I chose the title Physics from the Edge because the theory of inertia I have suggested (MiHsC) assumes that local inertia is affected by the far-off Hubble-edge. My webpage is here, I've written a book called Physics from the Edge and I'm on twitter here: @memcculloch

Saturday, 24 September 2016

MiHsC/QI vs New Galaxy Data

It is crucial for theorists like me to pay close attention to new data, and observational astrophysicist Prof Stacy McGaugh has a habit, like John Anderson, of writing nicely concise analyses of new data that are easy to test MiHsC/quantised inertia on. McGaugh and Lelli have just published a paper comparing the data from 153 different galaxies across a large range of scales from dwarfs to large disc galaxies, looking at two things 1) the acceleration within them as predicted using standard dynamics from their baryonic (visible) mass only (g_bar, x axis in the Figure), and the actual acceleration as determined from the observed motion of their stars (g_obs, y axis). The result is shown on a log-log plot here (from McGaugh and Lelli, 2016):
If Newton or Einstein's general relativity were right (without the crutch of dark matter) then you would expect the data to lie along the faint straight diagonal line, so that the predicted acceleration would equal that observed. It does not. As you can see the data lifts from the straight line on the left hand side because for lower accelerations the stars' orbital speed, and therefore acceleration, is greater than expected. This is the famous galaxy rotation problem noticed by Zwicky (1933) and Rubin & Ford (1980). The majority of astrophysicists fix this by adding dark matter to the galaxies where it's needed, but that is an ad hoc solution that no-one should have to resort to in a scientific age. In the paper, McGaugh and Lelli fit a curve to this data using this formula
and show it predicts well if the fitting parameter 'gt' is set to the value 1.2x10^-10 m/s^2 and both McGaugh and Lelli, and Milgrom in a note published soon after, say this looks like some of the possible variants of MoND (Modified Newtonian dynamics). Milgrom in his note even says 'nothing else can give this behaviour' but this is simply not true. MiHsC / quantised inertia also predicts this behaviour far more plausibly and simply than MoND, and without any adjustment at all. The MiHsC formula can be found in McCulloch (2012) (see Eq. 6 in the ref below, and note the comments on a' therein. The Eq. I showed before is based on Eq. 7 and is less general). The MiHsC formula (note, only strictly valid at a galaxy's edge) is
This formula is based on a specific physical model (MiHsC: Unruh waves harmonising with horizons) unlike MoND which is just an empirical relation with no physical model. It is obviously much simpler than McGaugh & Lelli's formula, and crucially there are no fitting parameters in MiHsC/quantised inertia at all! Its predictions of g_obs are exactly right see Figure below and are predicted from just the known quantities of the speed of light (c) and the size of the observable universe (Theta=8.8x10^26 m):

I do feel a bit like a broken record going on about this, but it's necessary because it has not yet been widely appreciated that a theory that predicts real data without any fitting parameters, like MiHsC does, is like a diamond in the mine. Like special relativity, it is a sign of something fundamental.


McGaugh, S.S, and F. Lelli, 2016. The radial acceleration relation in rotationally supported galaxies. Phys. Rev. Lett. (to be published). Preprint.

McCulloch, M.E., 2012. Testing quantised inertia on galactic scales. Astrophysics and Space Science, 342, 342-575. Preprint

Monday, 19 September 2016

Unruh Radiation Confirmed?

One of the criticisms of MiHsC (quantised inertia) is that "it relies on Unruh radiation which has not been seen". The fact that MiHsC predicts a whole range of important anomalies is some evidence for its use of Unruh radiation, but it is not direct evidence. In this blog entry I'd like to show you why I'm fairly confident that Unruh radiation has also been directly detected, based on papers by Beversluis et al. (2002) and Smolyaninov (2008).

Unruh radiation is usually difficult to see, to put it mildly, because its wavelength is L=8c^2/a, where a is acceleration and c is the speed of light, so for a typical acceleration of 9.8 m/s^2 the wavelength is eight light years. You'd have to wait eight years for the wave to pass by and be confirmed. It would make for a nice quiet sinecure, but there is a quicker way to see them. The formula shows that to produce Unruh waves near the more familiar EM spectrum you need to increase the acceleration, hugely!

