sign in sign up
<<
Home , John Archibald Wheeler

arXiv:0907.0902v1 [.class-ph] 6 Jul 2009 efudi h oko Janssen of can work approach the principle in a found of be favour in arguments Recent rnil racntutv hoy avyBonsre- Brown’s Harvey theory. book constructive cent a or principle when slower moving? tick force? it it a make see to of I clock result the the inside is changes the rod what unan- surely measuring are like: this STR of one’s contraction of understanding areas back of the other at difficulty questions the swered consequence, the a is, of As part That spe- theories. a “constructive” involving fields. as laws taught or are dynamical forces of physical basis cific the on deduced length and simultaneity of relativity contraction. many the student dilation, in the time introduce on to to relied STR is of stationary treatments which in introductory carriages) about railway bounces it moving as or of (often property par- of peculiar speed the is the is it it Therefore that unsatisfactory respectively. source postulate ticularly postulate relativity its light the by by the emitted out and ruled a apparently both or are in ether) propagating (the wave medium electromagnetic an a models natural either the as because light is of That counterintuitive or together. strongly taken are postulates when STR two of the case theory, the principle in a principles which as thermodynamics taught classical also of is the- case be- “principle” the problematic so-called unlike especially That is cause, a way that as them). STR taught Teaching of ory. usually versions is STR or the the is, in from proposed, (either deduction originally postulate by light forms the obtained and are postulate STR relativity of because is curricu- results This undergraduate the thermodynamics. the all classical in for from physics except in way lum of different both areas a taught, other in invariably the textbooks, almost and Another is classroom STR the contraction. that length simul- is and of factor relativity dilation of the time some especially taneity, of results, nature important counterintuitive the strongly One the acquaintance. is first on factor understand to difficult (STR) foefesta ti ett prahSRa princi- a principles from as STR STR of approach results to the best derive is and it theory that ple feels one if id ie nwehrSRi etrgre sa as regarded best is STR whether on differ are results physics, undergraduate of areas other In relativity of theory special the make factors Several 1 rusi aoro h osrcieapproach. constructive the of favour in argues fojcsadcokrtraini pca eaiiy ti It relativity. special in retardation clock and objects of rmtn tdn nesadn fseilrltvt and effects. relativity perspectival special purely of understanding student promoting ipepyia oeso esrn o n facokaeus are clock a of and rod measuring a of models physical Simple .INTRODUCTION I. osrcieapoc oteseilter frelativity of theory special the to approach constructive A etefrTm,Uiest fSde,Sde S 06 Aust 2006, NSW Sydney Sydney, of University Time, for Centre colo hsc,Uiest fNwSuhWls ynyNSW Sydney Wales, South New of University Physics, of School 2 n fNorton of and Dtd coe 0 2018) 10, October (Dated: 3 Even . .J Miller J. D. rbe thn,poie aslepaainada and explanation problem. the causal into the a insight of physical provides dynamics better the hand, con- on the at is Generally it problem as inadvertently readers. based some approach, might structive to which text- contrary best comments the very imply are in the of clear there some made in books, not even is and, point also textbooks can this many phe- Unfortunately, arguments the dynamical given. of on be most account based in STR constructive of a ground nomena that common literature is the it of arguments, kinematic by u nhnsgtta h ocpsaebs ent flow to ar- seen to best going are principles. is concepts the from the one directly that if hindsight even in approach STR gue constructive of concepts a the via introducing against impediment no iepicpeapoce oSRaepoie nSc III. Sec. construc- in the provided are on STR remarks to constructive Some approaches a tive/principle the from subject. obtained the how be to briefly can approach STR very of sketch results to main and example outset. convenient the at Newton’s anticipated of be form to relativistic needed the law that Another obscured second is solution. is it numerical result with a solved the for problem be need of the cannot impact by it method. the somewhat that the so is illustrate and model to analytically Bell’s used with be problem could A phys- that a analyzed model and ical relativity teaching to approach tive etr ae,w edntsaetoequalms. those share a not than need More we developments. modified later, contemporary century be of to con- light need was the might he in theory that was Maxwell’s STR that to cerned approach principle a ferred ihb T,i awl’ hoyo . of dealt theory problems appears Maxwell’s the at is It of was STR, some took by it also to with as who approach others now, constructive and Lorentz candidate a rods Fitzgerald, can obvious of measuring that time The of law the behaviour physical clocks. the a investigate with and pro- to should i.e. used way, STR same be to the conse- approach in relevant constructive ceed their A physics of of elaboration laws quences. the relevant by the proceeds of and proposal or discovery the nawl-nw ril,Bell article, well-known a In h anamo h rsn oki opeetamore a present to is work present the of aim main The h eeomn fms ra fpyisbgn with begins physics of areas most of development The ∗ ndsigihn ewe yaia and dynamical between distinguishing in rudta h oescudhl in help could models the that argued s I OSRCIEAPPROACH CONSTRUCTIVE A II. 5 dt eosrt h contraction the demonstrate to ed htoeo h esn htEnti pre- Einstein that reasons the of one that ai and ralia 2052 4 doae h construc- the advocated 1,4 hsteeis there Thus 2

