Philosophy of biology in the nineteenth centuryJagdish HattiangadiTHE PHILOSOPHY OF BIOLOGYThe emergence of biology as a unified subjectStudents of history and of biology share a common delight: as they study the details ofany subject, they find a fascinating diversity of cases which far exceeds any preconceivedexpectations. But that is not their sole delight. Some will also see unifying themestherein, with coincidences that beg for explanation and leitmotifs which please theaesthete. Some scholars choose to stress the diversity, perhaps even the perversity, to befound in events in history (or, in biology, of living forms). Other scholars may feelhappier following the motto
e pluribus unum.There is no right or wrong to it, that one is a unifier and another a divider of forms. Weascribe these differences in scientific or philosophical temperament to individual style ortaste: some like a tidy story and others prefer a wealth of detail. Nor is this a matter ofrespect or lack of it for detailed facts. A grand unifier may study facts painstakingly whiletrying to unify (perhaps bending the facts, or perhaps ruefully admitting failure) while theperson with a predilection for detail may believe in it only in principle, leaving it toothers to dig them up and record them. Those who grant that details are important may doso either to prove that they fit into a grand scheme or to disprove that very point.This difference is, perhaps not unsurprisingly, more than a matter of style. It is also anissue of substance in the nineteenth century, both in history and in biology. It became atopical question whether there is meaning in unfolding events: whether history exhibitsfundamental laws, and whether there is a grand design to explain adaptation among livingthings. Do some things just happen, or is everything determined by a deep plan?Philosophers and scientists alike were on both sides of this celebrated debate, and on eachside there are examples of intellectuals of both temperaments.But the issue of substancewas not merely a matter of temperament, it was a matter of doctrine, of theory and of thefuture direction of thought.The main issue of substance which animates the philosophy of biology in thenineteenth century has its root in the seventeenth, in the difficulties faced by Galileo’smechanical conception of the universe. This conception of a world ruled by mathematicallaws of motion, or mechanics, became the central feature of the so-called scientificrevolution of the seventeenth century. It was proposed by Galileo consciously inopposition to the Scholastic conception of the universe, which was dismissed withPtolemaic astronomy as false. Galileo proposed that it was superseded by the newCopernican solar system, and proposed his new mechanics to replace the older physics ofthe scholastics.The subsequent success of the mechanical conception of the universe cannot beequated with the success of any one of the mechanical systems proposed to describemotion. Galileo’s laws were followed in turn by those of Descartes, Huygens, Newton,d’Alembert, Euler, Lagrange, Poisson, Laplace, Fresnel, Hamilton, Maxwell. At the endof the period we are studying there followed entirely new mechanical systems fromEinstein (relativistic laws of motion) and Schrödinger and others (quantum mechanics).The accepted laws of motion keep changing, sometimes in matters of detail and at othersin more fundamental ways, but the general idea of a mechanical universe conjoinsconsistently with any of them.What is common to all the mechanical conceptions of the universe following Galileo iswhat they seemed to exclude: certain aspects of the universe which were readilyunderstood as far back as in Aristotle’s or even in Plato’s accounts of the world seemedto be incomprehensible within a thoroughgoing mechanistic scheme. The existence offorms, the prevalence of purposes and the realm of morality seemed to lie entirely beyondthe mechanical conception of the universe. If we regard mechanics as forming the basisof a new comprehensive philosophy to rival the old Scholastic philosophies (as modifiedfrom those of Plato and Aristotle), then form, purpose and morality had to be understoodsomehow as part of the new mechanical conception. But how? There is no readyexplanation for the existence of any of these three. To resolve this difficulty there weretwo basic ways to proceed: with ingenuity we could develop mechanical models toreproduce the effect of forms, purposes and morality artificially; or we could devise aconception of the world in which form, purpose and morality are quite real, and whereinmechanics has a diminished role to play in our understanding.A thoroughgoing mechanist could dismiss forms as residing in the eye of the beholder,or (invoking the medieval doctrine of nominalism) as residing in the act of naming.Purposes could be denied to all animals, and restricted to the human psyche, within whichcan also be located the free will, allowing us the luxury as moral beings, apparentlydenied to animals, of being naughty (domesticated animals being interesting exceptions).In this convoluted way the problems posed by forms, by purposes and by morality can bereduced to the mystery of the human mind. But having done so, we have only artificiallyisolated the recalcitrant Scholastic phenomena without having thus made any attempt atsolving the problems posed for mechanists. The thoroughgoing mechanist such asDescartes who throws all recalcitrant Scholastic phenomena into the category of mind isno better off as a mechanist than the one who, like La Mettrie, regards all thesephenomena as external, and explicable in material terms, without saying exactly how.Interaction between two substances is ruled out by Spinoza’s powerful argument that asubstance is by definition autonomous, and hence cannot be affected by anything whichis independent of it. The mechanical conception of the universe seemed to leave us withno option but to adopt one of these two alternatives: some form of materialism, or someform of idealism; we have either to seek an extension of mechanism to model andrecreate the effect of the recalcitrant Scholastic phenomena or to subsume all materialphenomena under the realm of ideas in a modified form.A titanic debate was touched off between Newton and Leibniz just prior to the latter’sdeath, which is found recorded in the Leibniz-Clarke correspondence [10.1]. Clarke, asNewton’s voice, describes a material world which is governed by mathematical laws, butwhich has many physical features for which there are no mathematical laws governingthem. Newton’s own view was that God acted upon the world from time to time topreserve it in its required form (the solar system, for instance, was unstable according toNewton, and continues without signs of collapse because God holds the planetsconstantly within their orbits). Leibniz, on the other hand, proposed a fully rationalistconception of the universe, in which everything and every event is determined by thePrinciple of Sufficient Reason. The world according to Leibniz consists of a communityof spirits (monads) each of which is completely determined in itself by its own nature.Each interacts successfully with other monads (or appears to do so, since any monad’sexperiences of other monads are completely predetermined by its own nature) onlybecause of the pre-established harmony by which God has established an order among allthings (monads).If we neglect the references to God, either because we no longer believe in any, or forthose who still continue to do so because we want to restrict ourselves to naturalphenomena, then the choice for us lies between these two schemes: Firstly, a world inwhich some things just exist and some events just happen without natural cause, and theyhave no natural explanation (Newton). Secondly, a world which is fully determined in itssmallest detail, and in which everything and every event within it is determined by aGrand Design which pre-establishes an otherwise inexplicable harmony (Leibniz).These two points of view—one materialist and the other idealist, one indeterministicand the other deterministic, one antithetical to Scholasticism and the other friendly tosome of its features, one seeking to understand what can be understood only in terms ofthe laws of matter and motion, and the other seeking to understand everything in terms ofa pre-established rational order—are not the only two possible ways to approach thesubject. But they did seem to many to offer the two most reasonable alternatives. Inbiology in the nineteenth century, however, are to be found some new ideas which cutacross these extremes. They resolved some of the difficulties which were raised withinGalileo’s mechanical conception of the universe.Thoroughgoing mechanists who avoided the relegation of all recalcitrant Scholasticphenomena to the mind, but accepted instead that forms and purposes were to be foundamong the phenomena (in