This was done unwittingly by Beverluis et al, 2002. They shone a 780 nm laser at a gold nanotip, a very sharp pin if you like, so that the 'plasmons' (quantised waves of free electrons) formed had to travel over the very sharp tip (radius of curvature, r=10^-6 m) at a very fast speed (v) and were therefore accelerated by

a = v^2/r = c^2/10^-6 = 9x10^22 m/s^2.

They noticed an anomalous photoluminescence from the nanotip with a wavelength of about 900 nm (see the Figure below, from Beversluis et al., 2002, which shows the intensity of the radiation emitted on the y-axis, as a function of wavelength, on the x-axis. See the peak at 900 nm). Beversluis et al. proposed a complex 3-stage process to explain it but also stated that each step was implausible.

Enter Smolyaninov (2008) who claimed that their suggested 3-step process for explaining this 900 nm radiation was indeed implausible, and showed how the conversion of 780 nm laser light to 900 nm photoluminescence could be due to Unruh radiation instead. It works as follows: the 780 nm laser light impacts the gold surface and creates 'plasmons' (quantised waves of free electrons) which zoom around the sharp bend at the tip (this is well known). The plasmons' acceleration generates Unruh waves of wavelength L = 8c^2/a = 8000 nm. The newly-incoming laser photons excite the gold molecules and normally they'd then drop back to the ground energy state and emit the same frequency photon (Rayleigh scattering), but the ground state is now higher because of the Unruh excitation so the molecules don't lose as much energy as they gained, and the light they emit now has a lower energy (a longer wavelength). In more typical, non-Unruh, cases this is called Raman scattering and is usually undetectable, but for rough surfaces it can be enhanced (a well-known process called Surface Enhanced Raman Scattering, discovered first by Martin Fleischman in 1973, well before he co-discovered cold fusion / LENR). So the emitted light has a lower energy and frequency as follows:

f(emitted) = f(laser) - f(Unruh)

Replacing frequencies with wavelengths, L, given f = c/L

c/L(emitted) = c/L(laser) - c/L(Unruh)

L(emitted) = 1/(1/L(laser)-1/L(Unruh)) = 864 nm.

This predicts the emission of light of wavelength 864 nm (or above, from the less curved parts of the nanotip), which agrees with the observed anomalous photoluminescence seen by Beversluis et al. of around 800 nm and above (see figure) and so this is good evidence that Unruh radiation has been indirectly seen, since other explanations have been implausible, so far.

This interests me not only because it provides more direct evidence for Unruh radiation, but also that it implies that Unruh waves, seen only in the reference frame of the accelerated object, can cause directly testable effects in the laboratory frame, and this makes me wonder about sonoluminescence (which MiHsC predicts) and LENR (which it doesn't yet, but might if high accelerations are involved).

One way to confirm that this experiment shows Unruh radiation would be to vary the curve of the nanotip and see if the emitted wavelength varies as expected (for example, a spherical nanotip, of uniform curvature, with a cylindrical stem should show a narrower Unruh-peak). Another more direct test of MiHsC (quantised inertia) itself would be to try to interfere with these shorter Unruh waves by shielding them, to see if the plasmons' inertia (trajectory) is affected.


Beversluis, M.R., A. Bouhelier, L. Novotny, 2003. Continuum generation from single gold nanostructures through near-field mediated intraband transitions. Phys. Rev.B., 68, 115433.

Smolyaninov, I.I., 2008. Photoluminescence from a gold nanotip in an accelerated reference frame. Preprint

Wednesday, 14 September 2016

MiHsC: from Hubble scale to Emdrive

A comment on my previous blog entry (by qraal) pointed out something that is obvious to those familiar with MiHsC, but needs to be more widely known: MiHsC does not just predict galaxy rotation like other models (eg: MoND and MOG) but, as I've shown in my papers, it can explain many other eclectic (varied) anomalies that have been seen, from cosmological scales right down to the lab scale (emdrive).

