Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Is Fermat's Least Time Principle Equivalent to a Least Information Approach?

Francesco R. Ruggeri

Published May 11, 2026• 0 citations
Publisher

Abstract

Fermat’s Least Time Principle clearly refers to a path in space which is associated with the least time of travel. In particular, one could apply this approach to derive Snell’s law, assuming a fixed initial (xo=0,yo), change in index of refraction below y=0 and a fixed end point (x0=L, -yo). Officially, the integral which is minimized is: Integral ds/v (1) where ds is a unit length on the curve traversed and v, the velocity. In this note, we point out that dt= ds/v is sometimes used in the literature as being proportional to the probability for a particle to be found in ds (see (2)). If a particle is moving fast, then dt is small and there is less probability of finding the particle (hence less information) than if it moves slowly. We argue that one may consider that Fermat’s Least Time Principle is equivalent to a minimization of information. This suggests the notion of probability instead of deterministic physics which might at first be implied by the notion of minimal time. This interpretation of information or least time may be fine for the derivation of the 2-D Snell’s law, but fails in the analysis of 1-D reflection-refraction at an n1-n2 interface. This suggests that something more may be occurring. The notion of dt representing the probability for a detection-measurement hit is based on the idea of an interaction occurring at the center-of-mass point of the particle. Given the photon example, and E=pc, we suggest that there might be a different interpretation of information. For fixed E, if c1=c/2, then p1=2p. By taking 1/p1, one obtains the factor ½ again. Thus, one might be able to consider 1/p as linked with information. Formally, t=distance/c = p* distance for light, but we suggest using this in general for both light and particles with rest mass. Then, 1/p is linked with information. High p means low 1/p and high information. One wishes to minize information. A function which enforces the 1/p resolution is exp(ipx) such that exp(ipx)exp(-ipx) =1 like dt=constant. This would explain the 2-D Snell’s law Fermat’s Least Time Principle case, but it also has implications for p= mo v = (2mo) (v/2). In this case, the particle’s have the same momentum, but different speeds. If one only thinks in terms of speed, then they carry different information, but in terms of an impulse hit, they carry the same information. If one insists that impulse hit is linked to information, then using 1/p as the information quantity seems to hold. As a result, one no longer follows the center-of-mass in space and time in such an approach. This suggests a different scheme which is ultimately equivalent to the statement that for dt=contstant, v=constant, one has the same probability for each dx. Mathematically, exp(ipx) creates periodic regions proportional to 1/p, but exp(ipx)exp(-ipx) =1 yields a constant for each dx which is akin to dt=constant in each of these dx regions. The form exp(ipx) allows one to solve 1D reflection-refraction, while minimization of information px (part of the action -Et+px) allows one to find Snell’s law.

DOI
10.5281/zenodo.20126702

Related papers

Fermat's Principle, Particles with Rest Mass and Quantum Mechanics

Francesco R. Ruggeri

Addendum Oct. 26, 2022 In ((3)) one should formally subtract i.e. d/dx{ -E1 t + (p1x) x +E2t + (p2x) x) = 0. Then p1x-p2x=0 and -E1 dt1/dx + E2 dt2/dx = 0. Thus conservation of momentum implies Fermat's principle. The minus sign between dt1/dx and dt2/dx follows from t1 being a variable of x and t2 of (L-x) where L is the length of the mirror such that initial and final y values have the same magnitude. (This does not change the conclusions made in the note we argue.) In a previous note (1) we argued that Fermat’s least time principle applied to light involved finding an expression for time in terms of spatial variables and the taking of d/dx where x is an invariant spatial variable in the problem. For example, for reflection from a stationary or constantly moving mirror in the y direction or refraction with different n(i) (indices of refraction in the y direction) x is invariant because there is no medium discontinuity in the x direction. We argued that Fermat’s principle is equivalent to a conservation law in the direction of the invariant variable and that this conservation is in fact that of momentum in this direction. In (2) we argued that E=pc with c=x/t i.e. the wave dispersion equation for light contains Fermat’s principle because one may write it as 0= -Et+px and apply d/dx using t(x) with x being the invariant variable. With conservation of momentum in the x direction and E1 = E2 for the two rays: dt1/dx + dt2/dx = 0 for a two ray problem. Fermat’s principle and E=pc pertain to light. The latter may be written as 0= -Et+px ((1)) for light with x/t=c which as we argued above leads to Fermat’s principle if one takes d/dx for two rays. One may note, however, that one also obtains Fermat’s principle for constant= -Et+px with momentum conservation in the x direction. Furthermore a particle with rest mass bouncing off a floor may behave like light. This we argue leads to the idea of considering two different frames. For light -E1 t1 + p1 x1 = 0 = -Et + px with x/t = x1/t1 = c. For a particle with rest mass -mo t1 + 0 = -Et+px with x/t=v where c=1 and mo is energy in a rest frame. (This is in fact the idea of a Lorentz invariant.) Thus Fermat’s principle for a free particle seems to appear in the form -Et+px = constant. The constant is 0 for light. The form -Et+px is consistent with a wavelength proportional to 1/p and a frequency to 1/E not only for light, but for the particle with rest mass. We argue that light moving in a medium is a complicated statistical process of scattering, but that the problem is treated in a “thermodynamical” manner by characterizing the medium by a single number i.e. single piece of information, namely the index of refraction i.e. c→c/n and p→pn for light. We argue that movement within the medium is like moving in a kind of constant potential for light. -Et+ (pn) x still holds and there is still conservation of momentum in the x direction for two media. Thus the idea of d/dx → p is already present in the photon form -Et+px which follows from Maxwell’s equations, but could also be derived from Fermat’s principle as argued in (2). In other words, ideas of quantum mechanics for a particle with rest mass could ultimately follow from Fermat’s principle, we argue. A free particle with rest mass is linked with constant= -Et+px and if one wishes to have an eigenfunction of momentum then: -id/dx exp(ipx) = p exp(ipx). For a constant potential, just like with light, p would change, but the form exp(ipx) would remain. Thus it seems that even in average situations (like refraction) one still wants to have a periodic type spatial function W(x) such that -d/dx d/dx W(x) /W(x) = <pp> such that <pp>/2m + V(x) = E for a particle with rest mass. This is the Schrodinger equation and it is found mathematically that only certain En are allowed if one wishes to have specific boundary conditions like W(x) →0 at +/- infinite for say an oscillator problem. Thus we argue that the fundamental idea of taking a derivative of an invariant spatial variable in Fermat’s principle which automatically yields conservation of momentum is a basic idea found in quantum mechanics even for a particle with rest mass, namely that d/dx is associated with momentum. Mathematically d/dx is a translation operator and so the link with Fermat’s principle shows how translation is linked with a spatially invariant spatial variable e.g. x. Furthermore Fermat’s principle shows that even complicated statistical scattering problems such as refraction (in which the photon scatters off of different molecules in different media) can be reduced to a periodic average problem with p→pn. This suggests that for a particle with a rest mass in a potential V(x) one should also have a periodic spatial variable such that d/dx d/dx is linked to average momentum, we argue.

