sort of acts he performs, as assessed by an outside observer. A moralist can be an ‘intentionalist’ in one or both of two ways. The intentionalism may concern what is judged ethically (object intentionalism). The object non-intentionalist considers that external acts alone are to be judged, whereas the object intentionalist considers that agents’ intentions must be judged as well as, or instead of, their acts. Or the intentionalism may concern the basis for ethical judgement (subject intentionalism). The subject intentionalist will hold that an agent’s own beliefs about what is right and wrong are an important element to be taken into account in reaching a judgement; the antiintentionalist will minimize their role or exclude them. Abelard is certainly both an object and a subject intentionalist. But his object intentionalism needs to be set alongside the different object intentionalism of his contemporaries, not contrasted with an imaginary externalism, and his subject intentionalism does not have the extreme consequences which it might at first seem to threaten. In their treatments of sin, Abelard’s contemporaries put great weight on the intentions accompanying a sinful act. The most popular theory envisaged them as mental acts preceding the external sinful act itself. Its exponents analysed the psychological stages of committing a sin, from contemplating the action, being tempted, indulging the temptation to performing the act. Although these theorists held that the performance of the external act added to the gravity of the sin, they considered that, already at an early stage of contemplating the sinful action with pleasure, the agent would be sinning to some degree, even if he went on successfully to resist the temptation. They held, then, that it is worse actually to sleep with a married woman than to be ready and about to do so but prevented by the unexpected arrival of the husband. But they would also consider that a man would have sinned to some degree if he merely thought with pleasure about sleeping with her, even if he would never have considered making any practical move to do so. By contrast, Abelard held that neither the performance of the external act itself, nor any of the thoughts or feelings preceding it but not directly linked to its real or planned performance need be considered in judging sin. Before we perform an action, he believed, we perform a mental act of willing it or (in the terminology he finally used, in Scito teipsum) ‘consenting’ to it. Consenting to an act means being entirely ready to perform it, and so I can consent to an act which I do not perform because it is thwarted. For Abelard, it is acts of consent—not any other type of mental events, and not external actions—which alone can be sins. The thwarted adulterer sins no less, having consented to adultery, than the successful one; whilst, if he is inflamed with passion for a married woman and incessantly imagines with pleasure the idea of sleeping with her, but he resists the temptation to do so, then not only does he not sin, he wins merit in the sight of God for his successful struggle.²³ The contrast made above between external acts and mental events preceding them omits what many philosophers now would consider the most important element in any theory of action: the idea that we act under a description, which is linked to various of our mental and external acts, both before and after the act in question. To some extent Abelard seems to have grasped this idea. He is very concerned to distinguish between consent and what he calls ‘willing’. When a man consents to adultery, he does not will to commit adultery if he would prefer it were the woman in question unmarried. When a man murders his feudal overlord in self-defence, knowing that his supporters will certainly try to take their revenge on him, he certainly does not will to commit murder, although he consents to it and, for Abelard, he would therefore be guilty of murder, just as, in the first example, he would be guilty of adultery ([7.21] 6:24–8:20, 16:16–32). What does Abelard mean by this notion of reluctant action, in which I do not will to perform what I do in fact perform? Although he does not develop the idea explicitly, he seems to have in mind that most acts fit a number of descriptions. The adulterer does not will to commit adultery since he would not choose to perform the act just under the description of ‘sleeping with a married woman’ (or ‘committing adultery’), nor the murderer his act just under the description of ‘killing one’s overlord’. In each case the agent consents to the act on account of other relevant descriptions of it such as sleeping with the woman one desires, or saving one’s life. What determines for Abelard which acts of consent are sinful and which not? We sin, Abelard believes, by showing contempt for God (e.g. [7.21] 4:31–2). He explains what it is to show contempt for God in two different ways: either as (a) doing what is not fitting (e.g. [7.21] 4:27–8) or as (b) not doing what we believe we should do for God (e.g. [7.21] 6:3–6). The juxtaposition of two so different accounts may seem puzzling. In any case, it seems that only (b) fits the equation of sin with contempt of God. If I do what is unfitting but believe that it is what I should do for God, I cannot be showing contempt for him (unless my contempt consists in not having found out what really is fitting and unfitting). The puzzle is solved, however, by Abelard’s beliefs about natural law.²⁴ Abelard considered that all mentally competent adults (who alone he held capable of sinning) at all periods of history naturally know the general moral precepts laid down by God, such as the prohibitions of murder, adultery and theft. He also believed that they all have the power of conscience, which he saw as an ability to see how particular actions fall under the general commands and prohibitions of moral law. There is then, for Abelard, no gap between what is fitting for moral agents to do and what they believe they should do for God. Although he is a subject intentionalist, he thus avoids the danger of having to allow that someone might not sin whatever action he performed simply by virtue of not believing that the action is sinful. There are, of course, difficulties about his view. Although, in his discussions of ethics in practice, Abelard is acutely aware of the possibility of moral conflict—where one and the same action is both enjoined and forbidden by divine law—for theoretical purposes he ignores such dilemmas. Moreover, Abelard has to account not just for natural law, but for the revealed laws of the Old and the New Testament. The Old Law raises a special difficulty for him. He considers, as in consistency he must, that a Jew who accepts the Old Law and breaks one of its special precepts, not contained in natural law, such as the dietary laws, commits a sin because he is showing contempt for God by his action ([7.16] 306:311–25). This, he grants, applies to a twelfthcentury Jew as much as to a biblical one. But Abelard also considers that the twelfth-century Jew is mistaken to believe that God now enjoins the dietary laws on him or on anyone. In practice, the point is of little importance. But in principle the gap which Abelard has allowed between moral belief and the truth about divine precepts upsets his whole theory, for it is hard to see what limit he could place on the beliefs which people or groups of people might sincerely hold about what are God’s special laws for them. GILBERT OF POITIERS Next to Abelard, the most profound and adventurous thinker of the twelfth century was Gilbert of Poitiers (1085/90–1154). Gilbert enjoyed the successful and comparatively undramatic career for which Abelard might have hoped. A native of Poitiers, he was taught at Chartres by Bernard and at Laon by Anselm. He became a canon and then chancellor of Chartres, and taught both there and in Paris. In 1142 he was made bishop of Poitiers. Like Abelard, Gilbert was the object of Bernard of Clairvaux’s suspicion and hostility. Gilbert was forced to defend his views on the Trinity and the Incarnation, first in front of Pope Eugene III (April 1147) and then at a consistory after the Council of Rheims (March 1148). As he had done with Abelard, Bernard used underhand tactics to try to ensure Gilbert would be condemned. But this time he was unsuccessful and Gilbert was allowed to return to his diocese without harm to his reputation.