As an introduction for those who are new to this blog. Unruh radiation is like Hawking radiation but is predicted to be seen only by accelerating objects. Its wavelength gets longer as accelerations reduce. MiHsC says that the long-taken-for-granted property of inertial mass is caused by the relativistic Rindler horizon that opens up behind objects when they accelerate. This horizon then damps the Unruh waves on that side but not in front, thus making a radiation pressure that resists the original acceleration and causes humans to say 'that's inertial mass'. This force is also weakened when objects have tiny accelerations because the Unruh waves get long enough to be damped by the spherically symmetric Hubble horizon leading to a new prediction: a weakening of inertia at very low accelerations. MiHsC is encapsulated by the equation below, which is an approximation to the real process of Unruh waves being deselected by horizons

where mi is the inertial mass, mg is the gravitational mass, c is light speed, |a| is the magnitude of the acceleration of the mass and the Theta is the Hubble scale (2.6x10^26 metres). The good thing about this equation is that it is inevitable, being based on an intuitive physical model and all these quantities are well defined and known, so any agreement with data that I talk about here has not occurred because I've 'tuned' the model. The model can only make one prediction, which happens to agree with nature.

On cosmic scales MiHsC correctly predicts the recently observed cosmic acceleration. This drops straight out when you put the above formula for mi into Newton's 2nd law and gravity law. You can try it yourself! At the same scale MiHsC predicts the recently-observed unexpected smoothness of patterns in the Cosmic Microwave Background (the so called low-L CMB anomaly). The Hubble horizon is damping long waves.

MiHsC predicts the rotation of all galaxies from tiny dwarfs, through discs to galaxy clusters, and just from the equation above, whereas MoND has to be 'tuned' by parameter a0, and dark matter has to be packed-in differently for each galaxy to make the rotation curve fit general relativity. MiHsC also predicts why the oddities start when accelerations go below a particular value: the Hubble horizon then starts deselecting the long Unruh waves (see diagram below).

MiHsC predicts the Pioneer, Galileo and Ulysses anomalies, with its weakening of inertial mass at low accelerations. The Pioneer anomaly has been also explained by a complex thermal model, but the emphasis here being on complex.. I distrust complex adjustable models. MiHsC predicts the flyby anomalies, close, but not perfectly, as being due to the weakening of inertia since mutual accelerations between passing spacecraft and the spinning Earth are lower near the poles, reducing the inertia mass of the craft and by conservation of momentum (mass times speed) boosting their speed.

It predicts the anomalous effects seen by Tajmar when he spun a supercooled disc and noted that a nearby accelerometer moved slightly with the disc (with no frictional contact).

It predicts the Emdrive, a truncated metal cavity with microwaves inside, which, In MiHsC, is rather like joining a big cosmos (the wide end) which allows more Unruh waves (more inertia) to a small cosmos (allows fewer waves, less inertia) so that objects, microwave photons in this case, going from the small end to the big end gain mass (ie: the mass shifts rightwards, see diagram) so to conserve momentum (mass x speed) the cavity has to go the other way, as it does. The numbers agree with the data quite well.

This list represents successes from a cosmic scale of 10^26 metres, right down to the lab scale of a 0.1m! I'm now working on the proton radius anomaly which may give me another 10 orders of magnitude of scale. I should say also, that MiHsC reduces to the standard model for high accelerations (in the equation above hen a is huge then mi = mg, the famous equivalence principle) so it is not contradicted by any empirical data.

MiHsC can also be called quantised inertia, which is a more accessible name. Another possible name is 'horizon mechanics', which was suggested recently by someone who's read my book: John Michael Dorman. More theoretically, it is another step consistent with the history of science, which has always progressed by debunking invisible quantities, like the Greek gods (Thales), epicycles (Newton), the aether (FitzGerald, Einstein). Now we can jettison dark matter (well, most of it). MiHsC also links together relativity and quantum mechanics in a natural way.

Mathematically, MiHsC is not complete, instead of the equation above, it ideally needs a formula to describe exactly how Unruh waves are allowed or disallowed by horizons of a complex shape, and then how the remaining Unruh waves push the object around. There lots of scope for mathematicians here..

Experimentally, the best way to prove MiHsC would be to try and accelerate an object so much that the Unruh waves it is assumed to see, become small enough (they're usually light years long) to be interfered with by our technology. For example if you set up a spinning disc or resonate light within a cavity, and then block the Unruh waves on one side only, MiHsC predicts the object should move towards the blockage.

