Whewell’s philosophy of science and ethicsStruan JacobsON SCIENCEIntroductionAmong the most prodigious of English minds of the nineteenth century, WilliamWhewell (1794–1866) was at various times, and among other things, philosopher,intellectual historian, scientist, educationist, theologian, economist, student of Gothicarchitecture, classicist. ‘Science is his [Whewell’s] forte and omniscience his foible’,quipped Sidney Smith. Born at Lancaster, son of a master-carpenter, Whewell won in1812 an exhibition to Cambridge University whose most famous College—Trinity—hewent on to serve continuously from 1817, initially as a Fellow then from 1841 as Master,to his untimely death from a riding accident.Whewell was intellectually eminent in his lifetime, then his reputation went through along eclipse. Interest in his work has, however, steadily increased over recent decades inan atmosphere more congenial to it, the intellectual shift—displacement of logicalempiricism by historically informed philosophy of science—associated with ThomasKuhn(The Structure of Scientific Revolutions)and others. Whewell believed, as now somany scholars believe, that the key to understanding the methods of science and thecharacter of its knowledge lies in history, rather than in formal analysis of propositionsand arguments.Whewellian scholarship has tended to concentrate on those texts that Whewell himselftook to be his most important,The Philosophy of the Inductive Sciences(1840 [2.3]) andits more concretely detailed companion,History of the Inductive Sciences(1837 [2.31]).1These were works for whose composition Whewell’s extensive interest in science—crystallography, mineralogy, geology, mathematics, astronomy, physics, biology,tidology, political economy—equipped him superbly.HistoriographyWhewell produced hisHistory of the Inductive Sciencesconcurrently with thePhilosophy.He envisaged a relation between the two works such that theHistorywould,besides providing a chronicle of scientific discoveries, empirically inform the otherinquiry. Philosophical study of the methods by which truth in science is discoveredrequires to be, which to date it has not been Whewell claims, ‘based upon a survey of thetruths’ already known to scientists ([2.1], 1:viii). In this respect Whewell sees theHistoryas a work without precedent, as he does also in its ‘point of view’ or ‘plan’.According to Whewell’s broad perspective, development of scientific disciplinestypically involves ‘preludes’, ‘inductive epochs’, and ‘sequels’. Certain disciplinaryhistories may reveal no epoch at all (thermotics, science of heat, for example); some areexhausted by a single case of the threefold sequence (about Newton’s law of universalgravitation revolves the solitary epoch of physical or explanatory astronomy, and roundthe undulatory theory of Young and Fresnel turns that of physical optics); others showtwo or more of these cases (formal astronomy, covering astronomical relations as distinctfrom causes, with epochal figures in Hipparchus, Copernicus and Kepler).How are inductive epochs, the periods chiefly constitutive of the history of science,identified by Whewell? Their most prominent feature is major intellectual breakthrough,achieved by the inductive method.No epochal discovery is ever made ‘suddenly and without preparation’ ([2.1], 1:10). Inthe course of preludes, by invention of hypotheses and by analysis, ideas gain in clarityand facts in precision; materials needful for discovery are so prepared. Among‘imperfect, undeveloped, [and] unconfirmed’ ([2.1], 2:370) preludial conjectures, someanticipate and ‘touch’ the great truths of epochs, as did Descartes’ vortex theory relativeto Newton’s theory of gravity.In sequels to inductive epochs champions of new theories overcome by argumentdefenders of traditional beliefs, winning over ‘the wider throng of the secondarycultivators’ ([2.1], 1:10). Implications of epochal discoveries are traced out, increasingtheir evidential support.Emphasizing ‘great discoveries’ in the development of disciplines, their concepts andvocabularies undergoing significant change, Whewell may be said to work in theHistorywith a concept ofscientific revolutions. And the term itself is in evidence, Whewellnoting the occurrence of ‘revolutions’ in ‘the intellectual world’ ([2.1] 1:9) and of‘revolutions in science’ ([2.1], 3:114).2 But his use of ‘revolution’ is attenuated, itsxconnotation including continuity between successive scientific theories, which, with thefact that ‘revolution’ commonly signifies sudden sweeping change, may explain whyWhewell is not at all times comfortable with the word. Discomfort, indeed denial,manifests in his remark that ‘earlier truths are not expelled but absorbed, not contradictedbut extended; and the history of each science, which may appear like a succession ofrevolutions, is, in reality, a series of developments’ ([2.1], 1:8). There coexist in thescience of Whewell’s historiographic depiction an element each of evolution andrevolution.3From Whewell’s account of scientific development as variations on the theme ofprelude, epoch, sequel, one is not to infer it is his view that science has undergonecontinuous development. He actually finds in the ‘history of human speculations’ ([2.1],1:11) times of stasis when science has languished. Book IV ofHistory of the InductiveSciencespresents as the major case the Middle Ages or, in Whewell’s phrase, the‘Stationary Period’. From the epoch of Hipparchus virtually to the time of Copernicus,science stood still. Inquiring into why this was so, Whewell notes how often during theseone-and-a-half millenniums ideas were left indistinct, and how rarely were theorieshauled before the tribunal of facts. These failures were results of deeper tendencies: rigidobeisance to intellectual authority, intolerance of dissent and an ‘enthusiastic temper’subjecting ‘the mind’s operations to ideas altogether distorted and delusive’ ([2.1], 1:81).Ideas and perceptionHistory of the Inductive Sciencescovers one side of the development of science, theobservational,Philosophy of the Inductive Sciencescovering the other, the ‘history of theSciences so far as it depends onIdeas’ ([2.1], 1:16). ThePhilosophycommences with abook on ‘Ideas in General’, books following on each of the better developed sciences:‘Pure Sciences’ (mathematics), ‘Mechanical Sciences’, ‘Chemistry’, ‘Morphology,including Crystallography’, and the like. Such sciences impress Whewell as richstorehouses of ‘unquestioned truths’. The second—theoretically much the moreinteresting—part ofPhilosophy,‘Of Knowledge’, chiefly consists of Books XI to XIII:‘Of the Construction of Science’, ‘Review of Opinions on the Nature of Knowledge, andthe Method of Seeking It’, and ‘Of Methods Employed in the Formation of Science’.It is contended by Whewell that sciences rest on ‘fundamental ideas’, each sciencehaving one idea or some combination of ideas uniquely its own. The ‘pure sciences’ (ofmathematics) are based on ideas of space, time and number; the ‘mechanical sciences’ onthat of cause; ‘palaetiological sciences’ (historical sciences, including geology andbiology) on historical cause. (An implication of this doctrine is denial of disciplinaryreductionism. It excludes the possibility of a fundamental bedrock science, of whose lawsthose of other sciences are functions and from which deducible. Each science is forWhewellsui generis.) To the development of each science ideas contribute ‘elements oftruths’. Elements of another kind also play a part, which dualism is nowhere brought outmore clearly by Whewell than in his account of perception. Analysing this will lead usinto his philosophy of science.Perception requiressensory impressionsof shape, surface, colour, and movement. Butthere must be in perception materials besides these, for sensations are formless,evanescent, disconnected and without assignable boundaries, whereas perception is ofenduringobjectswith specific properties, involved in definite relations. It is inferred byWhewell that mind is greatly involved, informing and fashioning the sensory flux.4Perception of objects with properties and in relations—discrete, spatially located, shaped,enduring, numbered, operated on by forces, resembling others—is the synthesis ofsensory presentations and mental emanations. Impressions on senses by phenomenawithout are materials to which form is given by ideas within.5 Sensations and ideas areone expression of that which Whewell describes as the‘fundamental antithesis’ofknowledge; elements which while mutually dependent are unlike in character and issuefrom opposite sources.Innate to the mind and developed over time through its intercourse with the world arepowers of forming fundamental ideas (space, time, number, likeness, causality, matter,force, substance, medium, symmetry and design) and of cognate conceptions. To exercisethese powers is to organize in definite ways the otherwise shapeless flux of sensations,impregnating it with assumptions. The product of the process is conceptually informedmaterial, sensations and ideas fused in perception.Whewellian ideas are nothing like the mental images or objects of thought to be foundin Descartes and Locke, but are close to Kant’s ‘forms of sensibility’ (space and time).Whewell’s account of mind and perception would appear to have been significantlyaffected by Kant. True, Whewell ([2.6], 334–5) describes his ‘main views’ as ‘verydifferent from Kant’s’ (instancing Kant’s denial of knowledge of noumena or things inthemselves), but he admits in the same breath to having ‘adopted some of Kant’s views,or at least some of his arguments. The chapters…on the Ideas of Space and Time in thePhilosophy of the Inductive Sciences,were almost literal translations of chapters in theKritik der Reinen Vernunft’. ‘Kant considers that Space and Time are conditions ofperception, and hence sources of necessary and uni-versal truth’ as indeed—except thathe recognizes more of such ‘conditions’ and ‘sources’—does Whewell ([2.6], 336).InductionWhewell’s philosophy of science builds on the foregoing analysis of perception,elaborating