CHAPTER XVI.Additional Remarks on Francis Bacon.Francis Baconand his works have recently been discussed and examined by various writers in France and Germany as well as England[197]. Not to mention smaller essays, M. Bouillet has published a valuable edition of his philosophical works; Count Joseph de Maistre wrote a severe critique of his philosophy, which has been published since the death of the author; M. Charles Remusat has written a lucid and discriminating Essay on the subject; and in England we have had a new edition of the works published, with a careful and thoughtful examination of the philosophy which they contain, written by one of the editors: a person especially fitted for such an examination by an acute intellect, great acquaintance with philosophical literature, and a wide knowledge of modern science. Robert Leslie Ellis, the editor of whom I speak, died during the publication of the edition, and before he had done full justice to his powers; but he had already written various dissertations on Bacon's philosophy, which accompany the different Treatises in the new edition.Mr. Ellis has given a more precise view than any of his predecessors had done of the nature of Bacon'sinduction and of his philosophy of discovery. Bacon's object was to discover the 'natures' or essences of things, in order that he might reproduce these natures or essences at will; he conceived that these natures were limited in number, and manifested in various combinations in the bodies which exist in the universe; so that by accumulating observations of them in a multitude of cases, we may learn by induction in what they do and in what they do not consist; theInductionwhich is to be used for this purpose consists in a great measure ofexcludingthe cases which do not exhibit the 'nature' in question; and by such exclusion, duly repeated, we have at last left in our hands the elements of which the proposed nature consists. And the knowledge which is thus obtained may be applied to reproduce the things so analysed. As exhibiting this view clearly we may take a passage in theSylva Sylvarum: "Gold has these natures: greatness of weight, closeness of parts, fixation, pliantness or softness, immunity from rust, colour or tincture of yellow. Therefore the sure way, though most about, to make gold, is to know the causes of the several natures before rehearsed, and the axioms concerning the same. For if a man can make a metal, that hath all these properties, let men dispute whether it be gold or no." He means that however they dispute, it is gold for all practical purposes.For such an Induction as this, Bacon claims the merit both of being certain, and of being nearly independent of the ingenuity of the inquirer. It is a method which enables all men to make exact discoveries, as a pair of compasses enables all men to draw an exact circle.Now it is necessary for us, who are exploring the progress of the true philosophy of discovery, to say plainly that this part of Bacon's speculation is erroneous and valueless. No scientific discovery ever has been made in this way. Men have not obtained truths concerning the natural world by seeking for the natures of things, and by extracting them from phenomena by rejecting the cases in which they were not.On the contrary, they have begun by ascertaining thelaws of the phenomena; and have then gone on, not by a mechanical method which levels all intellect, but by special efforts of the brightest intellects to catch hold of the ideas by which these laws of phenomena might be interpreted and expressed in more general terms. These two steps, the finding the laws of phenomena, and finding the conceptions by which those laws can be expressed, are really the course of discovery, as the history of science exhibits it to us.Bacon, therefore, according to the view now presented, was wrong both as to his object and as to his method. He was wrong in taking for his object the essences of things,—the causes of abstract properties: for these man cannot, or can very rarely discover; and all Bacon's ingenuity in enumerating and classifying these essences and abstract properties has led, and could lead, to no result. The vast results of modern science have been obtained, not by seeking and finding the essences of things, but by exploring the laws of phenomena and the causes of those laws.And Bacon's method, as well as his object, is vitiated by a pervading error:—the error of supposing that to be done by method which must be done by mind;—that to be done by rule which must be done by a flight beyond rule;—that to be mainly negative which is eminently positive;—that to depend on other men which must depend on the discoverer himself;—that to be mere prose which must have a dash of poetry;—that to be a work of mere labour which must be also a work of genius.Mr. Ellis has seen very clearly and explained very candidly that this method thus recommended by Bacon has not led to discovery. "It is," he says, "neither to the technical part of his method nor to the details of his view of the nature and progress of science, that his great fame is justly owing. His merits are of another kind. They belong to the spirit rather than to the positive precepts of his philosophy."As the reader of the last chapter will see, this amounts to much the same as the account which Ihad given of the positive results of Bacon's method, and the real value of that portion of his philosophy which he himself valued most. But still there remain, as I have also noted, portions of Bacon's speculations which have a great and enduring value, namely, his doctrine that Science is the Interpretation of Nature, his distinction of this Interpretation of Nature from