Tuesday, 6 September 2016

A marriage of relativity & quantum mechanics

I'm often accused of being a radical, but I'd like to point out that MiHsC is actually far less radical than dark matter. Consider dark matter. Its supporters believe that 96% of the universe is in the form of a new and invisible form of matter and energy that has a weird structure, as dark matter for example must cling to the edges of galaxies and stay away from their centres, so it needs an entirely new dark-sector physics to explain it.

In the meantime, MiHsC says only that we need to fully accept both relativity and quantum mechanics and therefore the horizons and Unruh waves that they have already predicted and, here's the crucial new bit: assume that inertial mass (never understood before, and Higgs only predicts 0.01% of it) is caused by the Unruh waves being modified by the horizons. The result is a dynamics far better than the usual one since it predicts all dwarf galaxies, spirals and clusters without any tuning, whereas the dark matter hypothesis needs a new type of matter and its physics to be added, and a different amount for each galaxy. This is why it is strange that dark matter-ists accuse MiHsC of being a weird theory out of the blue. The pot calling the kettle black.

MiHsC is simply a particular marriage of relativity and quantum mechanics that happens to predict inertial mass. Of course, people have been trying to marry these two theories off for a long time and have failed because they didn't want to break the equivalence principle or they worried about non-locality, but these breaks are theoretical and not necessarily forbidden by experiment.

The bride (quantum mechanics) and groom (relativity) were first introduced to each other by Hawking, Fulling, Unruh and Davies who predicted Hawking-Unruh radiation. People like Jennison, Milgrom and Haisch and Puthoff offered hints but without a plausible method. I disliked the complexity of the arrangement and suspected there was actually a baby coming (a better model of inertial mass), got my shotgun and rammed the two theories together, and damn the consequences. I've been able to continue because all the consequences that were feared for decades turn out not to disagree with experiment and in fact they predict many of the anomalies that the dark sector was invented to fudge. It just shows that sometimes you just have to suck it and see.

To make this point even more simple mathematically than before, I've recently submitted a paper showing I can derive MiHsC (with an annoying factor of 1.26 probably caused by my simplification of the shape of the Rindler horizon) in 10 lines just from Heisenberg's uncertainty principle and special relativity, similar to my approach to gravity in 2014 (see ref). So I'm just waiting patiently for the standard first rejection...


McCulloch, M.E., 2014. Gravity from the uncertainty principle. A&SS, 349, 957-959. Video

Friday, 26 August 2016

Dragonfly 44: a fudge too far.

Yesterday, a few people online kindly sent me the news that a galaxy called Dragonfly 44 has been found apparently containing 99.99% of the so-called 'dark matter' (see reference). Many, for example Sean Carroll, said that this supports the dark matter hypothesis. However, in reality it worsens the outlook for dark matter, which has to be added in different ad hoc amounts to each galaxy, in huge amounts to this one, and so it further supports MiHsC which works for all galaxies without any arbitrary tuning needed...

For each separate galaxy they find, the darkmatterists have to add dark matter in different amounts, 90% for the Milky Way, 99% for dwarf satellite galaxies, 99.99% for this one. This number is arbitrary, they chose it incestuously to make general relativity work for the data, and it has no reason behind it except to save GR. That means that the dark matter hypothesis is not only not falsifiable (you can look for dark matter for ever), it is also not predictive. Given the visible light distribution, dark matter cannot predict the velocity of stars. So dark matter is a bit like Peeves at Hogwarts: useless, but you can't get rid of him.

In contrast MoND and quantised inertia / MiHsC are both predictive. Given the visible mass M (it's best to base theories on visible stuff) MoND says the stellar velocity is v=(GMa0)^1/4 (a0 is a fitting parameter) and MiHsC says v=(2GMc^2/Hubblescale)^1/4 (no fitting parameter, and a slighly higher velocity, see equation below) and both predict the velocity dispersion of Dragonfly 44 within the uncertainty and without the need for any dark matter.

