Georges Canguilhem

Georges Canguilhem (1904–1995) was a French philosopher and physician whose work on the concepts of the normal and the pathological became foundational for twentieth-century philosophy of medicine and biology. His central argument — that life itself is a normative activity, and that health and disease cannot be reduced to quantitative deviations from a statistical mean — challenged the dominant positivist tradition inherited from Auguste Comte and Claude Bernard. Canguilhem belonged to a lineage of French philosophers concerned with the history and philosophy of science, running from Cavailles through Bachelard, and his influence extends through Michel Foucault and into contemporary debates about the philosophy of health, disease classification, and vitalism.

Life and Context

Foucault identified Canguilhem as standing outside the dominant philosophical currents of twentieth-century France — phenomenology, Marxism, philosophy of the subject — and located him instead in a distinct tradition concerned with the history and philosophy of science, running from Cavailles through Bachelard to Canguilhem himself (Canguilhem, 1978). Foucault drew a sharp division between a philosophy of experience, sense, and subject (Sartre, Merleau-Ponty) and a philosophy of knowledge, rationality, and concept (Cavailles, Bachelard, Canguilhem), arguing these two lineages were fundamentally opposed (Canguilhem, 1978).

Canguilhem became one of France’s foremost historians of science, professor at the Sorbonne and, until his retirement, director for many years of the Institut d’Histoire des Sciences et des Techniques de l’Université de Paris.(Canguilhem, 1978) Bachelard was his explicit mentor, and the imprint shows in his methodology: rather than reconstructing scientific change as the steady accumulation of results, Canguilhem focused on the structures of explanation through which novelty was developed.(Canguilhem, 1978) His own preface to the 1966 edition gives the corresponding self-description of the project: medicine seemed and still seemed to him “like a technique or art at the crossroads of several sciences, rather than, strictly speaking, like one science,” and the two problems that concerned him — the relations between science and technology, and that of norms and the normal — required a direct medical education for their precise position and clarification.(Canguilhem, 1978) He framed Le normal et le pathologique itself as an effort to integrate some of the methods and attainments of medicine into philosophical speculation, not as a normative judgment imposed from philosophy onto medical practice.(Canguilhem, 1978)

Foucault’s Introduction to The Normal and the Pathological

The 1978 English edition of On the Normal and the Pathological was translated by Carolyn R. Fawcett with editorial collaboration from Robert S. Cohen, and appeared with an introduction by Michel Foucault that became one of the principal texts through which Canguilhem’s work was communicated outside France.(Canguilhem, 1978) The translation of Le normal et le pathologique, first published by Presses Universitaires de France in Paris in 1966, appeared in the “Studies in the History of Modern Science” series edited by Robert S. Cohen, Erwin N. Hiebert, and Everett I. Mendelsohn.(Canguilhem, 1978)(Canguilhem, 1978) It brought together two distinct phases of the project: the original 1943 essay (reprinted without textual changes in 1950 and 1966) and the group of three essays written between 1963 and 1966.(Canguilhem, 1978)(Canguilhem, 1978)(Canguilhem, 1978)

Foucault’s introduction carries an argument that goes considerably beyond personal tribute. He begins with a claim about intellectual genealogy: without Canguilhem, he writes, large portions of post-1960s French thought become unintelligible. Althusserian Marxism, the sociology of Bourdieu, Castel, and Passeron, and the theoretical work of the Lacanians all bear his imprint, though Canguilhem’s name rarely appears in their pages.(Canguilhem, 1966)(Canguilhem, 1978) The dependence is specific enough to be verifiable: “take away Canguilhem,” Foucault writes, “and you will no longer understand much about Althusser, Althusserism and a whole series of discussions which have taken place.”(Canguilhem, 1978)

The essay’s deeper move is to articulate the two strands Foucault saw running through twentieth-century French philosophy since its engagement with phenomenology. One strand (Sartre and Merleau-Ponty) takes experience, sense, and subject as its primary concerns. The other (Cavaillès, Bachelard, and Canguilhem) takes knowledge, rationality, and concept.(Canguilhem, 1978)(Canguilhem, 1966) Foucault treats these as fundamentally opposed orientations, and positions Canguilhem as having opposed phenomenology’s philosophy of meaning and the experienced thing with a philosophy of error, concept, and the living being.(Canguilhem, 1966) The work that Cavaillès, Koyré, Bachelard, and Canguilhem carried out in France, Foucault notes, corresponds functionally to what the Frankfurt School was doing in Germany, despite the stylistic differences: both address a rationality that makes universal claims while developing in contingency, that authenticates itself through its own sovereignty but is perhaps not dissociated from inertias and despotisms.(Canguilhem, 1978)

Foucault also characterizes Canguilhem’s methodological contribution to the history of science. Canguilhem shifted the field’s attention from its traditional focus on formalized disciplines — mathematics, astronomy, Galilean and Newtonian physics — toward biology and medicine, where knowledge is much less deductive, much more dependent on external processes and institutional supports, and where it has remained tied much longer to the imagination.(Canguilhem, 1978)(Canguilhem, 1966) The history of science that Canguilhem practiced was not a history of the progressive discovery of truth inscribed in things or in the intellect. Marking discontinuities was for him neither a postulate nor a result but a “way of doing” integral to the discipline, because science proceeds through successive rectifications rather than gradual revelation.(Canguilhem, 1966)

Vitalism appears in the introduction not as a metaphysical doctrine but as a double indicator: theoretical, signaling unsolved problems about the originality of life; critical, warning against reductions that ignore the fact that the life sciences cannot do without certain value-positions regarding preservation, regulation, and adaptation. Foucault quotes Canguilhem’s own formula: “A demand rather than a method, a morality more than a theory.”(Canguilhem, 1966)(Canguilhem, 1978)