•Oct 25, 2022•0 citations

Association of X and P in Classical Physics and Information/Invariance Part II

Francesco R. Ruggeri

In part I we argued that Newton’s first law implies that constant momentum (i.e. momentum conservation) is associated with spatial invariance i.e. a lack of information in a certain direction. This is equivalent to stating there exists no force in that direction. We argued that one may apply the generator for translation in the spatial direction i.e. d/dx for x and apply it to a function f(x) such that df(x)/dx = (px). The simplest form is f(x)= (px)x. We also argued that it is precisely this function which appears in Fermat’s least time principle because time = distance/speed and one may multiply time by |p| = momentum magnitude. Given that momentum and displacement are parallel vectors for a light ray, p dot r = p distance = (px)x + (py)y for a two dimensional problem (e.g. reflection or refraction). “Distance d” is the key function to use because dd= xx + yy implies that for fixed y, d d/dx = x/d = sin(theta1) (angle measured from y axis) so p d d/dx = (px) which is a conserved quantity if x is invariant. Thus conservation of momentum for two rays is equivalent to extremization. Extremization, we argue, suggests there may be a built-in physical mechanism which allows a free particle to correct itself if perturbed slightly just like a Maxwell-Boltzmann gas which extremizes entropy subject to constant average energy does. This built-in mechanism should be linked to motion in space. Writing (px)x as ln(exp(ipx), one not only obtains an eigenvalue equation i.e. -d/dx exp(i (px) x) = (px) exp(i px x), but the notion of a relative change arises if exp(i px x) is interpreted as a kind of probability. A relative change is natural because an absolute one ignores the idea that probability may differ with x even though (px) is constant. Thus from the idea of invariance in space found in Newton’s first law, which is equivalent to conservation of momentum in that direction, the equation d/dx f(x) = (px) leads directly to -id/dx exp(i px x) = (px) exp(ipx x) where exp(i px x) represents a physical probability type of wave which stablilizes the constant (px) motion because this motion is equivalent to extremizing time or |p| time or p dot r / speed of light in Fermat’s principle. In other words, the quantum mechanics of a free particle is associated with stabilizing wave behaviour exp(ipx x) which was not seen experimentally at the time of Newton or Fermat, but seems to be part of their formalisms, with Fermat’s principle derivable from Newton’s first law.