In order to make the constructive derivation of length found from the condition contraction and clock retardation as transparent as pos- sible, it is desirable to construct a measuring rod and a 1 Q 2qx0 Ex (x0)= 2 3/2 = 0 (1) clock from a very simple system such as a system of point 4πǫo 4x − 2 2 " 0 (x0 + L ) # charges in equilibrium. A problem is that there are no stable configurations of charges that can be maintained where ǫo is the permittivity of free space. If Q = 2√2q, by the interaction of the charges with the electric and Eq. (1) is satisfied when x0 = L and so the equilibrium magnetic fields they produce. Diamagnetic materials are separation of the +Q charges is 2L in the of 6 an exception to this rule , which is an extension of Earn- the charges. 7 shaw’s theorem , but invoking diamagnetism is undesir- To consider what happens when the charge assembly in able because of the need for simplicity. As emphasized Fig. 1(a) moves speed v in the x-direction with respect to 8 by Swann in the present context, material objects are in S, we require the electric and magnetic fields of charges in equilibrium due to quantum mechanical effects but con- uniform motion. While some students in a course on STR sidering those is also ruled out by the need for simplicity. may not have met the required expressions, there are rel- 9 Fortunately there are simple qualitative arguments that atively simple derivations of them.10,11 Of course, one show that lengths of material objects transverse to the must not assume at this stage any of the results of STR direction of their motion are independent of their speed, so it is essential to the present approach that the required i.e. transverse length is the same for all inertial observers. expressions are obtained directly from Maxwell’s equa- We can rely on that to stabilize a system of charges which tions. Alternatively students could be asked to accept can then be used to derive and clock the expression for the electric field of a moving charge as retardation. given. Intuitively it is not unreasonable that the electric Our simple “measuring rod” is shown in Fig. 1(a). It field lines become “squashed” in the direction of move- is the distance between the +Q charges which are re- ment of a charge (and the magnetic field lines can be 2 pelled by each other and attracted by two charges q derived from those using B = v E/c where v is the × attached to the ends of a non-conducting bar of length− velocity of the source of E and c is the ). 2L with relative permittivity 1. We only consider motion When all four charges move with speed v in the x- along the x-axis which is perpendicular to the length 2L direction, the electric and magnetic fields at a vector dis- so, on the basis of the qualitative arguments mentioned tance r from the current (not retarded) position at time 12 above, all observers agree on the separation 2L (and do t of a charge q moving with velocity v = vxˆ are so independently of the definition of simultaneity which q rˆ 1 β2 is therefore not required at this stage). The two charges E r ( ,t) = 2 − 3/2 (2) +Q are in stable equilibrium for displacements along the 4πǫo r 1 β2 sin2 θ x-axis but unstable equilibrium for small displacements − perpendicular to the x-axis so it best to imagine that the  +Q charges are inside a transparent and frictionless tube -q -q lying along the x-axis. Perhaps it s worth noting that in order to write down Maxwell’s equations and perform the calculations, we 2L need a system of coordinates which, in turn, requires 2L +Q +Q +Q rods and synchronised clocks to set up. We therefore must first assume that rods and clocks exist and that the x0 clocks can be synchronised, say by slow clock transport, and then show that the existence of such rods and clocks -q -q is consistent with the theory. (a) (b)

FIG. 1: (a) The distance between the +Q charges in their equilibrium positions represents the measuring rod. The q charges are held in position by an insulating rod of length− A. Contraction of a moving measuring rod 2L. When Q = 2√2q and the charges are at rest with re- spect to the observer the separation of the +Q charges is also If the charges are at rest in an inertial frame S, only 2L. When the whole arrangement is set in motion in the electrostatic forces are involved. In equilibrium, the +Q horizontal direction, the distance between the +Q charges is contracted. (b) The periodic motion of the +Q charge after charges occupy the positions where the electric field is it slightly displaced from the center of the rod represents a zero and, by symmetry, these are equidistant from the clock. When the whole arrangement is set in motion in the rod at x0 along the x-axis which bisects the rod. Also ± horizontal direction, the period of the motion becomes greater by symmetry, only the x-component of the electric field so the “clock” becomes retarded. on the x-axis Ex (x0) can be non-zero. Thus x0 can be 3

