in producing the deposition of dew on the sur- “This law of causation, already se amply ese face.

tablished, admits, however, of efficient" addi “* But if we expose rough surfaces instead tional corroboration in no less than three ways. of polished, we sometimes find this law inter First, by deduction from the known laws of fered with. Thus, roughened iron, especially aqueous vapour when diffused through air or if painted over or blackened, becomes dewed any other gas, and though we have rot ye sooner than varnished paper : the kind of sure come to the Deductive Method, we will not face, therefore, has a great influence. Ex- omit what is necessary to render this specula. pose, then, the same material in very diversi- tion complete. It is known by direct isperi fied states as to surface' (that is, employ the ment that only a limited quantity of water rar Method of Difference to ascertain concomit- remain suspended in the state of vapou. ance of variations), 'and another scale of in- each degree of temperature, and abogt hus tensity becomes at once apparent; those sur maximum grows less and less as the empera faces which part with their heat most readily ture diminishes. From this it iollows deducby radiation, are found to contract dew most tively, that if there is already as much vapour copiously.'

suspended as the air will contain at its existing # The conclusion obtained by this new apo temperature, any lowering of that temperature plication of the method is, that cateris paribus will cause a portion of the vapour to be conthe deposition of dew is also in some propor- densed, and become water. But, again, we tion to the power of radiating heat; and that know deductively, from the laws of heat, that the quality of doing this abundantly.(or some the contact of the air with a body colder than cause on which that quality depends) is another itself, will necessarily lower the temperature of of the causes which promote the deposition of the stratum of air immediately applied to its dew on the substance.

surface; and will therefore cause it to part “ Again, the influence ascertained to exist with a portion of its water, which accordingly of substance and surface leads us to consider will, by the ordinary laws of gravitation or co that of texture ; and here, again, we are pre- hesion, attach itself to the surface of the body, sented on trial with remarkable differences, thereby constituting dew. This deductive and with a third scale of intensity, pointing out proof, it will have been seen, has the advansubstances of a close firm texture, such as tage of proving at once causation as well as costones, metals, etc., as unfavourable, but those existence; and it has the additional advantage of a loose one, as cloth, velvet, wool, eider that it also accounts for the exceptions to the down, cotton, etc.,sas eminently favourable to occurrence of the phenomenon, the cases ir the contraction of dew. The Method of Con- which, although the body is colder than the comitant Variations is here, for the third time, air, yet no dew is deposited, by showing that had recourse to; and, as before, from neces- this will necessarily be the case when the air is sity, since the texture of no substance is absoo so under-supplied with aqueous vapour, comlutely firm or absolutely loose. Looseness of paratively to its temperature, that even when texture, therefore, or something which is the somewhat cooled by the contact of the colder cause of that quality, is another circumstance body, it can still continue to hold in suspenwhich promotes the deposition of dew; but sion all the vapour which was previously sus. this third cause resolves itself into the first, pended in it: thus in a very dry summer there viz., the quality of resisting the passage of are no dews, in a very dry winter no hoar heat: for substances of loose texture are pre- frost. cisely those which are best adapted for cloth- “The second corroboration of the theory is ing, or for impeding the free passage of heat by direct experiment, according to the canon from the skin into the air, so as to allow their of the Method of Difference. We can, by outer surfaces to be very cold, while they re- cooling the surface of any body, find in all main warm within.'

cases some temperature (more or less inferior “It thus appears that the instances in which to that of the surrounding air, according to its much dew is deposited, which are very various, hygrometric condition) at which dew will begin agree in this, and, so far as we are able to ob to be deposited. Here, too, therefore, the serve, in this only, that they either radiate causation is directly proved. We can, it is heat rapidly or conduct it slowly: qualities be- true, accomplish this only on a small scale ; tween which there is no other circumstance of but we have ample reason to conclude that the agreement than that by virtue of either, the same operation, if conducted in Nature's great body tends to lose heat from the surface more laboratory, would equally produce the effect. rapidly than it can be restored from within. “And, finally, even on that great scale we The instances, on the contrary, in which no are able to verify the result. The case is one dew, or. but a small quantity of it, is formed, of those rare cases, as we have shown them :o and which are also extremely various, agree be, in which nature works the experiment for (80 far as we can observe) in nothing except in us in the same manner in which we ourselves mot having this same property.

perform it ; introducing into the previous “This doubt we are now able to resolve. state of things a single and perfectly denpito We have found that, in every such instance, new circumstance, and manifesting the effect the substance must be one which, by its own so rapidly that there is not time for any other properties or laws, would, if exposed in the material charge in the pre-existing circumright, become colder than the surrounding air. stances. It is observed that dew is uever The coldness, herefore, being accounted for copiously deposited in situations much screened independently of the dew, while it is proved from the open sky, and not at all in a cloudy that there is a connection between the two, it night; but if the clouds withdraw even for a must be the dew which depends on the cold-few minutes, and leave a clear opening, a ness ; or, in other words, the coldness is the deposition of deu presently begins, and goes on anne of the dew.