The important advantage of MiHsC over MoND is that MiHsC has no adjustability, at all! MoND was an inspiration for me, but it is not a theory, it is an empirical formula that happens to fit galaxy rotation if you 'tune' the adjustable parameter a0 against the data, so it is not surprising that it fits the data, because it has been fitted to it. The parameter a0 has no 'physical reason'. MoND is a bit like an engineering formula with its tune-able a0. In contrast MiHsC has been derived from first principles, says the velocity has to be one value with no tuning possible, and every parameter in the MiHsC formula below is known from observation and is there for a good physical reason, so it is remarkable that it works on all galaxies including Dragonfly 44:

In this way MiHsC resembles special relativity in that there is no arbitrariness about it, no input numbers, but it will need input from better mathematicians than me to properly describe higher order effects due to the interaction of quantum waves with relativistic horizons or cavities.

A loose analogy to this crisis in astrophysics would be holding a competition to see who can make a shoe to fit a foot they've only seen from a distance (like a theory that must predict galaxy rotation from observed quantities to be useful). Shoemaker 1 comes with his shoe, says 'It fits', but you see nothing. 'It's invisible!' he says (GR and dark matter). Shoemaker 2 comes with a shoe that has an adjustable strap and he adjusts the strap to fit it (MoND). This is better. Then shoemaker 3 comes with one unadjustable shoe, and it fits (MiHsC). Obviously, the third shoemaker is the best one.

I cannot emphasise enough the difference between tunable theories like dark matter, and to a lesser extent MoND, and a non-tunable theory like MiHsC. Only a non-tunable theory gives a real understanding of the physics.


van Dokkum, P., 2016. A High Stellar Velocity Dispersion and ~100 Globular Clusters for the Ultra Diffuse Galaxy Dragonfly 44. http://arxiv.org/abs/1606.06291

McCulloch, M.E., 2012. Testing quantised inertia on galactic scales. Astrophys. & Sp.Sci., 342 (2), 575-578. http://arxiv.org/abs/1207.7007

Tuesday, 23 August 2016

The Emperor's Naked!

Karl Popper made a statements that should by engraved above the entrance to every proud physics department:

"A theory can never be proven, but it can be falsified, meaning that it can and should be scrutinized by decisive experiments. If the outcome of an experiment contradicts the theory, one should refrain from ad hoc manoeuvres that evade the contradiction merely by making it less falsifiable."

In 1915 Einstein finished general relativity and this theory has been successful in regimes where the acceleration is high (such is in the inner Solar system and close-orbiting binary stars. Then in 1933 (by Fritz Zwicky) and in 1980 (by Vera Rubin) it was found that galaxies were orbiting so fast at their low-acceleration edges that, if they had any decency at all, they should explode centrifugally. Yet they don't: they generally persist in sensible bound/round shapes.

This means that most of the observed cosmos, the bit with a low acceleration, does not agree with general relativity. The theory has been hugely falsified, but tell that to a mainstream physicist at a conference (I have done) and you'll be pigeon-holed in a category somewhere below holocaust denier and they'll walk away in disgust. Speaking objectively, given GR's failure with 90% of the data, you can either claim that general relativity is wrong, and invent a new theory (for example I have suggested MiHsC, which reduces the outward centrifugal force in galaxies) or you can claim that there is a huge amount of invisible matter holding the galaxy together by gravity: dark matter.

There are several reasons why it is obvious that the problem is a failure of theory and not due to dark matter:

1) The galactic problems always start at a radius where the acceleration goes below about 10^-10 m/s^2 which is also the cosmic acceleration, so it is obviously linked (in a way that MiHsC explains simply).

2) Globular clusters and wide binary stars, which are far too small to have any dark matter, also show the same anomaly, starting at the same critical acceleration.

3) Many other well known anomalies, eg: cosmic acceleration, the low-l CMB anomaly, the flyby anomaly, the Pioneer anomaly, Tajmar's spinning disc results, the emdrive and the proton radius puzzle are all consistent with the galactic rotation anomaly if you look within the framework of MiHsC.

4) Dark matter is just a mess. It needs huge amount of new matter to be added in an ad hoc way and it also needs new physics to keep it spread out in a galactic halo.

5) There is no evidence for dark matter even after 40 years of very expensive looking.

6) Philosophically: dark matter defends a theory (GR) that failed to predict galaxy rotation properly, by inventing an ad hoc manoeuvre that makes it less falsifiable (see Popper's warning above). If a theory uses an ad hoc fix that is, even worse: as vague as a bad politician's promises, then beware!