1 B (r,t) = (v E (r,t)) (3) only a single frame which can be carried out by a variety c2 × of methods, e.g. slow clock transport, and which does not anticipate the relativistic results involving time which we where β = v/c and θ is the angle between r and the are in the process of deriving. In terms of the units of x-axis. Once again the equilibrium separation can be length used in S, the rod is contracted to 2 1 β2L determined from the condition that the fields at each of − using Eq. (5) while in S the rod is at rest and so x =2L the +Q charges are zero. There is no nett magnetic field p at the +Q charges and the separate conditions for each of from Eq. (1). From inspection of Fig. 2, the position of the +Q charges show that they are still equidistant from the +Q charge on the right is x = vt +2 1 β2L and x =2L, which leads to − the rod by a distance we can call xv. The equilibrium p separation 2xv of the moving +Q charges can be found from x = γ (x vt) . (6) − 2 1 β If we repeat a similar argument from the point of view Ex (xv)= − 14 4πǫo  of reference frame S, we find Q 2qx v = 0 (4) x = γ x + vt . (7)  2 3/2  4xv − 2 2 3/2 2 L2 (xv + L ) 1 β 2 2   − xv+L  Substitution of x from Eq. (6) into Eq. (7), leads directly     to which is satisfied when v x = 1 β2L = L/γ (5) t = γ t 2 x . (8) v − − c   p where γ =1/ 1 β2. Hence, when the charge assembly Together with the general argument about the co- is moving with constant− speed v, the moving “measuring ordinates transverse to the motion remaining unchanged, rod” representedp by the equilibrium separation of the +Q Eqs. (6) and (8) constitute the familiar Lorentz transfor- charges is reduced to 2 1 β2L from their separation, mation (LT). the length 2L of the “measuring− rod”, when at rest. Once the LT has been obtained in the constructive p approach to STR, the rest of STR follows, including, rel- evantly for us below, the relativistic form of Newton’s B. second law:

∂p d2x As shown in Fig. 1(b), the above arrangement can be F (x)= = γ3m . (9) modified to produce a clock by removing one of the +Q ∂t dt2 charges and moving the other +Q charge between the q charges (the vertical bar needs to have a suitable gap − drilled through it to accommodate the +Q charge). If S S -q the centre charge is given a horizontal displacement from the point midway between the q charges, it will expe- v rience a restoring force and execute− a periodic motion which can be used to measure time. In order to calcu- 2L late the period, it is necessary to have the correct form +Q +Q of Newton’s second law, i.e. the relativistic form of the second law. One can reach that result in a logical manner in the present approach by first noting that length con- traction is inconsistent with the -q x = x1 + x2 of co-ordinates between inertial frames. That motivates ! 2 a re-consideration of the transformation of co-ordinates x1 = vt x2 = 2 1 " L leading to the correct (Lorentz) transformation. The cor- rect (relativistic) form of the laws of mechanics then fol- x = 2L low as in any course on STR. A standard text-book argument13 can be adapted to the present case. Let the measuring rod be at rest in FIG. 2: The measuring rod apparatus is at rest in frame frame S with one end at the origin and the other at S which moves horizontally to the right with speed v with x =2L as shown in Fig. 2. At t = 0 in S, the origins of respect to frame S. At time t in S, the trailing edge of the S and S coincide with one end of the rod, and the other “rod” (left +Q charge) is at the origin of S and S and the end of the rod is located at x in S and x in S at the same leading edge (right +Q charge) is at x in S and at x in S. time. Note again this requires clock synchronisation in 4

C. Slowing of a moving clock the length contraction that occurs when a physical ob- ject changes frames somehow involves the relativity of Although clock retardation follows from the LT alone, simultaneity. the constructive approach requires us to show that a ma- The rod and clock models are reasonable because they terial clock can be constructed within the theory and that are analogous to the way one imagines the ion cores are it behaves in the expected manner. held in position in real materials. Of course, treatment When the middle charge +Q in Fig. 1(b) is dis- of real materials considered as an assembly of ion cores placed from its equilibrium position by xv, the relativistic and requires a quantum mechanical solution. restoring force F (xv), with the charges moving at speed On the other hand, the Hamiltonian involves only the v with respect to S, is electromagnetic field used in the proposed models. It recent times it has become possible to calculate the lat- Q 1 β2 tice constant of simple solids from genuinely ab initio F − (xv) = calculations.17 There does not appear to be such a cal- − 4πǫo  2qx culation for a solid moving at relativistic speeds. In any v x 3/2 ˆv (10) case, the necessary calculations are clearly too compli- 2 2 3/2 2 L2 (x + L ) 1 β 2 2 cated to be included