> So far as I can tell, there is no such authority for facts.
Huh? You just described it: reality is the authority.
> I think this situation is simpler if these all count as separate facts.
No, they are different theoretical descriptions of the facts. The facts are the particular experimental apparatus Michelson and Morley used and the particular numbers they recorded when they made their measurements.
> Do "the facts of gravity" include the predictions made by the theory of gravity?
Of course not. Predictions aren't facts. They're things that you compare with facts in order to test theoretical models.
> Speaking of Einstein, before he pubished his paper on special relativity, wasn't it a fact that the velocity of a bullet fired from a train (relative to the track) is the sum of the train's velocity and the bullet's velocity?
No. As far as I know nobody at that time had made accurate enough measurements of the velocity of anything to be able to test the Newtonian prediction vs. the relativity prediction. But that doesn't mean the Newtonian prediction was correct before relativity was discovered. It certainly doesn't mean that something changed about the world described by physical models when physicists switched from using the Newtonian model to using the relativity model. All that changed was the model: the world the model was describing didn't change.
> Given that we can't know which facts might later need to be relativized
Facts never need to be relativized. Relativizing is something that happens when you switch theoretical models. You seem to be consistently confusing facts with models.
Raw facts are meaningless without models, theories, or even, ideologies. Without meaning you can't do anything useful with it, or anything "that works." You always have to have a theory in order to predict the future based on facts and then have experiments to verify the predictions. No meaningful experiments can exist without a establishing theory.
All facts are alternative facts in the sense that facts are only accessible and made meaningful through theories and there is no one true theory.
> Raw facts are meaningless without models, theories, or even, ideologies.
No, they're not. They just don't allow you to make predictions without a model. But you don't need to agree on a theoretical model to agree that, for example, Michelson and Morley had a particular experimental setup and procedure and recorded particular numbers when they ran the experiment.
> You always have to have a theory in order to predict the future based on facts and then have experiments to verify the predictions.
A "theory" in this sense can be just about anything, from a simple guess to a detailed mathematical model with lots of ramifications. So I don't see that much is gained by making this observation.
Allow me to clarify: facts are meaningless in the literal sense that facts are devoid of meaning without interpretation. When you say facts be confirmed by experiments, the very confirmation already presupposes the existence of criteria by which you prove or disprove facts. Those criteria are precisely what constitutes theories.
Now back to the original question, it's true you can establish facts through experiments, but since facts are always mediated by theories, it's not very useful to discuss how one deals with purely raw facts and more useful to discuss how one approaches truth from mediated facts. I think OP meant facts in this sense.
I didn't say facts are confirmed by experiments. Facts are the experimental setup and results: what was done and what happened. Nothing has to be "confirmed"; the facts just are.
Theoretical models can be confirmed by experiments, but models are not facts. You appear to have the same confusion about this that MatrixMan has.
Suppose you have two theories, and you run an experiment whose results produce two assertions, each couched in the language of their respective theories, (I'd call these assertions facts). You say that reality is the authority on whether the two assertions represent the same underlying fact or different underlying facts, but it's not like there's some oracle that I can consult in the matter. So how does using reality as an authority help me individuate these facts?
And if it doesn't, then of what use is the idea that facts don't change when theories change, (except perhaps as a political too towards the discrediting of new ideas).
> The facts are the particular experimental apparatus Michelson and Morley used and the particular numbers they recorded when they made their measurements.
So you've got a bunch of mirrors and a pattern of light, and then later in the day you have the same pattern of light. That's not facts, that's just phenomena. Facts are supposed to convince me of something, and they work because they take advantage of my having already accepted the theory that supports them.
I have to suspect that anybody who presents me "facts" that don't have this quality has something to gain by distracting from their unwillingness to argue for the superiority of the their theory.
(Debate antics aside, thanks for the conversation. I've been having fun.)
Also, I think you missed my point about the trains. The statement that nobody bothered relativize until Einstein was equivalent to this:
> A train moves at 10 mph and a bullet fired from a window in the direction of the train's motion moves at 10000 mph. The speed of the bullet as measured by an observer standing beside the track is 10010 mph.
It was initially poor style to relativize this fact because its supporting theory had no competitors. It was only when Einstein proposed that simple addition was insufficient that relativizing it made sense.