Foucault’s reading of Canguilhem’s specific contribution to the life sciences centers on disease, death, monstrosity, anomaly, and error as categories that a science of the living being cannot bracket without losing its proper object. The living being involves self-regulation and self-preservation processes that mark a specificity the life sciences must take into account.(Canguilhem, 1978) The concept formation Canguilhem studies in the history of biology is strategic because a biological concept must simultaneously cut from the ensemble of phenomena those which allow analysis of processes specific to living beings, and give access to a structure of intelligibility through which elementary chemical or physical analysis can show up those same specific processes.(Canguilhem, 1978)(Canguilhem, 1978) Foucault draws from this a methodological principle: in the history of physics, the decisive moment is formalization and constitution of theory; in the history of the biological sciences, it is the constitution of the object and the formation of the concept.(Canguilhem, 1978)

Foucault’s introduction considers Le normal et le pathologique the most important and significant of Canguilhem’s works because it shows how the ancient problem of the normal and the pathological recently found itself at the intersection of the problems biology had borrowed from information theory: code, messages, messengers.(Canguilhem, 1978) At the most basic biological level, the play of code and decoding leaves room for chance, and what chance produces before it becomes disease, deficit, or monstrosity is something like a “mistake” in the information system.(Canguilhem, 1966) Life is what is capable of error, and this fundamental eventuality explains why the question of anomaly runs through all of biology.(Canguilhem, 1978)

Foucault names this position “philosophy of error” in a precise sense: “it is in starting from error that he poses philosophical problems, I should say, the philosophical problem of truth and life.”(Canguilhem, 1978) Forming concepts, Foucault writes in a formulation that captures what is at stake, “is one way of living, not of killing life; it is one way of living in complete mobility and not immobilizing life.”(Canguilhem, 1978)

Core Contributions

The Normal and the Pathological

Canguilhem’s 1943 medical thesis investigated whether pathological phenomena are qualitatively or merely quantitatively different from normal phenomena, insisting that this is a philosophical question not resolvable by physiology alone, since physiology already presupposes an answer.(Canguilhem, 1978) He identified the dominant medical ideology of the nineteenth century — the quantitative identity of normal and pathological — as deriving from Broussais’s physiological medicine, transmitted through Comte and Bernard.(Canguilhem, 1978) The thesis began as a doctoral dissertation in medicine and was always conceived as groundwork for a future thesis in philosophy; he acknowledged deliberately giving his conclusions “the appearance of propositions which were simply and moderately methodological” in order not to overload a medical thesis with philosophical ambition.(Canguilhem, 1978)(Canguilhem, 1978)

Canguilhem came to medicine from philosophy with a specific set of problems in mind. Philosophy, he observed, is “a reflection for which all unknown material is good, and we would gladly say, for which all good material must be unknown.”(Canguilhem, 1978) He expected medicine to provide an introduction to concrete human problems — a “technique or art at the crossroads of several sciences,” not a single science on the model of physics.(Canguilhem, 1978) The two problems that drove him — the relations between science and technology, and the problem of norms and the normal — required a direct medical education for their precise position and clarification.(Canguilhem, 1978) The clinic and therapeutics were essential to this education because clinical medicine is a technique for establishing or restoring the normal that cannot be reduced entirely to a single form of knowledge.(Canguilhem, 1978)

In the 1950 Preface, Canguilhem framed the normality/pathology distinction as a practical-philosophical problem inseparable from medical activity: the doctor cannot practice medicine without a concept of health, and that concept is not given by physiology but must be evaluated — making medicine irreducibly normative rather than purely descriptive.(Canguilhem, 1978) The 1966 Foreword established that the later essays addressed the extension of the normal/pathological question into molecular biology, sociology, and pathological psychology — disciplines that did not yet exist in their current form in 1943.(Canguilhem, 1978)

Canguilhem established a central distinction between anomaly and disease. Anomaly refers to morphological variation — structural difference from the species type — while disease refers to functional impairment experienced as suffering; an anomaly can exist without disease, and disease can exist without anatomical anomaly (Canguilhem, 1978). This entails that “normal” operates on two registers: the anatomical normal (species-typical morphology) and the physiological normal (functional adequacy for individual life), and pathology concerns the second, not the first (Canguilhem, 1978).

He critiqued the use of statistical averages as the basis for defining the normal in biology, drawing on Halbwachs’s work showing that Quetelet’s average man was a statistical fiction — no real individual embodies all average measurements simultaneously (Canguilhem, 1978).

The Arguments of The Normal and the Pathological

The thesis that pathological phenomena are identical to normal phenomena, differing only in quantitative degree, was, Canguilhem established, the generally adopted position of nineteenth-century medicine.(Canguilhem, 1978) This dogma was expounded in France by Auguste Comte and Claude Bernard, working under very different circumstances and with very different intentions.(Canguilhem, 1978) Its genealogy runs through Broussais, who had described all diseases as consisting in excess or deficiency of excitation in the various tissues above or below the degree established as the norm — Comte then elevated this into what Canguilhem calls Broussais’s principle, asserting that the pathological state is “not at all radically different from the physiological state” and constitutes “a simple extension going more or less beyond the higher or lower limits of variation proper to each phenomenon of the normal organism, without ever being able to produce really new phenomena.”(Canguilhem, 1978)(Canguilhem, 1978)

In Comte’s version, interest moves from the pathological to the normal: disease serves as a substitute for biological experimentation, which is often impracticable on human subjects, and the identity of normal and pathological is asserted as a gain in knowledge of the normal.(Canguilhem, 1978) In Bernard’s version, the direction reverses: interest moves from the normal to the pathological, with physiology serving as the foundation for a rationally grounded therapeutics, and the identity of normal and pathological is asserted as a gain in the ability to remedy pathological states.(Canguilhem, 1978)(Canguilhem, 1978) Bernard put it explicitly: “every disease has a corresponding normal function of which it is only the disturbed, exaggerated, diminished or obliterated expression.”(Canguilhem, 1978) “There is no case where disease would have produced new conditions, a complete change of scene, some new and special products.”(Canguilhem, 1978)

