How would one publish a non-mathematical theory?

[Jahee]

I was wondering that if there was a theory that didn't have a mathematical formula yet but had experimental data how would a scientist go about publishing it?

This isn't what I'm doing I'm just trying to see a possible process. Additional circumstances would be that the person publishing is a known but not necessarily famous theoretical physicist.

This is for a story. I wanted it to be relatively grounded in reality.

 

[DaveC426913]

Would be kind of unusual to have experimental data yet not have the mathematics to extract the principles from the data that support the theory.

There's got to be a way to connect the dots from data to theory - and that's usually the mathematical analysis.

IOW, I'm not really sure how one might achieve what you ask: publish a theory without the math to back it up.

Seems to me, what it would be is a hypothesis.

But I am no scientist and know nothing about publishing papers.

 

[mfb]

New experimental results that are not predicted by a theory are published somewhat regularly in several fields.

But a theory is inherently mathematics. "Theory without mathematics" is like "apples without fruits".

 

[jedishrfu]

Some years ago, there was a new AI system developed at Cornell U now named Eureqa that could discern the equations of motion just from data gleaned from measurements on a system liked a compound pendulum. It was quite impressive.

Later other scientists in Biology used the system to get equations from their cell measurement data. It worked it and they got the equations they wanted. But there was a problem. They couldn't publish anything because they couldn't explain the equations they had.

So in a sense this is what you are looking for where we have the data, we have the equations but we have no theory of how it works from first principles.

https://en.wikipedia.org/wiki/Eureqa

 

[Dale]

Several posts were deleted which did not address the OP's question. He wants to know the PROCESS for getting such a thing published, not asking whether or not such a thing has ever been published.

@Jahee can you specify if this story is a historical fiction where a historical process would be relevant or is it a modern/future story where modern/future processes would be desired

 

[Jahee]

Several posts were deleted which did not address the OP's question. He wants to know the PROCESS for getting such a thing published, not asking whether or not such a thing has ever been published.

@Jahee can you specify if this story is a historical fiction where a historical process would be relevant or is it a modern/future story where modern/future processes would be desired

This would be a modern story.

A little more information on the story: A new piece of evidence has come to light that represents how things truly work. This bit of evidence isn't originally in the form of any numbers, just a presentation. An example I could give without going into the story too much would be if someone proved to you time travel was always possible without a device and people just had to take a few steps to achieve it. And then proved it right in front of the scientist.

I was thinking that the scientist would try to quantify the range that they could influence. Like scale of how much they could bring to the past, would the amount of mass matter. Different questions like that. Then when they got as far as they could they would publish and try to bring more people into help figure out more about this new phenomena.

Would've posted a reply earlier but I just came back.

Hopefully it also answers why they would have data without a mathematical formula. Maybe a hypothesis would be a better word though. [I'm actually not too sure if hypothesis would be the right word either as I'm laying out a situation where one would have to accept something as fact before trying to build the explanation. That would mean that it isn't falsifiable. As with my time traveler example we have knowledge that a seemingly nonsensical fact is plopped into their laps and now as scientist they want to try to give it values that make some sense. I would want to say that they data they found isn't actually bearing enough weight on the phenomena but more on the possible effects it has on people. Maybe I could say that they have some tentative equations to start.]

@jedishrfu I don't know if the biologist situation would be the same as the one in my story. Because they have the effects and the scale. But not a mathematical formula to represent the phenomena.

 

[jedishrfu]

There's a lot of resources for future writers online. One such interview with John Malik, author of Dragon's Trail about self-publishing is interesting:

https://blog.publishdrive.com/self-publishing-sci-fi-fantasy/

There's also the Saves the Cat books that talk about writing screenplays and novels, how to construct the story, how to sell your creation, pitfalls...

I like the Saves the Cat book itself and the newest one Saves theCat Writes a Novel by Jessica Brody that transforms the notion of writing a screenplay to writing a novel using the Saves the Cat principles in the story.

Blake Snyder developed the original techniques in the book and developed the 10 categories of movies that helps you with story flow:

https://en.wikipedia.org/wiki/Blake_Snyder#Save_the_Cat.21_Screenwriting_Manual

 

[Jahee]

Would be kind of unusual to have experimental data yet not have the mathematics to extract the principles from the data that support the theory.