²⁵ Gilbert did not write prolifically. He produced biblical commentaries (on the Psalms and the Pauline Letters), where he kept close to the specifically Christian doctrinal themes, eschewing the opportunities for ethical speculation so eagerly followed by Abelard. A set of theological Sententie survives (in two versions). But they do not seem to offer a close account of his lectures, and certainly incorporate the views of other masters. Gilbert was an accomplished logician as well as a theologian, and he gave rise to a distinctively Porretan school of logic; but no logical work of his own is known. Gilbert’s contribution to philosophy emerges only from his long and intricate commentary on Boethius’ Theological Treatises (Opuscula sacra), probably written early in the 1140s. Since Gilbert’s technique as a commentator is to gloss every word of the original text, the character of Boethius’ treatises has a deep influence on his work, but in ways that are unexpected. Gilbert does not take over many of Boethius’ views or arguments directly, even though he is supposedly explaining them. Rather, he strains to fit Boethius’ words into his own often very different arguments, at the cost of an unwieldily profuse terminology and frequent obliqueness or obscurity in exposition. In his Opuscula, Boethius had been concerned in the main to use logical and metaphysical ideas rather straightforwardly as ways of elucidating and confirming orthodox Christian doctrine about the Trinity and Christology. Gilbert, however, insists that different principles of argument and ways of arguing must be used in different disciplines. He claims that arguments devised in connection with natural things (natural science), or in the course of analysing them into their in reality inseparable constituents (what Gilbert calls ‘mathematics’), cannot be used directly in talking about God (theology). But these arguments can be used indirectly, by ‘proportionate transumption’, a process in which some of what a natural or mathematical argument establishes is accepted as applicable to God, but not all.²⁶ This framework gives Gilbert the chance to develop his philosophical account of the natural world more fully than Boethius had done. But, in developing his natural and mathematical arguments, Gilbert is always at least in part concerned with how to ‘transume’ them proportionately so as to serve his, and Boethius’, ultimate theological aims. Gilbert’s main philosophical discussions—of topics such as predication, parts and wholes, individuation and the relation between body and soul—are all coloured in this way, and sometimes their rationale becomes clear only in the light of his doctrinal objectives. Yet it would be wrong to see Gilbert merely as a theologian propounding quasi-philosophical arguments to illustrate Christian doctrine. Parts of his thinking take up the type of philosophical questions which had been stimulated by the ancient logical texts and which had fascinated Abelard; and nowhere more clearly than in his complex and original treatment of the metaphysical structure of things.²⁷ Gilbert makes a fundamental distinction between what he calls quo est (‘from which it is’) and quod est (‘what it is’) [7.10] 91:51–8, 116:47–9).²⁸ (Driven by the requirements of exegesis, he also uses a bewildering variety of other terms to describe this distinction.) Examples of what Gilbert considers quod ests are Socrates, this man, that dog, this white thing. What he has in mind, it seems, are concrete wholes made of substances along with their accidents. Although a denominative word such as ‘white thing’ (album) is the word for a quod est, Gilbert assumes that in any given case when it is used its reference will be the same as that of a substance word and, often, of a proper name. So, for instance, the quod est in question might be this white thing, this labrador, Fido. (Gilbert does not envisage instances where a denominative might sort things differently for purposes of reference, for instance, ‘this white thing’ referring to Fido and the white ball in his mouth taken together; nor does he indicate how he thinks about man-made objects.) Quo ests are, for instance, whiteness, bodiliness, rationality, humanity, Socrateity. They are not, however, universal forms. Every quo est is singular (or one in number: Gilbert uses the two notions interchangeably), and every quod est is what it is from its own singular quo ests ([7.10] 144:58–60, 145:95–100). So, for instance, I am rational (supposing I am) and six-foot tall from a singular rationality and a singular being-six-foot-tall which are each quo ests numerically distinct from Socrates’ rationality and being-sixfoot- tall. As this account suggests, by ‘quo ests’ Gilbert means something very close to what Abelard and others had in mind when they spoke of particular forms (accidents or differentiae). So, for example, Abelard would talk of the particular being-six-foot-tall and the particular rationality attaching to Socrates by which he is six-foot tall and rational. Yet there are important differences between the two philosophers’ schemes. Abelard thinks of particular forms attaching to substances and, although one element of his discussion (the ‘bundle theory’) points in a different direction, he accepts that, were a substance per impossibile stripped of all forms, it would still retain an identity. For Gilbert, however, the relationship between quo ests and quod ests is causal and correlative. A quod est is made what it is by its quo ests and there can be no quo ests apart from a quod est ([7.10] 278:8–279:12). The notion of bare substance may be problematic, but that of a bare quod est would be simply ungrammatical, because a quod est must be a ‘what’ (a white thing, a rational thing, Socrates), made what it is by a quo est (whiteness, rationality, Socrateity). Gilbert’s scheme thus avoids some of Abelard’s problems (but at a price, since the notion of making or causing involved is thoroughly obscure: what the quo est makes into the quod est cannot be the quod est itself—so what is it?). Another important development in Gilbert’s scheme has already been indicated by mentioning the quo ests humanity and Socrateity. As well as all his simple quo ests, such as rationality, mortality and being-six-foot-tall, which make Socrates something which is mortal, rational and six-foot tall, there are also his complex quo ests, composed of two or more of the simple quo ests; so, for instance, his humanity—that by which he is a man—would be composed of the differentiae of the species man and of the differentiae of all the genera of that species (rationality, mortality, having senses, being alive, being bodily). The most complex of all these composite quo ests is called by Gilbert the ‘collected property’ or ‘whole form’ of Socrates (‘Socrateity’, for short). It is composed of all the quo ests ‘which both in actuality and by nature have been, are and will be’ those of Socrates ([7.10] 144:73–8, 274:75–95). Gilbert uses this idea of whole forms to make one of his most characteristic distinctions. As already mentioned, Gilbert holds that every quo est is singular; even composite quo ests, such as this humanity or Socrateity, are singular ([7.10] 167:7–19, 301:86–95). So too is every quod est singular. It is singular, Gilbert says, because the quo est which makes it into a quod est is itself singular ([7.10] 144:58–62). To be singular is not, however, for Gilbert to be individual. In his view everything which is individual is singular. But only those singulars which are not ‘dividuals’ are individual. Whatever is exactly similar (conformis) to something else, or could possibly be exactly similar to something else, is a dividual. Although the quo ests by which Socrates is rational and six-foot-tall are singular and distinct from the singular quo ests by which Plato is rational and six-foot-tall, each quo est of rationality and each quo est of being-six-foot-tall is exactly similar to every other such quo est. The same is true of almost every quo est, whether simple (mortality, whiteness) or complex (animality, humanity). Even if it should happen that as a matter of fact there is not, never has been nor ever will be a quo