The perpetual “is LENR even real” argument thread.

Quote from Alan Smith

Although I work from a much smaller store of knowledge and experience I suspect there is another factor beyond lack of interest. The 'fear of contagion' caused by the possibility od seeing something that cannot easily be explained away.

Is this something you have any evidence for? My experience is that all (good) scientists are very excited to find anomalous things that cannot be explained away. It is what they live for. I have never heard of this fear of contagion nor seen it prevent other anomalies from being of interest.

A good example would be the US guy who has now (twice) with different materials claimed much higher temperature superconductivity than anyone else has found. Both times there have been problems with replicability. The first time there was a problem with identified poor methodology. But lots of people are interested.

However, equally, good scientists will be skeptical and rigorous in their evaluation of evidence.

Here is an example

and here is a very fast (negative) replication:

Absence of near-ambient superconductivity in LuH$_{2\pm\text{x}}$N$_y$

Recently near-ambient superconductivity was claimed in N-doped lutetium hydride (ref. 1). This induces a worldwide fanaticism about the dream of room…

arxiv.org

the (scientific) jury is still out on this one but it looks more likely that this was a mistake.

Muted Response to New Claim of a Room-Temperature Superconductor

A research team says that they have made a material that conducts electricity without resistance at near-ambient conditions. The community has heard it before.

physics.aps.org

But even though there is as yet (and maybe never will be) no cheering I do not see fear of contagion.

Merely considerable interest tempered by the likelihood this is not what it is claimed to be.

Quote from Alan Smith

Yes, personal experience. I have invited quite a few people to my lab -or others to see a LENR system at first hand. The fear was very evident.

Yes. That is fair, because inviting somone to see a lab, or a system 1st-hand, is what you do to gain personal support, rather than what you do to publish groundbreaking scientific results.

That HTS guy published - and a lot of people were interested, at least one group (it will probably end up being many) tried best effort replication. And the conversation over that will go on until everyone (except possibly the guy making the claims) is sure what is going on.

As part of that - after the "I think I've got something" paper, people might visit labs. But mostly this is at the level of published (well - arxiv published) papers - multiple drafts of them.

Mostly, those confronted with a possible extraordinary result want to evaluate the evidence. How do you do that? From a detailed paper. Or, better, a detailed paper and then publish or even communicate informally what seem to be the gaps, followed by another paper with more data or better description. This can iterate, and should, if the results do not fit into any currently understood theory that would allow quantitative prediction, because in that case a lot more certainty is required from the experiments to go anywhere with them.

The point here is that you do not need people visiting your lab for them to evaluate results. In fact reflecting at length (e.g. several days or more) on a written paper is more likely to make progress than any such visit. Anyone who is interested and supportive can show that, informally or formally, by writing a critique that indicates what from their POV evidence is needed to make results stronger.

The time for lab visits is when parties are cooperating on an agreed research program - not when an extraordinary set of results is being evaluated to see whether cooperation is wanted.

Maybe the fear was because the people who came to see you knew that this was not quite a scientific visit, worried about non-science PR - and were not sure whether on balance it would be a good idea?

God knows - anyone aware of the Rossi saga and principles would want to be careful...

I am not saying that you are in any way not scrupulously honest: merely that no 3rd party can be sure, and lab visits (rather than evaluating written evidence in papers) imply a level of support which can be misconstrued.

THH

PS - bottom line - what I am suggesting - interest in data written up properly - does not suffer from this "contagion" issue.

Quote from Curbina

This is all the problem here, a majority of people working in science that are so comfortable with “what is known” that will refuse and react in denial towards any new information that challenges this paradigm. Storms has done everything by the book,

but as his results challenge the paradigm, he is shunned.

This conundrum is the entire topic of Thomas Kuhn’s “The Structure of Scientific Revolutions”. I strongly recommend reading it.

