The hypothesis of an Oxford—Paris split

Terminist logic grew to maturity in the period 1175—1250, a period that was also crucially important in the development of the universities of Paris and of Oxford. Scholars have recently focused their attention on divergences in the early development of the logical and semantic theories that constitute terminist logic, divergences suggesting that one cluster of doctrines is to be associated with Oxford, another with Paris. In view of the very marked differences between British and continental logic in the early fourteenth century, it seems important to investigate whether such differences can be traced backwards into the thirteenth and late twelfth centuries. As I hope to show, such divergences do in fact exist even if they are not so great as to make the traditions of Oxford and of Paris entirely independent.

Although by the turn of the thirteenth century terminist logic was acknowledged by all logicians as a common frame of reference, various interpretations of important issues were still being put forward. If divergences between the traditions of Oxford and Paris are to be established, the evidence is likely to be found in the discussions concerning the various properties of terms, such as supposition, appellation, ampliation, and restriction.

The school of the Parvipontani

Twelfth-century logicians seem to have basically agreed in claiming that an appellative (or common) name may vary its reference (appellatio) according to changes in the tense of the main verb of the proposition; they agreed further in describing this variation as either ‘restriction’ or ‘ampliation’ of the reference or appellation, and in providing rules associated with the three main tenses of the verb: present, past, and future.

The notion of consequentia

Etymologically, ‘consequentia’ suggests a following along. In medieval philosophical literature it was apparently quite proper to say that one concept follows another – e.g., that animal follows man – but more generally consequence was thought of as involving entire propositions.

There are, of course, many different relationships in which propositions can stand to each other. For instance, in a conditional proposition of the form ‘if p then q’ the proposition taking the place of ‘p’ is the antecedent of the conditional proposition, and the proposition taking the place of ‘q’ is the consequent. The relationship between that antecedent and the consequent in a true conditional is called implication by modern logicians. Again, two propositions may be related to each other in such a way that the first cannot be true unless the second is true also; and the relationship between the two propositions in that case is called entailment. Again, two propositions may constitute an argument. In the argument-form ‘p; therefore q’ the proposition taking the place of ‘p’ is the premiss and the proposition taking the place of ‘q’ is the conclusion. To employ an argument is to derive or infer the conclusion from the premiss (or premisses), and so the relationship between those two propositions is called derivation or inference.

Implication, entailment, and inference are all distinct from one another. For instance, a conditional proposition, like other propositions, is accepted or rejected by being classified as true or as false; an argument is neither true nor false but is accepted or rejected as valid or as invalid.

Spain as the bastion of late scholasticism

After the attacks of humanists, Ramists, reformers, and plain haters of philosophy over much of two centuries, it is amazing that scholasticism survived at all. Not only did it survive, it experienced a notable revival throughout much of western Europe towards the end of the sixteenth century and beginning of the seventeenth. Humanists and reformers were by no means unanimous in opposing the medieval scholastics. More important, the Iberian peninsula was comparatively unaffected by the intellectual and religious ferment of most of the rest of Europe. The schools of Spain and Portugal had a more or less continuous tradition of scholastic philosophy, and the leading figures in the general revival of scholastic thought round the end of the sixteenth century tend to be Spaniards like Bañiez, Vásquez, and Suárez. In northern Europe the scholastic revival looks more like a self-conscious and deliberate Aristotelian reaction to Ramists, humanists, and the like, but the northerners of whatever religious allegiance were happy enough to take guidance and inspiration from Spain.

New trends in late scholasticism

Although in obvious ways continuous with the main medieval tradition, late scholasticism, whether in its Iberian form or in its northern revival, shows certain very distinctive characteristics of its own which may be seen as marking a transition to some of the most prominent themes of early modern philosophy.

While these philosophers were nearly unanimous in rejecting medieval nominalism (indeed, in the north this was another of the things they were reacting against), Scotus, Ockham, and the later nominalistic tradition had a very powerful influence on them.