est exactly similar to a simple or complex quo est, then in almost every case it is possible that there might be one exactly similar ([7.10] 143:52–144:78, 270:73–271:82). No one, suppose, has ever had or will ever have a nose quite the same shape as mine; but ‘by nature’—hypothetically— there might be such a person. Or consider the complex quo est sunness which makes something into a sun. Gilbert thought (wrongly, of course) that there was and would be only one thing like this: sun is a species which contains only one member, the Sun. But by nature, he believed, there is nothing to prevent there being infinitely many suns, all made into what they are by quo ests of sun-ness, each singular but exactly similar to each other. The (as a matter of fact unique) quo est of sun-ness is therefore dividual just as the very many quo ests of humanity are dividual ([7.10] 273:53–71). There is just one type of quo est which, Gilbert claims, is not dividual because it is not actually or possibly exactly similar to any other quo est: the whole form of a quod est (for instance, Socrateity). Whole forms, then, are individuals, and so are their quod ests, that is to say, every quod est, since every quod est has a whole form. Where, then, for Abelard (who makes no distinction between particularity, singularity and individuality) a form such as this whiteness or that rationality is no less a particular thing than Socrates himself, Gilbert is able to discriminate more finely: this whiteness, that rationality and Socrates are each singular, but only Socrates (and his whole form, Socrateity) is individual. Gilbert’s idea of individuality also provides him with his approach to the problem of universals. Complex dividual quo ests fall into groups, the members of each of which, although themselves all singular, are completely similar to every other member of the group. In virtue of this complete similarity all the members can be regarded as one universal, a species: for example, humanity. Again, many complex dividual quo ests are completely similar not just to some other quo ests in every respect, but also to some in some respects; for instance, every quo est of humanity and every quo est of horse-ness are completely similar in respect of being bodily, being alive and having senses. They can therefore be further grouped into what is also regarded as one universal, the genus animality ([7.10] 269:34–50, 312:95–113). True to the emphasis of the discussion at the time, Gilbert considers just universals in the category of substance; presumably, though, he would also consider that groups of exactly similar simple quo ests (such as whitenesses or rationalities) are universals in other categories. Gilbert, therefore, is a realist over universals, but his real universals are all quo ests which cannot exist except in conjunction with quod ests which, because they are individual, cannot be universal. To the objection that his real universals must, like any real universal, be contradictorily both one and many, Gilbert could reply that, whereas it would be a contradiction to assert of many individuals that they are one, it is permissible to say of many singulars that they are one because of their complete similarity to each other.²⁹ All this rests on the presumption that whole forms are indeed individual: there is nothing else to which any of them is, or could possibly be, exactly similar. What entitles Gilbert to make this presumption? Gilbert gives no explicit answer, but it is worth looking carefully to see what, if anything, he had in mind. It seems obvious to relate the individuality of whole forms to the principle that no two bodily objects can be in the same place at the same time: they cannot therefore have exactly similar accidents in these respects. Gilbert’s comments in a different context ([7.10] 77:5–78:13, 148:88–92) show that he was highly aware of this point. Yet, at first sight, such an explanation seems not to fit Gilbert’s view (see above, p. 169) that the whole form of Socrates is composed, not only of all the quo ests that have, do and will make him what he is, but also of all those that he has ‘by nature’. The spatio-temporal dissimilarity principle does not rule out Plato having by nature the very same space-time accidents as Socrates actually has. Indeed, if Socrates’ complex quo est is composed of all the quo ests he has ‘by nature’, it seems that it must include every sort of quo est that can attach to a man, and that therefore the whole form of each member of a species is exactly similar to that of every other member of that species—and therefore, contrary to what Gilbert maintains, is dividual. What must be implicit, it seems, in Gilbert’s view is a distinction between quo ests which apply to alternative, different ways things might be (or, at least, a distinction between those quo ests which something has in actuality, and those which it has only by nature). This would accord with Simo Knuuttila’s view that Gilbert was one of the earliest thinkers who, influenced by the doctrine of divine omnipotence, was willing to admit synchronous alternative possible state of affairs, each belonging to different providential programmes, any of which God could put into effect although only one is the actual programme he chooses.³⁰ Gilbert would, then, be able to insist that, in any given providential scheme—and so in whatever scheme is the actual one— Socrates does not share his spatio-temporal quo ests with anyone else, and that therefore his whole form is not the same as anyone else’s. None the less, Gilbert did not in fact think out his views on modality this far; had he done so, he might not have wished to accept the many difficulties this view brings with it.³¹ THE PLATONIC CURRENT Most of the main twelfth-century thinkers owed something to Plato. Abelard, for instance, argued that the description of the World Soul in the Timaeus was an allegory of the Holy Spirit, and he used this surmise as evidence that the pagan philosophers knew of the Trinity before the coming of Christ. He also took the few comments about the Republic at the beginning of the Timaeus and used them as the basis for his own political ideal of cities where everything is done for the common good ([7.68] 304–7). Abelard’s Platonism, however, is opportunistic. He takes themes from Plato and transforms them for his own purposes, using them within a structure of thought which itself owes remarkably little to Plato. Some scholars have argued that there is an important Platonic element in Abelard’s treatment of universals and dicta (see [7.61] 149 and [7.56]). The interpretation offered here does not support that view. Similarly, historians have often held, contrary to the reading advanced here, that Gilbert rests his metaphysics on a notion of Platonic Ideas. There is, at any rate, room for dispute about the Platonism of Abelard and Gilbert, but neither continues the tradition of Bernard of Chartres in the direct way that was done, as John of Salisbury recognized, by William of Conches. William of Conches was already teaching and writing in the early 1120s and appears to have remained active until the 1150s; he taught perhaps at Paris or, so some have argued, at Chartres, and later taught at the court of the Duke of Normandy. Like Bernard, he was a grammar teacher. He wrote a commentary on Priscian’s Institutiones grammaticae (Principles of grammar), drawing extensively on the anonymous eleventh-century Glosule to Priscian; and he made detailed commentaries on a series of classical texts. These include poetry (he is known to have glossed Iuvenal) but he concentrated on Platonic writers: Boethius’ Consolation of Philosophy, Macrobius’ commentary on Cicero’s Dream of Scipio and, most important, the Timaeus itself. In interpreting these texts, William drew on the already well-established medieval tradition of reading pagan texts as allegories of Christian truth. He went beyond his predecessors in the thoroughness with which he applied this method. He would admit, if necessary, that, being pagans, the ancient authors could not be trusted in everything they said. Usually, however, he discovered a satisfactory reading. He shared with Abelard (and may have taken from him—but the chronology is not clear) the identification of Plato’s World Soul with the Holy Spirit. But unlike Abelard, for whom it carried important implications about the knowability of God, he was willing to drop the identification. For William, it was merely a convenient reading of an ancient text, which could if necessary be sacrificed. The ideas in his texts which William took most were scientific, rather than philosophical or theological ones.