the vicious and premature Anticipation of Nature which had generally prevailed till then; and the recommendation of a graduated and successive induction by which alone the highest and most general truths were to be reached. These are points which he urges with great clearness and with great earnestness; and these are important points in the true philosophy of discovery.I may add that Mr. Ellis agrees with me in noting the invention of the conception by which the laws of phenomena are interpreted as something additional toInduction, both in the common and in the Baconian sense of the word. He says (General Preface, Art. 9), "In all cases this process [scientific discovery] involves an element to which nothing corresponds in the Tables of Comparence and Exclusion; namely the application to the facts of aprincipleof arrangement, anidea, existing in the mind of the discoverer antecedently to the act of induction." It may be said that this principle or idea is aimed at in the Baconian analysis. "And this is in one sense true: but it must be added, that thisanalysis, if it be thought right to call it so, is of the essence of the discovery which results from it. To take for granted that it has been already effected is simply apetitio principii. In most cases the mere act of induction follows as a matter of course as soon as theappropriate ideahas been introduced." And as an example he takes Kepler's invention of the ellipse, as the idea by which Mars's motions could be reduced to law; making the same use of this example which we have repeatedly made of it.Mr. Ellis may at first sight appear to express himself more favourably than I have done, with regard tothe value of Bacon'sInquisitio in Naturam Calidiin the Second Book of theNovum Organon. He says of one part of it[198]: "Bacon here anticipates not merely the essential character of the most recent theory of heat, but also the kind of evidence by which it has been established.... The merit of having perceived the true significance of the production of heat by friction belongs of right to Bacon."But notwithstanding this, Mr. Ellis's general judgment on this specimen of Bacon's application of his own method does not differ essentially from mine. He examines theInquisitioat some length, and finally says: "If it were affirmed that Bacon, after having had a glimpse of the truth suggested by some obvious phenomena, had then recourse, as he himself expresses it, to certain 'differentiæ inanes' in order to save the phenomena, I think it would be hard to dispute the truth of the censure."Another of the Editors of this edition (Mr. Spedding) fixes his attention upon another of the features of the method of discovery proposed by Bacon, and is disposed to think that the proposed method has never yet had justice done it, because it has not been tried in the way and on the scale that Bacon proposes[199]. Bacon recommended that a great collection of facts should be at once made and accumulated, regarding every branch of human knowledge; and conceived that, when this had been done by common observers, philosophers might extract scientific truths from this mass of facts by the application of a right method. This separation of the offices of the observer and discoverer, Mr. Spedding thinks is shown to be possible by such practical examples as meteorological observations, made by ordinary observers, and reduced to tables and laws by a central calculator; by hydrographical observations made by ships provided with proper instructions, and reduced to general laws by theman of science in his study; by magnetical observations made by many persons in every part of the world, and reduced into subservience to theory by mathematicians at home.And to this our reply will be, in the terms which the history of all the Sciences has taught us, that such methods of procedure as this do not belong to theEpoch of Discovery, but to the Period ofverificationandapplicationof the discovery which follows. When a theory has been established in its general form, our knowledge of the distribution of its phenomena in time and space can be much promoted by ordinary observers scattered over the earth, and succeeding each other in time, provided they are furnished with instruments and methods of observation, duly constructed on the principles of science; but such observers cannot in any degree supersede the discoverer who is first to establish the theory, and to introduce into the facts a new principle of order. When the laws of nature have been caught sight of, much may be done, even by ordinary observers, in verifying and exactly determining them; but when a real discovery is to be made, this separation of the observer and the theorist is not possible. In those cases, the questioning temper, the busy suggestive mind, is needed at every step, to direct the operating hand or the open gaze. No possible accumulation of facts about mixture and heat, collected in the way of blind trial, could have led to the doctrines of chemistry, or crystallography, or the atomic theory, or voltaic and chemical and magnetic polarity, or physiology, or any other science. Indeed not only is an existing theory requisite to supply the observer with instruments and methods, but without theory he cannot even describe his observations. He says that he mixes an acid and an alkali; but what is an acid? What is an alkali? How does he know them? He classifies crystals according to their forms: but till he has learnt what is distinctive in the form of a crystal, he cannot distinguish a cube from a square prism, even if he had a goniometer and could use it. And the like impossibility hangs over all the other subjects. Toreport facts for scientific purposes