short, materialists), had to find a way to understand form,purpose and morality in a material world governed by mechanical laws. A livelydiscussion between materialists and their opponents characterizes what we may identifyin retrospect as the condition of biology as it coalesced into a subject in the nineteenthcentury.Biology or the science of life arose as a unified subject among the discussions betweenFrench materialists and their opponents in the last years of the eighteenth century.Buffon, La Mettrie, Lamarck and their free-thinking contemporaries conceived a unifiedscience of life, or of biology. Such was the need for recognizing this new subject that itwas quickly taken up by many scholars of different persuasion across the scientific world.There is no doubt of course that the unity of biology was strengthened by the remarkablebut later developments in cytology, genetics and evolutionary theory all of which cutacross earlier divisions within the previously known sciences of living things. The priorunification of the sciences of life depends on this: form and purpose are to be foundamong all living things on earth, and barely outside of them. The need for a unifiedscience of life arises out of the need to find mechanical models for these two categoriesof recalcitrant Scholastic phenomena, all of which seem to be found in and about livingthings. These issues are central to ‘Modern Philosophy’ as this is taught in universities tothis day. If we take that as our cue, we may say confidently that the unification of biologywas a philosophical attempt to solve some central problems of modern philosophy.
On mechanism and vitalismThe schism in modern thought between the thoroughgoing mechanists and those whosought to put mechanism in its proper (and diminished) place in the grand idealist schemebegins with the clash between Newton and Leibniz. Unlike the seventeenth century, theeighteenth marks a deepening separation between natural philosophy and moralphilosophy. By the nineteenth century this division became established. The word‘philosopher’ came to be reserved for an apologist for idealism or perhaps an opponent ofthoroughgoing mechanism in any form, and the expression ‘scientist’ came into use forthe thoroughgoing mechanist who followed the experimental method of investigation.These two professions came to inhabit different parts of the university, and came to adoptdifferent curricula, and thus and only so were able to keep the peace.The mechanists adhered to three things: some form of the laws of motion to understandeverything in the world; a conception of scientific method as experimental and inductive;and a healthy scepticism about Scholastic issues which were dismissed as superstitious.It is an unfortunate fact that heated debates between scientists on the question of howto accommodate form and purpose within mechanism often led them to accuse oneanother of becoming unscientific. Two schools of thought exist within the group ofthoroughgoing mechanists early in the nineteenth century concerning how toaccommodate purpose in nature. One group sought to explain it by a special life force(attached to a kind of fluid, vital matter); another group hoped to explain it entirely interms of other known forms of matter and force, such as the magnetic, the electric, thechemical, etc. The first sought to identify all living forms by an ingredient common tothem, and the other sought to understand life in terms of organization (of the organism)from ordinary or inert matter. In principle either of these ploys might have been true (orof course neither might be true). As it happens, by the end of the nineteenth century theodds had swung in favour of the latter point of view, and in the twentieth century,whatever a scientist accepts she will regard vitalism as an intellectual oddity.Nevertheless, in the nineteenth century it would be premature to describe vitalism asunscientific as is done all too often today.One of the unfortunate terminological ambiguities of this debate lies in the descriptionof the opponents of vitalism as ‘mechanists’. In a certain sense, a vitalist is also amechanist who happens to believe in an additional element with a force much likemagnetic or electric forces as they were conceived at the end of the eighteenth century.The mechanists were, therefore, not necessarily the only mechanists in that conflict ofopinion, if by mechanics we mean the well known laws of motion accepted at the time asgoverning all matter.Galvani’s experimental and theoretical contributions were regarded as unscientific byVolta late in the eighteenth century, but we may wish to differ in our assessment today.Perhaps Galvani was not correct in arguing that since the severed leg of a frog had beenseparated from its organisation (i.e. within the previously live frog) its twitching whenprobed by two metal prongs shows that there resides in the severed limb of the frog theprinciple of living matter. Perhaps Volta was right that this was not even a credibleargument. But if it had not been for Galvani’s argument, would Volta have sought toshow that the twitch in the severed leg of a frog arose from the ordinary metals separatedby the ordinary acidic fluid therein? Would he have otherwise looked for an apparatus toduplicate his model of a schematic form of a frog’s leg being probed by two metalprongs, by inventing the voltaic pile? And would we have discovered current electricity,or the decomposition of water by electrolysis, or any of those remarkable things inphysics and chemistry which followed Volta’s invention of the pile and the discovery ofthe electric current?Moreover, the methods used by vitalists could be and often were thoroughlyexperimental. Bichat made an excellent case for vitalism by conducting detailed studiesof anatomical phenomena. Perhaps it is this more than anything else which led ClaudeBernard, the great physiologist and methodologist, to recognize some fifty years later thatwhile the experimental method needs preconceived ideas, it needs also a healthyscepticism (see pp. 292–5 below). Between the time of Bichat and Bernard, the tide seemto have swung against vitalism in physiology, though the full demise of vitalism had toawait the twentieth century. The debate between vitalists and mechanists is a fineexample of a thoroughly scientific controversy, in which experimental results and goodarguments played an important role in the eventual outcome. As the difficulties forvitalism mounted and those for mechanists diminished, the tide of opinion swung againstthe vitalists. It is frequently but not invariably true that predominance of opinion amongexperts is a good gauge of the strength of the case made for and against the opinion, andin this instance the correlation seems to be quite good.The factors which led to the decline of vitalism in the nineteenth century lay bothwithin and without biology. In physics, there was an eighteenth-century consensus thatthe various forces of nature were distinct, and that each of them emanated from acharacteristic and dis-tinct type of matter. For example, physicists thought they haddiscovered electric fluids which supported electric forces, and magnetic fluids whichsupported magnetic forces, and caloric, which induced heat. This is quite compatible witha living material supporting a life force. But in the nineteenth century the growingexperimental confirmation of the interchangeability of forces (or the unity of all force,later redefined and identified as ‘energy’) came to undermine the vitalist idea of a specialforce shared by all and only living things. Within biology, arguments leading to thedemise of mechanism had much to do with the rise and predominance of physiology asopposed to anatomy in the study of form. The close connection between physiologicalfunction and evolution in Darwin’s account eventually made vitalism an outsider toscience as Darwin’s views, and their improved descendants, came to occupy centre stagein biology.Although vitalism had its share of friends and opponents in the nineteenth century, itwas only after Darwin’s conception of evolution by natural selection was grasped thatvitalism came to fall in favour very generally. The heated debate over the merits ofDarwin’s theory of evolution by natural selection, however, was not between twoversions of mechanism as was the case between the vitalists and the so-called mechanists.The issue debated by Darwin was the very different one philosophers had raised as theproblem of design. He provided us with a mechanistic alternative to a grand design, or apre-established harmony, or to some form of idealism. If we regard the Frenchmaterialists’ location of forms and purposes in matter as basically correct (incontradistinction to interactionists who find them in the mind), then we may say thatDarwin’s theory of evolution by natural selection solves the mind-matter problem (or themind-body problem) of the Cartesian philosopher.
