SOURCE: "The Canon of Avicenna," in Avicenna in Renaissance Italy: The Canon and Medical Teaching in Italian Universities after 1500, Princeton University Press, 1987, pp. 19-40.

[In the following essay, Siraisi outlines the main ideas of Avicenna's Canon of Medicine, focusing on the Latin translation by Gerard of Cremona, and comparing it with its Galenic sources.]

The encyclopedic medical work written by Avicenna (d. 1037) is far too lengthy and, as the massiveness of the Latin commentaries on short sections of it testifies, far too complex to be adequately characterized in brief. The following comments are intended only to draw the reader's attention to certain features of the organization and content of the Canon that seem particularly relevant to its reception in the schools of the West and the emergence of a tradition of Latin commentary and, especially, to the adoption of Canon 1.1 as a textbook for the teaching of medical theory. Beginning with a brief overview of the Canon as a whole, I shall then pass to a somewhat more detailed, but still highly compressed, account of the physiological treatise in Book 1, Part 1. In the absence of full studies of Avicenna's relationship to his sources and of the relationship of the Latin to the Arabic text of the Canon, a summary of Canon 1.1 can only be, at best, sketchy and impressionistic. The immediate goal is, however, merely to indicate some of the main features of that treatise as its appears in its Latin form. A few comparisons with other ancient and medieval treatments of somewhat similar material illustrate the distinguishing characteristics of Canon 1.1.

The following description refers to the Latin translation of the Canon attributed by his pupils to Gerard of Cremona (d. 1187), since, although revisions of the Gerard translation and retranslations of parts of the work began to circulate in the sixteenth century, the twelfth-century Latin text was the original basis of Western understanding of Avicenna's book. Furthermore, as we shall see, various of the later revised Latin versions retained much of Gerard's work. Given the length, complexity, and technical nature of the Canon, the translation attributed to Gerard of Cremona is a remarkable accomplishment. Certainly the long history of the use of this version attests that by and large Gerard or his pupils managed to produce a generally understandable rendering of Avicenna's opus. Nonetheless, even though comprehensive studies of the translation have yet to appear, various scholars in the sixteenth century and the twentieth have pointed out that it is characterized by a high proportion of both errors and obscurely rendered passages. It also retains numerous transliterated Arabic words, especially for names of plants and minerals in the pharmacological sections and for anatomical terms. But whatever its merits or defects, the Gerard translation was the means whereby the Canon reached all its readers in western European medical schools up until the early sixteenth century, and a good many of them thereafter.

Considered as an encyclopedia of Greco-Arabic medicine, the Canon is in many respects distinguished by comprehensiveness and good organization. Praise of Avicenna's work for these qualities became, indeed, a commonplace among Latin commentators. As G. B. Da Monte put it [in his 1554 commentary on the Canon], Avicenna was moved to write the Canon when he saw that "neither among the Greeks nor among the Arabs was there any single, complete, continuous book that taught the art of medicine," the Hippocratic writings being enigmatic and obscure, Galen extremely prolix, and Rasis confusing. [In his Preface to Galeni opera ex sexta Juntarum editione, 1586], the same author asserted that Avicenna "intended to reduce all the monuments of medical art scattered at large in various works of Galen into one, as it were, corpus.… Indeed, he collected much dispersed information into appropriate and defined places and arranged it in sequence." One may certainly agree that the Canon is comprehensive. Few aspects of traditional Greek and Arabic medicine are left untouched in its five books, which together amount to about a million words in length. However, the Canon is at least as much a collection of essentially separate and distinct manuals and reference works as it is an architectonic summa of medicine. Moreover, while individual sections are as a rule clearly organized and succinctly expressed, the overall arrangement is somewhat confusing and involves a certain amount of overlapping. The reader turns with gratitude to the lengthy analytic table of contents found in the early printed editions of Gerard's translation immediately after Avicenna's preface to the whole work and before each of its subsequent books.

Book 1 is divided into four main parts, or fen, to use the term adapted from the Arabic by the Latin translator. It opens with the statement of principles and handbook of physiology. The second fen of Book 1 classifies varieties, efficient causes, and symptoms of disease. Diseases are divided primarily into those caused by imbalance of the four elementary qualities of hot, wet, cold, and dry in the body, those caused by faulty composition, or conformation of bodily parts, and those caused by solutio continuitatis or continui, which may be rendered, more or less, as trauma. The efficient causes of disease are categorized as either connected with environment, regimen, and psychology (among them are included the factors embraced in the traditional scheme of the "non naturals"—air, food, and drink; repletion and inanition; motion and rest; sleep and waking; and passions of the soul) or as due to any of a wide range of individual physical events, some of which (for example, pain and swelling) the modern reader might be inclined to consider symptoms rather than causes. The section on symptomology both lists an array of individual signs of imbalance of complexion, of solutio continuitatis, of obstruction, and so on, and reviews the standard means of diagnosis by pulse and urine. The third fen of Book 1 concerns the conservation of health: separate sections on pediatric, adult, and geriatric regimen are followed by regimes for the delicate and complexionally imbalanced, and for travelers. The fourth and final fen of Book 1 deals with principles of therapy and forms of treatment appropriate for different conditions. Therapies discussed include emetic, cathartic, sedative, and other medications, bleeding, cauterization, blistering, and enemas.