the theme of the fundamental antithesis. The philosophy of science includesamong its main components an account of induction as the process of scientificdiscovery, and standards for measuring which theories are true. The most accessiblesource of all this is Book XI, ‘Of the Construction of Science’, forming as Whewell([2.3], 2:3–4) puts it the ‘main subject’ ofPhilosophy of the Inductive Sciences,all earlierchapters being in relation to it ‘subordinate and preparatory’.6The materials or ‘conditions’ of knowledge Whewell in opening his account of itexplains are ideas and conceptions emanating from mind and—the other side of theantithesis—facts originating from observation and experiment. Scientific knowledge isexpressed in ‘exact and universal’ propositions. It consists in the application ofconceptions or ideas, distinct and ‘appropriate’ to some science (as is symmetry tomorphology, force to mechanics, and vital power to physiology),7 to numerous factspossessed of clarity and certainty.8 Asystemof scientific knowledge is said by Whewell([2.3], 2:3) to exist ‘when, facts being thus included in exact and general propositions,such propositions are, in the same manner, included with equal rigour in propositions of ahigher degree of generality; and these again in others of a still wider nature, so as to forma large and systematic whole’.Necessary for achieving scientific knowledge are a twofold process of preparation—‘explication of conceptions’ and ‘decomposition of facts’—and a method Whewelldistinguishes as ‘colligation’. Explication is undertaken to clarify conceptions and togauge their suitability to research projects. (In the terminology ofHistory of the InductiveSciencesit is ‘preludial’ work.) Controversy is prominent in this process. Vague andunfamiliar, new conceptions are objected to, placing on protagonists the onus ofelucidation and integration. It is Whewell’s judgement ([2.3], 2:8), a combination of‘whiggishness’ and hindsight, that all controversies respecting scientific conceptions haveended in victory to ‘the side on which the truth was found’. Under the broad heading of‘explication’ Whewell also includes disclosure of necessary truths as inhering in ideasand functioning as presuppositions of sciences. Ideas must be clear before necessity inaxioms can be discerned and their implications be traced out: ‘the distinctness of the ideais necessary to a full apprehension of the truth of the axioms’ ([2.2], 170).As conceptions call for ‘explication’, in the other part of the process of preparingmaterials for discovery, facts have need of ‘decomposition’. Facts properly decomposedare ‘definite and certain…, free from obscurity and doubt’ ([2.3], 2:26). Reduced tosimple elements and considered in regard to clear ideas and conceptions, experience maybe rid of impurities—the likes of myth, prejudice and emotion. Among the original basisof decomposed facts for astronomy, for example, were the moon’s recurrent phases,rising and setting points of the sun, and those regular intervals between which the samestars become visible at the same time of the year.After explication and decomposition the way is clear for ‘colligation’ as the reallycreative work of science. By exercise of imagination, the scientist endeavours to hit upona conception capable of expressing a ‘precise connexion among the phenomena which arepresented to our senses’ ([2.3], 2:36). In the case of Kepler’s third law, ‘Squares of theperiodic times of planets are proportional to the cubes of their mean distances from thesun’, Whewell finds elementary facts about the planets’ solar distances and durations ofyears colligated (literally, ‘bound together’) by such conceptions as squares of numbersand proportionality.Explication, decomposition and, above all, colligation are the constructive elementsforminginduction,from which process Whewell ([2.3], 2:47) derives all‘real generalknowledge’of the world, by which he believes are discovered all laws of nature.Whewellian induction has little more than its name in common with the processtraditionally so called. It is not generalization from facts but invention of an hypothesisusing a new conception(s) or an old one(s) transposed from some other context; thehypothesis being superimposed on, so as to bring order to, facts hitherto not perceived asconnected. Successful hypotheses depend on suitable conceptions, for invention of whichthere is no methodological rule, being fruit of sagacity and serendipity.Whewellian induction also takes in negative testing of hypotheses to strike those outthat are false. In coupling colligation with critique, Whewell presages by one hundredyears in a most remarkable manner Karl Popper’s image of science as ‘conjectures andrefutations’. Writes Whewell:This constant comparison of his [the scientist’s] own conceptions andsupposition with observed facts under all aspects, forms the leadingemployment of the discoverer: this candid and simple love of truth, whichmakes him willing to suppress the most favourite production of his owningenuity as soon as it appears to be at variance with realities, constitutes thefirst characteristic of his temper.[2.3], 56–7).9There is for scientists a temptation to cling on to hypotheses, setting their faces againstempirical violations, and Whewell believes it helps explain the ‘obloquy’ into whichhypotheses ‘have fallen’ ([2.3], 2:58). This detraction, and more so the misuse, Whewellregrets. When, however, science proceeds properly, an episode of invention and testcontinues for as long as it takes for an hypothesis to be framed to bind ‘scattered factsinto a single rule’ ([2.3], 2:41). Whewell is able to find no clearer illustration of theprocess in the history of science than Kepler’s successive invention and disposal ofnineteen hypotheses (oval, combinations of epicycles, etc.) as he struggled to describe theorbit of Mars, which he eventually succeeded in doing through his colligation with theconception of ellipse.John Stuart Mill, prominent among Whewell’s adversaries, critically considered inASystem of LogicWhewell’s account of the case of Kepler. Mill did this under the heading‘Inductions Improperly So Called’, indicating that to his mind what Whewell terms‘colligations’ and identifies with inductions are strictly speaking not inductions at all.‘Induction’ for Mill ([2.17], 288) ‘is the process by which we conclude that what is trueof certain individuals of a class is true of the whole class’. It is generalization, theinferring of general propositions, whereas by colligation he ([2.17], 292) understands‘mere description’, there occurring no inference from, no going beyond, the facts.That which Mill ([2.17], 297) identifies as the ‘fundamental difference’ betweenWhewell and himself over how Kepler should be interpreted centres on whether theellipse colligation was logically equivalent to, only a description of, Kepler’s experience.Mill ([2.17], 295) demurs to Whewell’s suggestion that Kepler ‘putwhat he hadconceived into the facts’. In Mill’s account, having the concept of ellipse, Keplerconsidered whether observed positions of Mars were consistent with an elliptical orbit,and realized they were. ‘But this fact, which Kepler did not add to, but found in, themotions of the planet…was the very fact, the separate parts of which had been separatelyobserved; it was the sum of the different observations’ ([2.17], 297).So far as implications for his own philosophy of science are concerned, Mill, if I maydigress to note a most interesting historical fact, devotes undue attention to the lesser oftwo associated problems. Let us put to one side whether Kepler’s third law is no morethan a description, and focus on the fact that descriptions do not exhaust Whewell’s classof colligations. Universal propositions happen also, even more so, to be included andthese, going beyond the facts, are by Mill’s standards generalizations not descriptions.Mill ([2.17], 297) unreservedly agrees with Whewell that the ‘tentative method’ of tryingconceptions is ‘indispensable as a means to the colligation of facts for purposes ofdescription’. To be true to his inductive philosophy Mill must now restrict Whewell’shypothetical or conjectural method to colligation in the restrictive (Millian) sense,ascribing to induction (generalization) all universal propositions of science. But this hefails consistently to do. Ultimately he wishes also to admit hypothesizing as a furthersource of general propositions. So it is one finds in Mill’sSystem of Logictwophilosophies of science: one in essence inductive, the other hypothetical. Of Mill’srecognition and comprehension of this second method Whewell, as I have elsewhereargued, was almost certainly an important source ([2.15]; [2.16], 128–32).ConsilienceWhewell considers that scientific hypotheses are susceptible of proof, which is a majordifference from Popper in whose view they are falsifiable only. We may convenientlydenominate Whewell’s conditions of proof as ‘static success’ and ‘progressive success’.Each condition is required, and the two together are sufficient, for proof.An hypothesis is advanced to explain some property or behaviour of the members of aclass of phenomena. The ‘static’ condition of proof demands of a theory successfulpredictions of facts of the same nature within its subject class. The condition is minimaland non-definitive, being always satisfied by true, as well as sometimes by false, theories.While the theory of Ptolemy accurately predicted planetary positions along with solar andlunar eclipses, eventually it was judged an inaccurate representation of the structure ofthe heavens. Is the achievement of any so-far predictively successful theory eversufficient to insure it against refutation later on? In Whewell’s other condition of successlies the answer, and it we see is affirmative.Whewell notices two forms of ‘progressive’ explanatory success (‘successivegeneralization’ being his term for it). One concerns theories to which hypotheses areadded. It may be the case that the hypotheses successively ‘tend to simplicityandharmony;…the system…[becoming] more