On biology as a development of the science of GalileoWhen Leibniz proposed his idea of a divinely pre-established harmony, he had in mindthe extraordinary coincidence that two monads which ‘interact’ have complementaryexperiences. In his account, two monads have two aspects of the same world. Eachmonad determines its own inner nature, and two ‘interacting’ monads might well havenot been co-ordinated, and thus they may fail to ‘interact’ at all. (They may not possessaspects of the
same world.) But in creating the universe, God created the best of allpossible universes, and thus created a universe in which everything which exists andevery event within each monad is there for a reason: were anything other than as it is, thiswould have been a different possible universe and therefore a universe inferior to thisone. This being the best of all possible universes it determines for us what there is inevery minutest detail—a determinism as complete as can be imagined—based on theassumption that anything else would not have sufficient reason to exist.It is a remarkable fact of the history of ideas that Leibniz, the author of moderndeterminism (i.e. the Law of Sufficient Reason), was forgotten as its true author by theearly part of the nineteenth century. Instead, the determinism invented by Leibniz isattributed to Newtonian mechanics. Perhaps it is not so surprising that this is so, after all.As Koyré expresses it:the world-clock made by the Divine Artifex was much better than Newton hadthought it could be. Every progress of Newtonian science brought new force forLeibniz’s contention: the moving force of the universe, its
vis viva did notdecrease; the world-clock needed neither rewinding nor mending.The Divine Artifex had therefore less and less to do in the world. He did noteven need to conserve it, as the world, more and more, became able to dispensewith this service.Thus the energetic God of Newton who actually ‘ran’ the universe accordingto his free will and decision, became in quick succession, a conservative power,an
intelligentia supermundana, a ‘Dieu fainéant’.Laplace, who, a hundred years after Newton, brought the New Cosmology toits final perfection, told Napoleon, who asked him about the role of God in his
System of the World: ‘Sire, je n’ai pas eu besoin de cette hypothèse.’ But it wasnot Laplace’s
System, it was the world described in it that no longer needed thehypothesis God.([10.4], 276)But if Leibniz’s determinism had taken over the Newtonian system of the world, it wasnot by providing mechanism with a new pre-established harmony. Laplace’s determinismis one in which the particles at a given time, with their positions and momenta, determineonce and for all the future and past states of the world; but this is not any explanation forthe existence of forms or purposes (real or apparent) other than in the form of anunencashable promissory note.The positions taken on these issues by scientists in the nineteenth century are variedand complex. The variety spans many intermediate forms between the two extremeswhich are represented here as mechanism and teleology. In the nineteenth century, bothphilosophers and scientists had become determinists. Only the form of determinismseparated the two. The mechanists had come to adopt
physical determinism, adeterminism by mechanical forces alone, whereas the philosopher followed Leibniz inseeing in the system (or harmony) of the world such things as forms, their essences,purposes and purposeful beings (spirits). All of these were found by the philosopher in asimple ‘common-sense’ examination of the world. The pre-established harmony that wasonce postulated between minds or monads was now postulated to explain the adaptationof life forms. In this manner the pre-established harmony came to be regarded as a granddesign in which physical organisms were adapted one to another. Even the description ofdevelopment in an organism, in the work of von Baer, for instance, shows allegiance toboth a mechanical perspective and a ideological perspective, tempting some moderncommentators to call it ‘teleomechanism’, though this term would also apply totheologically inclined strict mechanists of the period as well (e.g. the geologist Hutton).William Paley likened a living organism to a watch: if we were accidentally to find awatch, would we not postulate a watchmaker? The wonderful way in which all of theparts of the watch fit together allows us to infer that there must have been a watchmaker.In much the same way, the extraordinary manner in which living forms are adapted totheir very particular surroundings, and often unable to survive in certain very slightlychanged surroundings, suggests a designer. But it is not the existence of a designer whichis the problem, but of a design or a plan. Many mechanists were faithful believers. Theymight still find a pre-established harmony to be unpalatable, given the success of modernmechanical science.The existence of forms on earth had thus taken a particular twist late in the eighteenthand early in the nineteenth century: the fact that Scholastic forms are evident amongphenomena is embarrassing enough for modern science. But the extraordinarycoincidence that the various forms were uniquely fitted or adapted to their surroundingsmade them a double embarrassment to a mechanist. In order to respond to this challenge,Lamarck proposed his theory of evolution, in which all living forms metamorphose intoother forms as they strive to make best use of the environment. Over the ages, thus, thevarious forms come to fit their surroundings with all the appearance of design, without adesigner. (Other precursors of evolutionary theory and/or metamorphosis, include DavidHume, Erasmus Darwin and Goethe.)There were two difficulties with Lamarckian evolution: the concept of ‘striving’ wasstill a difficulty for a thoroughgoing mechanist. But this was not such a difficulty if onewere also the kind of thoroughgoing mechanist who might be called a ‘vitalist’. A morefundamental difficulty than that was a methodological flaw which arose in the usualdefence of its main thesis. We turn to this difficulty of the theory of evolution, which wassolved by Darwin.