in an undergraduate course in STR. v − xv+L An alternative approach is to consider the measuring qQ   16 γ 3 xv for xv << L. (11) rod as an elastic solid. Davidon has dealt thoroughly ∼ − 4πǫoL with the kinematics and dynamics of moving and accel- Using the relativistic form of Newton’s second law in erated rods from that point of view and derived many Eq. (9), the centre charge executes simple harmonic mo- interesting results, including length contraction, by con- tion under the influence of the restoring force with a pe- sidering the dynamics in a single frame. Unfortunately, riod the analysis is not simple enough to be suitable for an introduction to STR. 3 4πǫ0L m Tv =2πγ = γT0 (12) s qQ A. Dewan-Beran 3 where T0 = 2π 4πǫ0L m/qQ is the period when the clock is at rest. This motion constitutes a simple clock. It is interesting to apply the thought experiment origi- p It runs more slowly by the factor 1 β2 when it is nally proposed by Dewan and Beran18 (also referred to in moving in S than when it is at rest and− it is reasonable Bell’s article4) to the configuration of charges in Fig. 1(a). to suggest that more elaborate movingp clocks will run Firstly remove the +Q charges from the spheres when slowly for similar reasons. they are in their equilibrium configuration at rest in S. The spheres remain in position. As measured in S, let all the components of the apparatus be simultaneously III. DISCUSSION subject to a uniform acceleration19 in the +x-direction for the same period so that their velocity changes from It would be possible to begin a course of STR with zero to v and they end up at rest in S. When they the above dynamical derivations of length contraction, reach rest in S, both the spheres remain separated by 2L and the LT. Having reached that point, as measured in S.18 Now replace the +Q charges. The the course would follow the more traditional development re-charged spheres will move to their new equilibrium although it would be more consistent to derive results position separated by 2 1 β2L as measured by S. using dynamical arguments wherever possible. All inertial observers21 agree− on some aspects of this p An advantage of the arrangement of charges in Fig. 1 is sequence of events and disagree on other aspects. All the that they could have macroscopic dimensions and could inertial observers agree that all parts of the apparatus be imagined to be on a laboratory bench top (the ar- started at rest in S and eventually ended at rest in S and rangement in Fig. 1 is reminiscent of the twin “pith” balls that all parts of the apparatus were given the same ac- which serve as an elementary demonstration of charge celeration for the same time. All inertial observers agree repulsion). Another significant advantage is that they that the pair of spheres move closer together when re- don’t involve expressly the invariance of the speed of charged and that the ratio of the initial (uncharged) to light for all inertial observers which is one of the coun- the final (charged) separations of the spheres in S is γ. terintuitive consequences of STR15 and therefore should Inertial observers disagree on the magnitudes of various be avoided in demonstrating other counterintuitive con- quantities: the velocities of S and S, the acceleration, sequences of STR. Perhaps the main advantage is that the period of acceleration and the lengths separating the the length contraction of an object that changes speed charged and the uncharged spheres in S and S. They is derived within a single frame and does not involve the also disagree on the timing of the beginning of the pe- comparison of the synchronization of clocks in two dif- riod of acceleration of the various parts of the apparatus. ferent inertial frames. This eliminates the notion that For example, from the point of view of S, who initially 5 observes the apparatus receding at speed v, the trail- der the influence of the dynamical forces within the rod, ing uncharged sphere is judged to begin accelerating− (in may produce an expansion or a contraction depending the opposite direction to the velocity) before the leading on the manner in which the rod is accelerated between uncharged sphere. As a consequence the separation be- frames. It is even possible to contrive the acceleration in tween the charges increases so that when both charges such a way that the uncharged spheres in our example reach rest in S, they are separated by 2γL according to (and the ion cores in a rod in the real world) reach S in S. When the +Q charges are replaced, S observes them their equilibrium positions in that inertial frame. Then contracting to a separation of 2L. re-charging the spheres would cause no change in their This thought experiment separates in time, and there- separation at all. Taylor and Wheeler describe such an fore makes explicit, the two dynamical processes that are acceleration of a measuring rod between frames.22 in operation when a physical object is transferred be- tween inertial frames: (i) the external force producing the required acceleration and (ii) the internal forces which B. Dynamical v perspectival effects maintain the object in equilibrium. Feinberg has consid- 20 ered this question in a more general way. Of course, in The treatment of the thought experiment