Canguilhem’s opening move in the 1943 thesis was to locate the normal/pathological debate within the long history of medical thought between two competing representations of disease: an ontological theory, in which disease is a foreign entity entering or leaving the body, and a dynamic theory, in which disease is an internal disequilibrium. He argued that the nineteenth-century quantitative thesis grew from the attempt to make disease scientifically tractable by collapsing this qualitative difference — and that deficiency diseases and infectious diseases favor the ontological reading, while endocrine disturbances and disorders prefixed “dys-” favor the dynamic one.(Canguilhem, 1966) The ontological theory of disease is not arbitrary: “the impetus behind every ontological theory of disease undoubtedly derives from therapeutic need,” since to act it is necessary to localize, and localizing disease reassures by suggesting that what has been lost can be restored and what has entered can leave.(Canguilhem, 1978)

Against the ontological tradition, Greek Hippocratic medicine offered what Canguilhem calls a dynamic and totalizing conception: nature within man and without is harmony and equilibrium, and disease is a disturbance of this harmony that “is not somewhere in man, it is everywhere in him; it is the whole man.”(Canguilhem, 1978) Disease, on this reading, is simultaneously a disequilibrium and an effort on the part of nature to achieve a new equilibrium: “the organism develops a disease in order to get well,” and medical technique therefore imitates rather than overrides the vis medicatrix naturae.(Canguilhem, 1978)

Broussais described all diseases as consisting in excess or deficiency of excitation above or below the normal degree in the tissues.(Canguilhem, 1966) Comte elevated this nosological principle to a general axiom.(Canguilhem, 1966) The pathological state is merely an extension of the physiological state, differing only in degree.(Canguilhem, 1966) Leriche defined health as “life lived in the silence of the organs.”(Canguilhem, 1966)

The semantic signature of the Broussais-Comte-Bernard position is that the pathological is designated as departing from the normal not so much by a or dys as by hyper or hypo: excess or deficiency of the same process, never a new kind of process.(Canguilhem, 1978) Canguilhem’s internal critique of this position begins with Broussais: Broussais “obviously confuses cause and effect” in defining the pathological state, since “a cause can vary quantitatively so that it nevertheless both continues and provokes qualitatively different effects.” A quantitatively increased excitation can bring about a pleasant state soon followed by pain — two states no one would want to confuse, though they differ only in the quantity of the cause.(Canguilhem, 1978) The logic is that to define the abnormal as “too much or too little” is already to recognize the normative character of the so-called normal state: “This normal or physiological state is no longer simply a disposition which can be revealed and explained as a fact, but a manifestation of an attachment to some value.”(Canguilhem, 1978)

Bernard’s version of the thesis holds only in limited cases: when the pathological phenomenon is limited to a single symptom abstracted from its clinical context (hyperchlorhydria, hyperthermia, reflex hyperexcitability), or when symptomatic effects are traced back to partial functional mechanisms (glycosuria in terms of hyperglycemia). Even in those cases it runs into difficulties.(Canguilhem, 1978) The deeper problem is that the pathological can be distinguished as an alteration of the normal only at the level of organic totality: “to be sick means that a man really lives another life, even in the biological sense of the word.”(Canguilhem, 1978) The reduction of quality to quantity required by the essential identity of physiology and pathology is a scientific operation that does not annul the quality it dismisses: “quantity is quality denied, but not quality suppressed.”(Canguilhem, 1978) And the conclusion follows: “it is completely illegitimate to maintain that the pathological state is really and simply a greater or lesser variation of the physiological state.”(Canguilhem, 1978)

Canguilhem situates the appeal of the nineteenth-century monist position: it expressed the rationalist conviction that “evil has no reality,” a posture that distinguished nineteenth-century medicine from the dualist medicine of the eighteenth century in which health and disease “fought over man the way Good and Evil fought over the World.”(Canguilhem, 1978)

One thread Canguilhem develops in the Hippocratic direction runs against the tendency of the Comte-Bernard tradition: the vis medicatrix naturae tradition understood disease not as disequilibrium simply but as “a generalized reaction designed to bring about a cure.” The organism develops a disease in order to get well; therapy must therefore tolerate and where necessary reinforce these spontaneously therapeutic reactions. Medical technique imitates natural medicinal action rather than overriding it.(Canguilhem, 1966)

René Leriche, whose surgical perspective Canguilhem examined alongside Goldstein’s neurological one, offered a definition that crystallizes the experience-centered approach to disease: “Health is life lived in the silence of the organs.”(Canguilhem, 1978) Leriche’s corollary was radical in its implications for medical epistemology: “if one wants to define disease, it must be dehumanized”; “in disease, when all is said and done, the least important thing is man.”(Canguilhem, 1978) Canguilhem refuses this dehumanization: “medicine always exists de jure, if not de facto, because there are men who feel sick, not because there are doctors to tell men of their illnesses.”(Canguilhem, 1978) Yet Leriche’s surgical perspective also contained a genuine insight: disease is not a “parasite living on and of the man it consumes” but “a new physiological order to which therapeutics must aim to adapt the sick man.”(Canguilhem, 1978) By studying the sick, Leriche observed, one discovers deficiencies that the most subtle animal experiment would fail to produce, and the complete study of disease becomes an essential element of normal physiology.(Canguilhem, 1978)

Canguilhem insisted that medicine exists because people feel sick, not because doctors tell them of their illnesses. The physician’s objective knowledge of disease historically derived from the patient’s experience of being different; clinical perspicacity overtakes the patient’s self-knowledge, but only because that self-knowledge originally called medical attention into being (Canguilhem, 1966)(Canguilhem, 1966).

Canguilhem’s treatment of neurology drew substantially on Hughlings Jackson, whose dissolution principle held that disease of the nervous system always has both a negative aspect (loss of higher functions) and a positive aspect (release of lower, hitherto suppressed centers). No disturbance can have a purely negative cause: a lesion in higher nervous centers frees the lower regulatory centers, and the disturbances of existing functions must be attributed to the “appropriate activity of henceforth insubordinate centers.” This principle was foundational for Goldstein’s later clinical work with head-wound patients (Canguilhem, 1966).

Drawing on Goldstein’s clinical work with World War I head wound patients, Canguilhem demonstrated that pathological reactions are not residues of previous normal behavior minus something destroyed — they are reactions that never appear in the normal subject in the same form or conditions (Canguilhem, 1966). His central thesis followed: disease does not consist in the absence of norms but is itself a norm of life, an inferior norm that tolerates no deviation from the conditions in which it is valid. The sick person is not abnormal because of the absence of a norm but because of incapacity to be normative (Canguilhem, 1966). Biological norms are individual, not statistical averages — Napoleon’s resting pulse of 40 was normal for his organism though aberrant relative to the average (Canguilhem, 1966).