Book 2 is on the subject of medicinal simples; most of it is given over to a list—arranged more or less in Latin alphabetical order, presumably by the translator—of individual substances, their properties, and the conditions for which they are supposedly remedies. In his third book, Avicenna provided twenty-one fen on ailments peculiar to each major organ of the body— arranged from head to toe—and their treatment. Most of the sections are preceded by one or two chapters on the anatomy and aspects of the physiology of the organ in question. Book 4 suveys diseases and injuries that either affect the whole body or may occur in any part of it, and their treatment; of its seven fen, the most notable is probably the first, on fevers, which provided a famous and influential account of the subject. The remaining fen of Book 4 concern the concept of crisis or critical days in illness; tumors and pustules; solutio continuitatis in the form of wounds, bruises, sprains, and ulcers; dislocations and fractures; poisons of mineral, vegetable, and animal origins (including animal bites and stings); and skin conditions. The fifth and final book of the Canon is an antidotarium, or manual on the preparation of compound medicines.

If one wishes to use the Canon as a reference tool, the arrangement of material just summarized works well for some subjects, but a good deal less well for others. A reader who is primarily interested in disease, for example, finds a sequence of more or less self-contained treatises offering complementary information. He or she can begin with classification, efficient causes, and symptoms of disease in general in Book 1, Fen 2, and proceed to the survey arranged by varieties of treatment in Book 1, Fen 4. Then, moving from the general to the particular, there is the account of diseases affecting different parts of the body in Book 3 and of diseases affecting the body as a whole in Book 4. Yet the division of the material into separate sections, and the provision of the analytical table of contents, means that anyone who is interested only in one particular topic—lepra, say, or diseases of the eyes—can easily turn to a few chapters on that subject alone.

The results are also fairly satisfactory if one considers the material available to a reader primarily interested in therapy. There is a self-contained survey of treatment methods (1.4) and a dictionary of medicinal simples (2); much discussion of treatments appropriate for disease of particular bodily parts is found in Book 3, and of ills affecting the entire body in Book 4; Book 5 provides guidance for compounding medicines. The arrangement has the merit of making it possible to approach the subject of therapy from a variety of different standpoints. One can, for example, relatively easily find an answer to any of the following questions: When and in what conditions is bleeding an appropriate treatment? What are the medicinal powers of cinnamon? What treatments are recommended for deafness? For various kinds of fevers? How is theriac compounded? However, although no systematic survey has been attempted for present purposes, it is clear that consistency throughout the whole, whether in terms of therapeutic recommendations or of properties attributed to substances, is not always achieved.

Least satisfactory, in some respects, is the arrangement of the physiological and anatomical material. Principles of physiology and the anatomy of bones, muscles, nerves, veins, and arteries are discussed in the first fen of Book 1; but the accounts of the anatomy and some further discussion of the physiology of other major organs (brain, lungs, heart, digestive and reproductive organs, for example) are scattered through the opening chapters of the twenty-one fen of Book 3. This separation, which is the result both of Avicenna's division of medicine into theory and practice and of the classification of bodily parts that he adopted, was, as we shall see, further accentuated by the curricula and practices of medieval university faculties of medicine. But in order to understand both the basis for Avicenna's treatment of physiology and anatomy and the principles underlying the organization of the Canon as a whole, it is necessary to turn to the content of the fīrst fen of Book 1.

The easiest way to describe Canon 1.1 is as a manual of Galenic physiology, although such a description is slightly misleading because the work also contains other elements of considerable importance. Furthermore, although the subject matter of the treatise is primarily physiological, the discipline with which it is concerned is not coterminous with even the loosest and most general modern definition of the science of physiology. Instead, Fen 1, along with other parts of Book 1, deals with the theoretical part of medicine, a subject that embraced the definition of medicine and of its place among the arts and sciences, and the fundamental principles of medical learning. Canon 1.1 is subdivided into six doctrinae, the last four of which are mainly physiological in content. The first two sections concern the definition, subdivisions, and subject of medicine and the theory of the elements, the last being considered to provide the basis in physical science for physiological theory.

The first section (doctrina) of Fen 1 opens with a definition of medicine: "Medicine is the science by which we learn the various states of the human body, when in health and when not in health, whereby health is conserved and whereby it is restored, after being lost." Medicine is thus presented from the outset as a science in the Aristotelian sense of a body of knowledge derived by demonstrative reasoning from true premises, but a science with a practical purpose. Implicit in the definition is the repudiation of any notion that medicine is merely a practical art, technology, or craft. Avicenna next asserts that medicine is divided into theory and practice, rebutting the counterclaim that any true science must of necessity be entirely theoretical with the example of philosophy, also, in his view, divided into theoretical and practical parts. Furthermore, it is not the theoretical part of medicine alone that is categorized as scientia; the practical part is also scientia, since it too involves formally organized knowledge, in this case the knowledge of how the tasks of healing should be carried out.