increasing. Dew formed in floar inter

vals will often even evaporate again when the therefore evade the obstacle ; and it is sky becomes thickly overcast.' The proof, here that the last key of nature appears therefore, is complete, that the presence or absence of an uninterrupted communication the Method of Deduction. We quit with the sky causes the deposition or non- the study of the actual phenomeron to deposition of dew. Now, since a clear sky is observe other and simpler cases; we nothing but the absence of clouds, and it is a establish their laws, and we connect known property of clouds, as of all other bodies between which and any given object nothing each with its cause by the ordinary in tervenes but an elastic fluid, that they tend methods of induction. Then, assuming to raise or keep up the superficial temperature the concurrence of two or of several of of the object

by radiating heat to it, we see at these causes, we conclude from their once that the disappearance of clouds will cause the surface to cool ; so that Nature in known laws what will be their tota! this case produces a change in the antecedent effect. We next satisfy ourselves as to by definite and known means, and the conse whether the actual movement exactly quent follows accordingly: a natural experinent which satisfies the requisitions of the coincides with the movement foretold, Method of Difference.”

and if this is so, we attribute it to the

causes from which we have deduced it. IX.

Thus, in order to discover the causes of These four are not all the scientific the planetary motions, we seek by simmethods, but they lead up to the rest. ple induction the laws of two causes : They are all linked together, and no first, the force of primitive impulone has shown their connection better sion in the direction of the tangeni ; than Mill. In many cases these pro- next, an accelerative attracting force. cesses of isolation are powerless; name. From these inductive laws we deduce ly, in those in which the effect, being by, calculation the motion of a body produced by a concourse of causes, submitted to their combined influence ; cannot be reduced into its elements and satisfying ourselves that the plane Methods of isolation are then impracti- tary motions observed coincide exactly cable. We cannot eliminate, and con- with the predicted movements, we consequently we cannot perform induction. clude that the two forces in question: This serious difficulty presents itself in are actually the causes of the planetary almost all cases of motion, for almost motions. “ To the Deductive Method," every movement is the effect of a con- says Mill, “the human mind is indebted currence of forces ; and the respective for its most conspicuous triumphs in effects of the various forces are found the investigation of nature. To it we so mixed up in it that we cannot sepa- owe all the theories by which vast and rate them without destroying it, so that complicated phenomena are embraced it seems impossible to tell what part under a few simple laws.” Our deviaeach force has in the production of the tions have led us further than the di. movement. Take a body acted upon rect path ; we have derived efficiency by two forces whose directions form an from imperfection. angle : it moves along the diagonal ; each part, each moment, each position,

X. each element of its movement, is the If we now compare the two methods, combined effect of the two impelling their aptness, function, and provinces, forces. The two effects are so com- we shall find, as in an abstract, the his. mingled, that we cannot isolate either tory, divisions, hopes, and limits of of them, and refer it to its source. In human science. The first appears at order to perceive each effect separately the beginning, the second at the end ve should have to consider the move- The first necessarily gained ascend. nents apart, that is, to suppress the ency in Bacon's time, and now begins actual movement, and to replace it by to lose it; the second necessarily lost others. Neither the Method of Agree ascendency in Bacon's time, and now ment, nor of Difference, nor of Resi- begins to regain it. So that science, dues, nor of Concomitant Variations, after having passed from the deductive which are all decomposing and elimi- to the experimental state, is now pass. native, can avail against a phenomenon ing from the experimental to the de waich by its nature excludes all elemi. | ductive. Induction has for its provina nation and decomposition. We must

• Mill's Logus, l.

