Translations · Text 108 · 1101100

Animadversiones, Part II, article 53: the figure comparing Descartes's rules with the true rules

Reference
GP IV 381–384
Date
1692
Original
Latin
Earlier English
this is a new translation (report tab 2)
Physics and dynamicsMetaphysics
View

Editor’s note

Article 53 is the graphical companion to the law of continuity. Leibniz fixes two equal bodies, B with a constant motion BW, and lets the motion of C run continuously from "same direction, same speed" through rest to "opposite direction, same speed". He then plots the motion of each body after the impact as an ordinate. The true outcomes lie on two straight lines; Descartes's rules give a broken "monstrous" drawing with jumps. Gerhardt prints the figure on a folding plate with a Latin key. The plate cannot be reproduced here, and the OCR of its small subscript labels is badly damaged, so the case labels are normalised as 1H to 5H (1H: C moves with B at equal speed; 2H: C moves the same way more slowly; 3H: H falls on A, C at rest; 4H: C moves against B more slowly; 5H: C moves against B at equal speed). Where the key cannot be read with confidence it is marked [?].

Original (GP IV 381–384)
English
1
Ad artic. (53). Usum regularum suarum Cartesius agnoscit difficilem esse, scilicet quod videret plane repugnare experimentis. Cum in veris motus regulis mirus sit consensus rationis et experientiae, neque ambientia tantum impediant regularum verarum successum, quantum ille videtur timere, ut scilicet exceptiones haberet ad effugiendum paratas. Quin potius, quanto magis corpora et dura et magna sunt, eo accuratius regulae ab observationibus exprimuntur. Sed quid dura aut liquida conferant mox videbimus, hoc loco placet ostendere in figura ad faciliorem captum, quomodo ope Criterii nostri adumbrari in rebus possit veritas velut in antecessum, et praeludio quodam, etiam antequam perfecta ejus delineatio obtineri queat: quod magnum usum habet pariter ad agnoscendos errores et ad appropinquandum veritati.
On art. 53. Descartes admits that the use of his rules is difficult, evidently because he saw that they flatly contradict experiments. Yet in the true rules of motion there is a wonderful agreement of reason and experience, and the surrounding bodies do not hinder the working of the true rules as much as he seems to fear, so as to have exceptions ready as a way of escape. On the contrary, the harder and larger the bodies are, the more exactly the rules are borne out by observations. What hardness or fluidity contributes we shall see shortly. Here I should like to show in a figure, to make it easier to grasp, how with the help of our criterion the truth can be sketched in things in advance, as it were, by a kind of prelude, even before a complete drawing of it can be obtained. This is of great use both for recognising errors and for approaching the truth.
2
Ponamus ergo corpora B ac C esse aequalia, et ipsius B celeritatem ac directionem repraesentari recta BW, ita ut tendat a B versus W, celeritate ut BW; sed ipsius C celeritatem atque directionem aliam atque aliam pro diversis casibus assumi AH, ita quidem ut in casu A1H vel A2H seu infra A directio ipsius C sit eadem quae ipsius B, adeoque in casu A1H quae aequalis est ipsi BW, celeritatem amborum esse aequalem, et directionem eandem, sed sumto H propius ad A, ut in A2H, directionem quidem ipsius C ab A ad 2H esse eandem quae ipsius B, quae est a B ad W, verum celeritatem ipsius C esse minorem quam ipsius B, quia A2H minor est quam BW, adeoque a B insequente attingi praecedens C. Si H incidat in A, ut 3H, tunc directionem et celeritatem ipsius C esse nullam, seu C quiescere; sin H sumatur supra A ut in A4H, A5H, tunc ipsius C directionem esse contrariam directioni ipsius B. Ducantur jam lineae PP et QQ, ita ut sit ordinatim applicata HP celeritas et directio ipsius B post ictum, et HQ celeritas et directio ipsius C post ictum hoc observato ut directio alterutrius vel utriusque quae post ictum est ad easdem partes ad quas ante ictum directio ipsius B fuerat, exprimatur per applicatam sinistram, et quae est ad partes contrarias directioni ipsius B signetur per applicatam dextram.
Let us suppose, then, that the bodies B and C are equal, and that the speed and direction of B are represented by the straight line BW, so that it moves from B toward W with a speed as BW. The speed and direction of C, however, are taken as different lines AH in different cases, in such a way that in the case A1H or A2H, that is, below A, the direction of C is the same as that of B. So in the case A1H, which is equal to BW, the two have equal speed and the same direction; but with H taken nearer to A, as in A2H, the direction of C, from A to 2H, is the same as that of B, which is from B to W, but the speed of C is less than that of B, because A2H is less than BW, and so the C in front is caught up by the B following. If H falls on A, as 3H, then the direction and speed of C are nothing, that is, C is at rest. But if H is taken above A, as in A4H, A5H, then the direction of C is contrary to that of B. Now let the lines PP and QQ be drawn so that the ordinate HP is the speed and direction of B after the impact, and HQ the speed and direction of C after the impact, observing this convention: a direction of either or both after the impact that is the same as B's direction before the impact is expressed by an ordinate to the left, and one contrary to B's direction is marked by an ordinate to the right.