in Sec. IIIA a normal object in which the charges are left in place, relies on the constructive or dynamical approach to STR. the internal forces act continuously so as to maintain, or It exposes clearly several aspects of STR which remain attempt to maintain, the charges in their instantaneous implicit or hidden in a purely principle or kinematic ap- equilibrium positions. The effect of the internal forces proach to STR. Firstly it distinguishes between perspec- depends on the mode of acceleration (the timing of the tival and dynamical effects. The changes in the sepa- start of the acceleration, whether the acceleration is con- rations of the spheres in their destination inertial frame stant, etc) involved in changing the velocity of the object. when they are uncharged and then re-charged, show that The direction of the acceleration in relation to the di- when a physical object in equilibrium changes frames, rection of the velocity makes a qualitative difference. We there are physical changes in it which can be calculated have just dealt with the case of the acceleration in the from a consideration of the forces which keep it in equilib- opposite direction to the velocity, which is always what rium and it is those changes which lead to the satisfaction is required from the point of view of the receiving frame. of the Lorentz transformation conditions. Some observers will judge that the assembly with un- On the other hand when an observer changes frames charged spheres has to speed up to reach S from S, i.e. (or when we compare the results of observers in differ- the acceleration is in the same direction as the velocity ent frames), there are no dynamical effects in the physi- of S. For those observers the trailing end of the ap- cal object being observed—the differences among the ob- paratus begins, as before, to accelerate first but now it servers are due to their different perspectives. In the is catching up with the other sphere so that the sepa- above thought experiment, the disagreements among the ration of uncharged spheres decreases over its original inertial observers about the magnitudes of quantities and value in S. Despite their disagreement on the magnitude the relative timing of the beginning and ends of the accel- of the separation of the uncharged spheres and whether eration period for the various parts of the apparatus are that separation increased or decreased as the result of perspectival. They have no dynamical explanation, nor the change of frames, as already mentioned, all observers do they require one. This point can be confusing because agree that the spheres move together when the charges the factors involved in the perspectival and dynamical ef- are replaced and that they move together by the same fects involve the same numerical values (determined by factor compared with their separation when uncharged. γ). When the measuring rods and clocks are moved be- This demonstrates that the fractional change in length tween inertial observers they suffer dynamical changes. with change in velocity caused by the forces that main- When each observer use their dynamically altered rods tain the equilibrium configuration are real in the sense and clocks to make measurements, it is not surprising that all inertial observers agree on them. that their results differ and that they differ by the same For completeness, consider the equal acceleration of all factors that are involved in the dynamical changes. parts of the apparatus originally at rest in S is begun si- While the point might appear trite and is obvious from multaneously from the point of view of S, the receiving the constructive approach to STR, failure to appreci- frame. The uncharged spheres reach S with their initial ate the differences between moving a physical object be- separation observed by S, namely 2 1 β2L. On re- tween frames and moving an observer between frames − charging in S, the +Q charges move further apart by a sometimes produces confusion. For example, it has been p factor γ to reach the points of equilibrium separated by used recently as argument against the constructive ap- 2L. All inertial observers agree on this movement (while proach to STR.23 The difference between transferring disagreeing, as before, about the magnitudes of the sep- physical objects between inertial frames and transferring arations involved). Thus if a rod is transferred between observers between inertial frames is summarised in Ta- frames, the movement of the charges in the rod from ble I where the length of a physical object in equilib- their old equilibrium separations in the original frame to rium is labeled a “connected” length to distinguish it their new equilibrium separations in the final frame, un- from the distance between two objects which are inde- 6