phenomena which are capable of being nature when we shall have deduced decomposed, and on which we can her millions of facts from two or threo experiment. Deduction has for its prov- laws. ince indecomposable phenomena, or I venture to say that the theory these on which we cannot experiment which you have just heard is perfect. The first is efficacious in physics, chem. I have omitted several of its cha-acter istry, zoology, and botany, in the earl- istics, but you have seen enough to rec ier stages of every science, and also ognize that induction has nowhere whenever phenomena are but slightly been explained in so complete and pie complicated, within our reach, capable cise a manner, with such an abundance of being modified by means at our dis- of fine and just distinctions, with sucb posal. The second is efficacious in as- extensive and exact applications, with tronomy, in the higher branches of such a knowledge of the practical physics, in physiology, history, in the methods and ascertained results of higher grades of every science, when science, with so complete an exclusion ever phenomena are very complicated, of metaphysical principles and arbi. as in animal and social life, or lie be- trary suppositions, and in a spirit more yond our reach, as the motions of the in conformity with the rigorous procedheavenly bodies and the changes of ure of modern experimental science. the atmosphere. When the proper You asked me just now what English method is not employed, science is at men have effected in philosophy; I a stand-still: when it is employed, sci- answer, the theory of Induction. Mili ence progresses. Here lies the whole is the last of that great line of philososecret of its past and its present. If the phers, which begins at Bacon, and physical sciences remained stationary which, through Hobbes, Newton, till the time of Bacon, it was because men Locke, Hume, Herschell, is continued used deduction when they should have down to our own times. They have used induction. If physiology and the carried our national spirit into philosmoral sciences are now making slow ophy; they have been positive and pracprogress, it is because we employ in- tical; they have not soared above facts; duction when deduction should be they have not attempted out-of-the used. It is by deduction, and accord. way paths; they have cleared the huing to physical and chemical laws, that man mind of its illusions, presumpwe shall be enabled to explain physio- tions, and fancies. They have emlogical phenomena. It is by deduction, ployed it in the only direction in which and according to mental laws, that we it can act: they only wished to mark shall be enabled to explain historical out and light up the already well-trodphenomena.* And that which has be- den ways of the progressive sciences. come the instrument of these two | They have not been willing to spend sciences, it is the object of all the their labor vainly in other than exothers to employ. All tend to become plored and verified paths ; they have deductive, and aim at being summed aided in the great modern work, the up in certain general propositions, from discovery of applicable laws; they which the rest may be deduced. The have contributed, as men of special atless numerous these propositions are, tainments do, to the increase of man's the more science advances. The fewer power. Can you find many philoso suppositions and postulates a science phers who have done as much requires, the more perfect it is become. Such a reduction is its final

XI. condition. Astronomy, acoustics, op. tics, present its models; we shall know You will tell me that our philoso

See chapter 9, book_vi. V. 2, 478, on the pher has clipped his wirgs in order to Physical or Concrete Deductive Method as

strengthen his legs. Certainly; and applied to Sociology ; and chapter 13, book iii., he has acted wisely. Experience limits for explanations, after Liebig, of Decomposition the career which it opens to us; it has Respiration, the Action of Poisons, etc. A given us our goal, but also var boun. whole book is devoted to the logic of the moral daries. We have only to observe the sciences ; I know no better treatise on the subbot.

elements of which our experience in composed, and the facts from which it we could explain every thing by there sets out, to understand that its range but we could not explain these themis limited. Its nature and its method selves. Mill says : confine its progress to a few steps. And, in the first place, * the ultimate originally and no others, or why they are com

“ Why these particular natural agents existed laws of nature cannot be less numer- mingled in such and such proportions, and dis ous than the several distinct species of tributed in such and such a manner throughou pur sensations. We can easily reduce space, is a question we cannot answer. More

than this : we can discover nothing regular in a movement to another movement, but the distribution itself; we can reduce it to a not the sensation of heat to that of uniformity, to no law. There are no means by smell, or of color, or of sound, nor

which, from the distribution of these cavees or either of these to a movement. We

agents in one part of space, we could conjeo

ture whether a similar distribution prevails in ca: easily connect together phenomena another." ** uf differe... degrees, but not phenomena diffenng in species. We find dis. And astronomy, which just now af tinct sensations at the bottom of all forded us the model of a perfect science, our knowledge, as simple indecompos science. We can predict the number

now affords us an example of a limited able elements, separated absolutely one {rom another, absolutely incapable of less positions of all the planetary being reduced one to another. Let ex

bodies : but we are obliged to assume, perience do what she will, she cannot

beside the primitive impulse and its suppress these diversities which consti- amount, not only the force of attraction tute her foundation. On the other hand, distances of all the bodies in question.

and its law, but also the masses and experience, do what she will, cannot es

We understand millions of facts, but cape from the conditions under which she acts. Whatever be her province, it it is by means of a hundred facts which is bounded by time and space; the fact

we do not comprehend ; we arrive at which she observes is limited and in- necessary results, but it is only by fluenced by an infinite number of other

means of accidental antecedents; só facts to which she cannot attain. She that if the theory of our universe were is obliged to suppose or recognize some

completed there would still remain two primordial condition from whence she great voids : one at the commencestarts, and which she does not explain.t ment of the physical world, the other Every problem has its accidental or

at the beginning of the moral world ; arbitrary data : we deduce the rest the one comprising the e.ements of from these, but there is nothing from being, the other embracing the ele. which these can be deduced. The sun,

ments of experience; one containing the earth, the planets, the initial im- primary sensations, the other primitive

“Our knowledge,” says Roypulse of heavenly bodies, the primitive agents. chemical properties of substances, are proceeded, of the laws of crystallisation, or

chemical composition, electricity, magnetism, such data. If we possessed them all

etc., points to various polarities, ultimately in* Mill's Logic, ii: 4.