3
Jam determinemus quaedam puncta linearum PP et QQ. Directio et celeritas seu (verbo) motus ante concursum ipsius B semper est BW, jam si ante concursum motus ipsius C sit ei aequalis et ad easdem partes nempe A1H (aequ. BW), utique ut constat non obstante contactu ambo B ac C retinebunt priorem celeritatem et directionem adeoque rectae 1H1P et 1H1Q motum ipsorum B ac C post ictum repraesentantes erunt aequales ipsi A1H seu BW et ad sinistram. Sed si motus ipsius C ante ictum sit nullus, seu A3H, ipso puncto 3H incidente in A, seu si C quiescat, tunc constat etiam quid fiat, nempe post ictum corpus B quiescet adeoque locus 3P incidet in A, C vero accipiet celeritatem quam habuerat B, et ad eas partes ad quas tenderat B; ergo habetur 3H3Q aequalis ipsi BW et ad sinistram, adeoque et puncta 3P et 3Q habentur. Denique si motus ipsius C sit aequalis motui ipsius B, sed ei contrarius, seu repraesentetur per A5H aequalem ipsi BW, sed 5H sumto supra A, hoc est si corpora proposita (aequalia ex hypothesi) aequalibus celeritatibus oppositisque directionibus concurrant, tunc etiam habetur eventus, nam eadem utrumque qua venerat celeritate reflectetur, adeoque dabuntur puncta 5P, 5Q. Nam 5H5P erit aequalis ipsi BW sed ad dextram, quia ita B reflectitur seu directione pristinae contraria movetur; et 5H5Q erit aequalis eidem, sed ad sinistram, quia C accipit directionem quam habuerat B.
Now let us determine some points of the lines PP and QQ. The direction and speed, in a word the motion, of B before the collision is always BW. If before the collision the motion of C is equal to it and in the same direction, namely A1H (equal to BW), then, as is evident, despite their contact both B and C will keep their former speed and direction, and so the lines 1H1P and 1H1Q, representing the motion of B and of C after the impact, will be equal to A1H, that is, to BW, and to the left. If the motion of C before the impact is nothing, that is A3H, the point 3H falling on A, in other words if C is at rest, then it is also evident what happens: after the impact body B will be at rest, so that the place 3P falls on A, while C will receive the speed B had, in the direction in which B was moving. So 3H3Q is equal to BW and to the left, and we have the points 3P and 3Q. Finally, if the motion of C is equal to that of B but contrary to it, that is, represented by A5H equal to BW but with 5H taken above A, in other words if the given bodies (equal by hypothesis) collide with equal speeds and opposite directions, then too we have the outcome: each will be reflected with the speed with which it came, and so we get the points 5P and 5Q. For 5H5P will be equal to BW but to the right, because B is thus reflected, that is, moves in the direction contrary to its former one; and 5H5Q will be equal to the same, but to the left, because C takes on the direction B had.
4
Habemus ergo tam puncta 1P, 3P, 5P, quae (quod notabile est) incidunt in rectam, quam et puncta 1Q, 3Q, 5Q, quae incidunt in aliam rectam et quidem rectae AH parallelam; puncta autem reliqua ut 2P, 4P etc. vel 2Q, 4Q equidem ex solo Criterio nostro, seu ex lege Continuitatis determinari non possunt, seu non constat ex his solis, quaenam sint lineae PP et QQ. Sufficitque omnia una linea continua connecti, quo locus sit Criterio nostro, omnesque adeo regulae istae incohaerentes jam tum ante plenam rei cognitionem aut inventionem speciei lineae excluduntur. Interim aliunde scimus, revera has lineas PP et QQ esse rectas, et ob permutationem in aequalibus corporibus celeritatum ac directionum HP semper aequalem esse ipsi AH, et HQ ipsi BW [?], adeoque tam HQ quam BW concipi hic posse constantes et simul motum significare ad easdem partes. Caeterum eas infra 1H, 1P, 1Q non produco, quia tunc B tardius moveretur quam C, adeoque non attingeret, neque adeo concursus intelligi posset. Eodem modo positis celeritatibus corporum aequalibus, et uno corpore retento, altero magnitudine variato, schema fieri posset ad eventum in utroque corpore duabus lineis designandum; imo quaecunque fieret hypothesis, una variabilitate excepta permanens, liceret adumbrare similem figuram. Sed sufficit specimen in uno dedisse, praesertim cum alia methodo omnia perfecte assequamur, quae per hanc tantum adumbrantur, quod ipsum tamen suum usum ad errores refutandos habere ostendimus. Et si nondum detecta tota res esset, ad quandam adumbrationem prodesset. Ex Cartesianis vero regulis non posset duci continua quaedam linea eventuum variatorum respondens lineae continuae hypotheseos variantis, et prodiret delineatio plane monstrosa et contraria nostro Criterio artic. 45 seu Legi Continuitatis. Placuit in figura comparare nostram et Cartesianam delineationem, unde hujus inconsistentia aut potius impossibilitas ad oculum apparebit.