You can read it for free here:

https://www.lri.fr/~mbl/Stanfo…papers/Kuhn-SSR-2ndEd.pdf

Yes, alas when it comes to philosophy of science I view Kuhn, Feyerabend, Lakatos, Popper (who all had different views) as all using philosophy when proper quantitative maths (Baysian probability theory) exists and needs to be considered. Philosophy is what you do when you cannot have quantitative science.

Since Bayesian methods have been shown to provide a quantitative way to compare hypotheses - even quite different ones - all those philosophers who were not aware of this are out of date.

Anyway - we will disagree about the majority of good scientists (not sure about people).

Good scientists are curious, fascinated by anomalies. But equally if they are experienced in experimental work they will be aware of the way results can be misinterpreted.

Who are you - or anyone in the LENR filed - to say you are right on your more positive evaluation of the published evidence base for LENR compared with those non-LENR scientists who do not find it persuasive?

I am not saying you must assume they are correct. Just that the correct scientific objectivity comes from arguing specifics. That, for something without a testable theory, is always the specifics of an individual replicable experiment which shows an anomaly.

When I try to do that here - just at the level of pointing out that the write-ups could be more complete. Everyone tells me that I am biassed and evil, etc. etc.

With such a reception I am less surprised that most non-LENR scientists do not see enough in those papers to have more than a small interest in LENR as a phenomenon.

You I expect fall into the trap - which Kuhn would share - but no-one understanding how Bayesian probability theory quantifies and corrects intuitive ideas of likelihood would share - of thinking that the evidence for LENR needs the same level of rigor and certainty as the evidence form some other hypothesis which makes quantitative predictions and can be disproved if they do not pan out.

Not every scientist around knows the application of Bayesian methods more broadly to hypothesis testing in science. They do however explain why anonalous results are treated with caution until they are very certain, or a theory comes along that predicts them.

The difference between predicting anomalies, and allowing anomalies, is key.

For example: "miracles happen". Is a valid scientific hypothesis (with a slightly more careful definition of miracle) that explains every single anonmalous results in physics. What's more, it is supported by all those anomalous results, since conventional theories cannot explain them but the miracles hypothesis can.

But scientists do not much like the miracles hypothesis - even if any objective scientsist must allow the possibility that it is true (even if they think the probability of this is low).

Appendix

Here is a philosopher's commentary on David Stove, who criticised Popper.

But Stove, and Popper, and this commentary, all argue from an incorrect understanding of probability theory which has now more of less completely been abandoned by mathematicians (though not statisticians). Hypotheses such as "All ravens are black" can be handled quantitatively by a proper probabilistic theory.

http://www.med.mcgill.ca/epidemiology/hanley/bios601/GaussianModel/JaynesProbabilityTheory.pdf

pp61

This result will be quite disconcerting to some schools of thought about the ‘meaning of
probability’. Although it is generally recognized that logical implication is not the same as
physical causation, nevertheless there is a strong inclination to cling to the idea anyway, by
trying to interpret a probability P(A|B) as expressing some kind of partial causal influence of
B on A. This is evident not only in the aforementioned work of Penrose, but more strikingly
in the ‘propensity’ theory of probability expounded by the philosopher Karl Popper.1
It appears to us that such a relation as (3.40) would be quite inexplicable from a propensity
viewpoint, although the simple example (3.38) makes its logical necessity obvious. In any
event, the theory of logical inference that we are developing here differs fundamentally,
in outlook and in results, from the theory of physical causation envisaged by Penrose
and Popper. It is evident that logical inference can be applied in many problems where
assumptions of physical causation would not make sense.
This does not mean that we are forbidden to introduce the notion of ‘propensity’ or
physical causation; the point is rather that logical inference is applicable and useful whether
or not a propensity exists. If such a notion (i.e. that some such propensity exists) is formulated
as a well-defined hypothesis, then our form of probability theory can analyze its implications.
We shall do this in Section 3.10 below. Also, we can test that hypothesis against alternatives

in the light of the evidence, just as we can test any well-defined hypothesis. Indeed, one of
the most common and important applications of probability theory is to decide whether there
is evidence for a causal influence: is a new medicine more effective, or a new engineering
design more reliable? Does a new anticrime law reduce the incidence of crime? Our study
of hypothesis testing starts in Chapter 4.