•Nov 22, 2022•0 citations

Spatial Invariance, Special Relativity and the Quantum Wave

Francesco R. Ruggeri

Newtonian mechanics (from the 1600s) defines a quantity called momentum p in terms of mv where m is mass and v=dx/dt with dx and dt tending to zero. Newton’s first law is equivalent to conservation of momentum in the absence of a force. The second demonstrates how p changes under a force. If one has equal action/reaction then overall momentum is conserved. As a consequence the idea of a “wave” associated with p does not arise. The main focus is on the idea of force. Fermat’s Least Time Principle for light (1600s), however, is based on minimizing the overall time traveled by light which reflects or refracts. Young’s interference experiments did not occur until the very early 1800s and Newton considered light to be a particle at this time. In fact light bounces off a foil creating an impact like a particle. Fermat’s principle does not directly make use of momentum or its conservation. It is focused on minimizing time. To do so, however, one must take a derivative with respect to a certain variable and set it equal to 0. This variable is the spatially invariant one in a problem. For example consider reflection (stationary or moving mirror) or refraction in two dimensions x,y. Let the mirror or medium face lie along the x axis at y=0. There exists a constraint on y and given a fixed starting point one cannot change y. This leaves x to be varied because it is invariant. Thus the extremization of time is really a statement of invariance of a spatial variable. In other words, Fermat’s principle is linked to the physical variables t and x while Newton’s focuses on p (momentum). This invariance in turn gives rise to conservation of momentum in the direction of the invariant variable i.e. (px). There is still the idea of Newton’s change in momentum as (py) abruptly changes in refraction as it hits the interface of media. The point we wish to make is that from Fermat’s least time principle one sees that conservation of momentum in the absence of a force is completely intertwined with invariance in space. The two cannot be separated in this approach and there is nothing unphysical about t or x. In fact, one may create a function A= -Et + (px)x + (py)y such that dA/dx partial = px = conserved quantity. In this function (px) and x are on the same footing. One does not have one without the other. This idea is carried over into special relativity as -Et+p dot r is a Lorentz invariant. Given that p momentum is physical (it manifests itself in impulse hits) and p and x are inseparable in (px)x we suggest there should exist a physical manifestation in x i.e. in space linked to the maintenance of conservation of momentum in this direction even in the absence of a force. This we argue leads to the physical idea of a wavelength which follows from A=-Et+px with A=0 in all constant velocity frames leading to a frequency proportional to 1/E and wavelength to 1/p. We have discussed these ideas in previous notes, but the main point we wish to make here is that equivalence of spatial invariance in the x direction in Fermat’s principle with conservation of momentum in the same direction suggests that if p is physical, the spatial invariance must also be manifested physically. It is not simply a mathematical convenience to link t to x to px, we argue. Furthermore this wavelength should be linked to establishing conservation of momentum as it does in quantum mechanics where the orthogonality of exp(ipx) i.e. Integral exp(-i p1 x) exp(ip x) dx automatically ensures p1=p without imposing momentum conservation a priori as in Newton’s first law. Both Newtonian mechanics and Fermat’s principle suggest that p may change abruptly due to a force or an immediate change in medium. The idea of a physical wavelength, however, suggests there may be interactions completely different from an abrupt change in p due to an impulse. In particular there may be interference or Bohm-Aharonov interactions. Newtonian Mechanics

•Nov 10, 2022•0 citations

Newton's First and Second Laws Leading to Different Spatial Probabilities One Being Quantum Mechanical?

Francesco R. Ruggeri

Newton’s second dp/dt = Force where p is momentum is equivalent to pp/2m + V(x) =E for a conservative force= -dV/dx with p=mo v(x), v being velocity. In general one may divide the x axis into equally spaced dx units with dt(x) (i.e. time spent) differing in each unit. A main point we wish to make is that dt(x) is the same whether the particle moves to the right or to left. Thus using dt(x) as a kind of density (proportional to C/v(x)) as done in (1) means this spatial density does not distinguish between motion to the right or left. This is one kind of spatial density we argue, one that considers only the magnitude of velocity. Heuristically, however, one might argue that motion to the right or left involves a skewed spatial density in the direction of motion. This we argue represents a second kind of spatial density which follows from Newton’s first law and is associated with quantum mechanical motion. Ideally the two should be linked. Newton’s first law indicates that a particle at rest remains at rest and one in constant motion in a direction continues in this state unless acted on by force. This law is usually not associated with a mathematical equation and so a spatial density is not derived. In (2), however, we argue that Newton’s first law is associated with invariance in a spatial direction if there is no force in that direction, for example x. In such a case, d/dx is the translation generator in that direction and momentum is kind of “information”. We argued in (2) that one may write d/dx f(x) = p to describe this with the simplest solution being f(x)= px for constant p. We further argued that this is in fact linked to Fermat’s least time principle for light. Time =distance/speed of light, but momentum and displacement are parallel. Thus for a two dimensional problem: p dot r = (px)x + (py)y = p distance = p sqrt(xx+YY) and p sqrt( (L-x)(L-x) + YY) for two rays, one being the incident and the other the reflected. Given that there is conservation of momentum in the x direction one may write (px) incident = (px) reflected or d/dx (p distance incident) - d/dx (p distance reflected) = 0 which is Fermat’s principle for reflection with one index of refraction. For refraction p1=pn1 and p2=pn2, but (p1x) = (p2x). f(x)=px satisfies d/dx f(x) = p. This expression depends on x, but also on the direction of motion i.e. one may have p or -p. In (2) we further argued that this expression may be written in terms of: i px = ln(exp(ipx)) with exp(ipx) being periodic, but neither growing or tending to a constant zero for a constant p. As a result we identified exp(ipx) with a second type of spatial density or probability, one which accounts for the direction of motion. exp(ipx) is also associated with a quantum mechanical description of a free particle. Thus Newton’s first two laws give rise to two different spatial densities, but they are related. In particular, the modulus of exp(ipx) equals constant which is P(x) or spatial density from Newton’s second law. In this note, we start with the assumption that a velocity (not velocity magnitude) spatial density should exist and examine what properties of such a density/probability arise.