The Cambridge History of Later Medieval Philosophy finds its natural place after The Cambridge History of Later Greek and Early Medieval Philosophy in the sequence that begins with Guthrie's History of Greek Philosophy. The sequence is not altogether smooth, however. At the beginning of The Cambridge History of Later Greek and Early Medieval Philosophy its editor, A. H. Armstrong, observes that although the volume ‘was originally planned in connexion with W. K. C. Guthrie's History of Greek Philosophy, … [it] has developed on rather different lines, and is not exactly a continuation of that work’ (p. xii). Similarly, although The Cambridge History of Later Medieval Philosophy was conceived of as the sequel to The Cambridge History of Later Greek and Early Medieval Philosophy, the relationship between the two is not so simple as their titles suggest; in fact, the fit between this volume and the Armstrong volume is less exact than that between the Armstrong volume and Professor Guthrie's plan. Many reviewers noted that the Armstrong volume seems misleadingly titled since it is really a study of only the Platonist tradition in later Greek and early medieval philosophy; but in concentrating in that way it does indeed complement Professor Guthrie's plan, which includes the Stoics and Epicureans as well as Aristotle while leaving out the Neoplatonists. On the other hand, The Cambridge History of Later Medieval Philosophy cannot be put forward as the full realisation of Professor Armstrong's expressed hope ‘that the philosophy of the thirteenth century and the later Middle Ages in the West, with later Jewish, Moslem, and Byzantine developments, will some day be dealt with in another Cambridge volume’ (ibid.).

Grammar in the early Middle Ages

Medieval speculative grammar grew out of the schoolmens' work with ancient Latin grammar as it had been transmitted in the canonical works of Donatus and Priscian. The efforts of early medieval glossators were directed towards explaining the authoritative texts, towards systematising the descriptional apparatus used by the authors, and towards harmonising the apparent or real contradictions which arose in a comparison of the grammatical and logical traditions. The results of their combined efforts were summarised in the famous Summa super Priscianum compiled about 1140 by Peter Helias. The grammarians' discussions, which had been influenced by the logicians, in their turn influenced and refined logical doctrine and played an important role in the emergence of the specifically medieval logical doctrines known collectively as terminist logic. The grammatical discussions about the meaning of substantive words, for instance, were crucial to the development of the theory of supposition. The twelfth-century grammarians emphasised the importance of explaining linguistic features causally, instead of just describing them as Priscian had done, and in this way attained a high degree of linguistic sophistication. But it would perhaps be too much to say that their efforts already inaugurated a new paradigm of linguistic description.

The general nature of ‘modistic’ grammar

Around 1270, however, a new theoretical framework was established. The phases of the development which brought this about are not yet known in detail, but the first representatives of the new doctrine seem to be Boethius of Dacia and Martin of Dacia.

Medieval Platonism

In the later Middle Ages philosophical questions on the ordinary sources of human knowledge attracted continuous though uneven attention. The fundamental problem for the discussions involved taking account of two commonly recognised extremes.

On the one hand, Augustine had in summary fashion heralded a unified philosophy. For him the best in all preceding Greek thought had been assimilated into the Platonism current in his epoch, a type now conveniently designated by the nineteenth-century term ‘Neoplatonism’. Within its own competence the perfected philosophy, as Augustine saw it, paralleled revealed biblical truth. His view set the framework for Christian intellectual tradition among the Latins for the eight ensuing centuries.

Medieval Aristotelianism

On the other hand, Boethius, from whose translations and commentaries medieval students learned their logic and received their general introduction to philosophy, had handed down an acquaintance with certain facets of Aristotle that resisted absorption into the Neoplatonic stream. By the mid twelfth century Aristotle had attained the status of the Philosopher par excellence. His thought, as enhanced by Islamic writers translated during the latter half of that century, deepened medieval inquiry into subjects significant for problems of cognition. During the thirteenth century Aristotle's major works became available in direct translation and were read with the commentaries of Averroes. They rapidly imposed their philosophical techniques upon the intellectual training in the newly established universities and guided it for the rest of the medieval period.