Bayesian hypothesis testing in medicine( and more generally data science) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8871131/

Popper's arguments are sound if logical inference must be binary (true and false only values allowed).

Bayesian techniques extend inference in a provably unique way to a wider range of belief values (that we might think of as probability of something being true").

Bayes and Popper

There has been little direct confrontation between Bayesians and Popperians. No doubt Popper views the Bayesian approach as a species of inductivism and…

link.springer.com

A good discussion of how Bayesian probability theory supercedes Popper's arguments, by providing a rigorous mathematical way in which belief update, to be consistent, must incorporate a variant of "proof by induction". In spite of this the Bayesian approach seems rigorous in the same sense that Popper's is also rigorous: but of course Popper did not have the tools now available:

There has been little direct confrontation between Bayesians and Popperians. No doubt Popper views the Bayesian approach as a species of inductivism and therefore not to be countenanced. But while Bayesians evaluate hypotheses primarily in terms of their probability, their position rests on no obscure ‘principle of induction’, but on Bayes’ rule. For a Bayesian, ‘learning from experience’ can only mean modifying prior probabilities by conditionalizing on observed experimental outcomes. On the other hand, the implications of Bayes’ rule for scientific method often have a distinctly Popperian ring, fostering the suspicion that many of the differences between the two schools are, at worst, merely verbal, and, at best, but differences of emphasis. We shall have something to say about this, but, in a more positive vein, we wish to explore here the extent to which Bayesian and Popperian viewpoints can be reconciled.

Quote from THHuxleynew

Is this something you have any evidence for? My experience is that all (good) scientists are very excited to find anomalous things that cannot be explained away. It is what they live for. I have never heard of this fear of contagion nor seen it prevent other anomalies from being of interest.

Quote from JedRothwell

That is a myth. The history of science shows that most scientists are conservative, and strongly opposed to new ideas and anomalies. For examples, see: http://amasci.com/weird/vindac.html#j38. Look that the history of laser and maser Townes' autobiography. Scientists only accept anomalies when two conditions are met:

When the anomaly confirms pre-existing ideas, or the scientist's own ideas.

When the anomaly does not suggest a new line of research that threatens existing funding. Money is by far the most important consideration to most scientists.

Quoting Townes:

One day after we had been at it for about two years, Rabi and Kusch, the former and current chairmen of the department—both of them Nobel laureates for work with atomic and molecular beams, and both with a lot of weight behind their opinions—came into my office and sat down. They were worried. Their research depended on support from the same source as did mine. "Look," they said, "you should stop the work you are doing. It isn't going to work. You know it's not going to work. We know it's not going to work. You're wasting money. Just stop!"

The problem was that I was still an outsider to the field of molecular beams, as they saw it. . . . I simply told them that I thought it had a reasonable chance and that I would continue. I was then indeed thankful that I had come to Columbia with tenure. (p. 65)

Opposition continued for several years after they demonstrated the masor:

Before—and even after—the maser worked, our description of its performance met with disbelief from highly respected physicists, even though no new physical principles were really involved. Their objections went much deeper than those that had led Rabi and Kusch to try to kill the project in its cradle . . .

Llewelyn H. Thomas, a noted Columbia theorist, told me that the maser flatly could not, due to basic physics principles, provide a pure frequency with the performance I predicted. So certain was he that he more or less refused to listen to my explanations. After it did work, he just stopped talking to me. . . .

. . . I visited Denmark and saw Niels Bohr . . . I described the maser and its performance. "But that is not possible," he exclaimed. I assured him it was. Similarly, at a cocktail party in Princeton, New Jersey, the Hungarian mathematician John von Neumann asked what I was working on. After I told him about the maser and the purity of its frequency, he declared, "That can't be right!" But it was, I replied, and told him it was already demonstrated. (p. 69)

There are descriptions of similar opposition in just about every honest scientific biography. Opposition is universal and it never has a rational basis. It is always emotional, just as your opposition to cold fusion is.