Health, then, is the physiological state that allows the transition to new norms; pathological constants have repulsive, strictly conservative value while physiological constants have propulsive value (Canguilhem, 1966). Cure is not a return to biological innocence but the acquisition of new norms of life, sometimes superior to the old ones — biological normativity is irreversible (Canguilhem, 1966). There is no biological science of the normal as such, only a science of biological conditions called normal, which is physiology (Canguilhem, 1966).

Canguilhem concluded this part of the argument by challenging the coherence of “objective pathology” as a phrase. One can carry out objective research in pathology — impartially, with controlled methods — but the object of that research cannot be conceived without reference to positive and negative qualification. Pathology’s object is not so much a fact as a value, which means that while pathological research can be methodical, it cannot be value-free in the way the positivist tradition demanded (Canguilhem, 1966).

The 1943 thesis concluded with what Canguilhem called biological normativity — the central concept synthesizing his critique. The principle that the morbid state is only a simple quantitative variation of the physiological phenomena defining the normal state was, he argued, narrow and inadequate.(Canguilhem, 1978) Types and functions can be qualified as normal “with reference to the dynamic polarity of life”: biological norms exist because life, as not only subject to the environment but also as an institution of its own environment, thereby posits values not only in the environment but also in the organism itself.(Canguilhem, 1978) The pathological state can be called normal insofar as it expresses a relationship to life’s normativity, but it involves other norms than the normal physiological state, and “the abnormal is not such because of the absence of normality. There is no life whatsoever without norms of life, and the morbid state is always a certain mode of living.”(Canguilhem, 1978) The physiological state is the healthy state because it allows the transition to new norms; man is healthy insofar as he is normative relative to environmental fluctuations — physiological constants have “propulsive value” while pathological constants have “repulsive and strictly conservative value.”(Canguilhem, 1978) Cure is the reconquest of a state of stability, but “no cure is a return to biological innocence”: to be cured is to be given new norms of life, sometimes superior to the old ones, reflecting an irreversibility of biological normativity.(Canguilhem, 1978)

In the 1963–1966 essays, Canguilhem extended the argument. He began by tracing the institutional history of the word “normal” itself: in French, it entered popular language from two professional vocabularies simultaneously — pedagogy (the écoles normales, which trained teachers) and medicine (hospital reform). Both were products of the French Revolution’s demand for rationalization, and their convergence is what Canguilhem connects to what was later called normalization.(Canguilhem, 1966) He demonstrated that the norm is a polemical concept: the Latin norma means T-square, and to set a norm is to impose a requirement on an existence whose variety presents itself as hostile.(Canguilhem, 1966)(Canguilhem, 1966)(Canguilhem, 1978) The normal in Canguilhem’s account is not a static or peaceful but a dynamic and polemical concept: a norm draws its meaning, function, and value from the existence, outside itself, of what does not meet the requirement it serves.(Canguilhem, 1978) The abnormal, as ab-normal, comes logically after the definition of the normal as its negation, but “it is the historical anteriority of the future abnormal which gives rise to a normative intention” — the normal is the effect obtained by execution of the normative project, which means the abnormal is logically second but existentially first.(Canguilhem, 1978) The Section II essays clarify what normativity means when applied to a person: Canguilhem calls “normativity” the biological capacity to challenge usual norms in case of critical situations, and proposes measuring health by the gravity of the organic crises surmounted through the establishment of a new physiological order.(Canguilhem, 1978) In the human case, norms are individual achievements rather than imposed templates: each person fixes his norms by choosing his models of exercise, the norm of a long-distance runner is not that of a sprinter, and each changes his norms with age and former norms.(Canguilhem, 1978) The corresponding picture of health is one in which the menace of disease is itself one of the components of health: the so-called healthy man is not simply healthy but maintains an equilibrium that he redeems on inceptive ruptures.(Canguilhem, 1978) Drawing on Garrod’s concept of “inborn errors of metabolism” (1909), he showed that molecular pathology reconceived disease as error in the information-theoretic sense — since DNA encodes protein synthesis via messages, any interpretation involves possible mistakes, and the substitution of one amino acid for another creates disorder through misunderstanding the command.(Canguilhem, 1966)(Canguilhem, 1966) This dissolved the boundary between errors of life and errors of thought: life’s capacity for error is what makes disease, knowledge, and their histories possible.(Canguilhem, 1966)

Section II also clarifies why there can be no purely objective science of the normal. The concept of the normal cannot be reduced to an objective concept determinable by scientific methods; there is only a science of biological situations and conditions called normal, which is physiology.(Canguilhem, 1978) Medicine is an activity rooted in the living being’s spontaneous effort to dominate the environment and organize it according to its values as a living being; it uses the results of all sciences in the service of the norms of life without itself being a science.(Canguilhem, 1978) “It is first and foremost because men feel sick that a medicine exists. It is only secondarily that men know in what way they are sick because a medicine exists.”(Canguilhem, 1978) The qualification of a biological phenomenon as pathological is not made by an objective method but through “the relation to the individual patient through the intermediary of clinical practice.”(Canguilhem, 1978)

The 1966 Foreword also makes explicit several positions Canguilhem had only implied in the 1943 thesis. He notes that if he were to write the essay again, he would devote a great deal of space to Selye’s work on the state of organic alarm, since the general adaptation syndrome could mediate between Leriche’s and Goldstein’s apparently divergent accounts of disease.(Canguilhem, 1978) He proposes more forcefully than in the original text that there is no a priori ontological difference between a successful living form and an unsuccessful one — the very category of an “unsuccessful” form depends on a determination of the obligations of the living being that biology alone cannot supply.(Canguilhem, 1978) And he records, with the cordiality he reserves for serious objections, Louis Bounoure’s charge that he had yielded to the “evolutionist obsession” by treating the living being’s normativity as a projection of the human tendency toward transcendence onto all of living nature.(Canguilhem, 1978) The Foreword does not retract the position; it acknowledges the cost of holding it.