As a result of these points I'm amazed that the rest of physics is spending most of its money (each dark matter detector is on the order of $100 million dollars) to defend general relativity by looking for dark matter and almost no-one is challenging the theory. I agree the dark matter option should have been looked at, a null result, like Michelson and Morley's is useful, but it must be in proportion to other options. Instead dark matter is tacitly assumed in journal papers and magazines articles, and there are even dark matter conferences which are inherently unscientific, since they pre-assume the solution they're looking for!

As Popper suggested, instead of trying to defend the status quo, it has always been more effective to attack the prevailing theory. It is easy to attack general relativity since the counter-evidence is already plentiful (see above), so why don't they?

After ten tears of fighting it, I think the obsession with dark matter is like the age-old story of the Emperor's new clothes. People are told that only the cleverest can see the clothes (dark matter), so everyone of course says that they can see them when the Emperor goes on walkabout awarding post-doc positions, but eventually some idiot comes along and says 'Ho Ho The Emperor's Naked!' and the spell is broken. Well, I'm willing to be the idiot, and I have suggested MiHsC, and, although I say it myself, MiHsC is far more successful than GR, absolutely beautiful and simple in form, joins quantum mechanics and special relativity for the first time, and offers a new way to get energy and propulsion (by learning to put horizons in the zero point field). Since we're talking about invisible fashions, it seems appropriate to quote Coco Chanel:

The most courageous act is still to think for yourself. Aloud.

Monday, 15 August 2016

Honey, I Shrunk the Proton!

Standard physics is having an increasingly embarrassing time. It failed to predict the galaxy rotation problem, then cosmic acceleration, both just about the biggest anomalies you could imagine, representing 96% of the whole cosmos. These embarrassments have been hidden under the carpet with the fudges of dark matter and dark energy (whereas MiHsC predicts the embarrassments). There have been other anomalies too: the low-l CMB anomaly, the alignment of quasars, the spacecraft flyby anomaly, the Tajmar effect, the emdrive (all of which MiHsC predicts), but these anomalies have mostly been ignored by the mainstream who are focusing on the internal consistency of a standard model ever more at odds with nature (Rearranging deckchairs on the Titanic). However, now comes an anomaly (the proton radius puzzle) that is so central to the standard model that it will be impossible for them to ignore.

The proton radius is well predicted by the standard model as 0.88x10^-15m and has been measured as such for many years. You can measure it by bouncing electrons off the hydrogen nucleus (a proton) or by firing lasers at electrons orbiting the nucleus in their circular train tracks (to use the simplified Bohr model) and seeing how far they jump between tracks, a jump that depends on the proton charge radius because of the Lamb shift (an effect of the quantum vacuum).

In 2010 a group at the Paul Scherrer Institute in Switzerland decided to see what would happen if they made a hydrogen atom, replacing the electron with its overweight twin the muon (identical to the electron, except 200 times heavier). The advantage of using a heavier muon was that it orbits much closer to the proton thus allowing a more accurate result when they track the maths back to predict the proton radius. To their surprise the muon jumped a bit more than expected between its orbital levels and the equations leading back to the proton radius implied it was 0.84x10^-15m: 4% smaller than before (this was confirmed in 2013 and 2016, see Pohl et al. below). This is an anomaly seven times larger than the uncertainty in the original proton radius measurement (a so-called 7 sigma anomaly), so it constitutes a significant discovery.

The trouble, or rather the opportunity, here is that there is nothing in the standard model to allow for a proton to shrink in the close presence of a muon. Cue MiHsC? I'm now reveling in the summer research period and I've just submitted two theoretical papers on MiHsC, one of them predicting the electron mass and showing that tight orbits can release mass-energy in a new way, accounting for gravity for example. It is interesting that this proton radius anomaly is wrapped up in the Lamb shift, a quantum vacuum effect. MiHsC is also a quantum vacuum effect.


Accessible report about it by John Timmer, Ars Technica: Report

 A more technical arxiv summary: http://arxiv.org/pdf/1502.05314v1.pdf

Latest paper by Pohl et al., 2016. Science.  http://science.sciencemag.org/content/353/6300/669