The subject of medical science, we learn in the following chapter, is the human body, insofar as it is subject to health and sickness. Since complete knowledge only comes with knowledge of causes, the physician must know the causes of health and sickness, these being classified in the Aristotelian manner into material, efficient, formal, and final causes. The subject of Fen 1 is announced to be the material, formal, and final causes of the human body as it is subject to health and sickness (efficient causes Avicenna identified with external factors affecting the body and discussed, as already noted, in Book 1, Fen 2). The medicus, who must also know regimen, symptoms, and medications, should investigate these causes by sense and anatomy, but he should take the fundamentals of physical science (for example, that the elements exist and are four in number) from natural philosophy without himself engaging in natural philosophical investigation.

The second doctrina of Fen 1, on the elements, is an admirable illustration of this last principle, since it provides a highly compressed account of those, and only those, parts of a mainly Aristotelian element theory necessary to establish the human body as an object in the physical world and to provide the foundations for an understanding of two central concepts of Greco-Arabic medical theory, namely humoral physiology and the idea of complexio, temperament, or krasis (that is, the balance of the elementary qualities of hot, wet, cold, and dry in living bodies). We read that the human body, like all other bodies, is compounded of the elements. The two heavy elements, earth and water, contribute chiefly to the formation of its members; and the two light elements, air and fire, contribute mostly to the generation of spiritus and motion of the members (although only the soul actually moves the members). In addition to learning the natural place and qualities of each element (earth in the center, cold and dry, and so on), the reader is also reminded that earth preserves shapes and forms; that water receives all shapes but does not preserve them, that air helps in rarefying and elevating substances; and fire assists in processes of maturation, mixing, and making subtle. When one recalls, for example, the notion frequently expressed by medieval authors that a good memory is due to the physical retention of images by relatively dry (that is, earthy) brain tissue, or the frequency with which the heating of mixtures plays a part in medical recipes, the medical and physiological relevance of the latter information becomes readily apparent.

The first two doctrina of Fen 1 thus constitute an important statement about the nature of medical learning. The definition of medicine is broad enough to include the study of all aspects of physical function. The status of medicine as a science is defended. Moreover, the subject is placed in the context of Aristotelian natural philosophy, through insistence on medicine's scientific character, the application to it of the concept of the four causes, and the assertion that the fundamental physical principles on which medicine rests are drawn from natural philosophy. The human body studied by the physician exists as part of the sublunary world of Aristotelian physics and is subject to its laws. At the same time the independence of medicine as a discipline is safeguarded by the warning that the medicus should not, qua medicus, investigate issues in natural philosophy. The significance of these assertions was not to escape Avicenna's Latin commentators.

In addition, the second of the two introductory sections, that on the elements, fulfills a transitional function. While it continues the Aristotelian and philosophical themes of Doctrina 1, it also leads directly into the mainly Galenic and physiological material of the rest of the treatise. The topics surveyed in the last four subdivisions of Fen 1 are complexio or temperament, the humors or bodily fluids, the parts of the body, and the virtues and operations, which include sensation, motion, pulsation, and so on. These categories, together with the elements and spiritus (discussed by Avicenna along with the virtues and operations), constitute the ancient scheme of the things natural. In composing Fen 1, Avicenna superimposed the Aristotelian scheme of the four causes upon the set of seven things natural, redefining the latter as a sequence that proceeded from the most remote to the most immediate material causes, and ended with the final causes of the human body. Thus Doctrina 1 informs the reader that the body's material causes are the elements (remote), the humors (somewhat remote), and the members and spiritus (immediate). The virtues and operations, which usually come last in the scheme of things natural, Avicenna regarded, appropriately enough, as the final or teleological causes of the human body. These definitions accounted for the things natural except complexio, which Avicenna proclaimed the body's formal cause. He discussed complexio in its accustomed place in the sequence of things natural between the elements and the humors. This position, somewhat awkward in terms of his arrangement according to the four causes, had perhaps to be retained because of the presence of another implied progression in the traditional order of the first four of the things natural (elements, complexio, humors, members), namely an ascending hierarchy of perceptibility to sense.