coherent as it is further extended’. ObservesWhewell ([2.3], 2:68), ‘The elements…we require for explaining a new class of facts arealready contained in our system’, and for him this amounts to an infallible sign of truth.His prize example is the addition to Newton’s theory of suppositions of: the sun’sattractionto satellites to account for the motions of the aphelia and nodes; mutualattractionof planets to account for perturbations;attractionbetween earth, sun, andmoon to account for tides, for the earth’s spheroidal form, and for precession. The patternis unmistakably that of progressive simplification, says Whewell ([2.3], 2:70), severalhypotheses resolving ‘themselves into the single one, of…universal gravitation’. In casesof the opposite kind, being historically the more numerous, assumptions happen neitherto be suggested by nor to be reconcilable with theories to which they are appended. ToDescartes’ vortex theory were added hypotheses to explain elliptical planetary and lunarorbits, perturbations, and earth’s gravitation. But they were independent, arbitrary, andadhoc,and brought the theory into disrepute. By scientists using it to account for weights ofchemical compounds, phlogiston theory was supplemented with, to the detriment of itscredibility, the barely coherent notion that phlogiston is an element at once heavy andlight, which upon entering compounds serves to reduce their weight.Ad hocassumptionsare to Whewell a sure sign of false theories.The other form of ‘progressive’ success is‘consilience of inductions’. This is achievedwhen a theory T (‘AllCshave propertyH’), advanced to causally explain a confirmedgeneralizationL1(‘AllP1s have propertyA’), is later discovered also to give a successfulexplanation of a confirmed law about a class of phenomena, L2 (‘AllP2s have propertyB’),differentfrom that for which it was framed.10Tis advanced to explainL1on theassumption thatP1s areCs. T’s explanation is confirmed if, in testing P1s in variouscircumstances,His found to cause propertyA. This is the ‘static’ condition above. If—the salient characteristic of consilience of inductions—theory T is successfully appliedbeyond its intended domain to an (originally)unintendeddomain, such thatHis shown tocause propertyB,it follows thatP2s, likeP1s, are members of the class ofCs.‘Static’ and ‘progressive’ success together, Whewell says, ‘irresistibly’ establish atheory’s truth. As Whewell uses the same theories to illustrate both types of ‘progressive’explanatory success, the thought occurs that perhaps his ‘consilience’ and ‘progressivesimplification’ are different names for a single process. In the words of one commentator:Actually however, Whewell was not that sure that… simplification is all thatdifferent from a consilience. If we get a consilience, then one hypothesis isbeing used to explain facts from two different classes. And this, in a sense, iswhat simplicity is all about, for we are using a minimum to explain a maximum.([2.19], 231)It is clear from what Whewell says that scientists can never hope to induce (generalize)theories and laws from others already established. Conception is unpredictable;imagination strikes out in different direc-tions. Progressive cognitive expansion is nonethe less suggested. While no one knows in advance how to explain established laws, anew theory is required deductively to so do as a condition of its acceptance. A successfulWhewellian scientific discipline resembles in the manner of its growth the construction inlayers (theories) of an inverted triangle from apex to base. Explanation is in the oppositedirection, deductively descending.Suffice in closing this section to note that Whewell’s paradigm cases of progressiveexplanatory success—universal gravity and the undulatory theory of light—physicistscame in the fullness of time to judge as false. There seems nothing for it but to sayWhewell fails to establish a criterion of proof: predictive successes, no matter howoutstanding, are no indicator of future performance of theories.NecessityWhewell’s writings on necessity as a property of (at least some) scientific theories wouldappear to have given rise to greater interpretative disagreement than has his treatment ofany other topic. It will be shown that he theorizes the property in more than one way,which fact, not widely appreciated, may help explain the variety of interpretation.Whewell first addressed necessity in a paper ‘On the Nature of the Truth of Laws ofMotion’ (1834), included later inPhilosophy of the Inductive Sciences. Focusing onmechanics, science of ‘motions as determined by their causes, namely, forces’ ([2.3],2:574), Whewell identifies three necessary axioms:‘Every change is produced by acause’; ‘Causes are measured by their effects’;and‘Action is always accompanied by anequal and opposite Reaction’([2.3], 2:574–6). Newton’s laws of motion are respectivelyrelated to the axioms, the first axiom applied to motion yielding the first law, ‘when noforce acts, the properties of the motion [direction and velocity] will be constant’ ([2.3],2:577), and so on. In the same essay Whewell distinguishes each law of motion into‘necessary’ and ‘empirical’ parts. The first law, continuing here with the same examplefor ease of exposition, in its necessary part affirms that ‘Velocity [as a property ofmotion] does not change without a cause’, the empirical part affirming that ‘The time forwhich a body has already been in motion is not a cause of change of velocity’ ([2.3],2:591). In what for Whewell consists the ‘necessity’ of the axioms and laws of motion?He describes the ‘necessary’ parts of the laws of motion, and, by implication, the axiomsfrom which they derive, as ‘inapplicable’. But this adjective, suggesting analyticnecessity, is not a happy choice, seeing that Whewell goes on to describe the axioms notonly as ‘absolutely and universally true’ and expressive of ‘absolute convictions’ but asserving also to regulateour experience.When ‘looking at a series of occurrences…weinevitably and unconsciously assume’ the axioms’ truth, for experience we we are unableto ‘conceive otherwise’ ([2.3], 2:575).11 Axioms being necessarily true of, andindispensable to the reception of, experience, so must be the necessary element in eachlaw of motion. Each such element makes a true, albeit highly general, assertion about theworld. While the first law of motiontruly assertsthat change of velocity requires a cause,it is silent on what kind of causation is involved (inherent or extrinsic?) and is on thislevel, as Whewell would have it, ‘inapplicable’. It isa posteriorithat we find out that thecausation in question is external, not an inner power. The kind of necessity of whichWhewell ([2.3], 2:593) here speaks is instructive and, at the same time, strictly universal(the ‘all possible worlds’ variety) for the necessary part of a law of motion cannot ‘bedenied without a self-contradiction’. The axioms of mechanics and the non-‘empirical’ (read non-‘contingent’) part of each law of motion appear in Whewell’s 1834essay as synthetic-necessary truths.InPhilosophy of the Inductive Sciencesthe analysis of necessity is along broadlysimilar lines except that the doctrine of ideas is now brought into play. Mechanicalsciences in general rest on the idea of cause and its modes or conceptions offorce(mechanical cause) andmatter(resistance to force), with the special branch of‘dynamics’ depending on the axioms of cause and the laws of motion. What the 1834essay presents as axiomssimpliciterinPhilosophyappear as axioms grounded in andexpressing the idea of cause. The axioms are described as ‘true independently ofexperience’, meaninga priori,and as true ‘beyond the limits of experience’, meaningnecessary and synthetic. That the truth of axioms is ‘necessary’ in the robust sense of thatword is evident from the imperative mood of the first axiom:‘Every Event must have aCause’([2.3], 2:452). To laws of motion also necessity is ascribed, yet they are said tohave been ‘collected from experience’ ([2.3], 1:247; [2.3], 2:453). The air ofcontradiction that surrounds this combination of claims Whewell dispels by noting ineach law the presence of a non-necessary element, thecontentbeing true but onlycontingently so. It is their universality and vocabulary (‘cause’, ‘effect’, ‘action’,‘reaction’, etc.), in a word their ‘form’, in virtue of which the laws ‘exemplify’ thecorresponding axioms, and in which their necessity consists. Laws so formed Whewell([2.3], 1:249) regards as necessary in yet another way, drawing attention to an‘indestructible conviction, belonging to man’s speculative nature’ that there are laws ofmotion, ‘universal formulae, connecting the causes and effects when motion takes place’.Philosophy of the Inductive Sciences,in that part of the third edition titledPhilosophyof Discovery,presents knowledge of necessary truth as ‘progressive’, Whewell largelydevoting himself to making sense of this fact. Part of his explanation is that necessarytruth proves accessible only to minds properly prepared. ‘To see the truth and necessityof geometrical axioms, we need geometrical culture’ ([2.6], 347), just as necessity inaxioms of mechanics remains opaque until, their discipline developed, scientists acquirethe right culture. While scientific disciplines (mechanics, chemistry, geometry, and thelike) could never have been raised other than on the basis of axioms, ‘internal conditions’of experience and knowledge as Whewell calls them, the knowledge that axioms arenecessary truths is acquired later.Whewell goes on to claim that in (but not only in) mechanics, his exemplar ofexcellent physical science, axioms encapsulate ‘in a manner’ the science in its entirety.