•Nov 23, 2022•0 citations

Fermat's Principle and Quantum Bound States

Francesco R. Ruggeri

In a previous note (1) we argued that Fermat’s principle of least time is present in the photon wave dispersion relationship E=pc which may be written as 0 = -Et+px with x/t=c. We argued that a nonzero constant = -Et+px applies to a free particle with rest mass, but that Fermat’s principle is still present. Furthemore the form -Et + px implies the existence of a frequency and wavelength for both a photon and free particle with rest mass. (Interestingly -Et+px is both the free particle relativistic action -mo t sqrt(1-vv/cc) and nonrelativistic .5mo t vv with c=1 and v=x/t. ) Fermat’s principle implies that d/dx → momentum (x component) as discussed in(1). An eigenfunction exp(ipx) demonstraties wave behaviour of a free particle with rest mass. In (1) we argued that if one knocks out a quantum bound particle it will be in a state exp(ipx). Thus this wave type information should also be included in the bound state so wepostulated a bound state represented by W(x) = Sum over p a(p)exp(ipx) (i.e. a wavefunction) such that the Fermat result d/dx → p still holds. In this note we wish to consider the idea of a statistical equilibrium and the concept of a constant energy. In the case of refraction it seems a complicated statistical scattering process occurs which slows down the motion of a photon i.e. c changes to c/n where n is the index of refraction. Thus one piece of information, namely n, describes the complicated statistical process as in thermodynamics. If E (energy) remains constant p→ pn. Thus in this complicated statistical scenario a cos(-Et+pn x) electric field pattern remains but p → pn i.e the wavelength has changed. The idea of a constant energy may even be applied to light scattering from a mirror moving with a constant v in the negative y direction. The actual energy and momentum of the incident and reflected rays are not equal, but for E1 special = p1 v(relative 1) and E2 special = p2 v(relative 2) with v(relative 1) = c- vcos(A) and v(relative 2)= c+vcos(B) where A and B are angles of incidence and reflection measured with respect to the y axis. E1 special t=E2 special. In this note we retain the idea that a particle knocked from a quantum single particle bound state is described by exp(ipx) in keeping with Femat’s principle. We consider the bound state to still have waves as Fermat’s principle suggests d/dx →p, but they cannot be of the same height and width because of the changing value of V(x) with x (i.e. the potential). According to Newton’s law one has conservation of energy i.e. pp/2m + V(x) = E, but we consider this to apply to a p (rms) = m v(rms) in the Schrodinger equation. If the forms exp(ipx)’s are to be retained within the bound quantum state (i.e. this information is not lost) and one has a constant E (energy) then the potential must be modified statistically to allow for the idea of conservation of energy for each p, but must also represent a constant potential to retain the form of exp(ipx). Thus in short the Schrodinger equation becomes a weighted sum of exp(i p(j) x) functions with each having a different wavelength (like refraction), but interacting with different constant potentials to ensure that E remains constant for each p(j). In other words we think of a potential V(x) as a statistical set of refraction discontinuities i.e. impulses which produce exp(ipx)’s which are like light moving through a homogeneous medium i.e. a momentum eigenstates. For this to happen, however, the various exp(ipx)’s need to combine with V(x) = Sum over k Vk exp(ikx) terms to create a Vmodified(p) exp(ipx) where Vmodiied(p) is a constant because each free particle exp(ipx) is associated with a different effective potential and a kinetic energy pp/2m, but the same energy En. We argue that this results from applying ideas of Femat’s principle to a particle with rest mass. Newton’s picture of a particle accelerating i.e. x(t) emerges if one considers rms values for which all information of exp(ipx) and crests/troughs of W(x) are ignored. In particular, W(x)=wavefunction may have crests and troughs, but -1/2m d/dx dW/dx/ W = KE(x) Newton which shows no crests or troughs, but simply balances V(x) in KE(x)+V(x) = E.

•Oct 28, 2022•0 citations

Is Fermat's Least Time Principle Equivalent to a Least Information Approach? Part 2

Francesco R. Ruggeri

In Part 1, we suggested that Fermat’s Least Time Principle is equivalent to a least information in space approach, where spatial information is given by length / (hbar/p) or px. Here we note that given the Lorentz invariant -Et+px, one actually has minimal information in both space and time. For a fixed E (E is the same in both n1 and n2 regions for refraction) in Et, it seems that one minimizes time, but one actually minimizes time/ (hbar/E). Independently in terms of time and space, one minimizes information. This leads to a consistent picture (which may use either space or time) we argue and also one that is statistical as opposed to the notion of minimizing time for no particular reason. Given that we use the notion of special relativity above, we try to see how this theory is linked to minimization of information. In another note (1), we showed that one may use F=dp/dt to derive the form g(v) = 1/sqrt(1-vv/cc). g(v) appears in the comparison of physics in a rest frame and one moving with -v, but it is F=dp/dt (a Newtonian vector equation) which fixes its functional form. Given F=dp/dt, one is forced to focus on conservation of momentum and projections in 3-space. We note, however, that special relativity yields E=[p] c for mo=0. Here [p] is the magnitude which is rotationally invariant like E. Given x=vt, one may write: -Et+px = 0 for a photon. Here p=[p]. As a result, one seems to have two very different ideas present, we argue. First, F=dp/dt, a vector equation, deals with components and conservation of momentum for various components. -Et+[p]x=0, however, is an equation which lies along the direction of motion, regardless of whether momentum is conserved in this direction. It is analogous to a wave equation with arguments -wt+x/lambda=0 If there really were a physical connection, then conservation of momentum in a particular direction would be akin to Newtonian mechanics, while -Et+[p]x=0 might be akin to wave mechanics. In Part 1, however, we treated the wave mechanics features as information without mentioning the word wave. It seems that -Et+[p]x =0 is linked with physical behaviour along the direction of motion, and conservation of momentum in one direction (not x) is a consequence. We suggest that a priori one might wish to follow a particle along its direction of motion and have a theory which is based only on numbers, not vectors. This, we argue, is the information theory introduced in Part 1, i..e minimization of information in either t or x.. Here we note that there are in fact two minimizations of information, one in time t / (hbar/E) and one in space x / (hbar/p). We apply this reasoning to two dimensional refraction (Snell’s law) and note that particle behaviour only appears when one considers projections to obtain conservation of momentum parallel to the n1-n2 medium interface.