A grammatical distinction between categoremata and syncategoremata

The paradigm of the categorical proposition with which medieval logicians were primarily concerned is a sentence of two words that serve as the subject term and the predicate term – e.g., ‘Socrates currit.’ Any word that can be used alone as a subject term or as a predicate term is classifiable as a categorematic word; all other words are classifiable as syncategorematic words, those that can occur in a proposition, whether categorical or hypothetical, only along with at least one properly matched pair of categorematic words – e.g., ‘Solus Socrates currit,’ ‘Socrates currit contingenter’, ‘Socrates non currit’, “Si Socrates currit, Socrates movetur.’ Drawing the distinction between categoremata and syncategoremata along this line, which seems to have been the original line of distinction, produces mutually exclusive and jointly exhaustive classes that coincide almost perfectly with certain groupings of the parts of speech (parts orationis) recognized by medieval grammarians: the names (both substantival and adjectivel), the personal and demonstrative pronouns and the verbs (excluding auxiliary verbs); the syncategoremate are all the others – e.g., the conjunction adverbs, and prepositions.

The logicians' notion of syncategoremata

The notice of synacategoremata that became important in medieval logic was, however, both narrower and broader than that comparatively orderly classification in terms of the parts of speech. Although more than fifty different words were considered in one or another medieval logicians' treatment of syncategoremate, by no means all non categorematic words in even the relatively small classes, such as conjunctions, were of enough interest to the logicians to be treated expressly among their syncategoremata.

Two sorts of questions for moderate realism

The vigorous early-fourteenth-century debate about universals was based on a rejection of Platonism, the theory that universal natures really exist independently of the particulars whose natures they are and independently of every mind. Fourteenth-century ‘moderate’ realists agreed that natures must be somehow common to particulars in reality, but Aristotle had convinced them that no one in his right mind could hold that the nature of a thing exists separated from it as Platonic forms were supposed to do. They insisted instead that the natures really exist in the things whose natures they are, as metaphysical constituents of them. But this contention had its own problems. Since there can be more than one particular in a given genus or species, natures cannot be the only metaphysical constituents of particulars; there must also be individuating principles that serve to distinguish one particular from another.

But what are these individuating principles? William of Champeaux's position that accidental properties individuate was denied by virtually everyone on the Aristotelian ground that substance is naturally prior to accidents but particular substances are not naturally prior to what individuates them. Thomas Aquinas held that prime matter, the ultimate property-bearer in composite substances, combines with quantitative dimensions to individuate. But Duns Scotus found this tantamount to conceding that accidents individuate after all. Besides, he argued, neither matter, quantitative dimensions, nor their combination was distinct and determinate in itself. Taking it as axiomatic that only what is distinct and determinate in itself can individuate, Scotus concluded that neither matter by itself, existence, nor any combination of accidents can do the job (Ordinatio II, d. 3, qq. 5, 3, and 4, respectively).

Abelard's conception of logic

Of all the scholastic logicians writing while the old logic (logica vetus) was still virtually the whole of the logical curriculum in the schools, Abelard is generally conceded to have been the most profound and original. He himself was keenly aware of the subtlety required of the logician and in one place says it depends on a divinely bestowed talent, rather than anything that can be developed by mere practice. Abelard treats dialectic (= logic) as an ars sermocinalis, i.e., like grammar a linguistic science. Its peculiar subject matter is arguments as expressed in language, whose validity it tries to judge in a scientific way. This linguistically oriented conception of the subject means that dialectic will overlap to some extent with grammar. In the first section below I shall selectively explore this overlap; in the second section I shall consider some of Abelard's views on more purely logical topics.

For Abelard logic also had a close relation to physica, i.e., the sciences of nature, since in explaining the ‘uses of words’ the logician must investigate in a general way the ‘properties of things’ which the mind uses words to signify. This relationship leads to a concern with the psychology of signification, to be explored in the third section below, and with ontology, the topic of the fourth section. This discussion is necessarily very selective and must omit consideration of many of Abelard's philosophical insights on relevant topics.