‘The whole science of Mechanics is only the development of the Axioms concerningaction and reaction, and concerning cause and its measures’ ([2.6], 357). For this boldclaim Whewell offers no effective support;12 his present interests not including analysisand defence of his doctrine of necessity.It has been indicated above that ‘Laws of Motion’ andPhilosophy of the InductiveSciencesboth treat the axioms of mechanics as wholly necessary and the laws of motionas necessary in form and contingent in content. In that account the laws are not deduciblefrom axioms. By implication, there being in what Whewell says nothing to suggestotherwise, the distinction between necessary and non-necessary truth is an ontologicaland absolute one, based on objective properties.13 ButPhilosophy of Discoverypresents adifferent doctrine altogether. As the entire science of mechanics is able to be exfoliatedfrom axioms (once the laws are discovered), the distinction between truths necessary andtruths non-necessary in mechanics must in this case be drawn onlyrelative tothe state ofknowledge (level of cultivation), all truths of mechanics in reality being, and eventuallyrecognized as being, necessary.As noted, Whewell on the subject of necessity inPhilosophy of Discoverywantsspecifically to explain the progressiveness of knowledge of necessity: how propositionswhose truth is in the first instance knowna posterioricome to be knowna priori(asnecessarily true). His explanation in its first part revolves around development ofdisciplines and cultivation of minds. Its second part presents the proposition thatstructural identity must exist between mind and the world. Ideas exist in God’s mind, andthese he has impregnated in the human mind as well as constituting the world accordingto them. Consequences of ideas in the world are laws of nature, which coincide withconsquences of ideas in the mind (propositions), consequences of both descriptions beingnecessary. This same general view Whewell inPhilosophy of Discoveryextends to ethics,postulating the existence of ethical ideas that coincide with ideas in God’s mind. We nowturn to Whewell’s ethical doctrines.ON ETHICSIntroductionWhewell has been credited with having revived in the nineteenth century ‘the study ofmoral philosophy at Cambridge’, being at the head of an impressive ‘line of Cambridgemoralists—John Grote, Henry Sidgwick, G.E.Moore, C.D.Broad—whose outlook,despite important differences, is strikingly similar on certain central issues’ ([2.21], 109).Their historical importance notwithstanding, Whewell’s ethical works have in thetwentieth century been almost entirely neglected.14Whewell began publishing on ethics in 1836 with a ‘Preface’ to James Mackintosh’sDissertation on The Progress of Ethical Philosophy.On the Foundations of Morals, FourSermonsappeared in 1837. The following year, appointed Knightbridge Professor ofMoral Philosophy (his second chair, that of Mineralogy having been held by him from1828 to 1832), Whewell delivered twelve lectures which later appeared asLectures onthe History of Moral Philosophy in England(1852, [2.5]). Certain themes of theselectures form a backdrop to theopus magnumof Whewell the moralist,Elements ofMorality, including Polity(1845, [2.4]).The leitmotif ofLectures on the History of Moral Philosophy in Englandis the historyof conflicting understandings of the foundations of morals, the basic principles of duty orright conduct. Theories of one class, distinguished by Whewell as ‘high’ or‘independent’, explain conduct as moral or immoralper se,right and wrong beingindependent (inherent) qualities of actions. According on the other hand to the theoriesWhewell designates as ‘low’ (the adjective serving to make plain his antipathy), moralityis reducible to facts more fundamental and real. The facts in question are consequences ofactions, notably pleasure and pain. The debate between theorists of these two proclivitiesarose in ancient Greece, Stoics pitted against Epicureans. Woven through Christianmoralizing, the independent perspective was long ascendant, up to the seventeenthcentury in fact. Then, at the same time as Hobbes and Locke were forcefully asserting thelow view of morality, developments in science and metaphysics were straining andbreaking the web of traditional belief. The high position began to look anachronistic,fewer people found it credible. For a new age, in a new climate of opinion, theindependent doctrine had to be appropriately reformulated. And while attempts have beenmade to adapt it to the altered circumstances, they have only been partial and never reallysatisfactory. So far as Whewell is concerned, the real work of reconstruction remains tobe done.From the early eighteenth century, in England generally and at Cambridge Universityin particular, dependent theorizing on morality formed ‘the general tendency’ ([2.5],165). Since the middle of that century, the successive main sources of the ethicalinstruction given at Cambridge have been Gay, Rutherforth and William Paley. Gay’sDissertation(1732) affirms God’s will as the determinant of virtue, and human happinessthe measure. On any occasion the action productive of greatest happiness is that whichGod would have us perform. InEssay on the Nature and Obligations of Virtue(1744)Rutherforth attempts to show that happiness as the ultimate end of action is a teaching ofreason, with revelation the source of our knowledge that God will reward with happinessin the next life right conduct towards others in this. The major production of Paley wasPrinciples of Morals and Politics(1785). Such has been its intellectual impact, Whewellputs it in another league altogether from the last mentioned works. Radiating influencefar beyond the walls of Cambridge colleges, Paley affected ‘the habits of thinking,reasoning and expression’ of English people ‘to at least as great an extent as any previousmoral doctrine has ever done’ ([2.5], 176–7). The central teaching of Paley is thathappiness for humanity is desired by God, by whom those of our actions that conduce tohappiness are deemed right and to whom they prove agreeable. We may know God’s willso Paley reasons, and reasoning thus he is close to Gay, by determining which actionsachieve most happiness for all concerned. There is for Whewell sad irony about Paley’slabours; of noble intentions and impeccable religiosity, his thought was a majorantecedent of secularized utilitarianism. Little did he, and profoundly would it haveaggrieved him to, know that his writings would in time ‘lead to dangerous and immoraltenets’ and ‘produce…evil’ ([2.5], 178). Whewell expresses inHistory of MoralPhilosophya resolve to deal with the problem as he recognizes it, seeking reform of theCambridge curriculum. The ‘system of [Paleyian] morals which is now taught among usis unworthy of our descent and office; and it will be my endeavour in future years…further to point out, and, if possible, to remedy the defects which I lament’ ([2.5], 184).He was as good as his word. Whewell’s ethical ‘writings became text-books atCambridge, and were naturally studied by young men reading for Trinityfellowships’ ([2.24], 1371).The reader ofHistory of Moral Philosophycatches glimpses of some of the likelysources of influence on Whewell’s constructive ethical thought. Grotius, Pufendorf andCumberland are approved of for having recognized in human nature principles besidesself-interest. He accepts Butler’s proposition that ‘The proper office of each of theprinciples of our nature assists us…to determine their limits, and to lay down rules fortheir direction, control, or restraint’, rules that is for framing ‘special moral duties’,although he can nowhere find in Butler a satisfactory ‘classification of the faculties andoperations of the human mind’ ([2.5], 112). And with Paley, Whewell ([2.5], 166) is ableto agree on at least this: the task of the moralist must include deductive ordering of ‘thecommonly-received rules of morality’. This leads on to Whewell’s constructive doctrine.Rational morality in contextThe diverse materials ofElements of Moralityare sorted into six books:‘Introduction.Elementary Notions and Definitions’, ‘Morality. Of Virtues and Duties’, ‘Religion. OfDivine Laws and their Sanction’, ‘Jus.Of Rights and Obligations’,‘Polity. The Duties ofthe State’ (‘Jus’treating law as it is,‘Polity’law as it ought to be), and ‘InternationalJus. Rights and Obligations Between States’. In its final (fourth) edition, the main bodyof the work runs to around 550 pages, with its longer books,‘Morality’and‘Polity’,inthe order of 200 and 130 pages respectively. There are two senses of ‘morality’ to benoticed in the titles. In that of the volume it connotes the five provinces of humanconduct and rules, of which each, while ‘intimately connected’, has its own questions andmodes of answering them. In the case of reasoned ‘Virtues and Duties’ (Book II),‘morality’ signifies the primary part of the whole.Elements of Moralityis a work of conservative intent. It is not designed to add to thefirst-order ethical knowledge we have already (which Whewell takes to be considerable),nor to inquire into how such knowledge is obtained—Whewell supposing himself to keepto the bare essentials. Whewell ([24], x) wrote theElementsprincipally to bring methodto bear on what we know about how we ought to act, ‘to construct a [moral]system’.15Our knowledge of rules, duties and virtues is seen as demonstrable by stepwisedeductions. The purely ethical part of the system, rational morality comprising Book II, isbased by Whewell on the supreme rule of human action or of ‘human nature’, whichnotion, as ‘universal standard’ of right, is redolent of natural law. From this supreme ruleof reason lesser moral rules are drawn, each conveying some part of its content.