So we have the points 1P, 3P, 5P, which (a notable thing) fall on a straight line, and the points 1Q, 3Q, 5Q, which fall on another straight line, one parallel to the line AH. The remaining points, such as 2P, 4P and so on, or 2Q, 4Q, cannot be determined from our criterion, the law of continuity, alone; from these alone it is not settled what the lines PP and QQ are. It is enough that everything be joined by one continuous line, so that our criterion has room to apply, and so all those incoherent rules are already excluded before the matter is fully known or the kind of line discovered. Meanwhile we know from elsewhere that these lines PP and QQ are in fact straight, and that, because equal bodies exchange their speeds and directions, HP is always equal to AH and HQ to BW [?], so that both HQ and BW can be taken here as constant and as signifying motion in the same direction. I do not continue the lines below 1H, 1P, 1Q, because then B would move more slowly than C and so would not reach it, and no collision could be understood. In the same way, with the speeds of the bodies taken as equal, one body held fixed and the other varied in size, a diagram could be made showing the outcome for each body by two lines; indeed, whatever hypothesis were made, if all but one variable is held fixed, a similar figure could be sketched. But it is enough to have given a sample in one case, especially since by another method we attain perfectly everything that is only sketched by this one; still, we have shown that it has its use in refuting errors, and if the whole matter had not yet been discovered it would serve for a kind of sketch. From the Cartesian rules, however, no continuous line of varied outcomes could be drawn answering to the continuous line of the varying hypothesis, and a downright monstrous drawing would result, contrary to our criterion of art. 45, the law of continuity. I have chosen to compare our drawing and the Cartesian one in the figure, so that the inconsistency, or rather the impossibility, of the latter will be plain to the eye.
5
[Plate.] Ad artic. 53 partis secundae. Repraesentatio Regularum Motus in casu aequalitatis corporum concurrentium. Secundum Cartesium: Delineatio monstrosa. Secundum Veritatem: Delineatio concinna. Explicatio Figurae. Ante ictum: Motus corporis B est BW constans, corporis C est A1H, A2H etc. Post ictum: Motus corporis B est Cartesio H1p, H2p etc., nobis H1P, H2P etc.; corporis C est Cartesio H1q, H2q etc. [?], nobis HQ constans. H infra [supra] A; sinisteriora [dexteriora] quam AH significant directionem eandem [contrariam] ei quam habuit B ante ictum. [Puncta ... ex reg. Cartes. 7; ... ex reg. Cartes. 3; 3p, 3q ex reg. ejus 6; 5p, 5q ex reg. ejus 1 [?].] Delineationes nostra et Cartesiana ex infinitis casibus habent duos solum convenientes, 1H et 5H. Delineatio nostra observat Legem continuitatis, Cartesiana interrumpitur et saltum habet quoad lineam pp in 3p et 5p, [quoad lineam qq] in 3q et 5q [?]; saltus notatus est punctis ad fingendam continuationem. Impossibile est ut a duarum linearum continuatarum pp et qq coincidentia excipi possit numerus punctorum determinatus, verbi gratia binarius, ut scilicet omnia p cum omnibus q respondentibus coincidant exceptis duobus solum casibus, scilicet 3p cum 3q et 5p cum 5q. Et hoc tamen fieri debet in delineatione Cartesiana.
[Plate.] On art. 53 of Part II. Representation of the rules of motion in the case of equal colliding bodies. According to Descartes: a monstrous drawing. According to the truth: a well-fitted drawing. Key to the figure. Before the impact: the motion of body B is BW, constant; that of body C is A1H, A2H, etc. After the impact: the motion of body B is, for Descartes, H1p, H2p, etc., for us H1P, H2P, etc.; that of body C is, for Descartes, H1q, H2q, etc. [?], for us HQ, constant. H below [above] A, and ordinates further left [further right] than AH, signify the same [the contrary] direction to the one B had before the impact. [The Cartesian points come from his rule 7 (C moving the same way), rule 3 (C moving against B), rule 6 (3p, 3q, C at rest), and rule 1 (5p, 5q) [?].] Of infinitely many cases, our drawing and the Cartesian one agree in only two, 1H and 5H. Our drawing keeps the Law of Continuity; the Cartesian one is broken and has a jump in the line pp at 3p and 5p, and in the line qq at 3q and 5q [?]; the jump is marked with dots to imitate a continuation. It is impossible that from the coincidence of two continuous lines pp and qq a fixed number of points, say two, should be excepted, so that every p coincides with every corresponding q except in only two cases, namely 3p with 3q and 5p with 5q. Yet this is what must happen in the Cartesian drawing.