Within this whole scheme, each section leads logically to its successor, and all major bodily functions are neatly fitted into the system. Having learned that the human body, like all other sublunary bodies, is fundamentally composed of the four elements, each with its pair of qualities, the reader proceeds next to Doctrina 3 on complexio. The balance of the elementary qualities produces, in plant and animal species, in human individuals, in parts of the body, and in medicinal compounds, the complexio, which Avicenna, as transmitted by Gerard, defined as itself a quality. Complexion, used in this special sense, is either balanced or unbalanced, temperate or intemperate; in reality, however, complexion never achieves perfect balance, although in man it can come very close. A given complexion can only be identified by comparison with one of a series of norms (such as with the ideal for the species or the individual). According to Avicenna, eight varieties of temperate and eight varities of distemperate complexion can be identified in this way. Medicines, too, can only be described as, for example, temperate, hot, or cold, with regard to their effect on a particular body or kind of body. The same medicine may be hot in regard to a scorpion and cold in regard to a man, or hotter for Peter than it is for Paul. Whereas the elements in the human body were conceived of as not individually or directly perceptible to sense, complexion was supposedly to some extent detectable by physical means; in human beings, the skin of the palm of the hand is the most temperate organ and hence serves as the measure of the complexion of the other parts of the body and the manifestation of the complexion of the body as a whole. As is apparent from the foregoing, all the organs and humors of the body were thought of as individually complexionate. Avicenna accordingly provided his readers with four lists of bodily parts, each list being arranged in descending order from the extreme of hotness, coldness, wetness, or dryness to the palm of the hand.

Since complexion in human beings was considered both as a kind of individual constitution and as sex-linked and subject to change in scientifically predictable ways over time (as well as in response to climatic and other conditions), discussion of complexion necessarily involved consideration of growth and aging as well as sex differences. Avicenna devoted a chapter to these topics, in which he considered some of the implications of the doctrine that life involves a process of diminution of heat and moisture from infancy to old age and ultimately death. A long section of this chapter weighs the relative complexional heat of children and youths; the problem was of interest because the theory just mentioned implied that infants ought to be the hotter, but the vigor of youth seemed to suggest the contrary.

Doctrina 4 defines the humors as liquids into which nutriment is converted. In addition to the four primary humors (blood, phlegm, red bile, and black bile), four secondary humors are identified. All are said to be found in both good and bad varieties. The good varieties of the primary humors are absorbed into and nourish the substance of the body; the bad are "superfluities," which are either excreted or damaging to the body if retained. Thus, discussion of the humors and their role in the body's economy involves attention to digestion, nutrition, and excretion, processes treated in a chapter on the generation of the humors. Topics covered include the stages of digestion and functions of mastication, saliva, stomach, intestines, liver, kidneys, and bladder.

The topic of the humors is succeeded in Doctrina 5 by that of the membra, or parts of the body, which are here described as generated by the humors and sustained by their nutritive activity. The introductory chapter of this section contains generalizations about the members paralleling those about the humors and complexions in the two preceding parts of the work; it is, however, followed by five summae containing a total of seventy-six chapters on the anatomy of the bones, muscles, nerves, arteries, and veins. Well over half of Fen 1, in fact, consists of these chapters on anatomy. However, … the only part of Doctrina 5 normally taught in European universities was the introductory chapter, and this is all that will be considered here. It is mainly concerned with various ways of classifying bodily parts. The first distinction made is between membra similia or simplicia (for example, bones, cartilage, nerves, ligaments, arteries, and veins) and composite or instrumental members (for example, the hands or the face); it will be observed that this distinction is not quite the same as that between tissues and organs, which it is sometimes said to resemble. Brief definitions of the nature and function of the more important membra similia follow. It is, of course, only such members that are treated in the subsequent anatomical chapters; for an account of the chief composite members as such the reader has to turn to the introductory chapters of each section of Book 3.

The next classification of members is according to whether they receive or emit virtue: examples of members that both receive and emit (suscipiens et tribuens) are the brain and the liver; a member that receives and does not emit is the flesh; one that emits and does not receive is, according to some, the heart, although others disagree. This allusion to the difference between the teaching of Aristotle and that of Galen on the role of the heart leads, appropriately enough, to consideration of the Galenic doctrine of the three, or four, primary or principal members. The discussion divides the parts into principal members and members that serve the principal members. The former are heart, brain, liver, and testicles; the subsidiary organs of the heart are the arteries and lungs, those of the liver the stomach and veins, and so on. Again, the members may be divided into those in whose generation the paternal sperm played a larger part (all membra similia other than flesh and blood) and those generated primarily from the retained menstrual blood of the mother (flesh and blood); discussion of this distinction leads into a brief excursus on embryology. Members may also be classified as fibrous (villosa) or nonfibrous in composition, the membra villosa being the stomach, intestines, urinary bladder, and gall bladder.