‘Moral Rules Exist Necessarily’ Whewell titles one of the chapters in the first book oftheElements,and it is a statement that underlies his explication of basic concepts. Thepredicate ‘right’ describes con-formity of action to rule. Rules indicate by what actionsobjects of designated classes may be properly attained. ‘Labour, that you may gainmoney’ prescribes labouring as ‘therightway to gain’ that object ([2.4], 48). Rules gainvalidity, and objects value, from superior cases. ‘The Rule,to labour,derives its forcefrom the Rule,to seek gain’ ([2.4], 48). In the same way, it is suggested, rules receivereasons: Why should I labour? To earn income to maintain myself and dependants.Climbing the scale eventually one comes upon an object of intrinsic value, source ofvalue in all other objects. The chain of reasons must if valid likewise have a terminus,one rule exerting sway. So it is that Whewell settles to his own satisfaction the existenceof the supreme rule and supreme good.In utilitarianism is recognized by Whewell the main alternative interpretation of ‘right’and cognate moral terms. He argues against it that pursuit of rightness is distinct from andapt to conflict with wanting to maximize pleasure, and in cases of conflict it is rightnessevery time that makes the stronger claim on us. To assess an action accurately as right isto provide for its performance a reason, writes Whewell: ‘paramountto all otherconsiderations. If the action be right, it is no valid reason against doing it, that it isunpleasant or dangerous. We are not to do what is pleasant and wrong. We are to do whatis unpleasant if it be right’ ([2.4], 6). ‘Right’ is in ethics an absolute term implying asupreme rule as the ultimate reason for action: above all we are to act rightly.16Whewell’s supreme rule extends beyond morality in the ordinary acceptation of theword. All feelings and conduct are regulated by it, ‘all intercourse of men, all institutionsof society’ ([2.4], 77). Rules acquire the property of rightness from expressing portions ofthe supreme rule of action, but whether Whewell regards as its components all rules thathave a defensible claim on conduct is uncertain. He distinguishes moral rules fromprudential, but by his ([2.4], 139) description of the latter as ‘not directly moral’ aconnection with the supreme rule is not necessarily excluded. What is clear however andfor Whewell’s purposes more important is thatlawsare part of the supreme rule’s contentno less than are moral precepts. These are respectively characterized as ‘Rules of external[bodily] action’ and rules of internal action or events—‘Will and Intention,… Desires andAffections’ ([2.4] 68)—which form ‘the only really human part of actions’ ([2.4], 66).17Laws create rights and correlative obligations; precepts impose duties.While ‘Every thing isrightwhich is conformable to the Supreme Rule of humanaction’ ([2.4], 54), on another level there exist what Whewell identifies as ‘national’moralities. ‘Nations and communities…have their Standards of right and wrong’,including positive laws and ‘current moral Precepts and Rules’ ([2.4], 199).From so devout a Christian as was Whewell (Anglican clergyman, doctor of divinityand theologian) it comes as no surprise to learn that ethics is religiously embedded.Natural religion furnishes the idea of the course of nature as in harmony with, as subjectto, divine moral government. God governs by laws he wills, that of which in humanaffairs is foremost being the selfsame supreme rule. Endowed with reason as part of thedivine providence individuals are able to, and of course should, act conformably to thedivine laws.The Supreme Rule of Human Action derives its Real Authority, and its actualforce, from its being the Law of God, the Creator of Man. The Reason for doingwhat is absolutely right, is, that it is the Will of God, through which thecondition and destination of man are what they are.([2.4], 141)By God’s appointment, violation of his moral laws is punishable by misery, andconformity rewardable with happiness. The apportionment, taking place in the next life,serves to complete the divine ‘moral government’, God determining the final, eternalcondition of the soul according to the level of its moral progress when embodied.The ‘history of the world’ recorded in the Scriptures Whewell presumes to be the ‘fact’corresponding to the ‘idea’ just noted. The part of the Scriptural record of chiefimportance concerns God’s transactions with humanity, ‘Revelationsof the Commandsand Promises of God, and of the Methods by which men are to be enabled to obey theseCommands, and to receive the benefit of these promises’ ([2.4], 257). The revelationmade through Jesus Christ confirms the claims of natural theology.Through the addition of Christian precepts and doctrines to those of rational ethics weare offered, Whewell believes, a more complete picture of the grounds of duty, providingfor virtue’s more effective pursuit. Christ embodied in human form the moral perfectionwe conceive in God, and Christian morality presents this‘Image of God in Christ’as the‘summit of the Moral Progress, which it is our Duty to pursue’ ([2.4], 258). It isimportant to our hopes and aspirations, and to the way we conduct our lives, that weknow whether sinners may be saved from eternal punishment. The answer, to be found inrevealed Christian doctrine, is that Christ’s crucifixion, burial and ascent into heavenformed a divine interposition to save humanity from ‘infliction of meritedpunishment’ ([2.4], 259). Conditions are revealed as necessary to fit the soul for futurelife: belief in Christ ensures participation in the Holy Spirit, and united with Christ thesoul receives its aliment.Law and moralityLeslie Stephen ([2.24], 1372) has described as ‘The most curious characteristic’ ofWhewell’s ethical writings ‘the prominence given to positive law in the deduction ofmoral principles’. It is possible to understand Whewell’s ethics only after studying howhe relates law to morality, and this requires the disentanglement of a number of threads inhis work.Of Whewell’s claims on this subject perhaps the most prominent is thatmoralitydepends on law. He opens with an argument that mental desires, our ‘most powerful’springs of action, can be gratified only if regulated, and on their gratification preservationof the fabric of society depends. Whewell ([2.4], 45–6) in the first instance describesthese rules as ‘moral’. This it may be thought is in conflict with the basic dependencerelation:‘that Moral Rules may exist, Men must have Rights’,where ‘rights’ signifiesabstract conceptions assigned to people by positive laws that are ‘subordinate to theSupreme Rule’ ([2.4], 51). But upon inspection the conflict turns out to be verbal only,apparent rather than real. ‘Moral’ is in the first sense inclusive of, and used in such a wayas to highlight, laws as a type of moral rules, having in the second its usual, morerestricted, meaning. At least three claims may be distinguished. There is a thesis, call it‘causal’, about the possibility of moral conduct: people cannot act morally unless alreadythey obey positive laws. Moral conduct and character presuppose legally ordered society.There is a closely related ontological claim about the possibility of moral rules: ‘We mustsuppose the Rights of Property, and the Laws of Property, before we can lay down theMoral Rules, Do not steal, or Do not covet another man’s Property’, etc. ([2.4], 55). Afurther claim, this made in regard to morality’s vocabulary, is that ‘Desires and Intentionscannot be defined or described in any way, without some reference to Things andActions; and therefore, cannot supply a basis of Morality independent of Law’ ([2.4],201). Without him saying it in so many words, the precedence of law for Whewell reallyboils down to it enabling mental desires to be satisfied. More than moral rules, suggestsWhewell, laws are directly concerned with constraining conduct and, being enforced,usually succeed in this.Whewell’s follow-up to this is a rather more informative statement that duties(morality) depend upon rights, not in general but of aspecific class. What ‘Rights mustexist’ are determined by—a common expression in eighteenth- and nineteenth-centuryphilosophical writing—springs of action([2.4], vii).18Acts Whewell believes are properly ours when reason directs the will. Through thefaculty of reason we discover truth and falsehood, conceive of general rules, recognizeactions as cases of rules, and discriminate right from wrong.19Will represents the final step of intention (purpose), a process of internal motion asWhewell conceives it, produced by springs of action. Springs of action he distinguishesby their ends. Bodily desires for physical objects areappetites,primordial among whichare ‘natural wants’ (hunger, thirst). Satisfaction of wants is a source of enjoyment, givingrise to the desire ofpleasures of sense. Employment of art to satisfy appetites stimulatesartificial wants. As appetites are directed to things,affections(love and anger) tend topeople. Most ‘universal and most powerful’ ([2.4], 32) among springs of action aremental desiresfor abstract conceptions of physical safety, private property,companionship of family and membership of civil society, kept promises and respectedagreements (‘mutual understanding’), ‘superiority’ (skill, strength, wealth, or power), and‘knowledge’ or reason.Moral sentiments(approval, esteem, and their opposites)accompany moral evaluations of actions, disposing us to treat people ‘as we approve ordisapprove their actions’ ([2.4], 41).Reflex sentimentsare the desire that others wouldlove and esteem us; and desire of honour, fame and self-approval.From mental desires as the predominant springs of action Whewell derivesfundamental rights. To have a satisfactory life each person is in need of: protection fromassault (right of person), control over physical objects (right of property), dependableconduct from others (right of contract), exclusive relations with certain others (familyrights), and an organization to establish rights (rights of government). The rights are toobjects of mental desires. (In‘Jus’,Book IV of theElements,rights in Roman andEnglish systems of law receive extensive classification under the foregoing heads.)These ‘primary and universal’ rights Whewell respects as absolutely valid, imperativesin all societies. They are rights thatmust be. ‘Family Rights’, as a case in point,‘necessarily exist’; while society without ‘Rights of Government…loses its socialcharacter; and the moral character of man cannot find its sphere of action’, and so on([2.4], 52). The most important objects in our lives, those of primary (mental) desires,cannot be secured outside a framework of law. Rights respected are realities; abstractconceptions