As for models, I've done a bit of model theory in geometry, and I've also studied models in physics (such as the Schrodinger model of the atom), but these don't resemble a pile of mirrors and a pattern of light--they're just the sort of thing theories are made of. If I'm missing some other sort of model, please let me know.
As near as I can tell, a fact is some correspondence between a model (which lives in a theory) and some phenomenon (which lives in reality). As such, facts without corresponding theories are nonsensical.
> Suppose you have two theories, and you run an experiment whose results produce two assertions, each couched in the language of their respective theories, (I'd call these assertions facts).
I wouldn't. An experiment will produce one set of "assertions", about what was done and what happened. Those are facts.
Different theories might add different "assertions" on top of that, but that doesn't change what was done and what happened in the experiment, and the theoretical "assertions" aren't facts. They're theories.
> So you've got a bunch of mirrors and a pattern of light, and then later in the day you have the same pattern of light. That's not facts, that's just phenomena.
No, they're facts. What you are calling "phenomena" are facts. At least, I call them facts.
> a fact is some correspondence between a model (which lives in a theory) and some phenomenon (which lives in reality).
No, I would not call that a fact. The phenomenon is the fact. The correspondence is a successful test of the model.
To go all the way back to what I quoted from you to start this subthread: the question you asked was "which facts to use?" And my answer was "the ones that are confirmed by experiments". Now that I see what you meant by "facts", I think the way you should have phrased your question is simply "which theoretical models to use?" And the answer is still "the ones that are confirmed by experiments", but with the added note that "confirmed by experiments" means "confirmed as correctly describing the facts".
If we water down the label "fact" so that it represents mere sense data--just the perception of an interferometer's output that happens to not vary based on time or orientation--then what good are facts anyway?
For them to be useful, they need to be in some way convincing, don't they?
> If we water down the label "fact" so that it represents mere sense data
Not just sense data, but the records of sense data: the records of what was done and what happened during the experiment. Those records make it possible for other people besides the experimenters to know what was done and what happened. Without that the facts would be useless because only the experimenters themselves would ever know them.
> It was initially poor style to relativize this fact
You're confusing facts with models again. The statement you give is not a fact: it's a prediction based on a theoretical model. Einstein proposed a different theoretical model, which we now know makes more accurate predictions. But no facts changed during this process. Thinking that switching models changed the facts is like thinking that switching from a less accurate map of a city to a more accurate map changes the layout of the streets in the actual city.
So the city has a layout, independent of me, and that layout is, as you say, a fact.
In order to recognize a particular patch of ground as a street, and not--say--an oddly shaped rock, I must recognize that it goes somewhere, so even though the facts are authoritative, I can only view them through their correspondence with a model (which is in this case, a map--either on paper or in my mind). Building that correspondence constitutes a successful test of the model.
So my belief about the best route to take to work is determined by my model which is more or less correct based on its correspondence with these facts.
Am I using the words your way now?
And if so, would you agree that "points A and B are 1 mile apart" is an example of a fact--or at least is a model that seems likely to be confirmed by contact with a fact?
> I believe that the facts you're referring to must be on page 685 of this document
The azimuths and wavelengths at the top of that page are facts, yes. Also, the description on the preceding pages of how the experiment was conducted contains plenty of facts.
The rest of p. 685 after the azimuths and wavelengths is all interpretation using a theoretical model.
I would call the first half of the page facts of optics, and the second half of the page facts of aether flow theory. The former are only true if you already accept some theory with sufficient machinery to map the device readout to a number corresponding with the concept 'wavelength'. And the latter are only true if you accept both optics and the aether theory that was popular at the time.
I see that you're trying to get something to stand on, some objective and unchanging reality to hold constant while new theories come and go--and I don't deny you its existence, but I don't see how statements about it can be encoded without making reference the theories in use by the experimenter.
Suppose that later these results need to be translated into the language of a new theory in order for them to be communicated with future generations (who may not know what a wavelength is). Once translated, would they not be new facts, made of new words and new numbers and true in a new theory?
There are alternatives, but they seem awfully problematic to me.
> I would call the first half of the page facts of optics, and the second half of the page facts of aether flow theory.