without some aid from theory, is not only useless, but impossible.When Mr. Spedding says, "I could wish that men of science would apply themselves earnestly to the solution of this practical problem: What measures are to be taken in order that the greatest variety of judicious observations of nature all over the world may be carried on in concert upon a common plan and brought to a common centre:"—he is urging upon men of science to do what they have always done, so far as they have had any power, and in proportion as the state of science rendered such a procedure possible and profitable to science. In Astronomy, it has been done from the times of the Greeks and even of the Chaldeans, having been begunas soon asthe heavens were reduced to law at all. In meteorology, it has been done extensively, though to little purpose, because the weather hasnot yetbeen reduced to rule. Men of science have shown how barometers, thermometers, hygrometers, and the like, may be constructed; and these may be now read by any one as easily as a clock; but of ten thousand meteorological registers thus kept by ordinary observers, what good has come to science? Again: The laws of the tides have been in a great measure determined by observations in all parts of the globe,becausetheory pointed out what was to be observed. In like manner the facts of terrestrial magnetism were ascertained with tolerable completeness by extended observations,then, and then only, when a most recondite and profound branch of mathematics had pointed out what was to be observed, and most ingenious instruments had been devised by men of science for observing. And even with these, it requires an education to use the instruments. But in many cases no education in the use of instruments devised by others can supersede the necessity of a theoretical and suggestive spirit in the inquirer himself. He must devise his own instruments and his own methods, if he is to make any discovery. What chemist, or inquirer about polarities, or about optical laws yet undiscovered, can make any progress by using anotherman's experiments and observations? He must invent at every step of his observation; and the observer and theorist can no more be dissevered, than the body and soul of the inquirer.That persons of moderate philosophical powers may, when duly educated, make observations which may be used by greater discoverers than themselves, is true. We have examples of such a subordination of scientific offices in astronomy, in geology, and in many other departments. But still, as I have said, a very considerable degree of scientific education is needed even for the subordinate labourers in science; and the more considerable in proportion as science advances further and further; since every advance implies a knowledge of what has already been done, and requires a new precision or generality in the new points of inquiry.
Francis Baconand his works have recently been discussed and examined by various writers in France and Germany as well as England[197]. Not to mention smaller essays, M. Bouillet has published a valuable edition of his philosophical works; Count Joseph de Maistre wrote a severe critique of his philosophy, which has been published since the death of the author; M. Charles Remusat has written a lucid and discriminating Essay on the subject; and in England we have had a new edition of the works published, with a careful and thoughtful examination of the philosophy which they contain, written by one of the editors: a person especially fitted for such an examination by an acute intellect, great acquaintance with philosophical literature, and a wide knowledge of modern science. Robert Leslie Ellis, the editor of whom I speak, died during the publication of the edition, and before he had done full justice to his powers; but he had already written various dissertations on Bacon's philosophy, which accompany the different Treatises in the new edition.
Mr. Ellis has given a more precise view than any of his predecessors had done of the nature of Bacon'sinduction and of his philosophy of discovery. Bacon's object was to discover the 'natures' or essences of things, in order that he might reproduce these natures or essences at will; he conceived that these natures were limited in number, and manifested in various combinations in the bodies which exist in the universe; so that by accumulating observations of them in a multitude of cases, we may learn by induction in what they do and in what they do not consist; theInductionwhich is to be used for this purpose consists in a great measure ofexcludingthe cases which do not exhibit the 'nature' in question; and by such exclusion, duly repeated, we have at last left in our hands the elements of which the proposed nature consists. And the knowledge which is thus obtained may be applied to reproduce the things so analysed. As exhibiting this view clearly we may take a passage in theSylva Sylvarum: "Gold has these natures: greatness of weight, closeness of parts, fixation, pliantness or softness, immunity from rust, colour or tincture of yellow. Therefore the sure way, though most about, to make gold, is to know the causes of the several natures before rehearsed, and the axioms concerning the same. For if a man can make a metal, that hath all these properties, let men dispute whether it be gold or no." He means that however they dispute, it is gold for all practical purposes.
For such an Induction as this, Bacon claims the merit both of being certain, and of being nearly independent of the ingenuity of the inquirer. It is a method which enables all men to make exact discoveries, as a pair of compasses enables all men to draw an exact circle.