There is also a personal note embedded in the 1966 Epilogue that Canguilhem himself drew attention to. He acknowledged that his 1943 conception of health as normative capacity was, as critics had pointed out, a conception of life as one might form it when young: “our conception of the normal is very archaic, while it is — undoubtedly because it is — a conception of life such as can be formed when one is young.”(Canguilhem, 1978) The normal is identified with “the previous euphoric state of the subject who has just fallen sick.”(Canguilhem, 1978) It took, Canguilhem allowed, “the temerity of youth to believe oneself equal to the task of a study of medical philosophy on norms and the normal.”(Canguilhem, 1978) Completing the new essays, he acknowledged both the difficulty of the undertaking and the passage of time: “The difficulty of such an undertaking makes one tremble. We are aware of this today as we complete these pages of resumption.”(Canguilhem, 1978) With this confession, Canguilhem noted that he had “in conformity with our discussion on norms, reduced our own with time”(Canguilhem, 1978) — enacting his own theory of the relationship between normativity and vitality.(Canguilhem, 1966)

Vitalism as Exigency

Canguilhem’s guiding thesis was that there is no fundamental conflict between knowledge and life; knowledge is a form of life and belongs to the activity of living beings (Canguilhem, Georges, 1952/2008). Knowledge is a general method for resolving tensions between humans and their milieu, not an autonomous activity divorced from life (Canguilhem, Georges, 1952/2008). The title Knowledge of Life carries a double meaning: the knowledge we have of life as object, and the knowledge that life itself produces (Canguilhem, Georges, 1952/2008).

For Canguilhem, vitalism is first and foremost a demand or exigency — not a specific doctrine — requiring that the question of the relation between life and science remain open (Canguilhem, Georges, 1952/2008). Living beings have a normative relation to life — they are not indifferent to conditions of health and disease — which gives biology and medicine an irreducible specificity (Canguilhem, Georges, 1952/2008). There is no pathological astronomy or physics because physical phenomena do not deviate from their natural type; pathology is a category unique to the life sciences (Canguilhem, Georges, 1952/2008).

The fundamental conflict in human experience, Canguilhem held, is not between thought and life but between man and the world, with knowledge serving as the instrument for overcoming obstacles and restoring equilibrium (Canguilhem, Georges, 1952/2008). Scientific theories do not proceed from facts but from prior theories; facts are only the indirect route by which theories evolve from one to another (Canguilhem, Georges, 1952/2008). The history of science reveals what Canguilhem called the “sense of possibility”: knowing is not running up against the real but validating the possible by rendering it necessary (Canguilhem, Georges, 1952/2008).

He argued that the persistence of vitalism cannot be compared to the persistence of phlogiston or geocentrism because vitalism has a vitality of its own that requires philosophical explanation (Canguilhem, Georges, 1952/2008). Biological theory throughout its history has been divided and oscillating between mechanism and vitalism, preformation and epigenesis, continuity and discontinuity — this oscillation is not mere failure but may express an undiscovered dialectic (Canguilhem, Georges, 1952/2008). Vitalism translates a permanent exigency of life within the living, which explains its characteristic vagueness compared to the precision of mechanism as method (Canguilhem, Georges, 1952/2008).

Vitalism and naturalism are indissociable in Canguilhem’s account: medical vitalism expresses a distrust of the power of technique over life, rooted in Hippocratic confidence in the natura medicatrix (Canguilhem, Georges, 1952/2008). He demonstrated that vitalism’s fecundity was proved by its role in founding modern embryology (Wolff), discovering mammal eggs (von Baer), and developing reflex theory (Willis) — vitalists made authentic scientific contributions (Canguilhem, Georges, 1952/2008).

The Organism-Milieu Relationship and Goldstein

Canguilhem drew extensively on Kurt Goldstein’s clinical and theoretical work to develop his account of the living being’s relationship to its milieu. Biological knowledge itself is a creative activity analogous to the organism’s own activity of interacting with its world; as Goldstein articulated, the cognitive movement of the biologist mirrors the organism’s own learning (Canguilhem, Georges, 1952/2008). Goldstein’s critique of mechanist reflex theory showed that a reaction is always a function of the organism’s orientation toward the whole situation; isolated stimuli have meaning for science but not for the living being’s own sensibility (Canguilhem, Georges, 1952/2008).

To live, Canguilhem wrote, is to radiate: to organize the milieu from and around a center of reference, so that the organism-milieu relationship is one of constitution, not reception (Canguilhem, Georges, 1952/2008). Health is a life of flexion and suppleness in relation to the milieu; a life forced to struggle against the milieu is already threatened, and the laboratory situation, where the milieu is imposed, is the archetype of a pathological situation (Canguilhem, Georges, 1952/2008). Man creates new milieus rather than passively submitting to environmental changes; there is no pure natural selection in the human species to the extent that cultural and technical activity constantly alters the conditions of human life (Canguilhem, Georges, 1952/2008).

Reception and Legacy

Foucault identified Canguilhem’s central philosophical move as treating life itself as a normative activity — not as neutral substrate but as value-positing force — making his vitalism not a biological metaphysics but an epistemological strategy (Canguilhem, 1978). This interpretation has been widely influential in connecting Canguilhem to contemporary debates about biological normativity and the philosophy of health.

Barthez’s definition of the vital principle explicitly claimed Hippocratic lineage while broadening the term beyond Hippocrates’ “impetum faciens” (Canguilhem, Georges, 1952/2008).

Canguilhem’s influence on contemporary philosophy of medicine extends in two directions. Foucault’s Birth of the Clinic (1963) traced how the normal/pathological distinction became entrenched in the epistemological framework of modern medicine, a project continuous with Canguilhem’s own critique (Unknown, unknown). In the analytic philosophy of medicine, Canguilhem’s argument that health cannot be equated with statistical normality — that health is instead adaptive capacity, or as he put it, the ability to fall sick and recover — has been positioned against Boorse’s Biostatistical Theory as the foundational statement of the normativist position on disease concepts (Unknown, unknown).