Discussion of the principal members leads naturally to consideration of the virtues, or powers, associated with each of them. These were divided into "natural," "vital" and "animal" virtues, a classification ultimately connected with the philosophical concept of the three-fold character of the soul. Hence, Doctrina 6, the concluding section of Fen 1, opens (after a brief introductory chapter) with an account of the natural virtues (or in other terminology, powers of the vegetative soul) associated with liver and testicles: growth, the ability to convert and assimilate nutriment, and reproduction. Avicenna then proceeded to a description of the role of spiritus in preparing the body for the reception of the animal virtues (that is, the powers associated with the sensitive soul possessed by all animals, including humankind) of motion and sensation. Spiritus in the Galenic tradition here followed by Avicenna was a highly refined bodily substance formed in part from inspired air; in Avicenna's version it was first generated in the liver from the vaporous part of the humors (to which, presumably, inspired air had contributed) and then received a second generation in the heart. Pervading the entire body, spiritus, which manifests itself in pulsation, serves as the vehicle of virtus vitalis, which keeps life itself in being. Specialized functions attributed to spiritus—the refinement of animal spirits in the brain and the role of visual spirits in seeing—were also briefly mentioned by Avicenna. The account of the animal virtues focuses mainly on the process whereby messages from the senses are received and interpreted in the brain. For the Latin Middle Ages and Renaissance, Avicenna was of course a leading authority on psychology, although his reputation in this field was chiefly based on his De anima rather than the Canon. Nonetheless, in the latter work, too, he weighed the validity of such classifications of brain function as sensus communis, phantasia, and imaginative, cogitative, estimative, and rational powers (the last of these being associated with the rational soul and reserved for humankind alone). Fen 1 finally concludes with a chapter in which it is pointed out that many of the body's actual operations require the cooperation of both natural and animal powers.

As will have become apparent from the foregoing summary, the first fen of Book 1 of the Canon is a compendium based almost entirely upon ancient Greek sources, chiefly Aristotle and Galen. Yet Avicenna's selection, organization, and interpretation of his material resulted in a work with a distinctive character of its own. Investigation of the relationship of Fen 1 to its sources is beyond the scope of the present work; however, consideration of the way in which a few somewhat arbitrarily chosen examples of Aristotelian and Galenic doctrines appear in Avicenna's treatise may serve to throw light upon the special characteristics and apparent goals of that work.

Avicenna was both physician and philosopher, although he distinguished quite sharply between the two roles. Repeatedly, he warned the readers of Fen 1 that the medicus should not himself investigate natural philosophy, but should rather accept the conclusions of experts in that field. However, information and ideas derived from philosophy in general and from Aristotle in particular in fact play quite a large part in Fen 1. The Aristotelian framework provided for the whole work in its opening sections has already been noted. In addition, Avicenna took every opportunity that offered itself to point out specific differences between the physiological and psychological views of Galen and his followers and those of Aristotle and his. The most important differences between Aristotle and Galen on the subject of physiology were those relating to the functions of heart and brain and to conception. Regarding heart and brain, Aristotle usually held the heart to be the primary organ of the whole body, whereas the Galenists, as already noted, considered the heart only one of several primary organs (namely, heart, brain, and liver, a triad to which the testicles were often added as a fourth). The differing doctrines had extensive ramifications in that they also entailed differing views on sensation and motion, the blood vessels, and other topics. As regards conception, Aristotle held that the active or formative principle was contributed entirely by the male, whereas Galen maintained that both sexes contributed actively to the formation of the fetus and that females as well as males therefore emitted sperm.

Hence, in the first general chapter on the members, Avicenna informed his readers, according to the version attributed to Gerard, that the brain was held to be the principium of sense absolutely by some (the Galenists), but only intermediately by others (the Aristotelians). A little farther on, he remarked that

… the medici differ from a great man among the philosophers. For this great man among the philosophers said that the heart is the member that gives out and does not receive. For it is the first root of all the virtues, and it gives to all the other members their virtues by which they are nourished and live and by which they know and move. But the medici and some of the first philosophers shared out these virtues among the members and they do not say that [the heart] gives out and does not take in. And the opinion of the philosopher is indeed more subtly proved and truer. But the opinion of the medici when it is attended to in the first place (in primis) is more obvious.

Later in the same chapter, referring to the simile used by "he who gave truth (verificavit) to the wise," likening the male contribution in conception to rennet and the female contribution to that of the milk passively acted upon by rennet, Avicenna observed:

This word is to no small, nay, to a large extent, contrary to the words of Galen. For it seems to him that both the power of coagulating and the power of being coagulated is in each of the two [that is, male and female] sperms.…

In several passages of Doctrina 6, on the virtues, Avicenna returned at some length to the topic of the differences between the physiology of Aristotle and that of Galen, explaining that their differences over heart and brain were reflected in varying accounts of the vital, animal, and natural powers (associated with heart, brain, and liver, respectively). In one of these passages the medicus was instructed that though the Aristotelian view was correct, it was not his business in his capacity as medicus to investigate the issue, since whether or not a particular organ was or was not the principium of one of the powers made no difference from the standpoint of medical treatment.