Now it is necessary for us, who are exploring the progress of the true philosophy of discovery, to say plainly that this part of Bacon's speculation is erroneous and valueless. No scientific discovery ever has been made in this way. Men have not obtained truths concerning the natural world by seeking for the natures of things, and by extracting them from phenomena by rejecting the cases in which they were not.On the contrary, they have begun by ascertaining thelaws of the phenomena; and have then gone on, not by a mechanical method which levels all intellect, but by special efforts of the brightest intellects to catch hold of the ideas by which these laws of phenomena might be interpreted and expressed in more general terms. These two steps, the finding the laws of phenomena, and finding the conceptions by which those laws can be expressed, are really the course of discovery, as the history of science exhibits it to us.
Bacon, therefore, according to the view now presented, was wrong both as to his object and as to his method. He was wrong in taking for his object the essences of things,—the causes of abstract properties: for these man cannot, or can very rarely discover; and all Bacon's ingenuity in enumerating and classifying these essences and abstract properties has led, and could lead, to no result. The vast results of modern science have been obtained, not by seeking and finding the essences of things, but by exploring the laws of phenomena and the causes of those laws.
And Bacon's method, as well as his object, is vitiated by a pervading error:—the error of supposing that to be done by method which must be done by mind;—that to be done by rule which must be done by a flight beyond rule;—that to be mainly negative which is eminently positive;—that to depend on other men which must depend on the discoverer himself;—that to be mere prose which must have a dash of poetry;—that to be a work of mere labour which must be also a work of genius.
Mr. Ellis has seen very clearly and explained very candidly that this method thus recommended by Bacon has not led to discovery. "It is," he says, "neither to the technical part of his method nor to the details of his view of the nature and progress of science, that his great fame is justly owing. His merits are of another kind. They belong to the spirit rather than to the positive precepts of his philosophy."
As the reader of the last chapter will see, this amounts to much the same as the account which Ihad given of the positive results of Bacon's method, and the real value of that portion of his philosophy which he himself valued most. But still there remain, as I have also noted, portions of Bacon's speculations which have a great and enduring value, namely, his doctrine that Science is the Interpretation of Nature, his distinction of this Interpretation of Nature from the vicious and premature Anticipation of Nature which had generally prevailed till then; and the recommendation of a graduated and successive induction by which alone the highest and most general truths were to be reached. These are points which he urges with great clearness and with great earnestness; and these are important points in the true philosophy of discovery.
I may add that Mr. Ellis agrees with me in noting the invention of the conception by which the laws of phenomena are interpreted as something additional toInduction, both in the common and in the Baconian sense of the word. He says (General Preface, Art. 9), "In all cases this process [scientific discovery] involves an element to which nothing corresponds in the Tables of Comparence and Exclusion; namely the application to the facts of aprincipleof arrangement, anidea, existing in the mind of the discoverer antecedently to the act of induction." It may be said that this principle or idea is aimed at in the Baconian analysis. "And this is in one sense true: but it must be added, that thisanalysis, if it be thought right to call it so, is of the essence of the discovery which results from it. To take for granted that it has been already effected is simply apetitio principii. In most cases the mere act of induction follows as a matter of course as soon as theappropriate ideahas been introduced." And as an example he takes Kepler's invention of the ellipse, as the idea by which Mars's motions could be reduced to law; making the same use of this example which we have repeatedly made of it.
Mr. Ellis may at first sight appear to express himself more favourably than I have done, with regard tothe value of Bacon'sInquisitio in Naturam Calidiin the Second Book of theNovum Organon. He says of one part of it[198]: "Bacon here anticipates not merely the essential character of the most recent theory of heat, but also the kind of evidence by which it has been established.... The merit of having perceived the true significance of the production of heat by friction belongs of right to Bacon."
But notwithstanding this, Mr. Ellis's general judgment on this specimen of Bacon's application of his own method does not differ essentially from mine. He examines theInquisitioat some length, and finally says: "If it were affirmed that Bacon, after having had a glimpse of the truth suggested by some obvious phenomena, had then recourse, as he himself expresses it, to certain 'differentiæ inanes' in order to save the phenomena, I think it would be hard to dispute the truth of the censure."