C. Constructive v principle approach TABLE I: The ratio of lengths involving two different inertial frames moving at relative speed v along the direction of the − 1 2 2 2 Finally, leaving aside pedagogic considerations, we lengths with γ = `1 v /c ´ . The “connected” lengths 2 refer to lengths of objects− in equilibrium and “unconnected” consider one of the arguments that STR should be pre- lengths refer to distances between independent objects. The sented as a principle or kinematic theory: the Lorentz first row is a perspectival effect: the ratio of the length ob- invariance of the non-gravitational laws of physics can be served from one inertial frame to the length observed from the traced back to the structure of Minkowski and other inertial frame by two different observers. The second so it is preferable and more logical to go directly from row is due to dynamical effects. The ratio of the unconnected Minkowski spacetime to the results of STR (kinemati- lengths is variable because it depends on the mode of accel- cally) than via the physical fields (dynamically). Put eration of the physical objects between the inertial frames. another way, if one insists on proceeding dynamically, Ratio of lengths when Connected Unconnected the Lorentz invariance of all the physical fields is vital but the latter has its origin in the Minkowski spacetime lengths lengths so proceeding from the latter is preferable.24 The argu- Observer changes frames γ γ ment that Lorentz invariance of the physical fields has Objects change frames γ Variable its origin in the Minkowski spacetime is justified by the principle of common origin inference (COI).2 Put briefly, either one proposes that the Lorentz invariance of the three non-gravitational fields (the electromagnetic, weak and strong interactions) is a remarkable coincidence or one seeks a common origin and finds it in Minkowski pendent of each other which is an “unconnected” length. spacetime. As Janssen mentions in passing25, an alter- One sees that for the perspectival changes resulting from native is that the three non-gravitational fields might be a change of inertial frame by an observer (or the com- different manifestations of a single field and it is that parison of observations of two observers), a distance ap- fact that is the COI for the Lorentz invariance of all the pears to change by γ whether or not the distance is a non-gravitational physical fields. There are quite strong connected or an unconnected length in the above sense. arguments that the differences between the strong, weak However when physical objects are transferred between and electromagnetic fields in the present era are due to inertial frames, the connected lengths change in the ratio symmetry breaking of a single field in an earlier era.26,27 γ but the unconnected lengths change by a factor which If the unification of the physical fields turns out to be cor- depends on the mode of acceleration involved in transfer- rect, then the COI argument for the role of Minkowski ring the objects between inertial frames. The last point spacetime in determining the Lorentz invariance of the has been illustrated by the measuring rod apparatus; e.g. physical fields involved in STR loses its force. in Sec. IIIA, the separation of the uncharged spheres is an unconnected length and on transfer from S to S, their separation remains at 2L but when charged spheres are IV. CONCLUSION transferred in an identical manner, and remained in equi- librium, their final separation is 2 1 β2L. − Presenting STR from a purely kinematic point of view p leaves students’ understanding of the subject deficient The LT can be considered either as an active or a and less able to deal confidently with non-text-book ex- passive transformation of co-ordinates (symmetry). The amples like the Dewan-Beran thought experiment. On passive sense, when the co-ordinate system is trans- the other hand, including the constructive or dynami- formed to a different inertial frame, corresponds to a cal approach in presenting STR enriches student under- perspectival change in the manner just discussed. The standing of the subject.4 The thought experiment involv- active sense involves the physical system under consider- ing the arrays of charges capable of acting as a measuring ation changing frames (given a “boost”). The previous rod and a simple clock described here facilitates the pre- discussion shows that an active boost cannot be consid- sentation of STR from the constructive point of view. ered to be the system as an entity acquiring the veloc- On the more general question of whether STR is bet- ity of the new frame because connected and unconnected ter viewed as a principle (kinematic) or constructive (dy- lengths would behave differently (except in the case when namic) theory of physics, there does not seem to be any the change in velocity was especially contrived but that compelling reason which justifies the almost universal restriction is not part of the concept of an active LT sym- emphasis on STR as a principle theory. As for the ques- metry). Therefore an active LT ought to be pictured (if tion of whether one should take a dynamical or kine- the process is physically pictured at all) as a transfer matical approach to the subject, it seems clear that the and re-assembly of the apparatus in the new frame. The changes in a connected object (object in equilibrium un- connected lengths of the objects comprising the system der internal forces) when it is transferred between in- then act as a guide for the placement of the unconnected ertial frames involves dynamical effects. On the other objects. hand, the different observations of a single connected ob- 7 ject by different inertial observers is a perspectival, not a dynamical, effect.