³² The Timaeus and Macrobius were regarded as important sources for natural science, and William developed this interest independently, in his Philosophia (c. 1125) and his later dialogue (part extension, part more cautious revision of the Philosophia), the Dragmaticon (between 1144 and 1149). In these works he also made use of, and sometimes combined or developed in an original way, medical and scientific sources translated from the Arabic and Greek. In the Philosophia and the Dragmaticon, and in the commentary to the Timaeus, William tried to show how all things came into being through the natural interaction of the four elements (fire, air, water and earth). Only the creation of the human soul required a separate intervention by God. William was firmly committed to this search for naturalistic explanations: he accused of pride and ignorance those who wished to explain everything by divine intervention, and claimed that he illustrated God’s power by his explanations of how God worked through nature ([7.31] 39–40). The other leading Platonist of the mid-twelfth century (he was described by a pupil, Hermann of Carinthia, as ‘the soul of Plato restored to mankind from heaven’) was Thierry ‘the Breton’, known also as Thierry of Chartres, where he was chancellor in the 1140s; previously he had taught both there and almost certainly at Paris. As well as an interest in rhetoric, logic and the various branches of mathematics, Thierry shared William of Conches’s penchant for naturalistic explanation of what many of their contemporaries would have described in terms of direct divine intervention. But Thierry gave his most distinctive philosophical teaching—and that which shows his Platonism most clearly—in the course of commenting on Boethius’ Theological Treatises.³³ A commentary known as Librum hunc (written late 1140s; incipit ‘Inchoantibus librum hunc’) can be shown to be fairly closely based on this teaching. Together with two other commentaries on the work from the same period (Incipit ‘Intentio auctoris’ and ‘Aggreditur propositum’), which contain close parallels to it and each other but also differ sharply in some of their doctrines, Librum hunc shows how Plato’s Theory of Ideas was adopted and transformed. The first stage in the transformation was Boethius’ doubling of the forms: ‘From those forms which are beyond matter’, writes Boethius, ‘come those forms which are in matter and make the body’—forms which therefore, more properly, should be called ‘images’. All three commentaries go on to argue that only the images which come into contact with matter are many, and that the forms from which they derive are really one form, God, the form of forms. The argument is developed in two main ways. Librum hunc asserts that the forms of all things ‘emanate’ (emanare) from the one, simple divine form. These forms have it in common with the one true form that they are an ‘equality of being’. By this he apparently means that a particular substance s is what it is (an s) from its form f, so that f can be called ‘the equality of being an s’. God’s form, ‘the wholeness and perfection of all things’, stands in the same relation to God as f to s. None the less, the commentator insists that all forms besides that of God are not really forms, but simply the images of form. Plurality, he goes on to explain, arises solely through the coming together of form, which is one, and matter, which is also in itself one.³⁴ In the two other commentaries, the talk is not of emanation but rather of God’s thinking.³⁵ When an artificer wishes to produce a mental exemplar of what he will make, he must think of the material of which it will be made: the exemplar is not itself enmattered, but it must be conceived in relation to matter. Similarly, God, who is himself the form of forms, conceives the forms of all things in relation to their matter. He then unites them with matter, at which point they cease to be forms and become images. In the second half of the century, the Platonic current (no longer closely connected with exegesis of the Timaeus or with natural science) intermingles often curiously with other influences. For example, there survives a fragment (itself book-length) of a very lengthy commentary on a work by the fifth-century Greek Neoplatonist known as ‘pseudo- Dionysius’ (because he issued his works under attribution to Dionysius, the Areopagite converted by St Paul). The commentary was written between 1169 and 1177 by William of Lucca. William was probably the author of a logical textbook based on Abelard’s teaching (see below, p. 175); he was certainly deeply influenced in his theology by Gilbert of Poitiers; and, like Thierry of Chartres, he formulated his thoughts about Platonic Ideas (which, in what remains of his commentary, are far from clear) with Boethius’ Theological Treatises in mind.³⁶ Alan of Lille (see below, pp. 177–8) provides another example of late twelfthcentury syncretic Platonism. THE LOGICAL SCHOOLS OF THE LATER TWELFTH CENTURY In the second half of the twelfth century, logicians divided themselves into a number of self-consciously distinct schools, all probably based in Paris.³⁷ Each of these schools derived from one of the leading logical masters of the preceding period. The Porretani (or Gilebertini) were the followers of Gilbert of Poitiers (who was called Gilbert Porretanus). Abelard’s followers were called the Nominales.³⁸ The Parvipontani (or Adamitae) were the followers of another influential logical, Adam of Balsham (d. 1159), called Parvipontanus because his school was at the Petit-Pont, whose Ars disserendi (c. 1132) offers a highly innovative approach to logic, both in its terminology and arrangement of material. The Meludinenses or Robertini were almost certainly the followers of Robert of Melun, although at present only Robert’s work as a theologian, not as a logician is known. Another important logician of the 1130s and 1140s was Alberic, a determined opponent of Abelard’s. Although no surviving text can be definitely assigned to him, his views are frequently and respectfully cited in the logical commentaries of the 1140s, some of which can be closely linked to him.³⁹ The logicians who called themselves the Albricani were certainly his followers; whether they can be identified with the Montani (from the Mont Ste Geneviève, where Alberic, but also Abelard and others taught) is unclear. The later twelfth-century masters who ran these schools remain anonymous, but a number of texts survive which show their sophistication and ingenuity. From the Porretans there is a substantial textbook of logic (called by its editors the Compendium logicae Porretanum), probably written between 1155 and 1170. William of Lucca’s Summa dialetice artis (where Abelard is throughout the supreme authority, called the Philosophus) illustrates the thinking of the nominalists; the Introductions montane maiores that of the Montani; the Ars meliduna—the longest and most sophisticated of all—the work of the Melidunenses.⁴⁰ The division of these logicians into schools is not a mere convenience of the historian: it reflects how the scholars thought of themselves. For each school there was a set of basic theses to which all those who belonged had to subscribe. The Porretan Compendium takes the form of a commentary of each of the Porretani’s theses, and there survive similar lists of theses with discussion for the Melidunenses and the Nominales.