There is no such thing as "facts of a theory". Theories are not facts. I don't understand why you persist in talking about these two different things as if they were the same.
> I don't see how statements about it can be encoded without making reference the theories in use by the experimenter.
Huh? The portions I already referred to as giving facts do just that. You don't have to have any particular theory to understand what those portions are describing. What's the problem?
> Suppose that later these results need to be translated into the language of a new theory in order for them to be communicated with future generations (who may not know what a wavelength is).
The word "wavelength" doesn't occur anywhere in the description of how the experimental apparatus was set up, how the observations were made, and what the observations were (as opposed to theoretical model-based descriptions).
I think you're making this a lot harder than it needs to be by refusing to distinguish between factual descriptions of an experiment and theoretical interpretations of the results.
Right below the word "results" is a table which correlates asmuths to "wavelengths" which I believe is a measure of how out-of-phase the arms of the interferometer were.
Yes, it's simpler to just assume that reality as you see it is full of facts that can be made sense of directly, but that's what prompted Planck's university adviser to claim that there was essentially nothing new to discover in physics. If you allow things to be "just true" without taking into account the things that make them true, you'll never take the kind of creative step that we see in Copernicus, Einstein, or Planck (despite his adviser's recommendations).
Scientific advancement happens when somebody faces a set of immutable facts and decides which one needs to become mutable in order for the rest to become more useful. If facts are flattened down to primitives, there is no underlying structure to aid in that decision.
And if course there is such a thing as "a fact of". '2 + 2 = 4' is a fact of arithmetic, 'things fall towards massive other things' is a fact of gravity, and 'Harry Potter is a Wizard' is a fact of that particular universe.
The dangerous notion that I'm trying to dispel is that there is any other sort--a raw fact that is portable across contexts.
You yourself seem like a pretty smart cookie, so I don't worry about you, but generally speaking I find that people who prefer to deal in absolutes are susceptible to manipulation by relativists such as myself. If I can manufacture a fact and insert it into your universe, you're kind of stuck with it because you treat them as out-of-scope for reconsideration.
If we allow science and authority to become synonymous--which the author of this article would have us do, we commit a similarly permanent act--one which shuts the door on advancement of the sort I've described to above.
I'm not trying to convince you that you are wrong. I'm trying to show you that your faith in the persistent and portable utility of facts is unjustified and represents both a creativity damper and an unnecessary attack surface--attack surface that would be exploited by maintainers of the status quo.
Or I would try to convince you of that, if I could get us past the unjustified part.
> Right below the word "results" is a table which correlates asmuths to "wavelengths"
Ah, I see; yes, the word "wavelengths" does appear there, and it also appears on the previous page, where they say: "the length of the path of a ray in our apparatus was 3224 centimetres, in which distance there are contained 5.5 x 10^7 wave-lengths of sodium light". So "wavelengths" here refers to the calculated wavelength of a particular energy level transition in sodium atoms that was used to provide the light in the interferometer. So the wavelengths in the table at the top of page 5 are measured numbers (the sizes of the interference fringes observed), translated into a particular choice of units (units of sodium energy level transition wavelengths). The numbers themselves are facts, but the choice of units is theoretically motivated, yes--a more theoretically neutral choice of units would be something like centimeters.
> it's simpler to just assume that reality as you see it is full of facts that can be made sense of directly, but that's what prompted Planck's university adviser to claim that there was essentially nothing new to discover in physics
I don't see this at all. It was precisely particular facts--facts that did not fit Newtonian theory--that prompted the development of relativity and quantum mechanics as better theories, that accounted for those facts. The whole idea of facts not fitting a theory requires you to have a concept of "facts" that does not depend on which theory you choose.
> Scientific advancement happens when somebody faces a set of immutable facts and decides which one needs to become mutable in order for the rest to become more useful.
Again, I think you are conflating "fact" with "theory" and hence you are confusing yourself. Here is how I would state what I think you are trying to say here: scientific advancement happens when somebody faces a set of facts that do not fit any currently accepted theory, and comes up with a new theory whose predictions match those facts. No facts change during this process; only theories change.