Doctrina 6 also warns of differences between philosophical and medical terminology. Galenic medical writers normally described motion and sense as animal, and growth and reproduction as natural, powers. Avicenna, as Gerard represented him, pointed out that "the philosophers when they say anima mean anima terrena, the perfection of the natural instrumental body. And they mean the principium of every power, that principle indeed from which motion and diverse operations come." Hence, the powers of growth and reproduction termed natural by medical writers might be described as animales in philosophical texts, although on other occasions animalis might be used to refer to the power of comprehending and moving. Furthermore, the reader is also instructed that philosophers use the term natural for "every power (virtus) through which operation occurs in a body according to the diversity of its form," and this usage refers to a higher faculty than the powers of reproduction and growth designated by the medical term virtus naturalis. Finally, yet other passages in the same doctrina point out differences between philosophers and medical writers over the proper classification of the internal powers of the brain.

In Fen 1, the various discussions of the relationship of philosophical and medical teachings carry a double message. On the one hand, the medicus is repeatedly warned off philosophizing. He is, in effect, told on a number of occasions that although the Aristotelian teaching is true, he should for all practical purposes follow Galen and not worry about the discrepancy. On the other, the reader gets a good many tantalizing glimpses of philosophical views, and hints are not lacking that the dual roles of philosopher and physician may be combined in one person. That Avicenna himself had functioned in both capacities was, of course, well known to medieval and Renaissance readers of the Canon in Latin. Furthermore, in Canon 1.1 Avicenna referred readers of Fen 1 to his own philosophical writings; and he drew attention to Galen's philosophical interests with the remark that if Galen is to be attacked for his views on the division of medicine "he ought not to be attacked insofar as he is a medicus but insofar as he is a philosopher."

For the strictly physiological, as distinct from the natural philosophical, material in Canon 1.1, Galen is by far the most important source. Almost all the physiological concepts schematized and abbreviated in Avicenna's treatise can be found in Galen's works. Except for the instances noted above, in which Galen's views are weighted against those of Aristotle, to the ultimate advantage of the latter, the Latin text of Canon 1.1 contains few or no examples of clearly intentional divergence from Galen's teaching. Moreover, Galen is several times cited by name, and is always mentioned with respect, even in instances when Aristotle's opinions are preferred from the standpoint of philosophy. Yet the spirit and approach of Avicenna in Canon 1.1 seems very different from that of Galen.

Avicenna's apparent goal in writing Canon 1.1 was to present a highly compressed yet comprehensive account of the whole of physiological theory; the work teaches a set of principles that have been, to a very large extent, effectively abstracted from any context of experience. Unified organization, clarity (usually), and brevity are achieved, but at a certain cost. Lacking, for the most part, are illustrative examples, admissions of uncertainty or limited knowledge, indications of divergence of views among ancient physicians, or acknowledgement of the presence of inconsistencies within the body of Galen's writings. By contrast, Galen's major treatises embody, as a rule, a leisurely and discursive investigation of a particular aspect of a unified science of medicine that is not sharply divided into theory and practice. Moreover, in Galen's writings physiological theory is generally interwoven into a context rich in detailed anatomical description, records of personal experience and observation, and, often, polemic against the holders of different scientific views. Without undertaking any detailed investigation of the way in which Avicenna used his Galenic sources (or the form in which these reached him), we may illustrate some of the ways in which the teaching of Galen is transformed by its presentation in the Latin text of Canon 1.1 by means of a comparison of a few passages therein with sections of three works of Galen that Canon 1.1 to some extent parallels in subject matter—namely, De complexionibus (De temperamentis), De naturalibus facultatibus, and De usu partium.

Let us take our first example from the treatment of the subject of complexio. Most of the main ideas expressed in Doctrina 3 of Canon 1.1 can also be found in Books 1 and 2 of Galen's De complexionibus. Among the themes present in Galen's work that are also treated in Canon 1.1 are the relation of complexion to the elementary qualities; the notion that the physician considers complexion not as a balance of weight (ad pondus) but ad justitiam, in terms of appropriateness for the species, individual, etc., under consideration; the idea that complexion can only be expressed in relative terms and that the point of reference must always be specified; the related idea that a given complexion may be cold in regard to one point of reference and hot in regard to another; the assertion that man is the most temperate of all complexionate beings and that the skin of the hand is the most temperate part of man; and discussions of the different complexions of various ages and sexes, of the inhabitants of different regions, and of the individual organs of the human body.