herent in the particles of which bodies are com “ There exists in nature a number of Per- posed; the comparative atomic weights of dilmanent Causes, which have subsisted ever since ferent kinds of bodies were ascertained by rethe human race has been in existence, and for solving, into more general laws, the uniformities An indefinite and probably an enormous length observed in the proportions in which substances of time previous. The sun, the earth, and combine with one another; and so forth. Thus, planets, with their various constituents, air, although every resolution of a complex uniwater, and the other distinguishable substances, formity into simpler and more elementary laws whether simple or compound, of which nature has an apparent tendency to diminish the nums made up, are such Permanent Causes. They ber of the ultimate properties, and really does have existed, and the effects or consequences remove many properties from the list; yet which they were fitted to produce have taken (since the result of this simplifying process is to plact (as often as the other conditions of the trace up an ever greater variety of different efproduction met), from the very beginning of our fects to the same agents), the further we ad experience. But we can give no account of the vance in this direction, the greater number of origin of the Permanent Eauses themselves."

- distinct properties we are forced to recognise ir Mill's Logic, i. 378.

cases.

er-Collard, "consists in tracing ignor- which we have found to be d universally on ou ance as far back as possible."

own planet. The uniform. ty in the succession Can we at least affirm that these irre- must be received not as a law of the universe,

of events, otherwise called the law of causation, ducible data are so only in appearance, but of that portion of it only which is within and in relation to our mind? Can we the rauge of our means of sure observation, with say that they have causes, like the de- a reasonable degree of extension to adjacent

To extend it further is to make a sup rived facts of which they are the position without evidence, and to which, in the causes ? Can we conclude that every absence of any ground from experience for er event, always and everywhere, happens timating its degree of probability, it would Da according to laws, and that this little idle to attempt to assign any." world of ours, so well regulated, is a We are, then, irrevocably driven back wrt of epitome of the universe ? Can from the infinite ; our faculties and we, by aid of the axioms, quit our

our assertions cannot attain to it; we narrow confines, and affirm any thing of remain confined in a small circle ; ou the universe ? In no wise; and it is mind reaches not beyond its experience; here that Mill pushes his principles to

we can establish nc uriversai and ne their furthest consequences : for the cessary connection between facts; such a law which attributes à cause to every connection probably does not even exist. event, has to him no other foundation, Mill stops here; but certainly, by carworth, or scope, than what it derives rying out his idea to its full extent, we froin experience. It has no inherent should arrive at the conception of the necessity; it draws its whole authority world as a mere collection of facts ; no from the great number of cases in internal necessity would induce their which we have recognized it to be true; connection or their existence; they it only sums up a mass of observations; would be simple arbitrary, accidentally. it unites two data, which, considered existing facts. Sometimes, as in our in themselves, have no intimate con- system, they would be found assembled nection; it joins antecedents generally in such a manner as to give rise to reg; to consequents generally, just as the ular recurrences; sometimes they would law of gravitation joins a particular be so assembled that nothing of the antecedent to a particular consequent; sort would occur. Chance, as Demoit determines a couple, as do all experi- critus taught, would be at the foundamental laws, and shares in their un- tion of all things. Laws would be the certainty and in their restrictions. result of chance, and sometimes we Listen to this bold assertion :

should find them, sometimes not It

would be with existences as with num. I am convinced that any one accustomed to bers-decimal fractions, for instance, abstraction and analysis, who will fairly exert which, according to the chance of their his faculties for the purpose, will, when his imagination has once learnt to entertain the two primitive factors, sometimes recur notion, find no difficulty in conceiving that in regularly, and sometimes not. This sume one, for instance, of the many firmaments is certainly an original and lofty concep; into which sidereal astronomy now divides the

tion. It is the final consequence of universe, events may succeed one another at random, without any fixed law; nor can any. the primitive and dominant idea, which thing in our experience, or in our mental nature, we have discovered at the beginning of constitute a sufficient, or indeed any, reason for the system, which has transformed the believing that this is nowhere the case. The theories of Definition, of Propositions, grounds. theref:re, which warrant us in reject. ng such a supposition with respect to any of and of the Syllogism; which has re the phenomena of which we have experience, duced axioms to experimental truths · must be sought elsewhere than in any supposed which has developed and perfected the necessity of our intellectual faculties."

theory of induction ; which has estab. Prac.ically, wc may trust in so well-es- lished the goal, the limits, the province, tablished a law; but

and the methods of science ; which “In distant parts of the stellar ragions, everywhere, in nature and in sciere, where the phenomena may be entirely unlike has suppressed interior connections ; those with which we are acquainted, it would be which has replaced the necessary by folly to affirm confidently that this general the accidental ; cause by antecedent : Law prevails, any more than those special ones and which consists in affirming that

• Mil's Logic, ii. 95.