•May 13, 2026•0 citations
Fermat’s Least Time Principle clearly refers to a path in space which is associated with the least time of travel. In particular, one could apply this approach to derive Snell’s law, assuming a fixed initial (xo=0,yo), change in index of refraction below y=0 and a fixed end point (x0=L, -yo). Officially, the integral which is minimized is: Integral ds/v (1) where ds is a unit length on the curve traversed and v, the velocity. In this note, we point out that dt= ds/v is sometimes used in the literature as being proportional to the probability for a particle to be found in ds (see (2)). If a particle is moving fast, then dt is small and there is less probability of finding the particle (hence less information) than if it moves slowly. We argue that one may consider that Fermat’s Least Time Principle is equivalent to a minimization of information. This suggests the notion of probability instead of deterministic physics which might at first be implied by the notion of minimal time. This interpretation of information or least time may be fine for the derivation of the 2-D Snell’s law, but fails in the analysis of 1-D reflection-refraction at an n1-n2 interface. This suggests that something more may be occurring. The notion of dt representing the probability for a detection-measurement hit is based on the idea of an interaction occurring at the center-of-mass point of the particle. Given the photon example, and E=pc, we suggest that there might be a different interpretation of information. For fixed E, if c1=c/2, then p1=2p. By taking 1/p1, one obtains the factor ½ again. Thus, one might be able to consider 1/p as linked with information. Formally, t=distance/c = p* distance for light, but we suggest using this in general for both light and particles with rest mass. Then, 1/p is linked with information. High p means low 1/p and high information. One wishes to minize information. A function which enforces the 1/p resolution is exp(ipx) such that exp(ipx)exp(-ipx) =1 like dt=constant. This would explain the 2-D Snell’s law Fermat’s Least Time Principle case, but it also has implications for p= mo v = (2mo) (v/2). In this case, the particle’s have the same momentum, but different speeds. If one only thinks in terms of speed, then they carry different information, but in terms of an impulse hit, they carry the same information. If one insists that impulse hit is linked to information, then using 1/p as the information quantity seems to hold. As a result, one no longer follows the center-of-mass in space and time in such an approach. This suggests a different scheme which is ultimately equivalent to the statement that for dt=contstant, v=constant, one has the same probability for each dx. Mathematically, exp(ipx) creates periodic regions proportional to 1/p, but exp(ipx)exp(-ipx) =1 yields a constant for each dx which is akin to dt=constant in each of these dx regions. The form exp(ipx) allows one to solve 1D reflection-refraction, while minimization of information px (part of the action -Et+px) allows one to find Snell’s law.
Addendum Oct. 26, 2022 In ((3)) one should formally subtract i.e. d/dx{ -E1 t + (p1x) x +E2t + (p2x) x) = 0. Then p1x-p2x=0 and -E1 dt1/dx + E2 dt2/dx = 0. Thus conservation of momentum implies Fermat's principle. The minus sign between dt1/dx and dt2/dx follows from t1 being a variable of x and t2 of (L-x) where L is the length of the mirror such that initial and final y values have the same magnitude. (This does not change the conclusions made in the note we argue.) In a previous note (1) we argued that Fermat’s least time principle applied to light involved finding an expression for time in terms of spatial variables and the taking of d/dx where x is an invariant spatial variable in the problem. For example, for reflection from a stationary or constantly moving mirror in the y direction or refraction with different n(i) (indices of refraction in the y direction) x is invariant because there is no medium discontinuity in the x direction. We argued that Fermat’s principle is equivalent to a conservation law in the direction of the invariant variable and that this conservation is in fact that of momentum in this direction. In (2) we argued that E=pc with c=x/t i.e. the wave dispersion equation for light contains Fermat’s principle because one may write it as 0= -Et+px and apply d/dx using t(x) with x being the invariant variable. With conservation of momentum in the x direction and E1 = E2 for the two rays: dt1/dx + dt2/dx = 0 for a two ray problem. Fermat’s principle and E=pc pertain to light. The latter may be written as 0= -Et+px ((1)) for light with x/t=c which as we argued above leads to Fermat’s principle if one takes d/dx for two rays. One may note, however, that one also obtains Fermat’s principle for constant= -Et+px with momentum conservation in the x direction. Furthermore a particle with rest mass bouncing off a floor may behave like light. This we argue leads to the idea of considering two different frames. For light -E1 t1 + p1 x1 = 0 = -Et + px with x/t = x1/t1 = c. For a particle with rest mass -mo t1 + 0 = -Et+px with x/t=v where c=1 and mo is energy in a rest frame. (This is in fact the idea of a Lorentz invariant.) Thus Fermat’s principle for a free particle seems to appear in the form -Et+px = constant. The constant is 0 for light. The form -Et+px is consistent with a wavelength proportional to 1/p and a frequency to 1/E not only for light, but for the particle with rest mass. We argue that light moving in a medium is a complicated statistical process of scattering, but that the problem is treated in a “thermodynamical” manner by characterizing the medium by a single number i.e. single piece of information, namely the index of refraction i.e. c→c/n and p→pn for light. We argue that movement within the medium is like moving in a kind of constant potential for light. -Et+ (pn) x still holds and there is still conservation of momentum in the x direction for two media. Thus the idea of d/dx → p is already present in the photon form -Et+px which follows from Maxwell’s equations, but