as objects are thus realized.As we have so far understood Whewell in this part of his thought, his essential thesesare two. There is a set ofidealrights corresponding to springs of action. The same rightsserve infactas the indispensable conditions of social order and morality.To this conjunction an objection immediately suggests itself: Is not the proposition thatideal rights necessarily exist contradicted by the variety of laws and rights found bothwithin (temporally) and between societies? The objection is one that Whewell ([2.4], 59)himself anticipates in order to obviate, claiming the positions to be ‘reconcilable’. In thisreconcilableness he sees a further case of the fundamental antithesis or distinctionbetweenideas(conceptions) andfacts. Ideas of the primary rights are derivations fromhuman nature. There can be no conceiving of people as members of society nor as moralagents except in so far as they have the (ideal) rights to personal safety, property, etc.That said, ideas of rights must be in society as facts. By positive laws the forms so tospeak are specifically defined and, on account of being historically conditioned, thedefinitions of ideas of rights differ between societies. The traveller whom law in onesociety permits to ‘pluck the fruits of the earth as he passes’ finds such fruits are inanother society ‘the Property of him on whose field they grew’. Explains Whewell ([2.4],59), the ‘Precept,Do not steal’is absolutely moral and ‘universal’, a corollary of primaryrights, whereas ‘the Law,To pluck is to steal’, is historically contingent or ‘partial’.According to the present thesisactual rights (in ‘national moralities’) are in all casesdefinitions of, and consistent with, the rights identified as ideal.Appearing inElements of Moralityis a further account of rights positive in relation torights ideal, almost certainly contradicting the account above. In a passage that meritsquoting at length we have Whewell (1864:68) saying:positive definitions of Rights for the moment, may be themselves immoral.Rights, as we have described them…, are arrangements not only historicallyestablished, but also established in conformity with the supreme Rule; that is,they are such as are right. The actual definitions of Rights at any moment, thatis, the state of the Law, may need improvement and reform: but in general, theLaw gives, for the moment, the definitions of Rights upon which Morality mustproceed.20Actual rights may be immoral, contradicting ideal rights. So, whilepositive rightsshould,the fact is theydo not, in all cases assign definite shape to some part or other of thecontent of ideal rights.21From his case for morality requiring a specific set of rights Whewell has veered away,apparently unaware of how damaging to his basic doctrine are the implications of suchchange. Societies may produce, and their ‘national’ moralities may exist on, immoralrights, conflicting with the ideal or ‘primary’ rights and failing to gratify primary desires.Closing coverage of this part ofElements of Moralityit may be commented thatWhewell’s procedure—grounding ideal rights in human nature—appears strange givenhis inclusion of the same rights in the supreme rule. To this, the first and fundamentalprinciple of morality, he might have been expected to go and exfoliate the ideal rightsdirectly.22Rational moralityIn ‘Morality. Of Virtues and Duties’ we confront the longest and most important of theElements’ six books. The core of this as of the other parts of Whewell’s ethical thought isthe ‘supreme rule’ coupled with the ‘supreme object’. These points have been noted.Whewell early in Book II equates the idea of the supreme object with moral goodness, towhich he otherwise refers as virtue or rightness in the soul. The soul attains this statewhen the faculties (consciousness, reason, will, imagination, and affections) are regulatedby virtue and habituated to duty. The supreme object—goodness or virtue—Whewell([2.4], 241) also designates as happiness, which he specifically says stands for ‘theSupreme Object of our Desires’ and is ‘identical with…the Ultimate and Supreme Guideof our Intentions’. Utilitarians have commonly equated happiness with pleasure,demonstrating to Whewell ([2.4], 254) how confused they are, for in happiness is acompound of ‘all other objects’, whereas pleasure is only thatsimplestate we experienceupon the satisfaction of our physical desires. The doctrine of absolutely right action (asaction in accordance with the supreme rule) denies that pleasure can be, while confirmingthat happiness is, the ultimate desire and end.The Desire of Happiness is the Supreme Desire. All other Desires, of Pleasure,Wealth, Power, Fame, are included in this, and are subordinate to it. We maymake other objects our ultimate objects; but we can do so, only by identifyingthem with this.([2.4], 241)23No doubt some utilitarians would approve of this, and imagine that Whewell on accountof it might be counted as one of them. But although Whewell takes it as true thathappiness ought to be increased, he goes on to suggest that there is no possibility ofanyone knowing how to go about it. A basket, or if you like an umbrella, concept, no‘special element’ is implied in happiness while ‘all…objects’ and ‘all good’ are ([2.4],243). As happiness, Whewell says, is too complex and indeterminate to be measurable,rules for attaining it are out of the question. Whereupon the possibility of Whewell’sproject of arranging existing rules-and-objects hierarchically boils down to this: theremust be some other way, more fertile of implications, in which to envisage the supremeobject.Whewell for his part typically discusses this object with reference to, and in such away as to identify it with, ‘goodness’, reducing it to five terms—benevolence, justice,truth, purity and order—each with subjective and objective denotations. Subjectively theyrefer to classes of dispositions or ‘cardinal virtues’, and objectively to ‘abstract mentalObjects or Ideas’ ([2.4], 75–6) to which dispositional desires and affections are directed.Benevolenceas virtue is the sum of all affections that bring together people, and theabsence of feelings that divide; of which the corresponding idea, ‘humanity’, is of thegood or well-being of all people.24Justiceas virtue is ‘the Desire that each person shouldhave his own’ ([2.4], 73), and as idea or object is ‘the Rule, To each his own’ ([2.4], 76).Truthas virtue or disposition has as its idea ‘Objective Truth,the agreement between thereality of things and our expressed conceptions of them’ ([2.4], 76). In the state ofpurityreason and moral sentiments govern bodily desires, the objective counterpart consistingin the idea of human nature ‘free from…mere desire’ ([2.4], 76).Orderdenotes thedisposition to conform willingly to positive laws and moral rules and, objectively, theidea of obedience.Goodness or virtue is the supreme object of action and of rules, subordinate objects alltaking their moral value from it. Enjoining us to love virtue (benevolence, justice, etc.)and to seek it ‘as the ultimate and only real object of action’ ([2.4], 77), the supreme ruledirects our ‘Affections and Intentions to their proper objects’ ([2.4], 96). Conforming tothis rule,characteris virtuous andactiondutiful, the twin elements serving to make aperson good. The supreme rule is possessed of a structure which for the most part reflectsits (supreme) object, with matching pairs—objective and subjective—of componentrules. Conveying aspects of the supreme rule’s content,express(objective) moralprinciples have objects in the form of ideas. The principles are those of humanity,‘Manis to be loved as Man’;justice,‘Each Man is to have his own’;truth,‘We must conformto the Universal Understanding among men which the use of Language implies’,desisting from lying; purity,‘The Lower parts of our Nature are to be governed by, andsubservient to, the Higher’;and order,‘positive Laws [are to be obeyed] as the necessaryConditions of Morality’([2.4], 95–6). Completing the list of express principles are:earnestness,‘The Affections and Intentions must not only be rightly directed, butenergetic’;and moral purpose, ‘Things are to be sought universally,not only insubservience to moral rules, butas means to moral ends’ ([2.4], 96). The supreme rule’sother set of principles Whewell distinguishes as ‘operative’. These have to be cultivated;express principles being incorporated into the character of an individual habitually toguide affections and purposes, functioning as springs of action.Specific virtues and duties, the bulk of the materials of rational morality, Whewellexplicates in terms of, and distributes according to, the plan above. Virtues are a subclass—those that are desirable—of the habits or dispositions of inner states (desires,affections and volitions), combining to form an agent’s moral character. Duties are rightactions, the ways in which we ought to conduct ourselves. Virtues exist and operateunconsciously, while duties are conscious. Between the two Whewell finds a relationapproaching to concomitance. Duties need not issue from virtues (motives to right actionsmay be amoral, even immoral), but virtues are manifested in and developed byperformance of duties, and by cultivation of virtues performance of corresponding dutiesis made more likely. ‘Acts of Duty are both the most natural operation of virtuousDispositions, and the most effectual mode of forming virtuous Habits’ ([2.4], 97).Distinguishing duties in relation to objects, Whewell’s basic categorization of them issimilar to that of cardinal virtues, with duties of affections (benevolence), property, truth,purity, and order. A duty of cultivating the affections is also recognized, reducible towhich is a duty of the affections: the ‘Duty of thus cultivating these Affections includesthe Duty of possessing such affections’ ([2.4], 115). It is unclear whether Whewellbelieves all duties owed to other people involve the duty of affection, but at least those ofjustice and truth in his view do.From these primary duties, enjoined by the ‘express’ principles of