Is it ever reasonable to conclude that the laws of nature have been violated based solely on testimony?

Question

This question assumes the existence of laws of nature. As I recognize that this premise may be controversial to some extent, frame-challenging answers are welcome. However, please note that there are already questions entertaining this possibility: Are there laws of Nature?, Is it possible that the laws of nature don't exist?.

With that said, let us suppose that laws of nature do exist and that a scientific consensus has been reached regarding certain heavily tested ones. Now, consider a scenario where person P is well aware of these established facts. Would it ever be reasonable for P to conclude that one or more of these rigorously tested laws of nature were violated solely on the basis of testimonies?

To clarify, P does not have direct access to technology, measurement devices, camera recordings, or other forms of physical evidence related to the alleged violation. The only evidence available to P in this thought experiment consists of testimonies—writings, speech, or reports produced by other individuals. These testimonies may vary in nature; for instance, they could originate from a single individual or from multiple people claiming to have witnessed the event. The witnesses might be alive and available for interviews, or they might no longer be alive, with their testimonies preserved through a tradition that values their accounts (as in the case of alleged "miracles" deeply valued by certain religious traditions).

The question is: under what conditions, if any, could testimonies alone reasonably lead P to conclude that a violation of the laws of nature occurred as a matter of fact? Or is this the type of conclusion that testimonies, regardless of their nature or number, could never justify?

Answer 1

There isn't a fixed set of propositions known as "the laws of nature." There have been many variants of these laws proposed over time, and often competing versions at any given time.

For example, Aristotelian physics is one set of "laws of nature."

Most individuals do not have direct experience with the relevant evidence to prefer one set of laws to another. They have limited access to equipment. Their time and attention are limited. Therefore, most individuals' relationship with the evidence that would cut against one set of laws of nature and support another is necessarily indirect. They rely on essentially the testimony of others.

Answer 2

It depends what you mean by testimony and violations of the laws of nature. If a person answers under oath in court that they have seen an apple fall upwards from the ground and reunite itself with a tree, then no, you would not believe it. If a physicists answers under oath in court that experiments at CERN, which she has supervised, have conclusively proved that our previous understanding of some aspect of the behaviour of muons is incorrect, then you might believe it.

The overall issue here is plausibility. Take a well established and accepted 'law' such as gravity, the effects of which we directly experience. We have theories that account for it with remarkable success. It is the cause of the fact that the Earth has gone around the Sun for eons. If a witness, or a thousand of them, swear it has been violated, you would want more than their hearsay to persuade you.

Answer 3

Anecdote is not the singular of data.

If there is no evidence, it isn't data, or a datum, yet. It may be suggestive, it may be worth investigating, but you don't know if the perceptions were accurate, you don't know if they were reported accurately, you don't know what wasn't perceived or reported that would put this into context.

There's nothing here yet that rises above "that's interesting, I wonder what really happened." It can be a valuable lead to further investigation, but no conclusions can be drawn yet.

If I describe a magic trick that I haven't figured out yet, it's premature to say more than that natural laws appeared to have been broken. Ditto for any other isolated observation. Seeking an explanation consistent with the model is usually a better place to start than immediately seeking outside the model, if the model has proven robust so far.

Answer 4

I mean it's almost certain that his has happened at some point. Like even if you allow measurements with machines, it's quite a recent development that these machines record something themselves, so practically the human scientist would make the measurement and note down their observations. So what is posted in a peer-reviewed journal might still have been barely more than testimony.

Sure in some cases you have a physical object or an experiment that other's can rebuilt or re-examine, but idk for all intents and purposes stuff like the LHC at CERN can't really be rebuilt so their data are also technically testimony.

Or idk picture ancient astrophysicist looking into space with a telescope and writing down their findings, even if they use a device that doesn't mean that anybody can reproduce that effect readily if they idk witnessed a meteor, Venus transit, solar or lunar eclipse and so on, which you would need to be there to witness.

So yes, that certainly has happened and can happen where the validity of the claim somewhat rests upon the credibility of the source. Thought rather than leading to a conclusion it might rather be the starting point of a journey to find evidence that is more reliable then that. So if an effect is witnessed it's likely that other measurement devices have witnessed it as well and that there is more data that can be gathered and evaluated.

So a one off might be discarded for know but for example if you read ancient texts detailing events that you now know could describe a real event, you're more likely to believe that testimony for example.

Or a highly ambitious scientists looking for that sweet Nobel prize might rush to conclusion and try to proof or disproof the testimony by assuming it's correct and testing it's consequences or whatnot.

Answer 5

Your question is in fact a part of a bigger question - should we at all trust a testimony. Not even about broken laws of nature but much smaller things. We know that people hallucinate and that people lie. We also know people misinterpret things just because they want something to be true even if they are mentally fine.

Any testimony is just some information you were given. It's not a fact, and there is big difference between information and facts.

And the more extraordinary a claim is, the more evidence we require. We also need to understand the context and reliability of information source. For instance if your Dad says it's raining outside and he was not noticed as randomly lying, there is no reason to disbelieve. But if a member of some cult tells you something from his "holy book" written thousands of years ago, or even a kid says he didn't take a candy when he was caught before with taking, well use common sense here.