could also be derived from Fermat’s principle as argued in (2). In other words, ideas of quantum mechanics for a particle with rest mass could ultimately follow from Fermat’s principle, we argue. A free particle with rest mass is linked with constant= -Et+px and if one wishes to have an eigenfunction of momentum then: -id/dx exp(ipx) = p exp(ipx). For a constant potential, just like with light, p would change, but the form exp(ipx) would remain. Thus it seems that even in average situations (like refraction) one still wants to have a periodic type spatial function W(x) such that -d/dx d/dx W(x) /W(x) = <pp> such that <pp>/2m + V(x) = E for a particle with rest mass. This is the Schrodinger equation and it is found mathematically that only certain En are allowed if one wishes to have specific boundary conditions like W(x) →0 at +/- infinite for say an oscillator problem. Thus we argue that the fundamental idea of taking a derivative of an invariant spatial variable in Fermat’s principle which automatically yields conservation of momentum is a basic idea found in quantum mechanics even for a particle with rest mass, namely that d/dx is associated with momentum. Mathematically d/dx is a translation operator and so the link with Fermat’s principle shows how translation is linked with a spatially invariant spatial variable e.g. x. Furthermore Fermat’s principle shows that even complicated statistical scattering problems such as refraction (in which the photon scatters off of different molecules in different media) can be reduced to a periodic average problem with p→pn. This suggests that for a particle with a rest mass in a potential V(x) one should also have a periodic spatial variable such that d/dx d/dx is linked to average momentum, we argue.
In part I we argued that Newton’s first law implies that constant momentum (i.e. momentum conservation) is associated with spatial invariance i.e. a lack of information in a certain direction. This is equivalent to stating there exists no force in that direction. We argued that one may apply the generator for translation in the spatial direction i.e. d/dx for x and apply it to a function f(x) such that df(x)/dx = (px). The simplest form is f(x)= (px)x. We also argued that it is precisely this function which appears in Fermat’s least time principle because time = distance/speed and one may multiply time by |p| = momentum magnitude. Given that momentum and displacement are parallel vectors for a light ray, p dot r = p distance = (px)x + (py)y for a two dimensional problem (e.g. reflection or refraction). “Distance d” is the key function to use because dd= xx + yy implies that for fixed y, d d/dx = x/d = sin(theta1) (angle measured from y axis) so p d d/dx = (px) which is a conserved quantity if x is invariant. Thus conservation of momentum for two rays is equivalent to extremization. Extremization, we argue, suggests there may be a built-in physical mechanism which allows a free particle to correct itself if perturbed slightly just like a Maxwell-Boltzmann gas which extremizes entropy subject to constant average energy does. This built-in mechanism should be linked to motion in space. Writing (px)x as ln(exp(ipx), one not only obtains an eigenvalue equation i.e. -d/dx exp(i (px) x) = (px) exp(i px x), but the notion of a relative change arises if exp(i px x) is interpreted as a kind of probability. A relative change is natural because an absolute one ignores the idea that probability may differ with x even though (px) is constant. Thus from the idea of invariance in space found in Newton’s first law, which is equivalent to conservation of momentum in that direction, the equation d/dx f(x) = (px) leads directly to -id/dx exp(i px x) = (px) exp(ipx x) where exp(i px x) represents a physical probability type of wave which stablilizes the constant (px) motion because this motion is equivalent to extremizing time or |p| time or p dot r / speed of light in Fermat’s principle. In other words, the quantum mechanics of a free particle is associated with stabilizing wave behaviour exp(ipx x) which was not seen experimentally at the time of Newton or Fermat, but seems to be part of their formalisms, with Fermat’s principle derivable from Newton’s first law.
Newtonian mechanics (from the 1600s) defines a quantity called momentum p in terms of mv where m is mass and v=dx/dt with dx and dt tending to zero. Newton’s first law is equivalent to conservation of momentum in the absence of a force. The second demonstrates how p changes under a force. If one has equal action/reaction then overall momentum is conserved. As a consequence the idea of a “wave” associated with p does not arise. The main focus is on the idea of force. Fermat’s Least Time Principle for light (1600s), however, is based on minimizing the overall time traveled by light which reflects or refracts. Young’s interference experiments did not occur until the very early 1800s and Newton considered light to be a particle at this time. In fact light bounces off a foil creating an impact like a particle. Fermat’s principle does not directly make use of momentum or its conservation. It is focused on minimizing time. To do so, however, one must take a derivative with respect to a certain variable and set it equal to 0. This variable is the spatially invariant one in a problem. For example consider reflection (stationary or moving mirror) or refraction in two dimensions x,y. Let the mirror or medium face lie along the x axis at y=0. There exists a constraint on y and given a fixed starting point one cannot change y. This leaves x to be varied because it is invariant. Thus the extremization of time is really a statement of invariance of a spatial variable. In other words, Fermat’s principle is linked to the physical variables t and x while Newton’s focuses on p (momentum). This invariance in turn gives rise to conservation of momentum in the direction of the invariant variable i.e. (px). There is still the idea of Newton’s change in momentum as (py) abruptly changes in refraction as it hits the interface of media. The point we wish to make is that from