Display More

Jed's case and mine are not (as Jed claims) in contradiction. I fully admit that not all scientists are "good". Indeed it may well be true that by some definitions not even "most" scientists are good.

New ideas of course face resistance - as they should - because there are 1000 new ideas for everyone that is actually more correct, in the end, than existing ideas.

And that resistance is not wholly rational. scientists are human, and therefore not totally rational.

You can see this for example in Jed's statements on this thread where he has very clear yes or no answers to a lot of questions - Popper would appreciate this (although not all the yesses, because Popper would see any hypothesis that could not be refuted as problematic).

But when Jed talks about sociology, people, etc, binary answers are almost never right, and almost always hurtful - because they generalise from the variety of individuals to some statement about the groups they belong to. And in Jed's case the doubly hurtful comments suppose these groups (here scientists) are 100% without exception - whatever Jed thinks they are.

This is wrong in fact - unevidenced - and morally wrong.

Scientists (good ones) are defined by their curiosity. That means they are open to new ideas. It does not however mean that they will agree with Jed about the absolute - or even on balance of probabilities likely cause of the phenomena documented in the various positive LENR experiments.

Some will. Some will not. Some won't think much about it because the modern evidence base is relatively poor, and the "old" evidence base was highly contentious. They will reckon that if LENR is real there will emerge some modern replicable experiment that shows it, and till then it is not worth the effort to follow.

From Jed's point of view two of those three groups are stupid or prejudiced.

A more objective person would not know (without a lot of details) who was stupid or prejudiced, but would reckon Jed's binary classification of everything was likely to be incorrect - many things are unknown. Such an unbiassed person would reckon that anyone having such definite and fixed ideas about science would be at great danger of being incorrect.

On Townes etc: and masers

The theoretical principles governing the operation of a maser were first described by Joseph Weber of the University of Maryland, College Park at the Electron Tube Research Conference in June 1952 in Ottawa,^[2] with a summary published in the June 1953 Transactions of the Institute of Radio Engineers Professional Group on Electron Devices,^[3] and simultaneously by Nikolay Basov and Alexander Prokhorov from Lebedev Institute of Physics, at an All-Union Conference on Radio-Spectroscopy held by the USSR Academy of Sciences in May 1952, subsequently published in October 1954.

Independently, Charles Hard Townes, James P. Gordon, and H. J. Zeiger built the first ammonia maser at Columbia University in 1953. This device used stimulated emission in a stream of energized ammonia molecules to produce amplification of microwaves at a frequency of about 24.0 gigahertz.^[4] Townes later worked with Arthur L. Schawlow to describe the principle of the optical maser, or laser,^[5] of which Theodore H. Maiman created the first working model in 1960.

For their research in the field of stimulated emission, Townes, Basov and Prokhorov were awarded the Nobel Prize in Physics in 1964.^[6]

So we have (masers):

• Theory worked out 1952
• First working device 1953
• Nobel Prize 1964

Whereas for LENR:

• Theory worked out (not yet)
• First working device (contentious)
• Not yet accepted hence no Nobel prize yet

12 years seems perfectly reasonable (actually very short) to go from first hypothesesed theory to Nobel Prize.

1 year seems pretty short to go from hypothesis to working model.

THH

Here is a comment on how Bayes updates Popper wrt hypothesis testing string theory.

Bayes does not remove the idea of falsification. it changes from a Jed-loved "yes or no" thing, to a probabilistic - "if a theory cannot sharply be falsified then its posterior probability is much lower than if it can"

Gathering of philosophers and physicists unaware of modern reconciliation of Bayes and Popper | Statistical Modeling, Causal Inference, and Social Science

But it seems that nobody in the room was aware of the “Bayesian Data Analysis” view in which Bayesian models can and should be checked. We have to do more communicating. Naive Popperianism is bad, but so is naive anti-Popperianism. Lakatos would be spinning in his grave.