> If I can manufacture a fact and insert it into your universe
I have no idea how you would go about doing that except by lying about something you observed. Which you could do--people have certainly lied to other people in the past and had their lies accepted, at least for a while--but I don't think that's a problem with people being too absolutist about facts. I think it's a problem with (a) some people lying, and (b) other people not being good at detecting lies.
And I certainly don't see how thinking of facts as theory-dependent helps a person to detect lies. Rather, it makes it harder to detect lies, because the whole concept of "lies" gets muddled: there is no objective truth, just different people's theories, so how can you say that this other person who manufactured a fact and inserted it into your universe was wrong? He was just telling his theory, and every theory is just as good as every other one.
The only way to escape this merry-go-round and defend yourself against this kind of manipulation is to have a concept of "facts" that is independent of your choice of theory, which allows you to have the concept of "this person is lying to me so I shouldn't believe what they say". How can you even have that concept at all if facts are theory-dependent?
> If we allow science and authority to become synonymous
Which is certainly not what I am advocating. In fact I'm advocating precisely the opposite, as you can see from other posts I've made elsewhere in this thread. Scientists trying to present science as an authority is a problem; and the fact that the author of the article wants to do that means he's wrong.
> a more theoretically neutral choice of units would be something like centimeters
Would it? A centimeter is defined in terms of the speed of light.
This isn't me being nitpicky, everything is defined in terms of something else, and to hold a justified belief (I won't call it a fact) requires the ability to place the belief in context with other justified beliefs to see:
- If any contracictions arise
- If the new belief can be used to justify other beliefs in the system.
These systems of belief form something like a directed graph (edge=justifies), and I suspect that they are kind of clumpy. Like, the majority of the beliefs only justify their immediate neighbors in the clump (which I won't call a theory). Rare are the edges in this graph that point from one clump to another, these are achieved by experiment. The experiment we have been discussing forms an edge between one clump (optics/electrodynamics) and another clump (motion). It was contact with "the facts" that allowed this bridge to be built.
I believe you would call whichever clump is more justified "the facts" and whichever one is less justified "the model being tested".
I suspect it is impossible, though it may just be devilishly hard, to know the difference between a fact, a very well justified belief, and a strongly held oppinion. None of the research I have seen indicates that humans are good at that making this kind of judgement. So we should always be a little suspicious that something that we perceive as a fact is actually one of the other two. If you encounter an apparent contradiction out there in the wild, one approach would be to tighten your grip on your justified beliefs, treat them as facts, and then call the newcomer a lie. But there are alternatives, which I will discuss in a moment.
When I claimed that people could manipulate absolutists by injecting "facts" into the absolutist's "universe", perhaps I should have said that what they are injecting is beliefs that are likely to be misapprehended as facts, and they get injected into the system of justified beliefs that is held by the victim. For more on these techniques, see: rhetoric. (Which isn't evil by the way, as some like to paint it. Whether or not the audience accepts the edits permanently, all effective communication involves the deliberate manipulation of this justification lattice in your audiences' mind.)
Relativists (or at least myself, since I have insight into only one mind), have more options when they run into something that is dresseed up like a fact, but causes contradictions. The interloper could be:
- A lie
- A justifiable belief, but one that is justifiable in a system that is inferior to mine
- A justifiable belief, but one that is justifiable in a system that is superior to mine
I'm constantly reworking what counts as true according to my belief system, so if a lie slips in it's not as big of a deal. Unless it's a very crafty lie, I'm likely to have edited it out in a few weeks anyway (since it neither justifies nore is justified by other beliefs in the system). Most of my cycles go into evaluating whether these things that appear to me as facts form a system that is better or worse than some alternative, and to which my system is itself an alternative (hence: alternative facts).
Here "better or worse" could mean a lot of things. It could have to do with contact with reality (i.e. I want a maximally coherent system) It could have to do with beauty It could have to do with something less lofty, like whether it helps me secure food for myself tonight
Both myself and the absolutist must take something on faith. The absolutist takes it for granted that his facts are the facts. I take for granted my ability to chose a "best" belief system (which I call a set of theories, but I'm open to a new term) There's a good article called "The Raft and The Pyramid" (https://www.andrew.cmu.edu/user/kk3n/epistclass/Sosa%20-%20R...) which shows that these two approaches are equivalent in their ability to bring us closer to the truth. But they are not equivalent in how they affect our behavior.