Darwin’s reconciliation of scientific metaphysics and methodNewtonian mechanism was not merely a set of ever-changing mathematical laws ofmotion; it came also with a distinctive method. Newton espoused an experimentalmethod, which was both championed and practised with great success in many fields ofscience, old and new. The success of modern science was attributed to the adoption ofNewton’s experimental method, which enjoined a close study of the facts and arepudiation of all hypotheses. If we study the great debate between evolutionists beforeDarwin and their opponents concerning evolution, we find however that it is theiropponents who were clearly scientific in their treatment of phenomena.The celebrated debate in 1830 in the French Academy between Geoffrey St Hilaire,who defended Lamarck, and Cuvier, who ridiculed evolutionary theory, was a cleardefeat for the evolutionists. Cuvier was a remarkable comparative anatomist. He hadminutely studied the skeletal structure of a great many animals. On the basis ofmeasurements he was able to establish ratios between skeletal limbs of a great variety ofspecies. The minute and exhaustive studies of bones allowed him to reconstruct entireskeletons from a few fossil bones, sometimes from a single bone. Cuvier came to beregarded as the Newton of comparative anatomy, for such were his accomplishments inthat subject.In 1812, Cuvier had come forward boldly to assert that a study of the fossil record fordifferent ages revealed that the earth had entirely different species from time to timeinhabiting it. This evidence, he claimed, was compatible only with the catastrophicdestruction of species and subsequent creation of a new collection of living forms duringsuccessive ages on earth. The facts he collected left him no other choice that he couldimagine but cycles of destruction and special creation.The evolutionists (e.g. Lamarck or his follower Geoffrey St Hilaire) could notchallenge Cuvier’s claims about the existence of these very different species insuccessive ages on earth. Cuvier’s scientific method was too rigorous to allow thatcounter-attack. His techniques continue to be useful in palaeontology and comparativeanatomy even today.The best that an evolutionist could do was to suggest a possible mapping betweenspecies in one age and the next to show how one set of species might havemetamorphosed into an entirely different one in the intervening period between twofossilizations. But so far apart were the species in different ages that the suggestedpathway from one form extant in one age to one extant in another was sometimesfantastic. Evolutionists seemed to be dreaming up their models of change, where thehard-nosed scientist who studied the established facts would be fully on the side ofCuvier, the consummate observer of minute and detailed facts about anatomy in differentspecies.For a philosopher of biology this debate is instructive not only because we now admirean idea that was once regarded as an unscientific speculation. It is also instructivebecause it is a clash of opinion between an ingenious metaphysical solution to a dilemmaof Newtonian mechanics and a rigorous application of Newtonian method. The facts are,as they were then known, clearly investigated by Cuvier in a scientific manner. Theevolutionist, on the contrary, appears to be a dreamer, a myth maker. The Lamarckianhypothesis defended by Geoffrey St Hilaire seems to be methodologically flawed,because it is an hypothesis (of the kind that Newton would not feign), and not anexperimentally established statement, such as Cuvier’s.This clash of opinion is of seminal importance. Lyell formulated a ‘uniformitarian’geology in contradistinction to Cuvier’s catastrophic theory of the history of the earth.This clash changed the way in which uniformitarians and evolutionists approached theirsubject. The vicious attack on evolutionism on methodological grounds made subsequentuniformitarians and evolutionists nervous for decades about proposing hypotheses.Darwin would not publish until he had investigated a great many particulars, perhapsmore than absolutely necessary for the purpose of promulgating or defending their views.Darwin himself, whose evolutionary theory was of a different kind altogether,nevertheless immersed himself in the study of minute details of living things beforecoming forward with his views, which made his work both immensely richer and muchless accessible to contemporaries than it might have been. Some controversies followingDarwin’s writing may be attributed to a lack of understanding of its main features, whichwere sometimes not clear to everyone, so great was the mountain of factual evidence inwhich it was lost. But an echo of Cuvier’s earlier critique remains to this day: the fossilrecord, they say, is too incomplete to warrant evolutionary theory. The case for evolutioncannot be made with so many gaps in the evidence. In the methodological developmentsof the nineteenth century it became evident that this demand is too great to make of anytheory. Suffice it to say that the fossil record cannot support Cuvier’s view either. Ifevolution is fantastic, then repeated extinction and creation are equally fantastic, withoutfurther evidence.The critical feature of the evolutionary theory of Geoffrey St Hilaire and Lamarckwhich made it unsatisfactory is that this form of evolution did not allow for the extinctionof species. Darwin’s theory of evolution by natural selection does. The fossil record fromone age to the next may be very different; and it may be futile to seek a one-to-onecorrelation between a species in one age and its descendants in another. Darwin suggeststhat even if most species in an age become extinct, a few which survive will produce allthe variations which populate the next age. To take a popular example, it would be futileto look for an evolved successor today to every dinosaur which once roamed the earth.But there may well be one form
(archaeopteryx) from which have evolved all the birds oftoday. Cuvier believed in mass extinction during the ice ages, and Lamarck in noextinction but only evolution. Darwin’s theory is successful because it allows for almostcomplete extinction, and for evolution as well. Darwin’s theory of evolution by naturalselection postulates that there is a large amount of small variation in offspring. Hisproposal of blind variation combined with natural selection resolved the scientific andmetaphysical issue of how form and purpose may arise—or may seem to do so—in amaterial and mechanistic universe.In one theory, he was able to defend the mechanistic conception of the world in amanner which was compatible with detailed study of fact. Newtonian method andNewtonian metaphysics were reconciled, and a major philosophical problem for Galileo’sscience was solved.There is nowadays some confusion about the form of gradualism which is necessarilyentailed by Darwin’s theory and a form of catastrophism which is compatible with it:Darwin’s theory is often described as gradualist in the sense that there are no catastrophesin the history of the earth. This form of gradualism is not appropriately attributed toDarwinian theory. The hypothesis that the extinction of a very large number of dinosaurspecies in a very short period of time by the catastrophic event of a meteor striking theearth is certainly not a critique of Darwinian theory. But the evolution of subsequent lifeformswould be described by a Darwinian as descent from the surviving life-forms thenextant, and this evolution would be gradual in the specific sense that the evolution woulddepend on small variation within species and their differential advantages. WhereasLyell’s geology denied the assumption that there are regular catastrophes, Darwin’stheory describes how to do without periodic bursts of creation
de novo. On Darwin’saccount, the evolution of species does not show leaps of creation, though it may wellundergo rapid destruction of species in what may be described as catastrophes.Two of the three Scholastic phenomena which could not be fitted into mechanistictheory were reconciled with it in Darwin’s model. Forms (species) were described asmutable, but apparent at a given time; and the appearance of a great purpose or a granddesign was also feigned in nature by the almost adaptive character of living forms whichhad been naturally selected in their environment in competition with variations whichbecome extinct.