∗ Email address: [email protected] 15 Sometimes the invariance of the speed of light for all iner- 1 Harvey R. Brown, Physical Relativity. Space-time Struc- tial observers is made one of the two postulates in a pre- ture from a Dynamical Perspective (Oxford University sentation of STR but that is not the minimalist form of Press, Oxford, 2005). the postulates as originally proposed by Einstein. 2 Michel Janssen, “Drawing the line between kinematics and 16 William C. Davidon, “Kinematics and dynamics of elastic dynamics in ,” Studies in History and Phi- rods,” Am. J. Phys. 43(8), 705-713 (1975). losophy of 40(1), 26-52 (2009). 17 Alex Zunger and Marvin L. Cohen, “First-principles 3 John D. Norton, “Why constructive relativity fails,” Brit. nonlocal-pseudopotential approach in the density- J. Phil. Sci. 59(4), 821-834 (2008). functional formalism. II. Application to electronic and 4 John S. Bell, “How to teach special relativity,” reprinted in structural properties of solids,” Phys. Rev. B 20(10), Speakable and Unspeakable in (Cam- 4082-4108 (1979). bridge University Press, Cambridge, 1987), pp. 67-80. 18 E. Dewan and M. Beran, “Note on Stress Effects due to 5 Ref. 1, pp. 69-71. Relativistic Contraction,” Am. J. Phys. 27(7), 517-518 6 M. D. Simon, L. O. Heflinger and A. K. Geim, “Diamag- (1959). netically stabilized magnet levitation,” Am. J. Phys. 69(6), 19 The acceleration could be achieved by immersing the ap- 702-713 (2001). paratus in a uniform gravitational field for the required 7 William T. Scott, “Who Was Earnshaw?,” Am. J. Phys. period of time (a gravitational field in which tidal effects 27(6), 418-419 (1959). are negligible over the dimensions of the apparatus). 8 W. F. G. Swann, “Relativity, the Fitzgerald-Lorentz Con- 20 E. L. Feinberg, “Can the relativistic change in the scales traction and Quantum Theory,” Rev. Mod. Phys. 13(3), of length and time be considered the result of the action 197-202 (1941). of certain forces?,” Sov. Phys.-Usp. 18(8), 624-635 (1976). 9 Edwin F. Taylor and , Spacetime 21 We consider only inertial observers whose relative velocity Physics, 2nd ed. (Freeman, New York, 1992), pp. 65-67. lies along the x-axis. 10 Valery P. Dmitriyev, “The easiest way to the Heaviside 22 Ref. 9, pp. 119-120.± ellipsoid,” Am. J. Phys. 70(7), 717-718 (2002). 23 Alberto A. Martinez, “There’s no pain in the Fitzgerald 11 Ben Yu-Kuang Hu, “Comment on ‘The easiest way to the contraction is there?,” Studies in History and Philosophy Heaviside ellipsoid,’ by Valery P. Dmitriyev,” Am. J. Phys. of Modern Physics 38(1), 209-215 (2007). 71(3), 281 (2003). 24 In the argument here, it is being tacitly assumed that if all 12 David J. Griffiths, Introduction to Electrodynamics, 2nd ed. the physical fields are Lorentz invariant, then all the laws (Prentice Hall, Englewood Cliffs, NJ, 1989), pp. 425-426. of physics involving them must be Lorentz invariant. 13 Ref. 12, pp. 460-463. 25 Ref. 2, p. 49. 14 In deriving Eq. (7), it has been assumed that the speed of 26 Savas Dimopoulos, Stuart A. Raby and Frank Wilczek, S with respect to S is v. This follows from the method of “Unification of couplings,” , 44(10), 25-33 clock synchronization− we have assumed. It is less desirable (1991). to rely on the principle of reciprocity which requires at- 27 Frank Wilczek, “Anticipating a new Golden Age,” Eur. tributing some properties to space and time which it would Phys. J. C 59(2), 185-196 (2009). be best to avoid in a constructive approach to STR.