De complexionibus 1, however, also contains extensive passages of exposition of differing views and polemic against them that have no parallel in Canon 1.1.3; furthermore, Book 3 of De complexionibus is devoted to detailed exposition of the complexion of various medicinal substances and the therapeutic applications of complexion theory, topics excluded from Canon 1.1. But the characteristic that most significantly distinguishes Galen's treatment from Avicenna's handling of some of the same material is the inclusion throughout Book 2 of De complexionibus of lengthy descriptions of the physical manifestations of different kinds of complexion. In Galen's treatise, the whole subject is closely related to anatomical experience, to observation of the patient, and to medical care; in Avicenna's, summary of the topic is explained by means of clearly stated but abstract and apparently arbitrarily determined rules, and examples are deliberately deferred until later in the Canon. Thus, Galen's remarks on the complexion of the various organs relate the complexional characteristics of each part to its physical characteristics and especially to its consistency; the flesh of the heart is drier than that of the spleen, kidneys, and liver because it is harder. The reader is urged to find out for himself that the heat of the heart is greater than that of any other organ by thrusting his fingers into an incision in the chest of a living animal. By contrast, Canon 1.1' s neatly schematized and readily memorable lists of organs in descending order of dryness and heat provide essentially the same information as that given by Galen—and a few gaps are fi̵lled in— but supporting references to physical experience are absent. In general, in Galen's much more than in Avicenna's account one is constantly reminded that complexion, and secondary characteristics dependent thereon, are to a significant extent accessible to sense.

Nowhere is this contrast more evident, perhaps, than in the treatment of the issue of whether the complexion of children or youths is hotter. In Canon 1.1, in one of the relatively few places in that treatise where a diversity of opinion among ancient physicians is acknowledged, otherwise unidentified medici antiqui are described as having been divided on the issue. Galen is said to have dissented from both views, and to have maintained that the heat of children and youths was in radice equal, but that the heat of children was greater in quantity and that of youths greater in quality. The rest of the passage is devoted to an explanation of Galen's supposed views, which turns on the two assertions: it is impossible to increase innate heat, and humidity declines throughout life. Thus, the heat of youths is conserved by less humidity than that of infants; they are at a stage of life in which the humidity is sufficient to preserve heat at a level that is qualitatively high, but not of such quantity as to permit further augmentation of the body in the form of growth.

Turning to the Galenic account as presented in De complexionibus, one is immediately struck by the much less abstract character of most of the discussion. Galen pointed out that the physician's evaluation of actual (as distinct from potential) complexional heat depended on touch, which with the aid of memory was indeed a satisfactory way of detecting changes in the heat of a single individual. However, if different individuals or whole categories of individuals were to be compared, then it was essential to compare like with like in regard to as many characteristics and conditions as possible—body type, diet, environment, and so on. These remarks, interesting for the adumbration of the idea of controlled observation of numerous exemplars, find no echo in the Latin Canon 1.1.3. Moreover, Galen's conclusion in De complexionibus was firmly rooted in his clinical experience; he stated that having touched the bodies of numerous children and adolescents, he found it impossible to assert on the basis of the sense of touch that either group was simply and unequivocally hotter than the other. However, the perspiration (a manifestation of heat, evidently) of children, was, owing to their greater humidity, greater in quantity, and that of adolescents more acrid in quality. This difference is because children have more innate heat, but the heat of adolescents is drier. Comparing this chapter with the passage in the Canon, one notes that the latter, characteristically, elaborates on Galen's conclusion, but does not contain those of his supporting arguments that depend upon clinical experience and sense perception. (Canon 1.1.3.3 does, however, also explain a few perceptible characteristics of small children in complexional terms.)

One more illustration of the way in which Avicenna handled Galenic material may be drawn from his treatment of the topic of digestion and the digestive organs. We have seen that in Canon 1.1 there is a short narrative of the digestive process as a whole in the chapter on the generation of the humors (1.1.4.2); to this must be added the description of the coats of the stomach and intestines in the section on membra villosa in the general chapter on the members (1.1.5.1), and the account of the four nutritive powers (attractive, retentive, digestive, and expulsive) in the third chapter of Doctrina 6. Among the most important Galenic accounts of this part of physiology are Books 4 and 5 of De usu partium on the digestive organs and digestive process, and the lengthy discussion of nutritive power in De naturalibus facultatibus.

In its main outline, the brief summary of digestion in Avicenna's chapter on the generation of the humors is fairly similar to the much longer account presented by Galen in De usu partium and with parts of the account in De naturalibus facultatibus. Missing from Avicenna's highly compressed version are, however, both the emphasis on functional teleology and the richness of detail characteristics of De usu partium and the context of polemic against Erasistratus and his followers found in De naturalibus facultatibus. It is really only the opening generalities of De usu partium 4.1-4 that are paralleled in Canon 1.1.4.1. Any attempt at more detailed description of the individual digestive organs and their functions, similar to that occupying the remainder of Books 4 and 5 of Galen's work, is postponed in the Canon until scattered chapters of Book 3. Furthermore, even in passages that do closely parallel one another in Galen's account and in the Canon, Avicenna's version is greatly simplified. In some instances, the process of simplification, and the stripping away or deemphasis of the metaphors freely used by Galen to clarify his descriptions and render them more vivid, resulted in a presentation of Galenic ideas in the Canon in a form that was simultaneously both cryptic and perhaps more literal than Galen intended. One example will suffice. In both De usu partium and De naturalibus facultatibus Galen compared the concoction of the chyle in the liver to the fermentation of wine. In the former work, he said:

Let us, then, compare the chyle to wine just pressed from the grapes and poured into casks, and still working, settling, fermenting, and bubbling with innate heat. The heavy, earthy part of its residues, which I think is called the dregs, is sinking to the bottom of the vessels and the other, light, airy part floats. This latter part is called the flower and forms on the top of light wines in particular, whereas the dregs are more abundant in heavy wines. In making this comparison, think of the chyle sent up from the stomach to the liver as bubbling and fermenting like new wine from the heat of the viscus and beginning to change into useful blood; consider too that in this effervescence the thick, muddy residue is being carried downward and the fine, thin residue is coming to the top and floating on the surface of the blood.