The relativist may consider structural details of proposed facts, while the absolutist is inclined to reject them outright. This has consequences all over the place, many of which are the problems the author (of the original article) indicates. Absolutists don't consider a candidate fact in the context of its enclosing theory--er, clump of justified beliefs--so they don't get a picture of the potenial richness of the world they are rejecting, lies are just lies and that's it. They are not accustomed to evaluating the merit of whole alternative systems, so if they are delusional they're likely to just keep calling things lies and dig there heels in further. It's not lies that we should be worried about, they're going to happen, it's persistent, contageous delusion of the sort whose entry is a one-way trip. As I'll argue later, familiarizing people with the equipment of relativism (i.e. my interpretation of Feyerabend--I'll confess to not having read the whole book) is the antidote.
To answer your question:
> How can you say that this other person who manufactured a fact and inserted it into your universe was wrong?
The interloper fact (unless it's a very bad lie) will have some set of beliefs that justify it. To use my original language for a moment, every statement has a theory where it is true (that's what theories are, systems for evaluating the truth or falsity of statements and assigning labels to phenomena)--but there is no guarantee that those theories are useful. So you set your worldview down--which is scary the first time you do it--and you start importing those beliefs, and you also import the beliefs that justify them, and you keep pulling on that thread untill you see just how big of a world you're dismissing by calling the interloper fact a lie. If that world is not good as the one you set down, you go back and decide that it's not for you.
So I guess the answer is that you can't, because although it's wrong in your universe, it's right in some other. But you can still reject it, and having gone in there and checked it out first, you may now understand more about this would-be liar. You may be able to show them a better way, or failing that you may be able to work out a compromise (in behavior) that doesn't require epistemic agreement. You may also come away with a bit of respect for the opposing theory, and rather than dismissing it outright, you might reconsider it from time to time.
This only works, of course, if you know you can make it back out. Otherwise there might be a system of thought so insidious that once you import it for examination, you then can't leave. So when I go exploring other belief systems (like yours, and like the author's) the only "fact" that I take with me is this: All facts are potentially mutable, except this one. It's that safety belt that gives me this reflex for opposing your position about facts.
> Would it? A centimeter is defined in terms of the speed of light.
It's defined in terms of the length of the path traveled by light in 9192631770/299792458 ticks of a cesium clock under particular conditions. In other words, it's defined in terms of facts that can be observed and measured without having to adopt any particular theory. This definition makes the "speed of light in vacuum" a universal constant equal to 299792458 meters per second, but that does not change anything about its being based on facts that can be observed and measured without having to adopt any particular theory.
> I believe you would call whichever clump is more justified "the facts" and whichever one is less justified "the model being tested".
Nope. Nothing you've said here has anything to do with the fact-model distinction. In each of your "clumps" there will be both facts and models.
> what counts as true according to my belief system
Your belief system is a theoretical model. Does that help?
The rest of your post is just more of the same confusion, and at this point I don't see much use in arguing about it any further. I've made my position as clear as I can.
Huh? You just described it: reality is the authority.
> I think this situation is simpler if these all count as separate facts.
No, they are different theoretical descriptions of the facts. The facts are the particular experimental apparatus Michelson and Morley used and the particular numbers they recorded when they made their measurements.
> Do "the facts of gravity" include the predictions made by the theory of gravity?
Of course not. Predictions aren't facts. They're things that you compare with facts in order to test theoretical models.
> Speaking of Einstein, before he pubished his paper on special relativity, wasn't it a fact that the velocity of a bullet fired from a train (relative to the track) is the sum of the train's velocity and the bullet's velocity?
No. As far as I know nobody at that time had made accurate enough measurements of the velocity of anything to be able to test the Newtonian prediction vs. the relativity prediction. But that doesn't mean the Newtonian prediction was correct before relativity was discovered. It certainly doesn't mean that something changed about the world described by physical models when physicists switched from using the Newtonian model to using the relativity model. All that changed was the model: the world the model was describing didn't change.
> Given that we can't know which facts might later need to be relativized
Facts never need to be relativized. Relativizing is something that happens when you switch theoretical models. You seem to be consistently confusing facts with models.