And here from wikipedia is a simple (lots of others available) description of Lindley's paradox - often taken as problematic for a Baysian hypothesis testing view, but in fact not at all (and would not be seen as that by a modern Bayesian):

https://en.wikipedia.org/wiki/…s_paradox#:~:text=Lindley's%20paradox%20is%20a%20counterintuitive,choices%20of%20the%20prior%20distribution.

The apparent disagreement between the two approaches is caused by a combination of factors. First, the frequentist approach above tests �0 without reference to �1. The Bayesian approach evaluates �0 as an alternative to �1, and finds the first to be in better agreement with the observations. This is because the latter hypothesis is much more diffuse, as � can be anywhere in [0,1], which results in it having a very low posterior probability. To understand why, it is helpful to consider the two hypotheses as generators of the observations:

• Under �0, we choose �≈0.500, and ask how likely it is to see 49,581 boys in 98,451 births.
• Under �1, we choose � randomly from anywhere within 0 to 1, and ask the same question.

Most of the possible values for � under �1 are very poorly supported by the observations. In essence, the apparent disagreement between the methods is not a disagreement at all, but rather two different statements about how the hypotheses relate to the data:

• The frequentist finds that �0 is a poor explanation for the observation.
• The Bayesian finds that �0 is a far better explanation for the observation than �1.

The ratio of the sex of newborns is improbably 50/50 male/female, according to the frequentist test. Yet 50/50 is a better approximation than most, but not all, other ratios. The hypothesis �≈0.504 would have fit the observation much better than almost all other ratios, including �≈0.500.

For example, this choice of hypotheses and prior probabilities implies the statement: "if � > 0.49 and � < 0.51, then the prior probability of � being exactly 0.5 is 0.50/0.51 ≈ 98%." Given such a strong preference for �=0.5, it is easy to see why the Bayesian approach favors �0 in the face of �≈0.5036, even though the observed value of � lies 2.28� away from 0.5. The deviation of over 2 sigma from �0 is considered significant in the frequentist approach, but its significance is overruled by the prior in the Bayesian approach.

Looking at it another way, we can see that the prior distribution is essentially flat with a delta function at �=0.5. Clearly this is dubious. In fact if you were to picture real numbers as being continuous, then it would be more logical to assume that it would impossible for any given number to be exactly the parameter value, i.e., we should assume �(�=0.5)=0.

A more realistic distribution for � in the alternative hypothesis produces a less surprising result for the posterior of �0. For example, if we replace �1 with �2:�=�, i.e., the maximum likelihood estimate for �, the posterior probability of �0 would be only 0.07 compared to 0.93 for �2 (Of course, one cannot actually use the MLE as part of a prior distribution).

Quote from JedRothwell

Note that I said, "experimental error." Finding theoretical errors does not count. This is an experimental observation. The only way to disprove it is to find errors in the instruments or techniques. You can never disprove an experimental observation by pointing to a theory, although many scientists such as 2004 reviewers tried to do that.

Jed. This is a fundamental misconception. Single experimental observations are not "proof". They are evidence the strength of which varies. They are, if you like, the Bayesian more sophisticated non-binary version of proof.

You therefore misunderstand what I am saying. I am not saying that "theory disproves" any experiment. Merely that when experiments are used to update the relative probability of hypotheses, other hypotheses, and the evidence for them, must be taken into account.

So your whole argument is based on a misconception of the nature of "proof" in science. And it is why continuously on this thread and elsewhere you express the view that I am either stupid or evil - when in fact I am neither of those.

To be more precise, and less general, we could look at an experiment (perhaps Ed's) and consider how all this works out. I could explain why the evidence in it, even if replicated by multiple labs, is not as strong as many would need in order to see it as making LENR likely. And, I could explain how it could be strengthened. I have already done this. You have not agreed, because, for you, 100% cannot be strengthened I guess.