So the answer is even broader - we can't rely on testimonials not even in extreme cases but in many other cases as well.

Answer 6

Peer-reviewed journal articles are testimony. They're the testimony of the journal editor, who trusts the testimony of the review board, who decided to trust the testimony of the authors, who trusted the testimony of the research assistants and laboratory technicians, who trusted the testimony of the manufacturer of the equipment, who trusted the testimony of the technicians and machinists who tested and certified each part.

Later, other journal editors trust the testimony of the other review boards, who trusted the testimony of other authors, who trusted the testimony of other assistants and lab-techs and manufacturers and techs and machinists, while attempting to reproduce or disconfirm the experimental results.

A few years go by. Other authors who trust the testimony of the journal editors who published the initial and reproducing experimentalists publish testimonies of their own, establishing whether or not the initial and reproducing experimentalists could have been mistaken... on the basis of hundreds of years of testimony from other scientists... which was conveyed to them by the testimony of professors and publishers.

If other scientists trust their testimony, it is added to the pile.

From these latest testimonies, testable predictions present themselves. Still more experimentalists get to work, trusting the same kind of chains of testimony as the first. Experimenting and trusting, publishing and trusting, reproducing and trusting, publishing and trusting.

Every once in a while, someone climbs this heaping mountain of testimony and nothing but testimony, and gets a little gold circle with Alfred Nobel's face on it.

At least... so I've heard.

Answer 7

The answer would depend in part on the kind of case in question. If someone claims to me to have found a heretofore impossibly large uranium sphere (presumably by deep-space telescopy), I would want to first check how they figured out that the sphere was made of uranium. If they used some standard process for evaluating this, and I applied that process and came to the same conclusion, then I might want to revise that method itself, rather than my theory about limits on the size of uranium spheres.

Alternatively, I could conclude that the sphere is embedded in a deviant spacetime region, or is subject to quantum fluctuations that have happened to line up in the form of a metastable unduly massive uranium mass. At the end of inquiry, I could just give up my theory that uranium spheres have a size limit in general. On no occasion would I have grounds for thinking that a force from outside the universe was intervening to stabilize the ultramassive sphere; any force detected in the process of such stabilization would be detected in the universe, one way or another, so the idea of this stabilization being a "violation" of a law of nature would be a non-starter.

As for, say, some human coming back to life, I would be much more disposed to accept the possibility that it had happened, if someone sincerely and non-maniacally told me of this. If the person telling me this is a religious maniac who worships the person who supposedly came back to life, well, it would be safer to conclude that they're either deluded or lying about the situation. Even if it weren't fair to think this, it would be another leap entirely from, "This human came back to life," to, "One specific paranormal entity resurrected this being." Maybe a demon resurrected that person, or possessed their corpse, and had them utter the words, "And I will one day return," so as to trick countless people over the ensuing millennia into wasting their emotional energy hoping that theirs would be the time of the return.

Or maybe an angel, or a God, resurrected them. How would I be able to tell? If I hear it secondhand that the resurrected man attributes his resurrection to God, despite this being impossible to perceive (especially for the dead man, while he is dead), I am inversely (exponentially, probably) less justified in accepting that specific statement over the demon-possessed-corpse hypothesis, etc.

Answer 8

In the question, we first decide something exists and then choose to break its definition.

If Laws of Nature exist, then what we are saying is that there are rules to how things happen and whether things can happen. There are methods and/or formulae that we can consistently, reliably and repeatedly use to predict how something happened, is happening, or will happen.

Let's now say that there is some event that was observed to not obey these Laws. We are being asked to decide:

1. Were the unbreakable (by definition) Laws of Nature actually broken?
2. Do the Laws remain unbroken and we simply misunderstood them in the context of this type of event?
3. Do the Laws remain unbroken and the observation of the event was mistaken?

The one answer we cannot give, given the setup of the question, is 1.

We can't create a definition for the thought experiment and then suppose that definition is incorrect. What must be true in this question, is that either our interpretation of the Law(s) is mistaken (and we would need to find out the more general rule), or our evidence is faulty (regardless of the quantity or quality of it).

If the question is aimed at what level would we move from deciding 2 is more likely than 3, then there are already standards for this type of supposition. Science goes through this process often and the higher the repeatability of the Law, the more evidence required to assume 2 over 3.

What this doesn't do is allow us to sneak in a being that can either create or break those Laws, because that is actually a 1 answer.

Answer 9

The laws of nature are indirectly based on testimony and are based upon observations of outcomes that you only find out indirectly.

Thus, you’re actually comparing one form of testimony vs. another. And given that the laws of nature have stood tests through time, it makes sense to trust the testimony in regards to the behaviour of those laws over the testimony of experiences that break them.

The exception would be if the testimony of law breaking events becomes widespread, repeatable, and regularly occurring where outcomes continue to match new predictions without exception (i.e. the modern scientific experiment). But then that would just mean we’ve came across a new law and that would then be included in the laws of nature.