Fermat’s least time principle one sees that conservation of momentum in the absence of a force is completely intertwined with invariance in space. The two cannot be separated in this approach and there is nothing unphysical about t or x. In fact, one may create a function A= -Et + (px)x + (py)y such that dA/dx partial = px = conserved quantity. In this function (px) and x are on the same footing. One does not have one without the other. This idea is carried over into special relativity as -Et+p dot r is a Lorentz invariant. Given that p momentum is physical (it manifests itself in impulse hits) and p and x are inseparable in (px)x we suggest there should exist a physical manifestation in x i.e. in space linked to the maintenance of conservation of momentum in this direction even in the absence of a force. This we argue leads to the physical idea of a wavelength which follows from A=-Et+px with A=0 in all constant velocity frames leading to a frequency proportional to 1/E and wavelength to 1/p. We have discussed these ideas in previous notes, but the main point we wish to make here is that equivalence of spatial invariance in the x direction in Fermat’s principle with conservation of momentum in the same direction suggests that if p is physical, the spatial invariance must also be manifested physically. It is not simply a mathematical convenience to link t to x to px, we argue. Furthermore this wavelength should be linked to establishing conservation of momentum as it does in quantum mechanics where the orthogonality of exp(ipx) i.e. Integral exp(-i p1 x) exp(ip x) dx automatically ensures p1=p without imposing momentum conservation a priori as in Newton’s first law. Both Newtonian mechanics and Fermat’s principle suggest that p may change abruptly due to a force or an immediate change in medium. The idea of a physical wavelength, however, suggests there may be interactions completely different from an abrupt change in p due to an impulse. In particular there may be interference or Bohm-Aharonov interactions. Newtonian Mechanics
Newton’s second dp/dt = Force where p is momentum is equivalent to pp/2m + V(x) =E for a conservative force= -dV/dx with p=mo v(x), v being velocity. In general one may divide the x axis into equally spaced dx units with dt(x) (i.e. time spent) differing in each unit. A main point we wish to make is that dt(x) is the same whether the particle moves to the right or to left. Thus using dt(x) as a kind of density (proportional to C/v(x)) as done in (1) means this spatial density does not distinguish between motion to the right or left. This is one kind of spatial density we argue, one that considers only the magnitude of velocity. Heuristically, however, one might argue that motion to the right or left involves a skewed spatial density in the direction of motion. This we argue represents a second kind of spatial density which follows from Newton’s first law and is associated with quantum mechanical motion. Ideally the two should be linked. Newton’s first law indicates that a particle at rest remains at rest and one in constant motion in a direction continues in this state unless acted on by force. This law is usually not associated with a mathematical equation and so a spatial density is not derived. In (2), however, we argue that Newton’s first law is associated with invariance in a spatial direction if there is no force in that direction, for example x. In such a case, d/dx is the translation generator in that direction and momentum is kind of “information”. We argued in (2) that one may write d/dx f(x) = p to describe this with the simplest solution being f(x)= px for constant p. We further argued that this is in fact linked to Fermat’s least time principle for light. Time =distance/speed of light, but momentum and displacement are parallel. Thus for a two dimensional problem: p dot r = (px)x + (py)y = p distance = p sqrt(xx+YY) and p sqrt( (L-x)(L-x) + YY) for two rays, one being the incident and the other the reflected. Given that there is conservation of momentum in the x direction one may write (px) incident = (px) reflected or d/dx (p distance incident) - d/dx (p distance reflected) = 0 which is Fermat’s principle for reflection with one index of refraction. For refraction p1=pn1 and p2=pn2, but (p1x) = (p2x). f(x)=px satisfies d/dx f(x) = p. This expression depends on x, but also on the direction of motion i.e. one may have p or -p. In (2) we further argued that this expression may be written in terms of: i px = ln(exp(ipx)) with exp(ipx) being periodic, but neither growing or tending to a constant zero for a constant p. As a result we identified exp(ipx) with a second type of spatial density or probability, one which accounts for the direction of motion. exp(ipx) is also associated with a quantum mechanical description of a free particle. Thus Newton’s first two laws give rise to two different spatial densities, but they are related. In particular, the modulus of exp(ipx) equals constant which is P(x) or spatial density from Newton’s second law. In this note, we start with the assumption that a velocity (not velocity magnitude) spatial density should exist and examine what properties of such a density/probability arise.