morality, Whewellgoes on to deduct rationally many specific rules of duty, but their enumeration shall notdetain us. The general classification of duties reappears to shape discussion in other partsofElements. The book‘Polity’has chapters on the State’s duties of ‘Justice and Truth’,‘Humanity’, ‘Purity’ and ‘Order’. Explicating religious ethics, Christian precepts arearranged by Whewell under such headings as ‘Duties of the Affections’, ‘Property’,‘Truth’, ‘Purity’, ‘Obedience and Command’.One further duty to be noted is that of moral progress, incumbent on States andindividuals alike. The State must be recognized as an agent given that it, among otherthings, ‘makes war and peace, which it may do justly or unjustly; keeps Treaties, orbreaks them; educates its children, or neglects them’ ([2.4], 208). Continuously existing,purposive and active, States are moral beings endowed with life, and ‘During this Life, itis their Duty to conform their being more and more to the Moral Ideas’ ([2.4], 208). Byits duty of moral progress the State is obliged to acquire more virtue and to perform itsduties better. It must become more just and act more justly (observe treaties with, andrespect possessions of, other States; make laws that remedy inequalities from the past),become more benevolent and act more benevolently (liberate slaves, relieve poverty), andso on. What meaning is to be given to Whewell’s talk of the State as a subject of virtue,and of it being by the enhancement of its own virtue disposed to perform duties better? Itseems probable he regards this virtue and its increase as a function of citizens:And, as the condition of other Duties being performed, the moral Education ofits citizens,and consequently of itself,is a Duty of the State. It is its Duty toestablish in the minds of its children, and to unfold more and more into constantand progressive operation, the Moral Ideas of Benevolence, Justice, Truth,Purity, and Order.([2.4.], 208, emphasis added)Correspondingly, to the individual is ascribed a ‘reflex’ duty of moral progress or ofmoral self-culture. The character should according to this duty be developed as far aspossible towards the ideal centre of morality at which, the point of goodness, cardinalvirtues represent tendencies (‘operative principles’) of conduct so firmly established thatno duty can ever be transgressed. Among the requirements of this duty,benevolentaffectionsof gratitude to benefactors, compassion to the afflicted and love betweenfamily members are to be made steadier and more earnest; themalevolent feelingsofviolent anger, peevishness and captiousness are to be suppressed.Justiceis to bedeveloped as an operative principle, the character being cleansed of stains of greed,covetousness, and partiality; the agent desiring only possessions to which she or he isentitled and only for moral purposes. The individual is likewise called on to foster theother virtues—truth, purity, order—while eliminating desires running contrary to them.‘We have to form our character, so that these principles [benevolence, justice, truth,purity, and order] are its predominant features. We have to seek not only todo,but tobe;not only to perform acts of Duty, but to become virtuous’ ([2.4], 142). The duty of moralprogress is itself performed, just as the several virtues are most effectively cultivated, byperformance of the other, more particular, duties. And in opposite fashion whenevertemptations are succumbed to and rules of duty transgressed, when affections aremalevolent and intentions fraudulent, moral progress ceases and the character goes down‘a retrograde moral course’ ([2.4], 143).While the present study of Whewell’s view of ethics has concentrated on that which helabels ‘rational morality’ as the truly distinctive, striking and substantial part of histeaching, it is as well to remind ourselves that this department depends for him in a mostfundamental manner on religion. The cardinal virtues, as noted, are conceived of byWhewell as perfectly realized in God and as having had embodiment in human form inChrist. The supreme rule of rational morality is the rule of God which fact in conjunctionwith his imposition of sanctions provides the compelling reason for obeying it. Thesupreme rule as understood by Whewellderives its Real Authority, and its actual force, from its being the Law of God,the Creator of Man. The Reason for doing what is absolutely right, is, that it isthe Will of God, through which the condition and destination of man are whatthey are.([2.4], 141)The rule commends to us, and commands us to emulate, the character of Jesus, moralprogress having the goal of ‘a godlike being and a heavenly life’ ([2.4], x). On oursuccess in following the supreme law and in making moral progress depends our prospectof happiness in this life and, more so, the next.PostscriptIt is natural to ask whether in its structure and method Whewell’s analysis of moralityresembles his philosophy of science. There is a difficulty to be faced, however, beforeanswering this, the two main works in question,Philosophy of the Inductive SciencesandElements of Morality, including Polity,each being predominantly on a different levelfrom the other. Rehearsal of some basic distinctions will enable us to identify moreclearly the source of the difficulty. Concerning the subject-matter of physical objects andprocesses (level 1), scientists advance prepositional laws and theories (level 2), withphilosophers theorizing scientific practice and products (level 3). Whewell’sPhilosophyis a contribution on the third of these levels. The corresponding levels of ethics putsimply are: human agents and actions, rules dictating how agents ought to conductthemselves, and meta-ethical theories of the discovery and evaluation of right rules. MostofElements of Moralityoccupies the second level, systematizing occurrent rules, dutiesand virtues. Philosophical analysis of ethical knowledge and its methods is not its majorconcern. We are not altogether without material for comparison, Whewell producingsome philosophy in theElements,but for comparison to be full and rigorous more of suchmatter would be needed. We proceed bearing in mind this caveat.Whewell himself ([2.5], 168) speaks confidently of the ‘analogy between the progressof the science of Morals and other sciences’. Concepts and knowledge in both subjectsform part of the flux of history. InElements of Morality([2.4], 202), for example, it isobserved generally that ‘As the intellectual culture of the nation proceeds, abstract wordsare used with more precision; and in consequence, the conceptions, designated by suchwords, grow clearer in men’s minds’.Whereas progress of scientific theories turns for Whewell on wider colligations, that ofethical rules is accounted for by him differently,25 the key apparently being morality’sinteraction with law. Definitions that laws give of rights vary socially and historically,and the virtues and duties supported in society depend upon its laws. At the same timemorality exercises authority over law and, as a nation’s morality improves, laws shouldbe and, Whewell it would appear believes, typically are brought into line. Exactly howimprovement in morality is effected and how people have in the past been able to tell thatmorality was changing to the good are matters which in Whewell remain obscure.Perhaps Christianity’s revealed morality has acted as both inspiration for and criterion ofimprovement in national moralities. But whether (as I am inclined to believe) or not thisspeculation constitutes Whewell’s opinion on the matter, the main point at the moment isthatElements of Moralityhas no doctrine corresponding to that of colligatory trial anderror.Whewell as explained sees a corpus of real moral knowledge having formed throughhistory. It is to this knowledge, expressed in rules, that under the rubric of ‘rationalmorality’ he imparts systematic shape. The axioms employed by him to this end are, likethose he locates in science, understood by Whewell to be presuppositions of knowledge,embedded in ideas. Yet their roles are appreciably different: axioms are in rational ethicsused forprovingclaims to knowledge (whether rules are right), which in science falls tothe lot of thea posterioritest of consilience, axioms transmitting necessity to, andassisting deductive ordering of, propositions less general.The presence in science and ethics of the ‘fundamental antithesis’ provides what is forWhewell perhaps their most striking similarity. The antithesis is recognized inPhilosophy of the Inductive Sciencesin its second edition ([2.3], x) as ‘suited to throwlight upon Moral and Political Philosophy, no less than upon Physical’. Now without inany way gainsaying that Whewell often employs the terms ‘idea’ and ‘fact’ inElementsof Morality,one does find that phenomena to which he applies them are in their naturesmost diverse. He for example writes of ideas and conceptions of primary rights, and oftheir ‘definitions’ by positive laws as facts. In another part of the work, as already wehave had cause to comment, he finds in natural religion the idea ‘of the course of theWorld’, whose corresponding fact is God’s manifestations in history (revealed religion).But Whewell’s principal usage, of ‘idea’ at any rate (for he is silent on what the term‘fact’ might in this case mean),26 is that which we find in his explication of rationalmoral-ity. And the way in which he understands ideas here is nothing like how heunderstands them in his philosophy of science. Whewellian ideas in rational ethics areidealsformingsubjectsof inquiry. To know their content is to have moral knowledgeabout what rules are right. In science, by contrast, ideas are conditions of and materialsfor knowledge, of whichphysical realityis thesubject.Gaining knowledge of that reality,a scientist at the same time gains knowledge of ideas, but the latter is more properlyregarded as a secondary effect of enquiry rather than as its purpose.Whewell has been described as having ‘written with an explicit view not merely toembrace the world of learning but to synthesize it and