Quote from JedRothwell

f you have good reasons to doubt the experimental results, what are they? Where are they published? Show me a paper laying out specific, quantifiable, falsifiable reasons why every single major positive experiment is wrong.

Just to reiterate. I know this is your requirement for, in your view, LENR to be unproven. So if it cannot be done (obviously it cannot be done) LENR is proven.

It is however not what anyone else would think.

So lets not argue about other stuff. if you maintain this, you and mots of the scientific community will not agree.

And that lack of agreement would extend to anything similar to LENR which rests on a diverse collection of well-attested anomalies. The phenomenon of UFO sightings provides a good counter-example.

Let is stick with your sentence above which I believe gets to the heart of the different views we have. I could say a lot more about why no scientist would in general agree with that proposition. Particularly your placement of the word falsifiable.

Quote from JedRothwell

So, the notion that I think scientists are "100% without exception" opposed to cold fusion is bonkers. It is preposterous. ~5,000 scientists have stuck their necks out and published papers that show cold fusion evidence. Many of them suffered because of that. They were blackballed, fired, threatened with deportation; they faced trumped up charges of fraud, and their reputations were shredded in the mass media. So, there is such a thing as the noble scientist who sacrifices for the truth. I know most of them by name.

Quote from THHuxleynew

But when Jed talks about sociology, people, etc, binary answers are almost never right, and almost always hurtful - because they generalise from the variety of individuals to some statement about the groups they belong to. And in Jed's case the doubly hurtful comments suppose these groups (here scientists) are 100% without exception - whatever Jed thinks they are.

This is wrong in fact - unevidenced - and morally wrong.

Jed's criticism of me is technically correct. The "groups" I give a paragraph later in the same post show what I mean:

Quote from THHuxleynew

Scientists (good ones) are defined by their curiosity. That means they are open to new ideas. It does not however mean that they will agree with Jed about the absolute - or even on balance of probabilities likely cause of the phenomena documented in the various positive LENR experiments.

Some will. Some will not. Some won't think much about it because the modern evidence base is relatively poor, and the "old" evidence base was highly contentious. They will reckon that if LENR is real there will emerge some modern replicable experiment that shows it, and till then it is not worth the effort to follow.

From Jed's point of view two of those three groups are stupid or prejudiced.

A more objective person would not know (without a lot of details) who was stupid or prejudiced, but would reckon Jed's binary classification of everything was likely to be incorrect - many things are unknown. Such an unbiassed person would reckon that anyone having such definite and fixed ideas about science would be at great danger of being incorrect.

So I object to Jed's godlike assumptions that (a) he understands why other people believe things and (b) when a lot of competent people have looked at something, and they reach different views, that a small percentage who are strongly on one side of this thing are necessarily correct.

Quote from Curbina

You are indirectly suggesting that Storms did not properly present his results, to which I disagree, and probaly him, too.

Not in the way you mean. Again, you like Jed are using binary reasoning. If I have a new idea, I can present it in many ways, with more or less evidence. The stronger I present it, the more people will agree with me.

I am merely saying that Ed's paper could be strengthened, and that in its current from it is not likely to convince many non-LENR scientists (well - Ed has said that and more, so it is hardly contentious).

Saying that a paper could be strengthened is not saying that it is improper. Equally, there is no requirement for somone reading a paper to find its thesis persuasive.

I was maybe using "proper" in the sense of "in a way that might convince others". I think in context it was clear enough, but it was a poor choice of word - most of the time here I have not used that word about Ed's experimental write-up. So I apologise to Ed if I implied that this time (can't remember exactly how I used it).

Quote from JedRothwell

And, as Townes said, if he had not had tenure, the maser would have been "killed in the cradle" by Rabi and Kusch.

Right, but no-one else said that, and it is extremely unlikely. It took Townes only 1 year from publication of the theory to get a working prototype. Other people looking at that same theory would also have found working examples. Different ones.

Townes may like to think of himself as a once in a Millenium genius inventor able to do things no-one else can - but I doubt that.

We know it is quite plausible for a working maser to lead to the underlying theory being accepted - because in spite of Townes's complaints that happened in this case!