Continuing issues in the philosophy of biologyAfter Darwin successfully promulgated his theory of evolution by natural selection, anentirely new set of issues in the philosophy of biology came to light which were onlydimly realized before.Form and speciesOne of these issues concerns the classical idea of form. There may be a certain sense inwhich the mechanical conception of the universe challenged the idea of forms as the basisof all knowledge. Certainly, in some subjects in which mechanics was successful, formsceased to play an important role. Many were the subjects, however, which resisted theadvent of mechanical theories. The study of form in animals, plants and minerals(particularly crystals) left forms as a fundamental category for our knowledge.In classifying species of plants, for instance, Linnaeus and Buffon provided rivalschemata. Buffon’s nominalist scheme was more general and philosophical, whereas thatof Linnaeus, which stressed the essential qualities of species, was found much moreuseful in the practice of classification. Whichever system of classification one adopts, itis necessary for the practising natural historian, in order to classify things according totheir form, to presuppose that each species
has a characteristic form, which may becaptured in a typical specimen. Abnormal individuals may be found, of course, but thenatural historian had to guard against choosing one of them as a specimen, whichdifficulty prompted Buffon’s doubts about essential properties. An elaboratemethodology had been developed to pursue natural history to respond to these concerns,the recounting of which falls outside the scope of this essay.Darwin’s theory of evolution by natural selection undermines the theoretical basis forthis enterprise. Species, according to Darwin are not fixed but constantly changing. Thenormal situation is an abundance of variety in offspring. Thus in any species thecharacteristic form is not one but a multiplicity. Are there such things as species at all?(This is not quite the traditional problem of natural kinds and realism, though perhapsrelated to it.)Darwin did not deny that an examination of the flora and fauna around the earth wouldyield a knowledge of identifiably different species. As a young man he delighted incollecting beetles. He did not need to be reminded that identifying forms is what makesthe practice of natural history possible. His claim is a historical one: Darwin envisagedliving organisms as belonging to a single tree. As the branches fanned out, some lineswould come to an end (the forms would become extinct) but some would continue toflourish and would produce numerous varieties. Given a cross section of time the treewould project on a plane the characteristic grouping together of living things into species,genera, etc. as we find these in our records. But there are two provisos: Firstly, there arealways some variations, and these are the source of evolutionary change. Secondly, overtime, we recognize that species are mutable, and organisms from one species will be seento have a common ancestry (and therefore share formal similarity) with the most remoteof living organisms if only we are willing to go back far enough on the tree of life.Darwin’s theory eats its cake and has it, too. Species have characteristic propertiesmore often than not, because the process of natural selection may well isolate a form oflife as a species. This is what makes Linnaean natural history possible. But within anyspecies there are many small variations, which will, in the course of time, speciate.Because species change in this manner, each existing variation has equal right to beregarded as characteristic of the species—or better still, it is the variety whichcharacterizes the species. So Buffon is perhaps right after all in denying the existence ofessences to living forms.This conception of mutability challenged the idea of ‘sorts’ or ‘kinds’ as fundamentalto our understanding of living organisms. In the theory of collections or aggregates orclasses, it is possible to take any aggregate and regard it as a class. If any collection is aclass, is there anything special about a species considered as a class? Is there some way inwhich it is natural, and not artificial?From the old conception of morphology, it is only their form which binds similarorganisms into a species. But when we consider Darwinian evolution, a species must beunderstood as a class of organisms which share the ability to generate common offspring.Thus we find in Darwinian theory a criterion for describing a class, when it is a species,as a natural kind. There are no doubt difficulties with this. For one thing, inability togenerate offspring may be due to separation in time or by geography, which leaves it anopen question whether two separated groups belong to the same species when they do notgenerate common offspring, even though we may suppose that they could. On the otherhand some combinations of animals may have only sterile offspring, or have offspringwhich are sterile after one or two or more generations. For all these and still more reasonsthe notion of a species has become both richer and more troublesome since Darwin. Thecriterion of form or essence to identify specimens, however practical and useful, isusually undermined by the existence of variety, and has been seriously undermined as afundamental tool of biological thought.Quite recently, this issue has been raised again under the slogan ‘species asindividuals’, i.e. the idea that any one species is not a class of objects similar in somerespect but an organic unity. This twentieth-century discussion seems to contribute littleto what Darwin had already considered apart from obscuring the perfectly good notionsof class and of individual. The theory of classes allows any collection to be alsoconsidered as an individual if we so wish; one might even wish to say that that is itswhole point. Considering a species as an organic unity does not deprive it of its status asa class of organisms, any more than the class of cells within an organism is denied statusas a class because they are all part of one organism. In both cases the defining property ofthe relevant class may be a historical one. All (or almost all) the living cells in the bodyof Georg Cantor have the property of having descended from one fertilized egg from hisparents. They form a class none the less, as defined by that property, and Georg Cantorwas an individual all the same.History and determinismAnother fundamental issue of interest to philosophers to emerge from evolutionary theoryis the conception of an ‘open’ history. Darwin’s account of evolution included an idea ofsmall variation in great abundance which has been variously described as ‘blind’ and‘random’. The inability of the environment, or of the organism, to direct the variation inthe offspring is a very fundamental feature of Darwinian evolutionary theory. In order todistinguish the first view which was proposed by Darwin from a theory which allowsorganisms to have offspring which inherit the good acquired characteristics of the parent,the latter is often called ‘Lamarckian’. Darwin himself vacillated between givingLamarckian and what we may wish to call Darwinian accounts of adaptation. Climateinducedor environment-induced variation would be a third variety, which we may wishto call Lyellian evolution, to commemorate Lyell’s views on it in his