⁴¹ Most of the theses concern controversies which arose in connection with the Isagoge, Categories and On Interpretation over topics such as universals, predication, parts and wholes and entailment. Often they are stated in a deliberately paradoxical fashion; for instance, according to the Melidunenses, ‘Socrates and Plato are not Socrates and Plato’; according to the Nominales, ‘Nothing grows’. The divisions between the schools emerge very clearly in the differing solutions each proposed to the objection Alberic had raised to Abelard’s theory of conditionals (see above, pp. 157–8). Alberic’s argument begins from the principle that, if p implies q, then p and r implies q, and it chooses as an exemplification of this argument-pattern one in which the conjunction of p and r is an impossibility. The Porretani rejected this argument-pattern, because it leaves one of the conjuncts without a role in the implication, whilst the Montani refused to accept conditionals where the antecedent is impossible. The Nominales, Abelard’s own followers, seem (not surprisingly) to have been left in some confusion; the Melidunenses argued that nothing follows from a false statement; whilst the Parvipontani alone did not treat Alberic’s argument as a reductio, but accepted it along with its conclusion and therefore the paradoxes of strict implication: from an impossibility anything follows, and a necessity follows from anything.⁴² It would be very wrong, however, to imagine that the logicians of the later twelfth century did no more than react to and systematize the ideas of their founders. First, even on the most closely discussed questions of the previous decades, the new masters had their own thoughts. So, for example, the Ars meliduna proposes a sophisticated Platonic theory. Universals are all ‘intelligible things’. What the mind grasps when it considers a universal is not, though, a relation of similarity, but the ‘coming together’ (communio) of things which—in the case of genera and species (Animal, Man), but not that of other universals (white, rational)—brings it about that what participates in it is something.⁴³ Second, the later twelfth century was a time when previously unknown Aristotelian logical texts (the logica nova) first became known: the Prior Analytics, the Topics and On Sophistical Refutations. All three, for instance, are used in the Ars meliduna. This development is less important, however, than it might seem. It was only the third of these texts, Aristotle’s treatise on sophisms (arguments which are incorrect but superficially plausible) that was studied enthusiastically. On Sophistical Refutations was known before the others, by the 1120s; Alberic and his followers were greatly impressed by it; and it continued to fascinate logicians for the rest of the century. Although its value in detecting logical fallacies in theological arguments may have initially recommended the treatise, it came to fuel a growing interest in the principles of deductive argument.⁴⁴ The third and most important development of the years 1150–1200 may help to explain the generally slight impression made by the logica nova. It was in this period that medieval logicians broke away from the framework of study set by the ancient authorities. For instance, whereas Abelard’s Dialectica had mainly followed the pattern of the textbooks by Aristotle, Porphyry and Boethius, both the Porretan Compendium and the Ars meliduna reorganize the whole subject-matter of logic into four parts: terms, statements (propositiones), what terms signify, what statements signify (see [7.77] II, 1, 539). The new approach went beyond matters of organization. Almost all the branches of what would be called the logica modernorum—those parts of logic not covered in the ancient texts—began to be developed at this time: besides the theory of conditionals (which Abelard had already begun to develop), the theory of the (semantic) properties of terms; the study of sophisms and of words such as ‘only’, ‘except’, ‘begins’, ‘ceases’; the treatment of semantic paradoxes and the special sort of logical disputation called ‘obligations’.⁴⁵ PHILOSOPHICAL THEMES IN LATER TWELFTH-CENTURY THEOLOGY The divisions between schools were less clear-cut among the theologians than the logicians. There were, certainly, those who followed Gilbert of Poitiers either very closely or with more freedom. Abelard, too, was highly influential; but his more controversial theological doctrines were usually rejected and his distinctive ethics adopted piecemeal, if at all. Although there are a number of references to Nominales in theological contexts, they seem not to be to any distinctively nominalist theology, but rather to nominalist logical positions which were used in a theological discussion. There was, by contrast, far more variety in the manner of pursuing theology than was the case for logic, even leaving aside the monastic theologians such as William of St Thierry, Ailred of Rievaulx and Bernard of Clairvaux himself, who distanced themselves from the schools. At the Abbey of St Victor in Paris, where William of Champeaux had founded a theological tradition, which was carried on in the 1130s and 1140s especially by Hugh (of St Victor), Richard (of St Victor) wrote, sometime after about 1150, a long and carefullyworked De trinitate. Its aim is to show (rather as Anselm had tried in the Monologion) that there are strong rational grounds for holding, not merely that God exists, but that he is triune. Richard, however, writes to illuminate the faithful, not to convince non-believers. If he figures less in the history of philosophy than some of his contemporaries, it is not because his arguments lack sophistication but rather because his views are not easily detached from their theological context. In the theology of the schools—mainly the schools of Paris—there were two main approaches to method. Gilbert of Poitiers’ idea that each branch of knowledge, including theology, has its own fundamental rules, combined with the axiomatic method used in the third of Boethius’ Theological Treatises (and Gilbert’s own use of it in his commentary on that treatise), led to the attempt to produce an axiomatic theology. Peter of Vienna (or of Poitiers) places near to the beginning of his Summa (c. 1150) a set of rules which apply to created things (many of them read like the theses of the logical schools) but according to some of which we may also gain knowledge about God. In his Regulae caelestis iuris (c. 1170–80) Alan of Lille sets out no fewer than 134 special theological rules, which he expounds and attempts to justify in the work, sometimes deriving one from another.⁴⁶ Axiomatic theology turned out, however, to be a passing fashion. The most influential of all twelfth-century theological works turned out to be the Sentences written by Peter the Lombard in about 1155–7. Glosses began to be written on the Lombard’s Sentences in the later twelfth century and, from the thirteenth to the fifteenth centuries, commentary on what were simply called ‘the Sentences’ was the vehicle for much of the most important theological and philosophical work (see pp. 194– 5). The Lombard probably based his Sentences on the discussion of doctrinal difficulties which took place in his lectures on the Bible, but he drew together his material in a systematic way, to provide an orderly consideration, problem by problem, of the whole area of theological debate. The Lombard’s Sentences were valued, above all, for their orthodoxy—although many passages show a powerful logical mind, fully abreast of the subtleties of an Abelard or a Gilbert of Poitiers. Some of those who followed his methods were far more openly enthusiastic for logical analysis; few more so than his pupil, Peter of Poitiers (not to be confused with the Porretan Peter, also from Poitiers). Peter’s discussion of divine omnipotence ([7.23] 48–68) in his Sentences (c. 1176) provides a good example of this logically intricate approach to doctrinal problems, since Peter borrows some of the best ideas of theologians from the past two decades and also adds his own.⁴⁷ Like Abelard, he argues that ‘God is omnipotent’ does not mean that he can do all things, since he cannot walk or eat or sin, but rather (following Augustine) that God can do whatever he wills. Peter adds to this the requirement, mentioned by Peter the Lombard, that ‘nothing whatsoever can be done to God’ (which would seem to entail that he can not do whatever he does not will to do). Peter does not explain why, but this extra requirement would overcome the objection that Augustine’s definition is too weak since it makes omnipotent whoever limits his wishes to his capabilities. He goes on to consider the position (Abelard’s, but he is not named) that God can do only what he does, which is supposedly entailed by a variety of considerations of the form: God does only and all what is good (what is fitting, what his justice requires). Peter’s solution is to distinguish between ‘good’ predicated of men and of God. Men are good because what they do is good, but what God does is good because it is done by God. He can then— making a similar distinction to that used by Abelard between de dicto and de re modalities—distinguish two senses of ‘God can do only what it is good to be done by him’. They are a composite (de dicto) sense: ‘God can do only that-which-if-done-by-him-is-good’; and a divided (de re) sense: ‘God can do only that which is good: i.e. he cannot do that which is now bad.’ Only the composite sense yields a true statement, and the composite sense does not limit what God can do, but merely affirms that whatever he in fact does is good (because he does it). Peter then works through a number of more purely logical fallacies which seem to limit God’s power. Peter finishes the section with a long discussion of God’s power to alter natural necessities. God, he argues, can bring about what is impossible according to natural causality, for instance, that a man is an ass. But ‘the man is an ass’ is true only if it is interpreted as saying that, according to natural causes, he is man, but according to a higher cause, he is an ass. CONCLUSION: OLD SOURCES, NEW SOURCES AND THE ACHIEVEMENTS OF TWELFTH-CENTURY PHILOSOPHY The preceding sections will have given the impression that, in most important respects, twelfth-century thinkers used only a narrow range of ancient and late antique texts, most which had been available since the ninth century: the logica vetus, Plato’s Timaeus, Boethius’ On the Consolation of Philosophy and Theological Treatises, complemented by Latin philosophical and scientific texts by Cicero, Macrobius and Martianus Capella; the only exception seems to be Aristotle’s On Sophistical Refutations, the one work of the logica nova which was taken up with enthusiasm in this period. The twelfth century thus appears to present a stark contrast to the thirteenth, when philosophy in the Latin West was transformed by contact with the whole range of Aristotle’s writings, and with work by the great medieval Arab and Jewish thinkers, such as Avicenna, Averroes and Maimonides. In one way, this impression is misleading. A number of the translations which would be influential during the thirteenth century were made in the years from 1150, especially in Toledo (see pp. 226– 7). One of the most important of these translators, Dominicus Gundissalinus (d. after 1181), wrote a number of independent works which combine the influence of Avicenna and Latin authors such as Boethius.⁴⁸ There were writers such as Adelard of Bath (writing between c. 1110 and 1145) and Hermann of Carinthia (fl. 1138–43) who learned Arabic and exploited Arabic sources—though their interests were more scientific than philosophical.⁴⁹ The Liber de causis (Book about Causes), a translation of an Arabic adaptation of the late Neoplatonist Proclus’ Elements of Theology, was known to Alan of Lille. Some of Aristotle’s non-logical works were being used in Salerno late in the century; around 1200 or shortly afterwards, David of Dinant, who had travelled in Greece, translated and used passages from Aristotle’s scientific writings, and John Blund wrote a De anima (On the Soul) using Aristotle and, especially, Avicenna.⁵⁰ Despite these reservations, it would still be right to conclude that the main achievement of twelfth-century philosophy was not related to newly available ancient or Arabic material. Nor, indeed, despite many of these scholars’ great reverence for the ancients, should it be seen in terms of a deeper assimilation of the ancient texts previously known, or even of a new approach to them. The great writers of the first half of the century—Abelard, Gilbert of Poitiers (and some would wish to add William of Conches and Thierry of Chartres)—posed and tackled philosophical questions with an originality which makes the model of assimilation inappropriate. The second half of the century did not produce any philosophers of the same stature, but it saw two important developments, largely unrelated to ancient sources: the development of a systematic, argumentative method of theology, and the elaboration of sophisticated logical techniques for semantic analysis and the study of argument. It would be these, along with the effects of the new Aristotelian and Arabic material, that would provide the framework for the impressive philosophical developments in thirteenthand fourteenth-century Paris and Oxford. NOTES 1 For a fuller sketch of the historiography of twelfth-century philosophy, see Marenbon [7.66] 101–6. 2 See Chapter 1, pp. 14–15 for fuller discussion. 3 See Marenbon [7.42] 80–4; for a wider study of the themes in early medieval commentaries and glosses on Aristotle and Porphyry, see Marenbon [7.67]. 4 For the identification of Garlandus and material on in voce exegesis, see Iwakuma [7.52] 47–54; for this interpretation, see Marenbon [7.68] 108–16. 5 See Jolivet [7.57] where the material is collected and this interpretation advanced; cf. de Libera ([7.61] 142–5) and M.Tweedale, ‘Logic: to the time of Abelard’ in Dronke [7.49] 204–5. 6 John of Salisbury [7.13] III, 2, pp. 124–5; trans. in [7.34] 151–2. 7 The arguments for attributing the Glosae to Bernard, along with a full account of Bernard’s life and the testimony (mainly from John of Salisbury) to his teaching, are given by Dutton, in his edition of the Glosae [7.7] 21–45, 239–49. 8 The dating of the Dialectica is controversial: see Mews [7.70] 74–104 and Marenbon [7.68] 41–3. 9 On impersonal statements, see Jacobi [7.53]; on the copula, see de Rijk [7.76] (acute discussion and full bibliography on the question). 10 Christopher Martin (see especially [7.69]) has been the first modern scholar to explain Abelard’s theory of entailment. He also brings out the parallels between Abelard’s approach and that in modern ‘relevant’ logics. The following paragraph depends entirely on his work. On the theory of topics, see above, Chapter 1, pp. 14–15. 11 On the close connection between the principle observed by Boethius (see above, Chapter 1, p. 14) that it is not possible that p implies q and p implies not-q, and the principle that it is not possible that p implies not-p (arguably the two principles are equivalent), see Martin [7.69] 381. 12 See esp. Abelard [7.19] 153:33–160:36 and [7.20] 365:13–370:3 and cf. de Libera, ‘Abélard et le dictisme’ in [7.44] 59–92; Martin, ‘The logic of the Nominales’, in Courtenay [7.48] 110–26; de Rijk, ‘La signification de la proposition (dictum propositionis) chez Abélard’, in [7.74] 547–55; Marenbon [7.68] 222–9. 13 See Abelard [7.19] 199–210 and Abelard [7.14] (entirely on modality). On this subject, see Knuuttila [7.59] 82–96 and Marenbon [7.68] 221–5. 14 He presents these at Abelard [7.20] 10:17–16:18, 513:15–522:9; cf. Tweedale [7.81] 89–132. 15 Abelard develops this ontology in the Dialectica and the Logica: for a full discussion and references, see Marenbon [7.68] 117–37. 16 See Abelard [7.20] 129:33–6 and (on differentiae: see below) 84:14–21, 92:22– 9; cf. Marenbon [7.68] 119–22. 17 See esp. Abelard [7.17] 342:2434–344:2532; the whole question is discussed and other texts are given in Marenbon [7.68] 138–61. 18 There is a large literature on Abelard’s theory of universals. Among the important modern discussions are Tweedale [7.81], de Rijk, ‘The semantical impact of Abailard’s solution of the problem of universals’, in Thomas [7.80] 139–50 and Jolivet [7.57]. The following paragraphs summarize the rather different view proposed in Marenbon [7.68] 174–201. 19 On imposition, see Abelard [7.19] 595:11–31. 20 The fullest treatment of various types of difference is in the Theologia Christiana, Abelard [7.17], 247:1677–255:1936; cf. Marenbon [7.68] 150–5. 21 In Marenbon [7.68] 213–331, the three levels of Abelard’s ethical theory are examined. The following section draws especially on chapters 11 and 12 (pp. 251–81). 22 Valuable discussions are given by Blomme [7.46] and M.de Gandillac, ‘Intention et loi dans l’éthique d’Abélard’ in [7.74] 585–608. 23 Abelard [7.21] 10:28–14:25. Abelard puts forward his analysis of the ethical act in a number of works, including the commentary on Romans and the Sententie. References here are made, wherever possible, to Scito teipsum, both because it contains Abelard’s latest formulation of his ideas and it is easily accessible in good translation. 