Compare this with Cold Fusion. People go on constructing cold fusion experiemnts and obtaining new data. That is like thousands of new working masers. Except that in the case case of cold fusion (now LENR) there is still disagreement about whether any of those experiments in particular, or the corpus as a whole - shows that low energy nuclear reactions of the type claimed exist.

Quote from JedRothwell

This is the heart of the scientific method. All discoveries start with experiments. Only an experiment can prove or disprove a claim. When theory conflicts with experiment, or when theory cannot explain an experiment, theory always loses; experiments always win. This is the ironclad basis of science defined by Francis Bacon in 1620:

Nature is only to be commanded by obeying her.

If the scientific community does not believe that experiments are the sole source of authority, and that replicated experiments are the only standard of truth, then the scientific community is no longer scientific. It has become a debased religion.

Quote from JedRothwell

If you have good reasons to doubt the experimental results, what are they? Where are they published? Show me a paper laying out specific, quantifiable, falsifiable reasons why every single major positive experiment is wrong.

So - sticking to the topic (difficult sometimes!).

Experiments are indeed the sole source of evidence. Authority? I think I'd rate maths above experiment. Where something is mathematically proven the causality is 100% - all experiments are uncertain - some more so than others. Both both are needed to get anywhere in physics.

So science is about linking together many experiments - finding the common themes - calling them theories. Showing (maths) that the different results are consistent.

We do not really disagree about the primacy of experiments - they are what tell us what the real world is.

Where we disagree is that when we have a set of say 1,000 experiments (from jed's 5000 papers, say at least 1 in 5 is experimental).

Jed states that every single one of these experiments must have specific, quantifiable, falsifiable reasons why they are wrong.

Or else - just one of them not obviously wrong - the claim is proven.

That is untrue.

It is true that experiments are all the evidence we have. But experiments are prone to human and other errors. It is the nature of experimental error that we cannot always detect it, even when afterwards we are sure it was in error. That is why scientists take great care to try and make results robust - and even then no-one believes an unusual experiment until it is (a) robust (no room for multiple interpretations) and (b) replicated.

Jed thinks the burden of proof is on skeptics to prove an experiment could be wrong.

In fact the burden is on experimenters, to prove their results are correct.

It is a very large difference.

It is true that a single, clear, replicable experiment (for example a way to get excess heat clearly beyond possible errors and chemical bounds) would be enough to prove LENR (or at least make it look more likely than other possibilities).

Clarity - means a bomb-proof write-up, and results that are clear.

Run once - it provides interesting results, and clear methodology, enough for many to want to replicate

Replicated - we have most accepting LENR.

It is what team google hoped they would be able to find.

It is what we here have asked for.

I'm still waiting for it. I have not given up. But it makes me depressed when people in the field argue its not needed.

I point out - if Ed's experiment is replicable, and his results as Jed says - then this would be a good example of that. I think.

I don't understand - on the one hand everyone says results are clear, and some (e.g. Ed) say they are replicable.

On the other - no-one seems to advocate from the set of such experiments one that google etc could run with. there is funding now. there are people who given a decent prospect will juts replicate it for fun!

Replicability - the Big Excuse

I have heard here, and I accept, the argument that it is difficult to control NAE density. That is why an LENR experiment must have a methodology where multiple samples (cathodes, whatever) are tested, and results recorded for every one. The expected distribution (Ed's graph) can be checked but in any case where the experiment calls for 10 or so runs we can presume if the two modern results (Ed, Staker) we have looked at here are correct, that we will gst positive results from some suitable protocol eliminating the ones that do not pass it.

Such an experiment, with protocol, with all details - would be a replicable LENR experiment - even though the individual samples were all different.

That also means that without characterising the distribution of results - single "good" runs mean very little, because they do not speak to replicability.

I don't understand why modern experiments (such as ed, Staker) provide a single result, where than a set of results for multiple cathodes that would show in what way the experiment could be replicated.

THH