Principles ofGeology.What is interesting from a philosophical perspective about Darwin’s distinctive theory(even if he sometimes used other theories also) is that in his account there is an elementof chance in the evolution of species which cannot be eliminated. A chance event hereand now could have a profound influence upon the course of the future history of life onearth. Indeed, the entire history of life is a history of many chance events which producethe appearance of a pre-established harmony (or what is more neutrally called‘adaptation’).It may seem at first that this is a peculiarity of biology that it raises chance to such animportant level of fundamental principle, though we have seen it repeated latterly inthermodynamics and quantum mechanics. Although Darwin was not technically astatistician, his conception of the biosphere was a fundamentally statistical one. Histheory gave the statistical view, itself adumbrated earlier in the nineteenth century,considerable scope for development, although this development had to await the ‘newsynthesis’ of Darwinian theory of selective fitness with Mendel’s theory of geneticinheritance, and Waismann’s theory of the eternal or at any rate long-living germ line.Among nineteenth-century philosophers, Peirce is perhaps the only philosopher toadopt this indeterminist consequence of Darwinian evolution. There are manyphilosophical problems concerning indeterminism, statistics and probability, and chancethat are of interest to a philosopher of biology, though only in forms evident in thetwentieth century.MethodologyWhen Darwin wrote his
Origin of Species, there was almost universal agreement that anyscientific theory, to be successful, must describe the world as fully deterministic. Theindeterministic theory of Darwin with a prominent place for chance within it creates amethodological difficulty. Whether it is a methodologically satisfactory theory or not canbe asked while assuming that a theory must fit the model of theories in physics as thenconceived. In the nineteenth century this was not as great an enigma as it became in thetwentieth, when methodologists frequently worried about the methodological status andexplication of Darwinian evolutionary theory.In the nineteenth century, and indeed to this day, the central methodological difficultyraised about Darwin’s theory of evolution by natural selection is that the record of facts(i.e. fossils) is incomplete. Darwin’s theory may be described for that reason asunfounded, or poorly founded. Alternatively we may dismiss the methodology whichdemands so much of Darwinian theory or of any other, for that matter, as an unrealisticmethodology to adopt.Compared to methodologies propounded in the nineteenth century, theories of methodin the twentieth are much more varied and much less demanding. Foundationalism is indoubt today more than ever. It would be a mistake, however, to think that Darwin’stheory had a great deal to do with this change. All the evidence seems to suggest, rather,that methodology has developed more in response to problems in mathematics and inphysics, and less in response to those in biology, even though the present scepticismconcerning foundations fits Darwinian science extremely well.Although the influence on methodology of reflections upon the development ofevolutionary theory is minimal, the same is not true for reflections on the content ofevolutionary theory. Darwin was among those who realized that his theory implied thatall life including human life has evolutionary origins. In his
Evolution of the Emotions inAnimals and Man, Darwin sought to extend his theory explicitly to human feeling. Wealso know from his diaries that he regarded human intelligence and some critical ideas tohave been inherited, too. In fact a case can be made that Darwin was a follower ofWhewell until he became a Darwinian in 1839 when he realized that a substitute for whatWhewell had called fundamental ideas (which are not derived from experience) could beunderstood as having been inherited from our simian ancestry [10.3]. An evolutionaryepistemology promises to be one of the most fundamental and profound philosophicalconsequences of Darwinian evolutionary theory, though what it is exactly remainsundecided.MoralityThe mechanical conception of the universe still fails to accommodate one class ofScholastic phenomena, concerning morality. Where evolutionary epistemology is clearlyan interesting subject with much to teach us, evolutionary morality is, like a mechanisticconception of purposes in the seventeenth century, still enigmatic. Whereas Darwin’saccount shows how to do without a grand design, and how to explain the existence offorms among living organisms, there is in evolutionary theory as yet no satisfactorytheory of the existence of morality.There is of course ample room to account for the fact of the existence of mores amonggroups of people. Just as we can study different animals to study their mating, nesting orfeeding behaviour, so too we can observe humans in different groups and study themmoralizing. This might lead us to think that we have an evolutionary understanding ofmorality if we have some explanations of how they come to acquire their moralizinghabits, but that would be a mistake. The characteristic feature of morality is not that webehave in some way, or moralize in some way, but that we regard some behaviour asimmoral or wrong even as we practise it. What needs to be understood is how it comesabout that some things are wrong, or immoral, and not just why we so regard them.The understanding of morality from an evolutionary perspective certainly had acontroversial and well publicized attempt in the nineteenth century. One of Darwin’smost ardent admirers, Herbert Spencer, proposed a doctrine called Social Darwinism. Inthis doctrine the lesson for us from competing living forms as Nature evolves (red intooth and claw), is that the fittest survive, and the weak perish. Applied to the socialsphere it led to what was roundly attacked as an amoral and callous view of humansociety. Its popularity with some despicable political movements in the twentieth century(e.g. with the National Socialists, or Nazis) has left many intellectuals with a horror ofsocial theoretical biologists.But the issue of morality is squarely one which remains unresolved within the Galileanrevolution, and, however distasteful and misguided Spencer’s Social Darwinism, onemust give him and other intellectuals of the nineteenth century credit for recognizing thisas a fundamental difficulty of modern science which needs to be addressed. Indeed it isbecause an entire society under the Nazis, in the name of their entire society, espousingSocial Darwinist slogans, was so immoral as to practise systematic murder and genocidethat we have to ask not only how individual immorality is possible but also howcollective immorality is possible. No account of how actual mores are acquired orpropagated can explain this.As opposed to Spencer, who tried to extract a morality from the natural course ofevents as he interpreted them, G.E.Moore argued early in the twentieth century that anyattempt to derive a claim that something must be so based on the claim that it is so, is afallacy (what he called the Naturalistic Fallacy). His argument is that of whatever isdescribed as a fact we may still ask meaningfully whether it is good that it is so. Since wecan always meaningfully ask that question, we cannot identify the meaning of ‘good’with what is the case. Moore’s argument purports to make the realm of morality (and ofnorms and prescriptions generally according to later philosophers) independent of therealm of nature. How it may have come about that these realms are independent is adifficulty for naturalists.The situation at the turn of the twentieth century was that naturalism in ethics wasopposed to normativism, and the matter was unresolved, and so it remains to this day.