24 See esp. Marenbon [7.64]; Abelard develops his ideas about natural law especially in Book II of the Theologia Christiana and in the Collationes. 25 On Gilbert’s life, see Nielsen [7.73] 25–39. 26 For Gilbert’s distinction between the different disciplines, see esp. [7.10] 79:43– 88:69; on proportional transumption, see e.g. 143:42–7, 170:87–93. On the whole question of Gilbert’s method, see Marenbon, ‘Gilbert of Poitiers’, in Dronke [7.49] 330–6. 27 In my piece on Gilbert (cited in the previous note—esp. 329–30, 351–2) I lay strong (in retrospect too strong) emphasis on the extent to which Gilbert’s doctrinal aims shaped his arguments, without bringing out how Gilbert was also contributing to the philosophical debate of his times. De Rijk’s criticism [7.75] 34–5, though mistaken in attributing to me a hostile intention towards Gilbert, is in this way very just. 28 Valuable analyses of Gilbert’s theory of quo est and quod est are provided in de Rijk [7.75] and Gracia [7.50] 155–77. 29 Many scholars consider (on the basis, especially, of [7.10] 195:100–7) that, for Gilbert, quo ests are images of disembodied Platonic Ideas. In ‘Gilbert of Poitiers’, in Dronke [7.49] 349–51, I argue that this is a misinterpretation: Gilbert introduces disembodied forms merely in his account of the creation of the elements, not as the archetypes of quo ests. For a different view again, see de Libera [7.58] 170–5. 30 See Knuuttila [7.59], esp. 211–17; Knuuttila discusses the passage in question at pp. 216–17. 31 On this view, ‘Socrates’ would have to be a rigid designator, picking out the individual Socrates in every possible world where he exists, however different he is from one world to another. But Gilbert’s statement about the whole form comprising every quo est that belongs to the thing by nature seems to leave the room for variation disturbingly wide. There seems, for instance, no guarantee that Socrates must have the same parents from one possible world to another, and it becomes unclear how at all we should identify Socrates. 32 See esp. Elford, ‘William of Conches’, in Dronke [7.49] 308–27; more generally on William, see Gregory [7.51]. 33 For reconstructions of Thierry’s underlying ideas, see Dronke, ‘Thierry of Chartres’, in Dronke [7.49] 358–85 and Gersh, ‘Platonism-Neoplatonism- Aristotelianism: a twelfth-century metaphysical system and its sources’, in Benson and Constable [7.45] 512–34. 34 [7.27] 81:1–7, 82:25–33; this passage is analysed and its sources discussed by Gersh in the article cited in the previous note, pp. 517–24. 35 ‘Aggreditur propositum…’ (printed as Thierry’s Glosa), [7.27] 275:11–276:39 and ‘Intentio auctoris…’ (printed as Thierry’s Lectiones), [7.27] 168:76–170:33, 176:45–50, which in this discussion puts forward, somewhat less clearly, almost exactly the same view as ‘Aggreditur propositum…’. 36 See William of Lucca, commentary on ps–Dionysius, ed. F.Gastaldelli, Florence, 1984, xcii–xciii for dating, and xxi–xxvii for the authorship of the Summa dialetice artis. For Platonic Ideas, see especially pp. 100–2. For further discussion of this and a related, unpublished text, see Marenbon [7.66] 114–17. 37 The various articles in Courtenay [7.48] provide the best guide to what is known about these schools. See especially Iwakuma and Ebbesen, ‘Logico-theological schools from the second half of the twelfth century: a list of sources’ (pp. 173– 210), which I follow closely here. A very intelligent discussion of the material is given by de Libera [7.61] 132–7. A wealth of material is collected in de Rijk [7.77] II, 2. For a survey, see Jacobi, ‘Logic: the later twelfth century’, in Dronke [7.49] 227–51. 38 This has been disputed, but two contributions to Courtenay [7.48]: Normore (‘Abelard and the school of the Nominales’, pp. 80–96) and Iwakuma (‘Twelfthcentury Nominales: the posthumous school of Peter Abelard’, pp. 97–109), put the identification beyond reasonable doubt. 39 See de Rijk [7.78] and Marenbon, ‘Vocalism, nominalism and commentaries on the Categories from the earlier twelfth century’, in Courtenay [7.48] 51–61, at 54–5. 40 Both the Introductiones montane maiores and the Ars meliduna remain unpublished; de Rijk discusses and prints extracts from them in [7.78] 12–22 and [7.77] II, 1, 264–390. For the Summa dialetice artis, see [7.32]. 41 For the Melidunenses, the so-called Secta Meliduna (see de Rijk [7.77] II, 1, 282– 6, where the list of these is printed); for the Nominales, the so-called Positio ‘nominalium’, ed. Ebbesen [7.4] 430–2. 42 These observations are taken from Martin [7.69] 394–400, where much fuller details are given. 43 The most important parts of the discussion are printed by de Rijk [7.77] II, 1, 306–9; I am grateful to Dr Yukio Iwakuma for supplying me with a transcript of further material from the Ars. De Libera ([7.61] 158–67) discusses the treatment of universals in the Ars at length. He suggests that the theory had a further refinement, in that universals are complex intelligible structures, which are expressed not by common names but by complex expressions. 44 The reception of On Sophistical Refutations is examined in detail in de Rijk [7.77] I. 45 See below, Chapter 17, where most of these areas are discussed. 46 For a brief guide to the extensive bibliography on Porretan theologians, see my ‘A note on the Porretani’, in Dronke [7.49] 353–7. There is no space here to do justice to the varied work of Alan of Lille (c. 1120–1203), which includes philosophical allegories, sermons and two more straightforward theological textbooks; the best guide is in the introduction to Alan of Lille [7.3]. 47 There is a fine analysis of discussions of this subject (and of Peter of Poitiers) in Boh [7.47] ; I follow Boh in some of my discussion in the next paragraph. 48 See Jolivet, ‘The Arabic inheritance’, in Dronke [7.49] 134–45 for a detailed discussion and full bibliography. 49 See Burnett, ‘Hermann of Carinthia’, in Dronke [7.49] 386–404. 50 On David of Dinant, see Maccagnolo, ‘David of Dinant and the beginnings of Aristotelianism in Paris’, in Dronke [7.49] 429–42; on John Blund, see Jolivet, ‘The Arabic inheritance’, in Dronke [7.49] 146–7. BIBLIOGRAPHY Original Language Editions 7.1 Adam of Balsham Ars disserendi, in L.Minio-Paluello (ed.) Twelfth-century Logic, Texts and Studies I, Rome, 1956. 7.2 Alan of Lille Regulae caelestis iuris, ed. N.Häring, Archives d’histoire doctrinale et littéraire du moyen âge 48 (1981): 97–226. 7.3 Alan of Lille Textes inédits, ed. M.-T. d’Alverny (Etudes de philosophie médiévale 52), Paris, 1965. 7.4 Anonymous ‘Two nominalist texts’, ed. S.Ebbesen, CIMAGL 61 (1991): 429–40. 7.5 Anonymous Compendium Logicae Porretanum, ed. S.Ebbesen, K.Fredborg, L. Nielsen, CIMAGL (1983): 46. 7.6 Anonymous, twelfth-century logical works on sophisms and on properties of terms, in de Rijk [7.77], vol. I and vol. II, 2 respectively. 7.7 Bernard of Chartres (?) Glose super Platonem, ed. P.E.Dutton (PIMSST 107), Toronto, 1991. 7.8 David of Dinant Quaternuli (fragments), ed. M.Kurzialek (Studi mediewistyczne 3), Warsaw, 1963. 7.9 Garlandus (of Besançon) Dialectica, ed. L.M.de Rijk, Assen, 1959. 7.10 Gilbert of Poitiers Commentaries on Boethius, ed. N.Häring (PIMSST 13), Toronto, 1966. 7.11 Hermann of Carinthia De essentiis, ed. C.Burnett, Leiden and Cologne, 1982. 7.12 John Blund Tractatus de anima, ed. D.A.Callus and R.W.Hunt (Auctores Britanni Medii Aevi 2), London, 1970. 7.13 John of Salisbury Metalogicon, ed. J.B.Hall (CC c.m. 98), Turnhout, 1991. 7.14 Peter Abelard, authentic ending of De interpretation commentary, in L. Minio- Paluello (ed.) Twelfth-century Logic II: Abaelardiana inedita, Rome, 1958. 7.15 ——Collationes, ed. R.Thomas, Stuttgart and Bad Cannstatt, 1970. 7.16, 7.17, 7.18 published in Petri Abaelardi opera theologica (CC c.m. 11–13), Turnhout, 1969–1987: 7.16 Peter Abelard Commentary on Romans, ed. E.Buytaert (11). 7.17 ——Theologia Christiana, ed. E.Buytaert (12). 7.18 ——Theologia summi boni, Theologia scholarium, ed. E.Buytaert and C.Mews (13). 7.19 ——Dialectica, ed. L.M.de Rijk, 2nd edn, Assen, 1970. 7.20 ——Logica and Glossulae in B.Geyer (ed.) Peter Abaelards philosophische Schriften (BGPTMA 21), Münster, 1919–31. 