Ideology and Rationality in the History of the Life Sciences

The 1988 collection refines and extends Canguilhem’s earlier work by introducing the concept of “scientific ideology,” a category distinct from both false science and genuine science. False science, Canguilhem argues, is characterized by an inability to encounter falsehood: it never renounces anything, never has a history, and its assertions cannot be falsified [cang-ir88-ch01-002]. Scientific ideology is different. It has a history, it imitates the norms of some already-constituted scientific discourse in an adjacent field, and it comes to an end when a genuine discipline operationally demonstrates validity in the same domain, displacing the ideology by exclusion [cang-ir88-ch01-001]. Scientific ideology always presupposes the prior existence of scientific discourses in an adjacent field: it cannot arise without there already being some science in a neighboring domain whose prestige it recognizes and whose style it imitates [cang-ir88-ch01-003]. The key move is the distinction from Marxist “ideology”: where Marx’s concept involves the inversion of the knowing relation (directing attention away from its proper object), scientific ideology involves presumption about conclusions not yet proven and displacement as to where genuine science will eventually emerge [cang-ir88-ch01-007]. When the genuine science arrives, it does not occupy the location the ideology had predicted; the science discovers more complexity where the ideology had found simplicity [cang-ir88-ch01-004].

The historian of science who fails to recognize and incorporate scientific ideology into the account risks producing a history that is itself ideological (a false consciousness of its object), because critical knowledge requires the historian to actively construct, not simply approach, the object of study [cang-ir88-ch01-008]. Canguilhem’s relationship to Bachelard’s epistemology is explicit here: Bachelard’s distinction between obsolete and valid science confirms the need to study scientific ideology precisely because what is now obsolete was once considered objectively true [cang-ir88-ch01-009]. Canguilhem adds a further distinction that is methodologically necessary: scientific ideologies must not be confused with the ideologies of scientists, meaning the philosophical doctrines propounded by scientists speaking outside their domain of competence. Eighteenth-century concepts of Nature and Experience, for example, were ideological concepts of scientists rather than scientific ideologies in the strict sense [cang-ir88-ch01-010].

Mendel’s genetics did not extend the trail blazed by pre-Mendelian hereditary ideology; that ideology followed multiple social and legal concerns that had no bearing on Mendel’s scientific problem [cang-ir88-ch01-005]. Spencer’s evolutionist ideology justified industrial society against traditional society and workers’ demands, was partly antitheological and antisocialist, and served as a Marxist ideology that concealed its true purpose [cang-ir88-ch01-006].

Chapter two applies this framework to John Brown’s Elementa Medicinae (1780) as its principal case study. Brown’s system reduced the medical art to two therapeutic acts (stimulation and debilitation) based on the single principle of organic incitability [cang-ir88-ch02-001]. Canguilhem traces the intellectual derivation of the system: Brown took from Cullen’s concept of neurosis the idea that the nervous system is the source of vitality and that nearly all human diseases are diseases of the nerves, but simplified Cullen’s framework by asserting that every organism possesses a finite and fixed quantity of incitability as an axiomatic force, without justification or evidence [cang-ir88-ch02-003]. By this axiom, Brown then unified all vital phenomena under the claim that “the same powers produce all the phenomena of life,” collapsing the animal and the vegetable, medicine and agriculture, nerves and muscles, health and sickness into one [cang-ir88-ch02-004].

In Germany, Schelling and Roschlaub incorporated Brown into Naturphilosophie by identifying excitability with cosmic magnetism, giving the system metaphysical amplification it could not have generated on its own [cang-ir88-ch02-002]. France resisted it: the Montpellier tradition’s vitalism, Stahlianism, and Bichat’s teaching that life is a set of functions that resist death were philosophically incompatible with Brown’s claim that “life is not a natural but a forced state” dependent entirely on external powers [cang-ir88-ch02-005]. This differential reception shows how a medical ideology can be amplified by philosophical alliance and resisted by philosophical incompatibility [cang-ir88-ch02-002][cang-ir88-ch02-005].

Canguilhem notes that the concept of irritability from Glisson (1654) through Haller and Bichat was ambiguous and contested [cang-ir88-ch02-007]. Brown, by reverting to Glisson’s simple equivalence of irritare, incitare, exstimulare, and vigorare, ignored this historical complexity [cang-ir88-ch02-007]. The system’s success is attributed to its extreme simplicity and beneficial practice changes, while its inadequacies include excluding organ states and neglecting external causes [cang-ir88-ch02-008]. Some contemporaries noted that a physician could be trained under Brown’s system in four weeks, and they compared Brown to ancient methodist physicians [cang-ir88-ch02-009].

Canguilhem also traces the afterlife of Brown’s core principle (the identity of normal and pathological phenomena) through Broussais, Magendie, and eventually Claude Bernard, where it hardened into what Canguilhem calls a medical ideology of unlimited therapeutic power, hostile to Hippocratism [cang-ir88-ch02-006]. Canguilhem’s own earlier study on Brown’s excitability theory, presented at the 1974 International Congress of the History of Science, had already traced the significance of Brunonianism for the normal/pathological debate, making chapter two of Ideology and Rationality a continuation of a long engagement (Canguilhem, 1966).

The distinction between biological fact and pathological value appears in one of Canguilhem’s notes with notable clarity in the context of infection. An “inapparent infection” is detectable by antibodies in the serum (a biological fact, a modification of the humors), but it is not an inapparent disease. Infection modifies the humors; disease requires clinical manifestation. This distinction reinforces the broader thesis: not every biological deviation from type constitutes a pathological state in Canguilhem’s normative sense (Canguilhem, 1966).

Chapter three moves from a single system to a structural argument: the nineteenth-century medical systems were not displaced by superior medical theories but succumbed to a revolution in the art of healing brought about by chemistry, a science that, in Berthelot’s phrase, created its own objects rather than describing pre-existing ones [cang-ir88-ch03-001]. Canguilhem begins with the fall of Broussais, whose physiological medicine reached its peak around 1830 but did not survive the cholera epidemic of 1832: statistical methods introduced by P.-C.-A. Louis had already demonstrated that antiphlogistic therapy was no more successful than rival treatments, and the death of prime minister Casimir Perier finally discredited the systematic use of antistimulant physiological treatment [cang-ir88-ch03-002].