Intellectual and religious reaction to the French Revolution

The upheaval of the French Revolution destroyed many academic and ecclesiastical organisations belonging to the old order, but a reaction was not slow in coming. As Sainte-Beuve noticed in 1854, the number of Le Moniteur for Easter Sunday 1802, which published news of the Peace of Amiens and of the Concordat between Napoleon and the Pope, also published a review of a recent book: Le Génie du Christianisme, by Chateaubriand (1768–1848). An appeal to tradition against the excesses that had followed from rationalism found romantic expression in Chateaubriand's book, but the concept of tradition was to receive a more philosophical cast from others – De Maistre (1763–1852), De Bonald (1754–1840), and Lamennais (1782–1854). All three had idiosyncratic views on the role of language, and the opposition they provoked influenced the form taken by the revival of scholasticism. Writing in 1809, De Maistre claimed that the content of language depends upon the life and customs of those who use it; it eludes arbitrary enactments; it was not invented by men, nor can its diversity be attributed to human means. For De Bonald, the disagreements of philosophers oblige us to seek for moral science what physical sciences have already: a fixed point, a criterion of truth, something that will be public, readily accessible, and evident. He believed this to be ‘the primordial and indispensable gift of language, bestowed upon the human race’. Reason and experience in individuals need the setting and tradition of society, and of the language it hands on, to reach truths that go beyond particular facts.

The old and the new responses

The obligations-literature appears to have entered a new phase with the Oxford Calculators, centred at Merton College in the 1320s and 1330s. Although the Mertonian Thomas Bradwardine seems to have contributed little to the development of obligations, his contemporary Richard Kilvington was more innovative in this regard, in ways described in Part A of this Chapter.

We are in a somewhat better position to assess the contribution of Roger Swineshead to obligations. Swineshead certainly appears to have been part of the intellectual circle with which Kilvington and Bradwardine are associated, and he may well have studied with them. Probably sometime after 1330 and before 1335, Swineshead wrote his pair of treatises on obligations and insolubles. There is reason to speculate that some of the most characteristic features of Swineshead's Insolubilia grew out of reflection on Kilvington's Sophismata. It is possible that this is true of Swineshead's Obligationes as well, but that remains to be established.

Swineshead's Obligationes is markedly different from earlier treatises in the genre. So true is this that Robert Fland, writing some time between 1335 and 1370, distinguishes two separate traditions in the obligations-literature. One of these traditions he calls the ‘old response’ (antiqua responsio); it conforms to the views of Burley, to those of the treatise attributed to William of Sherwood, and to those found in most if not all of the other early treatises. The second tradition Fland calls the ‘new response’ (nova responsio); it appears to have originated with Swineshead.

Clerical and Aristotelian science

The question of the interpretation of Aristotle in the Middle Ages must be dealt with within the context of the medieval conception of science. Medieval learning was characterised by an attitude which was dominant – though in varying degrees and varying circumstances – from the time of Alcuin to that of Bellarmine. For the Middle Ages it was not the individual who taught, but the Church, through the clergy. Clerical science was accordingly the corporate transmission of traditional wisdom. The cleric was a ‘master’ chosen by God to teach his people the way of salvation, as Rhabanus Maurus put it at the beginning of his De institutione clericorum. His authority as a teacher was guaranteed by a divine call within the ecclesiastical hierarchy; the authority of his teaching was guaranteed by Scripture and the Church Fathers. But his authority extended even beyond the sacred sciences, in a way which reveals the relationships between this clerical attitude towards knowledge and the structure of medieval society. The relationship between the clergy and the laity is clearly symbolised in Alcuin's Dialectica, a dialogue in which Alcuin as magister instructs Charlemagne himself as discipulus, in one of the profane sciences of the trivium and quadrivium.

Within this conception of the scientific enterprise a standard method of interpretation was developed based on the presumed concordance of the fundamental authorities, and schools evolved whose function was the training of masters who should transmit traditional learning to God's people.

The first translations

During the Middle Ages the Nicomachean Ethics received less attention than Aristotle's writings on natural philosophy and metaphysics, and still less than his logical writings. Although the Ethics was never condemned in any form, it was apparently not until the second half of the fourteenth century that it was adopted as a regular textbook in the Arts faculties; and it was only in the fifteenth century, as the number of commentaries shows, that it began to be studied really intensively.