7.21 ——Scito teipsum (Ethics), ed. D.Luscombe, Oxford, 1971. 7.22 ——Sententie, ed. S.Buzzetti, Florence, 1983. 7.23 Peter of Poitiers Sentences, I and II (only these two vols published), ed. P.S. Moore and M.Dulong, Notre Dame, Ind., 1943, 1950. 7.24 Peter of Vienna (Poitiers) Summa, ed. N.Häring, as Die Zwettler Summe (BGPTMA n.f. 15), Münster, 1971. 7.25 Peter the Lombard Sentences, 2 vols (Spicilegium Bonaventurianum), Grottaferrata, 1971, 1981. 7.26 Richard of St Victor De trinitate, ed. G.Salet (Sources Chrétiennes 63), Paris, 1959. 7.27 Thierry of Chartres and others, Commentaries on Boethius, in N.M.Häring (ed.) Commentaries on Boethius by Thierry of Chartres and his School (PIMSST 20), Toronto, 1971. 7.28 William of Champeaux Sententiae, in O.Lottin, Psychologie et morale aux XIIe et XIIIe siècles, V, Gembloux, 1959, pp. 189–227. 7.29 William of Conches, Commentary on Timaeus (Glosae super Platonem), ed. E. Jeauneau, Paris, 1965. 7.30 ——Dragmaticon, ed. W.Gratarolus (as Dialogus de substantiis physicis), Strasbourg, 1567; repr. Frankfurt, 1967. 7.31 ——Philosophia mundi, ed. G.Maurach, Pretoria, 1980. 7.32 William of Lucca Summa dialetice artis, ed. L.Pozzi (Testi e saggi 7), Padua, 1975. Translations 7.33 Anonymous Abbreviatio montana, in N.Kretzmann and E.Stump (eds) Cambridge Translations of Medieval Philosophical Texts: Logic and the Philosophy of Language, Cambridge, 1988, pp. 39–78. 7.34 John of Salisbury Metalogicon, trans. D.D.McGarry, Gloucester, Mass., 1971. 7.35 Peter Abelard, discussion of universals from Logica in P.Spade, Five Texts on the Mediaeval Problem of Universals, Indianapolis, Ind. and Cambridge, 1994, pp. 26–56. 7.36 ——Theologia Christiana (extracts), trans. J.R.McCallum, Oxford, 1948. 7.37 ——Collationes, trans. J.Payer (as Dialogue between a Jew, a Christian and a Philosopher) (Mediaeval Sources in Translation 20), Toronto, 1979. 7.38 ——Scito teipsum, in the edition by Luscombe [7.21]. 7.39 ——selections (in French) in J.Jolivet, Abélard ou la philosophie dans le langage (Vestigia 14), Freiburg, Switzerland, 1994. 7.40 ——Letters and Historia calamitatum, trans. B.Radice, Harmondsworth, 1974. Bibliographies, Catalogues and Biographies 7.41 Barrow, J., Burnett, C. and Luscombe, D. ‘A checklist of the manuscripts containing the writings of Peter Abelard and Héloïse and other works closely associated with Abelard and his school’, Revue d’histoire des textes 14–15 (1984–5): 183–302. 7.42 Marenbon, J. ‘Medieval Latin commentaries and glosses on Aristotelian logical texts before c. 1150 AD’, in C.Burnett (ed.) Glosses and Commentaries on Aristotelian Logical Texts (Warburg Institute Surveys and Texts 23), London, 1993, pp. 77–127. 7.43 Mews, C. and Jolivet, J. ‘Peter Abelard and his influence’, in Contemporary Philosophy: a New Survey, 6/1, Dordrecht, 1991, pp. 105–40. Rich bibliographical information will be found in Dronke [7.49], especially in the bio-bibliographies, pp. 443–57. For biographies, see the bio-bibliographies in Dronke [7.49] 443–557; for Abelard, see Marenbon [7.68] 7–35 and Mews [7.71]. Studies 7.44 Abélard: Le ‘Dialogus’, la philosophie de la logique (Cahiers de la revue de théologie et de philosophie 6), Geneva, Lausanne and Neuchâtel, 1981. 7.45 Benson, R.L. and G.Constable (eds) Renaissance and Renewal in the Twelfth Century, Oxford, 1980. 7.46 Blomme, R. La Doctrine du péché dans les écoles théologiques de la première moitié du XIIe siècle (Universitas catholica Lovaniensis. Dissertationes ad gradum magistri…consequendum conscriptae, series III, 6), Louvain and Gembloux, 1958. 7.47 Boh, I. ‘Divine omnipotence in the early Sentences’, in T.Rudavsky (ed.) Divine Omniscience and Omnipotence in Medieval Philosophy (Synthese historical library 25), Dordrecht, 1985, pp. 185–211. 7.48 Courtenay, W. (ed.) a collection of articles on twelfth-century nominalism, Vivarium 30 (1992). 7.49 Dronke, P. (ed.) A History of Twelfth-Century Western Philosophy, Cambridge, 1988. 7.50 Gracia, J. Introduction to the Problem of Individuation in Early Medieval Philosophy, Munich and Vienna, 1984. 7.51 Gregory, T. Anima mundi: La filosofia di Guglielmo di Conches e la Scuola di Chartres, Florence, 1955. 7.52 Iwakuma, Y. ‘“Vocales”, or early nominalists’, Traditio 47 (1992): 37–111. 7.53 Jacobi, K. ‘Diskussionen über unpersönlichen Aussagen in Peter Abaelards Kommentar zu Peri Hermeneias’, in E.P.Bos (ed.) Mediaeval Semantics and Metaphysics (Artistarium supplements 2), Nijmegen, 1985, pp. 1–63. 7.54 Jeauneau, E. Lectio philosophorum, Amsterdam, 1973. 7.55 Jolivet, J. Arts du langage et théologie chez Abélard, 2nd edn (Etudes de philosophie médiévale 57), Paris, 1982. 7.56 ——‘Non-réalisme et platonisme chez Abélard: Essai d’interprétation’, in J. Jolivet (ed.) Abélard en son temps, Paris 1981; repr. in Jolivet, Aspects de la pensée médiévale: Abélard: doctrine du langage, Paris, 1987. 7.57 ——‘Trois variations médiévales sur l’universel et l’individu: Roscelin, Abélard, Gilbert de la Porrée’, Revue de Métaphysique et de Morale 1 (1992): 111–55. 7.58 Jolivet, J. and Libera, A. de Gilbert de Poitiers et ses contemporains: aux origines de la ‘logica modernorum’ (History of Logic 5), Naples, 1987. 7.59 Knuuttila, S. Modalities in Medieval Philosophy, London and New York, 1993. 7.60 Landgraf, A. Introduction à l’histoire de la littérature théologique de la scolastique naissante, French edn prepared by A.-M.Landry (Publications de l’institut d’études médiévales, Montreal 22), Montreal and Paris, 1973. 7.61 Libera, A. de La Querelle des universaux: De Platon à la fin du Moyen Âge, Paris, 1996. 7.62 Lottin, O. Psychologie et morale au XIIe et XIIIe siècles, V, Gembloux, 1959. 7.63 Luscombe, D. ‘From Paris to the Paraclete: the correspondence of Abelard and Heloise’, Proceedings of the British Academy 74 (1988): 247–83. 7.64 Marenbon, J. ‘Abelard and natural law’, in A.Zimmermann (ed.) Miscellanea Mediaevalia 21(2): Mensch und Natur im Mittelalter, Berlin and New York, 1992, pp. 609–21. 7.65 ——‘Abelard’s ethical theory: two definitions from the Collationes’, in H.J. Westra (ed.) From Athens to Chartres, Leiden, New York and Cologne, 1992, pp. 301–14. 7.66 ——‘Platonismus im 12. Jahrhundert: alte und neue Zugangsweisen’, in T. Kobusch and B.Mojsisch (eds) Platon in der abendländischen Geistesgeschichte: neue Forschungen zum Platonsimus, Darmstadt, 1997, pp. 101–19. 7.67 ——‘Glosses and commentaries on the Categories and De interpretations before Abelard’, in J.Fried (ed.) Dialektik und Rhetorik im früheren und hohen Mittelalter, Munich, 1997, pp. 21–49. 7.68 ——The Philosophy of Peter Abelard, Cambridge, 1997. 7.69 Martin, C.J. ‘Embarrassing arguments and surprising conclusions in the development of theories of the conditional in the twelfth century’, in J.Jolivet and A.de Libera (eds) Gilbert de Poitiers et ses contemporains (History of Logic 5), Naples, 1987, pp. 377–400. 7.70 Mews, C. ‘On dating the works of Peter Abelard’, Archives de l’histoire doctrinale et littéraire du moyen âge 52 (1985): 73–134. 7.71 ——Peter Abelard (Authors of the Middle Ages II, 5: Historical and religious writers of the Latin West), Aldershot, 1995. 7.72 ——‘Nominalism and theology before Abelard: new light on Roscelin of Compiègne’, Vivarium 30 (1992): 4–33. 7.73 Nielsen, L.O. Theology and Philosophy in the Twelfth Century: A Study of Gilbert Porreta’s Thinking and the Theological Expositions of the Doctrine of the Incarnation during the Period 1130–1180 (Acta theologica danica 15), Leiden, 1982. 7.74 Pierre Abélard, Pierre le Vénérable (Colloques internationaux du Centre National de la Recherche Scientifique 546), Paris, 1975. 7.75 Rijk, L.M. de ‘Semantics and metaphysics in Gilbert of Poitiers: A study in twelfth-century metaphysics’, Vivarium 26 (1988): 73–112; 27 (1989): 1–35. 7.76 ——‘Peter Abelard’s semantics and his doctrine of being’, Vivarium 24 (1986): 85–128. 7.77 ——Logica modernorum I and II, Assen, 1962, 1967. 7.78 ——‘Some new evidence on twelfth-century logic: Alberic and the school of Mont Ste Geneviève’, Vivarium 4 (1966): 1–57. 7.79 Southern, R.W. ‘Humanism and the school of Chartres’, in Medieval Humanism and Other Studies, Oxford, 1970, pp. 61–85. 7.80 Thomas, R. (ed.) Petrus Abaelardus (1079–1142): Person, Werk und Wirkung (Trier theologische Studien 38), Trier, 1980. 7.81 Tweedale, M. Abailard on Universals, Amsterdam, New York and Oxford, 1976.