Notes to text 108

  1. Descartes, Principia II.53, says the rules are hard to apply because no body in the world is ever free of surrounding bodies; arts. 54–63 then deal with hard and fluid bodies. Leibniz reads this as an escape clause and answers that experiments with large hard bodies confirm the true rules best.
  2. The figure: the vertical axis carries the motion of C before impact (H below A means C moves with B, above A against it, A itself rest). Horizontal ordinates from that axis give each body's motion after impact, leftward for B's original direction, rightward for the reverse. For equal elastic bodies the bodies simply exchange motions, so B's outcome line PP is a straight line through A (HP = AH) and C's outcome line QQ is a vertical straight line at distance BW (HQ = BW). This is a very early instance of plotting outcome against a continuously varied parameter to test a physical law.
  3. Descartes's drawing: for equal bodies, rules 3 and 7 make the two bodies move on together after impact, so his p and q points coincide everywhere except at rest (rule 6: B rebounds with 3/4, C goes on with 1/4) and at equal and opposite speeds (rule 1). Leibniz's objection in the key is that two continuous lines cannot coincide everywhere except at two isolated points. The reconstruction of the key's rule numbers follows the argument, since the OCR of the plate's legend is damaged [?].
  4. "HQ ipsi BW" is a reconstruction; the OCR reads "/Q ... LM", which fits no line named in the text, and the sense (C always leaves with B's former motion BW) requires BW [?].
  5. "Alia methodo omnia perfecte assequamur": the exact rules derived from conservation of vis viva and of "progress" (relative velocity and the common centre of gravity), as in the Dynamica de potentia and the Specimen dynamicum.
This text in the whole · 1692, a year from which 198 pieces are catalogued166316901716