render it a unified intellectualwhole’ (Fisch [2.13], 31), and this may well be true. But if he so sought synthesis andunification, what has been said in this paper must cast doubts on whether he attainedthem.NOTESA significant portion of the research for this study was undertaken in 1992 during aVisiting Fellowship at The Australian National University’s Research School of SocialSciences. The author gratefully acknowledges the support of Professor Geoffrey Brennan,the School’s Director, and that of Professor Eugene Kamenka, chair of the ResearchSchool’s History of Ideas Unit. For helpful comments on draft versions of the paper hethanks: Mr Kerry Cardell, Professor David Walker, Dr George Zollschan, and Dr FrankMaher.1 ‘Whewell himself regarded his History of the Inductive Sciences…and Philosophy of theInductive Sciences…as both the crowning achievement of his career and the unifying “hardcore” of his entire system of thought’ ([2.12], 2). I shall in what follows also refer to theseworks simply as History and Philosophy. References unless otherwise indicated are to thereprint editions (1967).2 Cohen ([2.7], 528ff.) is illuminating on Whewell’s notion of revolution.3 Schipper ([2.20], 49), after noting several respects in which Whewell’s historiography ofscience anticipates that unfolded by Kuhn in The Structure of Scientific Revolutions, arguesthat in this—cognitive continuity (Whewell) as against discontinuity (Kuhn)—consists theirmajor difference. This leads on to a further difference: science from Kuhn’s standpoint isultimately aimless while Whewell ascribes to it the aim of truth.4 As Richard Gregory ([2.14], 219), distinguished psychologist of perception, states: ‘objectshave a host of characteristics beyond their sensory features. They have pasts and futures; theychange and influence each other, and have hidden aspects which emerge under differentconditions’ (emphasis added). Whewell’s account of perception (and of facts) has a readilydiscernible counterpart—‘theory-dependence of observation’—in writings of such twentiethcenturyfigures as Karl Popper, N.R.Hanson, and Kuhn.5 As explicitly defined, ‘ideas’ refers ‘to certain comprehensive forms of thought,—as space,number,…—which we apply to the phenomena which we contemplate’. Whewell ([2.3], 2:5–6) then proceeds to define the important related term—‘conceptions’—as ‘specialmodifications of these ideas’, giving as examples ‘a circle, a square number, an acceleratingforce, a neutral combination of elements’.6 A fine-grained, more technical and more manifestly prescriptive rendering by Whewell ofscientific methods is in Book XIII of the Philosophy. Ducasse ([2.9], 183–217) digests thisparticular book; Todhunter ([2.25], 139–42) detects several discrepancies between itsstructure and that of Book XI.7 With reference to their disregard of this condition Whewell explains the Greeks’ failure fullyto cultivate science, whereas in the ‘Stationary Period’, as noted, the accent was the opposite,on ideas at the expense of facts.8 This suggests that facts are reports of perceptions of concrete objects and events. ‘Fact’ forWhewell may also cover theories or inductions as candidates for explanation. In the latterdistinction theories and facts are of the same nature, have the same properties, the distinctionbeing time-dependent and relative.9 Niiniluoto [2.18] deals at length, in most interesting fashion, with this and other similaritiesbetween the images of science of Whewell and Popper.10 This explication of consilience owes a great deal to that of Fisch [2.11].11 According to Fisch ([2.10], 287; [2.12], 158) Whewell’s axioms are true only of ideas. But if,as quite clearly Whewell maintains, axioms are true ‘beyond… experience’ they must also betrue within and of it. He is to be understood as saying that axioms (1) express the content ofideas, and (2) are propositions about physical reality.12 Some would say (e.g. Fisch [2.10]; [2.12]) that Whewell’s Mechanical Euclid providesample evidence for this claim. But Whewell there on my reading presents the distinctionbetween inductive (non-necessary) and non-inductive (necessary) truths in mechanics asabsolute rather than as epistemologically relative (to the state of knowledge). So no matterwhat he may say or suggest to the contrary, on his account of it in that work, the non-BIBLIOGRAPHYWorks by Whewell2.1——History of the Inductive Sciences,London: Parker, 3 vols, 1st edn, 1837, 2nd edn,1847, 3rd edn, 1857; 3rd edn, reprinted by Cass, London, 1967.2.2——The Mechanical Euclid,London: Parker, 1837.2.3——The Philosophy of the Inductive Sciences, Founded Upon Their History,London:Parker, 1st edn, 2 vols, 1840, 2nd edn, 1847, 3rd edn, 3 vols, 1856–60; 2nd edn,reprinted by Cass, London, 1967.necessary truths cannot and can never be logical derivatives of the necessary. That which istrue (but could be false) is not among the implications of that which must be true.13 Again, aside from those passages already referred to, Whewell should be seen as operatingwith just such a distinction in his opening account of necessity in Philosophy of the InductiveSciences ([2.1], 54–73).14 The paucity of exceptions serving to prove this rule are Schneewind ([2.21] and [2.22], 101–17) and Donagan [2.8].15 About this Whewell is clearest in the preface of the first edition. See also the letter in StairDouglas ([2.23], 326), and the preface of the fourth edition ([2.4], 3 and 6).16 For further discussion of happiness and of utilitarianism see Whewell [2.5], x, 170ff., 174ff.,and 188ff.; [2.4], 125–6, 254ff.17 Not always faithful to this characterization, Whewell ([2.4], 244) may on occasion subsumeexternal action under precepts.18 The language of necessity (‘must’), freely employed in the Elements, is ambiguous insignifying what ought to be (ideal morality), or else what cannot but exist (in ‘national’moralities) in that human nature demands and cannot survive without it. Arguing that certainrights ‘must’ exist, Whewell conflates both senses.19 For extensive enumeration of the operations of the faculty of reason see Whewell [2.4], 23–4.20 Other relations between rights and morality are delineated later in the work ([2.4], 201f.,229ff., 337ff.).21 The same contradiction between this and his doctrine about necessary rights is latent in a notein the second and subsequent editions ([2.4], 62).22 This calls into question the categorization of Whewell as an ‘intuitionist’. Tracing rights fromhuman nature his method is a combination of empiricism and deduction.23 This embracing conception of happiness is noteworthy for being akin to that which one findsin John Stuart Mill’s ‘utilitarianism’.24 Not the least of the difficulties facing the reader of Elements is that a number of key termshave different denotations which Whewell fails to mark. ‘Ideas’ is a case in point: usuallyreferring to a select class of ideals or objects of action, Whewell ([2.4], 94–5) may also applyit to dispositions to seek those objects.25 Similarity of methods is suggested, it is true, in the Lectures ([2.5], 168) but not shown, andcareful examination of the texts uncovers no evidence of it.26 Schneewind ([2.21], 120) speaks in this context of ‘Human Nature’ as the corresponding‘fact’, but he provides no evidential support to show Whewell has this in mind and, so far asthis author can ascertain, none exists.2.4——The Elements of Morality, including Polity,London: Parker, 1st edn, 2 vols,1845, 2nd edn, 1848, 3rd edn, 1854, 4th edn, 1 vol., 1864.2.5——Lectures on The History of Moral Philosophy in England,London: Parker, 1852;reprinted by Thoemmes, Bristol, 1990.2.6——On the Philosophy of Discovery,London: Parker, 1860; reprinted New York,Burt Franklin, 1971 (vol. 3 of 3rd edn,Philosophy of the Inductive Sciences).Other works cited2.7 Cohen, I.Revolution in Science,Cambridge, Mass.: Belknap Press, 1985.2.8 Donagan, A. ‘Whewell’sElements of Morality’,The Journal of Philosophy,71(1974):724–36.2.9 Ducasse, C. ‘William Whewell’s Philosophy of Scientific Discovery’, in E. Madden(ed.),Theories of Scientific Method: The Renaissance through the Nineteenth Century,New York: Gordon and Breach, 1989, 183–217.2.10 Fisch, M. ‘Necessary and Contingent Truth in William Whewell’s AntitheticalTheory of Knowledge’,Studies in History and Philosophy of Science,16 (1985): 275–314.2.11——‘Whewell’s Consilience of Inductions: An Evaluation’,Philosophy of Science,52 (1985):239–55.2.12——William Whewell Philosopher of Science,Oxford, Clarendon Press, 1991.2.13——‘A Philosopher’s Coming of Age: A Study in Erotetic Intellectual History’, inM.Fisch and S.Schaffer (eds)William Whewell: A Composite Portrait,Oxford:Clarendon Press, 1991, 31–66.2.14 Gregory, R.Eye and Brain,New York: McGraw-Hill, 1981.2.15 Jacobs, S. ‘John Stuart Mill on Induction and Hypotheses’,Journal of the History ofPhilosophy,29 (1991):69–83.2.16——Science and British Liberalism,Aldershot: Avebury, 1991.2.17 Mill, J.A System of Logic Ratiocinative and Inductive,Toronto: University ofToronto Press, 1981.2.18 Niiniluoto, I. ‘Notes on Popper as Follower of Whewell and Peirce’ in I. Niiniluoto(ed.),Is Science Progressive?,Dordrecht: Reidel, 1984, 18–60.2.19 Ruse, M. ‘The Scientific Methodology of William Whewell’,Centaurus,20(1976):227–57.2.20 Schipper, F. ‘William Whewell’s Conception of Scientific Revolutions’,Studies inHistory and Philosophy of Science,19 (1988):43–53.2.21 Schneewind, J. ‘Whewell’s Ethics’ in N.Rescher (ed.),Studies in Moral Philosophy,Oxford: Blackwell, 1968, 108–41.2.22 Schneewind, J.Sidgwick’s Ethics and Victorian Moral Philosophy,Oxford:Clarendon Press, 1986.2.23 Stair Douglas, J.The Life Selections from the Correspondence of William Whewell,D.D,London: Kegan Paul, 1881. Reprinted Bristol: Thoemmes, 1991.2.24 Stephen, L. ‘Whewell, William’,The Dictionary of National Biography,London:Oxford University Press, vol. 20, 1967–8, 1365–74.2.25 Todhunter, I.William Whewell, D.D. Master of Trinity College Cambridge: AnWhewell’s philosophy of science and ethics 49Account of his Writings with Selections from his Literary and Scientific Correspondence,London: Macmillan, vol. 1, 1876.