Claude Bernard’s experimental medicine was itself, on Canguilhem’s account, a medical ideology whose four components (theories, progress, determinism, and action) mirrored the progressive ideology of mid-nineteenth-century industrial society [cang-ir88-ch03-003]. Magendie provides an instructive limiting case: this pioneer of experimental physiology failed to understand anesthesia, vehemently opposed its use in surgery, returned from London in 1832 convinced that cholera was not contagious, and approved anticontagionist reports. Genuine experimental achievement and scientific ideology, Canguilhem shows, can coexist in the same career [cang-ir88-ch03-004]. Bernard’s conviction that diseases were fundamentally nervous phenomena prevented him from taking cellular pathology (Virchow) or germ pathology (Pasteur, Koch) seriously [cang-ir88-ch03-005]. The real therapeutic advances came from outside the trajectory Bernard’s ideology had predicted, originating in crystallography and histological staining rather than in physiology [cang-ir88-ch03-006][cang-ir88-ch03-007].

Ehrlich’s project, Canguilhem observes, depended on prior developments in adjacent sciences that no medical theory could have predicted: aniline dyes became available only in 1856 (Perkin), and structural organic chemistry only in 1865 (Kekule). Without these, chemotherapy was not merely impossible but inconceivable in Magendie’s time [cang-ir88-ch03-008]. Surgery was transformed in the same period by antisepsis: Lister recommended systematic use of carbolic acid in 1867, only twenty years after Semmelweis was forced to resign from the Vienna obstetrical clinic for the simple act of requiring students to wash their hands [cang-ir88-ch03-009]. The chapter closes with a principle Canguilhem considers general in the history of science: discord within the scientific community never totally impedes communication, since every discipline draws on the same stock of instruments and raw materials. Bacteriology could not have been produced in the nineteenth century without owing something to the theories it helped relegate to ideology [cang-ir88-ch03-010].

Chapter four offers a genealogy of the concept of biological regulation from Lavoisier through Bernard to Cannon and Wiener. The argument is that “regulation” carried fundamentally different meanings at each stage. For Leibniz, regulation was an inherent property built into a system from the beginning: regularity, not the triumph over instability [cang-ir88-ch04-002]. Lavoisier was the first to apply the term to the animal machine, identifying three main regulators (respiration, perspiration, digestion) but framing them within a strictly conservative framework in which nature had “set regulators everywhere” in a physical order already in equilibrium [cang-ir88-ch04-003][cang-ir88-ch04-004]. Comte represents a transitional stage: for him, regulation came from the exterior. The stability of the solar system stabilizes living systems via their environments, and madness occurs when “the outside cannot regulate the inside” [cang-ir88-ch04-005]. Bernard inverted Comte’s schema entirely: for Comte, regulation meant stabilization by a stable external environment; for Bernard, it meant an internal mechanism enabling the organism to compensate for environmental deviations [cang-ir88-ch04-006]. In 1867, Bernard reported a discovery he considered “unparalleled in physiology,” the identification of a “nervous autoregulator” determining the work of the heart, which completed the physiological picture of a feedback mechanism [cang-ir88-ch04-007]. The conceptual line Bernard → Cannon → Rosenblueth/Wiener is what eventually produced cybernetics from a physiological concept, with “cybernetics” itself coined by Ampere in 1834 but lying dormant until the theory arrived to use it [cang-ir88-ch04-001].

After Bernard, “regulation” entered the common vocabulary of physiology, and within three decades expanded from cardiac and calorific mechanisms to the entire domain of biological functions, a broadening marked by Driesch’s Die organischen Regulationen (1901), the first book to give the concept title-level treatment in the plural [cang-ir88-ch04-010]. The Malthusian concept of population regulation is a revealing lateral case: Malthus borrowed the Hippocratic vis medicatrix metaphor explicitly, describing the “desire to improve one’s lot and the fear of making it worse” as “the great vis medicatrix rei publicae” and “a powerful principle of health,” conflating biological and social order under one conservative schema [cang-ir88-ch04-009]. Canguilhem notes that this entire genealogy occurs against the backdrop of a fact all eighteenth-century medical theories except Brown’s acknowledged: the vis medicatrix naturae, the organism’s inherent power of restitution, which pre-scientific medicine recognized before any experimental proof of regulatory mechanisms [cang-ir88-ch04-008].

Chapter 5: On the History of the Life Sciences since Darwin

The fifth essay surveys the life sciences from Darwin to molecular biology, deploying the framework of scientific ideology to diagnose the uneven and non-linear reception of Darwinism. Canguilhem opens with a structural observation about intellectual history: Newtonian cosmology, though the first major defeat of anthropocentrism, paradoxically delayed the second defeat (Darwinism), because it framed natural phenomena in terms of providential adaptation. The idea that species could be transformed by random adaptation was inconceivable until the idea of a preordained adaptation of each species to its way of life had first been destroyed [cang-ir88-ch05-001].

Darwin’s theory of natural selection was initially regarded as an ideology by many, not all of whom were fools [cang-ir88-ch05-002], because it was exploited by Spencer’s social evolutionism [cang-ir88-ch05-002]. Darwin’s theory was only later confirmed, in ways unforeseen even at the end of the nineteenth century, by population genetics [cang-ir88-ch05-002].

One of Darwin’s most original contributions, on Canguilhem’s reading, was the introduction of time as a biological agent. Fossils were petrified time, embryos operative time, rudimentary organs retarded time: together, these constituted the archives of biological history. Classification ceased to be a static portrait of coexisting forms and became a synoptic canvas woven of threads of time; the common ancestor replaced the archetype [cang-ir88-ch05-003]. Yet Darwin’s failure on the question of heredity illustrates a general epistemological principle: theory must guide practice, not the reverse. Darwin confounded generation and heredity and never used hybridization to study transmission, because he lacked the theoretical specification that would have told him what data to collect. On this question, Canguilhem judges, Darwin remained a man of the eighteenth century, the distance from Maupertuis to Darwin being smaller than it appears [cang-ir88-ch05-004].