METHODOLOGY OF BIOLOGYOrigins of the subject and of some termsWhether there was any philosophy of biology in the nineteenth century is debatable: thereis as good reason to deny it as to assert it. The expressions ‘philosophy of science’ and‘philosophy of biology’ were invented in the nineteenth century by William Whewell.Were we to rely on that alone we would have to allow that there is such a subject by1840. But if we were to seek practising philosophers of biology, none comes to mind, atleast none who would self-consciously describe any of their work as belonging to such afield. The name invented by Whewell for this field came to designate something whichclearly exists only in the latter half of the twentieth century, with some writings ofT.Goudge and J.H.Woodger. Later, the writings of D.Hull, still later followed by a hostof interesting writers on the subject (Ghiselin, Ruse, Wimsatt, Sober), all of whom wouldbe happy to describe their relevant works as belonging to the philosophy of biology.To a modern historian of the philosophy of biology in the nineteenth century thiscreates an interesting question of choice: lacking a clearly defined field in the nineteenthcentury, one could dismiss it as non-existent. This implies that there is no philosophy ofbiology until concerns arising out of logical empiricism (a unique intellectual movementof the twentieth century) led to the birth of this subject. But this would belie the fact thatmany of the issues taken up today did arise earlier, as we have seen, however differentthe context in which they arose in the nineteenth century.The strategy which suggests itself is to pick out issues in the philosophy of biologytoday and to seek to present these very issues as they once emerged or developed in thenineteenth century. This is the strategy which has been adopted in this chapter. Theobvious difficulty with this strategy is that it may be prone to anachronism: how do weprevent our criteria of choice of issue from imposing our own concerns for those of thepast? To a certain extent this is unavoidable. In writing a history of philosophy in thenineteenth century, or in writing a history of the subject of history, there would be agenerally accepted sense
at the time in the period being studied that some things werewithin the field, even if today they were to be classified as belonging elsewhere. Issues inpsychology or in sociology, for instance, which arose in the early part of the nineteenthcentury would have to be classified as part of philosophy because they were so regardedthen. In this sense we cannot find a bench mark or a criterion of what would have beenpart of the philosophy of biology in the nineteenth century as seen by a contemporarythen. But to be forewarned was to be forearmed.The issues of the philosophy of biology may be divided into two kinds:methodological, and substantive. There is, as I shall soon suggest, an overlap there aswell.The substantive questions within nineteenth-century biology which are of concern tophilosophers of biology today may be classified into three categories: those connectedwith problems of evolutionary theory and related developments; those connected with theproblem of reduction of life sciences to physics and chemistry; and those related to theunderstanding of human beings in the light of modern biology. There are of course a hostof issues which may fit within or across these categories. All these issues were alreadycontroversial in the latter part of the eighteenth century and continue to attract interest tothe present day, and some of them have been sketched in the first section above.In studying the methodology of biology in the nineteenth century one could include thecommentators on science (Whewell, Bernard) or those involved in the practice ofscientific research who exhibit or are obliged to pronounce upon method (Cuvier,Pasteur, Darwin): the discussion of methodology in the practice of scientific research isgenerally a sign of a clash between defenders of different theoretical perspectives, all ofwhom attempt to use methodological considerations to buttress their respective cases.The second kind of methodological pronouncement is usually controversial, because itis made in the interests of controversy. The substantive issues in biology which stand outtoday as worth discussing are just those that were once the subject of controversy.Practical methodology is therefore closely bound up with the same substantive issues thatwe have identified as part of the philosophy of biology in the nineteenth century.There may also be a connection between the writings of abstract methodology and thecontroversies of the nineteenth century: Whewell’s work may be related to thecontroversies arising from substantive issues in physics (empiricism versus
a, prioriknowledge, for instance) and Bernard’s from those in medicine (anatomical as opposed tophysiological considerations in medical research). Nevertheless, the form of these selfconsciouslywritten methodological tracts differs from the others: the former must betaken literally as methodologies. The others are more casual and less systematic remarksuttered in the interests of other argumentation. We may sometimes disregard themethodological apologia and prefer instead to analyse the science in action. For thisreason there has been included, in the section below, a brief account of twomethodological treatises of the nineteenth century of particular interest to the philosophyof biology.Since the unification of biology is an important part of the story recounted here,perhaps some comments are in order about each of the subjects of history, philosophyand biology as they are found in the nineteenth century. The first two of these subjectstrace their origin to an era which is at least as early as that of the ancient Greeks—Herodotus and Socrates respectively being cited as their originators. (The words wereinvented then, but it is always possible to suggest that there were predecessors in oraround ancient Greece, or in another civilization prior to the Socratic invention of theword.) Philosophical history, a particularly influential conception of time and events inhuman history, seems to be an especially noteworthy product of the nineteenth century(Hegel, Comte, Marx). It is an open question which will not be taken up here what director indirect influence philosophical history might have had on biology.Unlike philosophy and history, biology is a comparative beginner. It is recognized as aunitary and integral subject worthy of a separate designation for the first time only late inthe eighteenth century. Many of the fields which are now part of biology as weunderstand it have a hoary history: zoology, botany, physiology, anatomy, as well ashosts of sub-disciplines like ornithology, entomology. They were well developed subjectsfor a long time before they came to be regarded as component parts of a single subjectidentified as the science of life, or of biology. It is an interesting fact about this newsubject, biology, that there is a philosophy of it according to us. In contrast, we wouldfind a philosophy of entomology or of botany to be unnecessary without furtherargument. It seems that there is a unity to biology which warrants a philosophy of it.Perhaps the thesis that the unity of biology is a unity of philosophical approach, promptedby the fundamental problems of modern philosophy, is not the whole story. But if it ispart of the story, it still makes philosophy much more central to the development ofbiology, and vice versa, than is generally supposed.