In a previous note (1) we argued that Fermat’s principle of least time is present in the photon wave dispersion relationship E=pc which may be written as 0 = -Et+px with x/t=c. We argued that a nonzero constant = -Et+px applies to a free particle with rest mass, but that Fermat’s principle is still present. Furthemore the form -Et + px implies the existence of a frequency and wavelength for both a photon and free particle with rest mass. (Interestingly -Et+px is both the free particle relativistic action -mo t sqrt(1-vv/cc) and nonrelativistic .5mo t vv with c=1 and v=x/t. ) Fermat’s principle implies that d/dx → momentum (x component) as discussed in(1). An eigenfunction exp(ipx) demonstraties wave behaviour of a free particle with rest mass. In (1) we argued that if one knocks out a quantum bound particle it will be in a state exp(ipx). Thus this wave type information should also be included in the bound state so wepostulated a bound state represented by W(x) = Sum over p a(p)exp(ipx) (i.e. a wavefunction) such that the Fermat result d/dx → p still holds. In this note we wish to consider the idea of a statistical equilibrium and the concept of a constant energy. In the case of refraction it seems a complicated statistical scattering process occurs which slows down the motion of a photon i.e. c changes to c/n where n is the index of refraction. Thus one piece of information, namely n, describes the complicated statistical process as in thermodynamics. If E (energy) remains constant p→ pn. Thus in this complicated statistical scenario a cos(-Et+pn x) electric field pattern remains but p → pn i.e the wavelength has changed. The idea of a constant energy may even be applied to light scattering from a mirror moving with a constant v in the negative y direction. The actual energy and momentum of the incident and reflected rays are not equal, but for E1 special = p1 v(relative 1) and E2 special = p2 v(relative 2) with v(relative 1) = c- vcos(A) and v(relative 2)= c+vcos(B) where A and B are angles of incidence and reflection measured with respect to the y axis. E1 special t=E2 special. In this note we retain the idea that a particle knocked from a quantum single particle bound state is described by exp(ipx) in keeping with Femat’s principle. We consider the bound state to still have waves as Fermat’s principle suggests d/dx →p, but they cannot be of the same height and width because of the changing value of V(x) with x (i.e. the potential). According to Newton’s law one has conservation of energy i.e. pp/2m + V(x) = E, but we consider this to apply to a p (rms) = m v(rms) in the Schrodinger equation. If the forms exp(ipx)’s are to be retained within the bound quantum state (i.e. this information is not lost) and one has a constant E (energy) then the potential must be modified statistically to allow for the idea of conservation of energy for each p, but must also represent a constant potential to retain the form of exp(ipx). Thus in short the Schrodinger equation becomes a weighted sum of exp(i p(j) x) functions with each having a different wavelength (like refraction), but interacting with different constant potentials to ensure that E remains constant for each p(j). In other words we think of a potential V(x) as a statistical set of refraction discontinuities i.e. impulses which produce exp(ipx)’s which are like light moving through a homogeneous medium i.e. a momentum eigenstates. For this to happen, however, the various exp(ipx)’s need to combine with V(x) = Sum over k Vk exp(ikx) terms to create a Vmodified(p) exp(ipx) where Vmodiied(p) is a constant because each free particle exp(ipx) is associated with a different effective potential and a kinetic energy pp/2m, but the same energy En. We argue that this results from applying ideas of Femat’s principle to a particle with rest mass. Newton’s picture of a particle accelerating i.e. x(t) emerges if one considers rms values for which all information of exp(ipx) and crests/troughs of W(x) are ignored. In particular, W(x)=wavefunction may have crests and troughs, but -1/2m d/dx dW/dx/ W = KE(x) Newton which shows no crests or troughs, but simply balances V(x) in KE(x)+V(x) = E.
In Part 1, we suggested that Fermat’s Least Time Principle is equivalent to a least information in space approach, where spatial information is given by length / (hbar/p) or px. Here we note that given the Lorentz invariant -Et+px, one actually has minimal information in both space and time. For a fixed E (E is the same in both n1 and n2 regions for refraction) in Et, it seems that one minimizes time, but one actually minimizes time/ (hbar/E). Independently in terms of time and space, one minimizes information. This leads to a consistent picture (which may use either space or time) we argue and also one that is statistical as opposed to the notion of minimizing time for no particular reason. Given that we use the notion of special relativity above, we try to see how this theory is linked to minimization of information. In another note (1), we showed that one may use F=dp/dt to derive the form g(v) = 1/sqrt(1-vv/cc). g(v) appears in the comparison of physics in a rest frame and one moving with -v, but it is F=dp/dt (a Newtonian vector equation) which fixes its functional form. Given F=dp/dt, one is forced to focus on conservation of momentum and projections in 3-space. We note, however, that special relativity yields E=[p] c for mo=0. Here [p] is the magnitude which is rotationally invariant like E. Given x=vt, one may write: -Et+px = 0 for a photon. Here p=[p]. As a result, one seems to have two very different ideas present, we argue. First, F=dp/dt, a vector equation, deals with components and conservation of momentum for various components. -Et+[p]x=0, however, is an equation which lies along the direction of motion, regardless of whether momentum is conserved in this direction. It is analogous to a wave equation with arguments -wt+x/lambda=0 If there really were a physical connection, then conservation of momentum in a particular direction would be akin to Newtonian mechanics, while -Et+[p]x=0 might be akin to wave mechanics. In Part 1, however, we treated the wave mechanics features as information without mentioning the word wave. It seems that -Et+[p]x =0 is linked with physical behaviour along the direction of motion, and conservation of momentum in one direction (not x) is a consequence. We suggest that a priori one might wish to follow a particle along its direction of motion and have a theory which is based only on numbers, not vectors. This, we argue, is the information theory introduced in Part 1, i..e minimization of information in either t or x.. Here we note that there are in fact two minimizations of information, one in time t / (hbar/E) and one in space x / (hbar/p). We apply this reasoning to two dimensional refraction (Snell’s law) and note that particle behaviour only appears when one considers projections to obtain conservation of momentum parallel to the n1-n2 medium interface.