We are not so well informed about this period as we are about the beginnings of philosophical ethics in the Middle Ages. According to the latest historical research, the first translation of the Nicomachean Ethics appeared in the twelfth century; but it covered only the second and third books (ethica vetus.) A second translation, of which only the first book (ethica nova) and a few fragments remain, came at the start of the thirteenth century.

The new conception of philosophical ethics

The texts thus made available to Latin readers make two claims for philosophy: (a) happiness and human perfection are a legitimate object of philosophical concern; (b) virtue, or good human character, can be rationally discussed without recourse to theology.

These philosophical claims were not easy to accept. The problem does not lie in the conception of a natural virtue. Such an idea was already present in twelfth-century theology, and so the reception of the Aristotelian concept of virtue was not a revolutionary step.

Sources of the concept of intention

The concept of intention played a key role in the discussions of epistemological, logical, and semantic questions in later medieval philosophy. The significantly different use of ‘intention’ in other fields such as ethics or natural philosophy is not at issue here.

‘Intention’ in the relevant sense is associated with two concepts that occur already in the writings of Al-farabi and Avicenna, where they are associated with the words “m'qul” and “ma'na”, both of which were translated into Latin as ‘intention’. In his commentary on the first chapter of Aristotle's De interpretation Al-farabi understands by “ma'qui” — his translation of the Greek word ‘noēma’- a concept or a thought that has to be examined by the logician in two respects: in its relation to things outside the soul and in its relation to words. “Ma'qul” means nearly the same as “ma'na”, which appears already in Al-farabi's De intellectu et intellecto and was later used by Avicenna to signify the reality of the known considered as known. Thus “ma'qul”, ma'na, or intentio is that which is immediately before the mind, whether the object of the intention is outside the mind (in which case the intention is a first intention) or itself an intention (in which case the intention is a second intention). The distinction between first and second intentions was prefigured in Al-farabi's theory of abstraction.

Introduction to the medieval Latin Aristotle

All of Aristotle's works were translated into Latin in the Middle Ages and nearly all were intensely studied. The exceptions are the Eudemian Ethics, of which no complete translation survives, and the Poetics, which, although translated by William of Moerbeke, remained unknown. Most of the works were translated more than once, and two of them, the Physics and Metaphysics, were translated or revised no fewer than five times. The translations we are concerned with spanned a period of about 150 years; some were made from the Arabic, but the majority directly from the Greek. Some translations became popular and remained so; some became popular but were then superseded by other translations; others barely circulated at all.

An examination of the medieval Latin Aristotle cannot consider only the genuine works of Aristotle, but must also deal with works credited to Aristotle in the Middle Ages although now believed to be spurious. It is also essential to consider translations of Greek and Arabic commentators on Aristotle. (All these translations – of genuine and spurious works and of commentaries – are listed for easy reference in a single table below.)

The basic source for our knowledge of medieval Latin translations of Aristotle is a corpus of over 2,000 manuscripts dating from the ninth to the sixteenth century, most of which are distributed among the major libraries of Europe. They contain the texts of the translations and in some cases constitute the only documentation we have.

The prevalence of issues involving infinity and continuity

Natural philosophy in the fourteenth century is, when compared to that of the preceding century, more extensive, less repetitious, and more varied in the problems it treats, the solutions it sets forth, and the approaches and methods it employs in reaching those solutions. However, if one examines in some depth not merely the expositions and questions dealing with the relevant works of Aristotle but also the numerous non-commentatorial works constituting this literature, one cannot but be impressed by the unusual amount of time and effort spent in dealing with problems involving in one manner or another the infinite and the continuous.

Often these problems concern infinity or continuity from the outset; but equally often the problems are extended or developed by the fourteenth-century scholar to take into account some aspect of the infinite or the continuous in a manner that was not apparent in the problem as initially stated. A discussion of the way in which one should measure a quantity that varies in intensity throughout its subject might, for example, be carried so far as to accommodate ‘infinite values’. Alternatively, a discussion of angelic motion might involve one in a rather full investigation of the composition of all continuous quantities. Indeed, the prevalence of issues involving the infinite or the continuous in later medieval natural philosophy is such that an exhaustive history of these two notions in the later Middle Ages would constitute a very large part of the history of natural philosophy during this period.