In De naturalibus facultatibus the same idea is presented as follows:

What else, then, remains but to explain clearly what it is that happens in the generation of the humors, according to the belief and demonstration of the Ancients? This will be more clearly understood from a comparison. Imagine then some new wine which has been not long ago pressed from the grape, and which is fermenting and undergoing alteration through the agency of its continued heat. Imagine next two residual substances produced during this process of alteration, the one tending to be light and air-like and the other to be heavy and more of the nature of earth; of these the one, as I understand, they call the flower and the other the lees. Now you may correctly compare yellow bile to the first of these, and black bile to the latter.…

In the Canon the presentation of its idea runs as follows:

In every concoction of this sort there is to be found foam and sediment.… The foam is red [yellow] bile; and the sediment is melancholy [black bile].

Avicenna's drastic condensation of Galenic material is accompanied on occasion not only by oversimplification, as in the instance just noted, but also by inconsistency, the latter sometimes merely echoing Galen's own. Thus, in the account of the membra villosa in the Gerard version of Canon 1.1, the urinary bladder is said in passing to have a single tunic; but in the account of the bladder's anatomy in Book 3 it is described as having two tunics. This discrepancy presumably stems from the fact that Galen had assigned the bladder one tunic in De usu partium and in one passage of De naturalibus facultatibus, but in a second passage of the latter work had given it two. Moreover, while the details of Galen's—and Avicenna's own— experience were excised from Canon 1.1, certain concepts adumbrated in Galen's works in Canon 1.1 were elaborated to the point where they may be regarded as original contributions. A case of this kind that has been the subject of modern scholarly attention is Avicenna's development of the concept of so-called radical moisture as an explanation of the process of aging. It must be emphasized, however, that the focus in the physiological sections of Canon 1.1 upon the systematic presentation and explanation of principles, at the expense of detail and supporting evidence, is unquestionably part of the deliberate strategy of that treatise; other parts of the Canon are rich in material pertaining to the experience of the physician and the treatment of patients.

Avicenna was neither the first nor the earliest available in Latin among authors who had written in Arabic to produce a schematic general summary of Greco-Arabic medicine structured around a division into theory and practice. The collection of brief medical opinions attributed to Joannitius and known as the Isagoge begins with the assertion that medicine is divided into theoretical and practical branches and proceeds to a consideration of the naturals, the nonnaturals, and the things against nature (that is, diseases). Avicenna has been shown to have drawn some of his material from the medical encyclopedia of Rasis (d. 925); another predecessor and possible partial source of the Canon was the survey of medicine written by Haly Abbas (d. 994). As compared with these other works, the Canon offers a fuller and more intellectually demanding account of its subject matter than the highly abbreviated Isagoge, and its organization is somewhat easier to grasp than that of the work of Rasis. The work of Haly Abbas is fully comparable to the Canon in scope and sophistication as well as structure. It is divided into ten books on theory and ten books on practice, of which the first four books on theory roughly parallel Canon 1.1 in content. In organization, the Liber totius medicinae is in at least one respect superior to the Canon: Haly's entire account of physiology and anatomy is presented as a coherent unit in his opening books, whereas in the Canon, as we have seen, the treatment of internal organs is excluded from the physiological and anatomical material in Book 1, Fen 1. Like Canon 1.1, of which it may well have been one of the sources, the Liber totius medicinae contains lengthy passages summarizing Galenic concepts. But the Aristotelian element, which plays so marked a part in Canon 1.1 seems not to be found in the corresponding parts of the Latin version of Haly's work. It was precisely Avicenna's philosophical interests that gave Canon 1.1 its distinctive cast. In Canon 1.1, as in Avicenna's philosophical writings, elements of diverse origin were drawn together. It seems likely that, indeed, Avicenna's well-known readiness to incorporate both Aristotelian and Neoplatonic elements into his philosophical system stimulated his interest in considering the relationship between Aristotelian and Galenic thought, which also contained Platonic elements.

Such, in barest outline, was the Canon as translated by Gerard of Cremona or his associates. The scope and organization of the work, the mass of practical information it contained, the author's summaries of Galenic thought (inadequate though these might sometimes be in detail), and the philosophical context in which Avicenna succeeded in placing medicine while preserving its independence as a distinct science—all initially secured for the Canon a place of honor in medieval western European medicine.…