Two important methodological treatisesThe origin of the expression ‘philosophy of science’ may be traced to Whewell, whoproposed it in his book
Philosophy of the Inductive Sciences, Founded upon TheirHistory (1840). Book IX is entitled ‘The Philosophy of Biology’, which is part of thephilosophy of science.The advances which have, during these last three centuries, been made in thephysical sciences;—in Astronomy, in Physics, in Chemistry, in Natural History,in Physiology;—these are allowed to be real, to be great, to be striking: may itnot be then that these steps of progress have in them something alike?—that ineach advancing movement there is some common process, some commonprinciple?Then a little later he says, ‘if we can, by attending to the past history of science, discoversomething of this common element and common process in all discoveries, we shall havea
Philosophy of Science’ ([10.5], vi). In the opening section of Book I, philosophy ofscience is said to offer nothing less than a complete insight into ‘the essence andconditions of all knowledge, and an exposition of the best methods of the discovery of alltruths’.It is evident that a philosophy of science would include at least a methodology, and aninformed analysis of the history of science to exhibit that methodology. In addition itwould have to exhibit an insight into all real knowledge—a tall order indeed. Whewell’sown writings are remarkable for the insight he exhibits into diverse subjects and theirhistory, which few have matched.But when we turn to his account of the philosophy of biology, the reading isdisappointing (but no more than Mill, Comte or Spencer). There he lists five schools ofbiological thought, and a cursory account of some developments in physiology. We donot find any especially remarkable insight into the essence and conditions of biologicalknowledge, or even of the particular methods which may have made them successful.Instead, we find that when he can he applies the paraphernalia of a philosophy arrived atfrom the study of physics and mathematics to biology—his conception of a fundamentalidea not derived from experience, for instance, is inspired by Kant’s conception of
apriori synthetic judgement, introduced to show how mathematical knowledge is possible.Searching for a nineteenth-century figure who actually studied what Whewell mayhave called the essence and conditions of biological knowledge and who reflected uponthe process of discovery to extract some insight from it, we find only one book whichmerits our attention, Claude Bernard’s classic,
An Introduction to the Study ofExperimental Medicine (1865).Claude Bernard was one of the great physiologists of his day. His most memorableachievement perhaps was the discovery of the internal environment of animals, whichallows for an explanation of the comparative autonomy of some animals even thoughthey remain in constant interaction with environment. He had also made numerous andbrilliant discoveries in physiology before that, such as the function of the pancreas,animal glycogenesis, experimental production of diabetes, the existence of vasomotornerves, which are mentioned among other discoveries in Paul Bert’s introductory eulogy([10.2], v-xii).Claude Bernard’s work does not address biology as an integral subject. He dealsexclusively with physiology, and mentions anatomy. But his work is so centrally inphilosophy of biology as we now understand it that it cannot be left out of account.Bernard argues forcefully for the need to study not just form as in comparativeanatomy but function as well. And he suggests that in order to do so it is necessary notjust to observe organisms but to experiment with them. What, we may ask, is thedifference between observation and experiment?Bernard provides us with one of the most lucid and brilliant accounts ofexperimentation and experimental reasoning ever given. He begins by distinguishing theprocess of experimentation from that of observation. We take observation to be a passivegathering of facts, where in experiments there is an intervention into the process beingstudied, ‘a variation or disturbance that an investigator brings into the conditions ofnatural phenomena’ ([10.2], 5). But in distinguishing an observation from an experimentand both from experimental reasoning, he notes that the objective of an experiment is tounderstand a phenomenon from a perspective under our own control, ‘to reasonexperimentally, we must usually have an idea and afterwards induce or produce facts, i.e.observations, to control our preconceived idea’. ([10.2], 20).Bernard’s account of the experimental method in observations as well as inexperimentation is anything but passive. ‘Of necessity, we experiment with apreconceived idea. An experimenter’s mind must be active, i.e. it must question nature,and must put all manner of queries to it according to the various hypotheses whichsuggest themselves.’ And his account of the experimental method is this:the metaphysician, the scholastic, and the experimenter all work with an
a,priori idea. The difference is that the scholastic imposes his idea as the absolutetruth which he has found, and from which he then deduces consequences bylogic alone. The more modest experimenter, on the other hand, states an idea asa question, as an interpretative, more or less probable anticipation of nature,from which he deduces consequences which, moment by moment, he confrontswith reality by means of experiment.([10.2], 27)Bernard’s brief for a study of experimental medicine is an attempt to bring science to bearon a subject which he saw then asstill in the shades of empiricism and suffers the consequences of its backwardcondition. We see it still more or less mingled with religion and with thesupernatural. Superstition and the marvellous play a great part in it. Sorcerers,somnambulists, healers by some virtue of a gift from Heaven, are held as theequal of physicians. Medical personality is held above science by the physiciansthemselves; they seek their authority in tradition, in doctrines or in medical tact.This is the clearest of proofs that the experimental method has by no meanscome into its own in medicine.([10.2], 45)The conception of experimental medicine proposed by Bernard suggests thatexperimentation has exactly the same character whether we experiment on inorganicchemicals or on living tissue. Thus he argues that there is just one method for the study ofall living and non-living things, which is in direct contrast to the claims of some vitaliststhat living organisms provide exception to the general rules governing the study of dead(non-living) matter.While Bernard argued forcefully and lucidly for the unity of experimental method, healso pointed out that living objects must be treated differently from inorganic things.So far we have been explaining experimental conditions applicable to bothliving and inorganic bodies; for living bodies the difference consists merely inthe greater complexity of the phenomena…. But in the behaviour of livingbodies we must call the reader’s attention to their very special interdependence;in the study of vital functions, if we neglected the physiological point of view,even if we experimented most skilfully, we should be led to most false ideas andthe most erroneous deductions.([10.2], 87)Living organisms must be treated as a harmonious whole. And in this manner he arguesfor the need to do not only comparative anatomy but experimental medicine as well.The greatness of Bernard’s suggestions lies not only in the profound changes that heforesaw and helped advance in the profession of theoretical medicine but also his genuinecontributions to methodology, or to the philosophy of science as this subject had beenconceived by Whewell. Bernard’s analysis of the sceptical doubt which is used by theexperimenter without letting it get out of control, his defence of the need for preconceivedideas together with the injunction that we must be ready to abandon them as soon asnature turns recalcitrant—all these are so remarkable in capturing the essence of scientificmethod that it is one of the few books on methodology which continues to be read asprofitably now as when it was first printed.Compared to Bernard’s brilliant work, there is nothing else of interest in the nineteenthcentury, and perhaps even since then, in the form of a sustained methodological treatiseon the topic of experimental method in biology.
NOTEMy thanks to Professor Margaret Schabas and Mr David Clingingsmith for theircomments and assistance with the paper; the errors which remain are of course my
BIBLIOGRAPHY10.1 Alexander, H.G., ed.,
The Leibniz-Clarke Correspondence, Manchester: ManchesterUniversity Press, 1956.10.2 Bernard, C.
An Introduction to the Study of Experimental Medicine, 1865, trans.from the French by H.C.Greene, 1927, reprint New York: Dover, 1957.10.3 Curtis, R.
Charles Darwin and the Refutation of Whewellian Metascience: How ThePhilosophy of Science Learned from the History of Science, Ph.D. Dissertation, YorkUniversity, 1982.10.4 Koyré, A.
From the Closed World to the Infinite Universe, Baltimore: JohnsHopkins University Press, 1957.10.5 Whewell, W.
Philosophy of the Inductive Sciences, 2nd edn, 1847, reprint London:Frank Cass, 1967.