Another of the Editors of this edition (Mr. Spedding) fixes his attention upon another of the features of the method of discovery proposed by Bacon, and is disposed to think that the proposed method has never yet had justice done it, because it has not been tried in the way and on the scale that Bacon proposes[199]. Bacon recommended that a great collection of facts should be at once made and accumulated, regarding every branch of human knowledge; and conceived that, when this had been done by common observers, philosophers might extract scientific truths from this mass of facts by the application of a right method. This separation of the offices of the observer and discoverer, Mr. Spedding thinks is shown to be possible by such practical examples as meteorological observations, made by ordinary observers, and reduced to tables and laws by a central calculator; by hydrographical observations made by ships provided with proper instructions, and reduced to general laws by theman of science in his study; by magnetical observations made by many persons in every part of the world, and reduced into subservience to theory by mathematicians at home.
And to this our reply will be, in the terms which the history of all the Sciences has taught us, that such methods of procedure as this do not belong to theEpoch of Discovery, but to the Period ofverificationandapplicationof the discovery which follows. When a theory has been established in its general form, our knowledge of the distribution of its phenomena in time and space can be much promoted by ordinary observers scattered over the earth, and succeeding each other in time, provided they are furnished with instruments and methods of observation, duly constructed on the principles of science; but such observers cannot in any degree supersede the discoverer who is first to establish the theory, and to introduce into the facts a new principle of order. When the laws of nature have been caught sight of, much may be done, even by ordinary observers, in verifying and exactly determining them; but when a real discovery is to be made, this separation of the observer and the theorist is not possible. In those cases, the questioning temper, the busy suggestive mind, is needed at every step, to direct the operating hand or the open gaze. No possible accumulation of facts about mixture and heat, collected in the way of blind trial, could have led to the doctrines of chemistry, or crystallography, or the atomic theory, or voltaic and chemical and magnetic polarity, or physiology, or any other science. Indeed not only is an existing theory requisite to supply the observer with instruments and methods, but without theory he cannot even describe his observations. He says that he mixes an acid and an alkali; but what is an acid? What is an alkali? How does he know them? He classifies crystals according to their forms: but till he has learnt what is distinctive in the form of a crystal, he cannot distinguish a cube from a square prism, even if he had a goniometer and could use it. And the like impossibility hangs over all the other subjects. Toreport facts for scientific purposes without some aid from theory, is not only useless, but impossible.
When Mr. Spedding says, "I could wish that men of science would apply themselves earnestly to the solution of this practical problem: What measures are to be taken in order that the greatest variety of judicious observations of nature all over the world may be carried on in concert upon a common plan and brought to a common centre:"—he is urging upon men of science to do what they have always done, so far as they have had any power, and in proportion as the state of science rendered such a procedure possible and profitable to science. In Astronomy, it has been done from the times of the Greeks and even of the Chaldeans, having been begunas soon asthe heavens were reduced to law at all. In meteorology, it has been done extensively, though to little purpose, because the weather hasnot yetbeen reduced to rule. Men of science have shown how barometers, thermometers, hygrometers, and the like, may be constructed; and these may be now read by any one as easily as a clock; but of ten thousand meteorological registers thus kept by ordinary observers, what good has come to science? Again: The laws of the tides have been in a great measure determined by observations in all parts of the globe,becausetheory pointed out what was to be observed. In like manner the facts of terrestrial magnetism were ascertained with tolerable completeness by extended observations,then, and then only, when a most recondite and profound branch of mathematics had pointed out what was to be observed, and most ingenious instruments had been devised by men of science for observing. And even with these, it requires an education to use the instruments. But in many cases no education in the use of instruments devised by others can supersede the necessity of a theoretical and suggestive spirit in the inquirer himself. He must devise his own instruments and his own methods, if he is to make any discovery. What chemist, or inquirer about polarities, or about optical laws yet undiscovered, can make any progress by using anotherman's experiments and observations? He must invent at every step of his observation; and the observer and theorist can no more be dissevered, than the body and soul of the inquirer.
That persons of moderate philosophical powers may, when duly educated, make observations which may be used by greater discoverers than themselves, is true. We have examples of such a subordination of scientific offices in astronomy, in geology, and in many other departments. But still, as I have said, a very considerable degree of scientific education is needed even for the subordinate labourers in science; and the more considerable in proportion as science advances further and further; since every advance implies a knowledge of what has already been done, and requires a new precision or generality in the new points of inquiry.