Pasteur’s refutation of spontaneous generation paradoxically impeded Darwinism: by insisting that like produces like and nothing else, Pasteur heightened skepticism about any doctrine of species transmutation [cang-ir88-ch05-007].

Mendel is the chapter’s central epistemological puzzle. He fits none of the standard historical categories: he was not a precursor (he followed the trail all the way to the end himself) and not a founder (a founder cannot be unknown to those who erect an edifice on the foundations he laid). His scientific work was, Canguilhem writes, like a premature infant that died because the world was not ready to receive it [cang-ir88-ch05-005]. The success of medical microbiology created a parallel blockage: because microbes were stamped with negative medical value, their positive value as objects of theoretical research for studying the laws of life was not recognized, and the birth of a biochemistry of microbes was retarded [cang-ir88-ch05-008].

The chapter’s account of molecular biology frames the Watson-Crick discovery as requiring not just genetics and biochemistry but x-ray diffraction crystallography, electron microscopy, and radioisotope tracing, because without these technologies researchers could not have shown that the conservative and innovative functions of heredity are embodied in DNA macromolecules [cang-ir88-ch05-009]. The DNA crystal itself, Canguilhem argues, is not encountered ready-made in nature but is a “superreal,” non-natural object: the product of considerable technical and theoretical labor, and the latest in a series of new scientific objects invented since the end of the nineteenth century (cellular extract, intermediate metabolite, Drosophila gene) [cang-ir88-ch05-006]. The chapter closes with the epistemological consequence of the entire sequence: because physicists and chemists had “dematerialized” matter, biologists were able to explain life by “devitalizing” it. Darwinism, which began as a descriptive science, became a deductive one; physiologists became mathematicians [cang-ir88-ch05-010].

Chapter six extends the normality argument of The Normal and the Pathological into the full history of biological thought, introducing what Canguilhem calls the principle of “thematic conservation.” The chapter begins with the invention of biology as a disciplinary name: Lamarck’s first use of “biology” appeared in his Hydrogeology (1802), and when he returned to the word in the preface to his Zoological Philosophy (1809), it was still conceived around an Aristotelian hierarchical series from most imperfect to most perfect animals. The new biology’s object was the same as Aristotle’s [cang-ir88-ch06-001]. This Aristotelian inheritance extended into anatomy and physiology through the term “organon”: Aristotle borrowed it from the lexicon of artisans and musicians (where it means tool and instrument) to designate a functional part of the animal or vegetal body, embedding an analogy between nature and art, life and technics, that the biological sciences preserved until at least the end of the eighteenth century [cang-ir88-ch06-002].

Normality is not a concept that biology invented; it is one biology cannot do without. Its history runs from Aristotle’s teleology, where form expressed through rough constancy constitutes a norm against which the exceptional is abnormal [cang-ir88-ch06-003], through Descartes’s mechanism, which could not eliminate self-preservation even when it tried [cang-ir88-ch06-004]. Stahl’s animism, which defended the irreducibility of the organism and the cooperation of parts governed by “natural autocracy,” left its mark on Bichat and through him on Claude Bernard, who acknowledged his debt to Bichat despite his published criticism of vitalism [cang-ir88-ch06-005]. Buffon’s theory of “internal molds” and “organic molecules” represents another moment in the sequence: organic molecules, Buffon maintained, were indestructible and accumulated from generation to generation in forms shaped by internal molds, but Buffon could not avoid associating anomalies with degeneration. It was left to Darwin to discover a natural mechanism for normalizing minor anomalies [cang-ir88-ch06-011]. Darwin’s reformulation introduced normality as a transitive relational capacity linking organism to environment and descendants rather than a fixed rule [cang-ir88-ch06-006].

Twentieth-century biochemistry reached, paradoxically, a conclusion opposite to what nineteenth-century organic chemistry had tended toward. Rather than abolishing the distinction between living and nonliving, it established that living things exist in unstable dynamic equilibrium, requiring the concepts of regulation and homeostasis for their unique understanding [cang-ir88-ch06-007]. The proliferation of “auto-” prefixed terms in modern biology (auto-organization, auto-reproduction, auto-regulation, auto-immunization) reflects this epistemological finding: living systems are open, non-equilibrium systems that maintain organization both because they are open to the external world and despite that openness [cang-ir88-ch06-008]. Molecular genetics grounded the normal/abnormal distinction at the molecular level: gene mutations that block chemical syntheses by altering enzyme catalysts are “errors in copying a text,” grounding the contrast between normal and abnormal in the structure of living things while confirming its irreducibility to purely physical concepts [cang-ir88-ch06-009]. The principle of thematic conservation (the persistent return of self-preservation, regulation, and normality across all epochs) reflects that sickness and death are biological problems, not physical ones, and that biology is consequently structurally different from physics in ways that no reductionist program has dissolved [cang-ir88-ch06-010].

See Also

• Normal and Pathological
• Normalization
• Vitalism
• Vis Medicatrix Naturae
• Michel Foucault
• Gaston Bachelard
• Claude Bernard
• Auguste Comte
• Kurt Goldstein
• Rene Leriche
• Biological Normativity

Sources

All claims cite evidence cards from:

• Canguilhem, G. (1966/1991). The Normal and the Pathological. Trans. C. R. Fawcett & R. S. Cohen. New York: Zone Books. [Source ID: canguilhem-normal-and-pathological-1966]
• Canguilhem, G. (1978/1991). On the Normal and the Pathological (Foucault introduction). [Source ID: canguilhem-normalpath-1978]
• Canguilhem, G. (2008). Knowledge of Life. Trans. S. Geroulanos & D. Ginsburg. New York: Fordham University Press. [Source ID: canguilhem-knowledgeoflife-2008]
• Canguilhem, G. (1988). Ideology and Rationality in the History of the Life Sciences. Trans. A. Goldhammer. Cambridge: MIT Press. [Source ID: canguilhem-ideology-rationality-life-1988]

Editorial Notes

Gaps the encyclopaedia compiler flagged for future evidence work, collected from inline markers in the body and frontmatter.

Life and Context

Reception and Legacy