There's got to be a way to connect the dots from data to theory - and that's usually the mathematical analysis.

IOW, I'm not really sure how one might achieve what you ask: publish a theory without the math to back it up.

Seems to me, what it would be is a hypothesis.

But I am no scientist and know nothing about publishing papers.

I posted a general reply below but thank you for what you mentioned. I may have been a bit liberal with calling it a theory but I'm not sure it could fall under theory or hypothesis. What they have is more something along the lines of a fact (I know we steer clear with calling something as hard lined as a fact but this is the best word I guess I could come up with. Law doesn't really work because the phenomena hasn't been extensively tested by multiple people yet).

 

[DaveC426913]

Hopefully it also answers why they would have data without a mathematical formula. Maybe a hypothesis would be a better word though. [I'm actually not too sure if hypothesis would be the right word either as I'm laying out a situation where one would have to accept something as fact before trying to build the explanation. That would mean that it isn't falsifiable. As with my time traveler example we have knowledge that a seemingly nonsensical fact is plopped into their laps and now as scientist they want to try to give it values that make some sense. I would want to say that they data they found isn't actually bearing enough weight on the phenomena but more on the possible effects it has on people. Maybe I could say that they have some tentative equations to start.]

This sounds like a phenomenon. It's observed, can't be denied, but is yet to be explained.
Another term might be The <Discoverer's Name> Effect.

 

[Jahee]

How about a conjecture then?

*retracted

That probably works the best. Hmm. I guess the question then becomes can you request for other scientists to take a look at something that is mere conjecture. That is for me to figure out. Thank you for helping me find the most appropriate word. I think that is the choice my character would make.

 

[DaveC426913]

That probably works the best.

Sorry, I retracted that suggestion, after going back and reading your post #6.

 

[Jahee]

This sounds like a phenomenon. It's observed, can't be denied, but is yet to be explained.
Another term might be The <Discoverer's Name> Effect.

I see. I'll do a bit of looking around regarding the use of that.

 

[Jahee]

There's a lot of resources for future writers online. One such interview with John Malik, author of Dragon's Trail about self-publishing is interesting:

https://blog.publishdrive.com/self-publishing-sci-fi-fantasy/

There's also the Saves the Cat books that talk about writing screenplays and novels, how to construct the story, how to sell your creation, pitfalls...

I like the Saves the Cat book itself and the newest one Saves theCat Writes a Novel by Jessica Brody that transforms the notion of writing a screenplay to writing a novel using the Saves the Cat principles in the story.

Blake Snyder developed the original techniques in the book and developed the 10 categories of movies that helps you with story flow:

https://en.wikipedia.org/wiki/Blake_Snyder#Save_the_Cat.21_Screenwriting_Manual

Thank you for the writers reference material.

 

[Stephen Tashi]

Additional circumstances would be that the person publishing is a known but not necessarily famous theoretical physicist.

There are non-mathematical disciplines that use symbols and diagrams. Examples are linquistics and the grammatical analysis of languages. An interesting fictional situation would be that a scientist advocates a theory that some aspect of nature is communicating via a language and uses linguistics to support his argument.

I don't know how linguistics works. I once took a logic class that met in a classroom just after a linguistics class. The linguistics teacher did not erase the blackboard before he left, so all the symbols and diagrams were left for our class to see. I didn't understand them.

 

[jedishrfu]

Something like these linguistics symbols for phonetic sounds:

http://www.englishbiz.co.uk/grammar/main_files/definitionsn-z.htm

https://www.ling.upenn.edu/courses/Fall_2014/ling115/phonetics.html

 

[Jahee]

There are non-mathematical disciplines that use symbols and diagrams. Examples are linquistics and the grammatical analysis of languages. An interesting fictional situation would be that a scientist advocates a theory that some aspect of nature is communicating via a language and uses linguistics to support his argument.

I don't know how linguistics works. I once took a logic class that met in a classroom just after a linguistics class. The linguistics teacher did not erase the blackboard before he left, so all the symbols and diagrams were left for our class to see. I didn't understand them.

My bad. I didn't put in commas. He is a theoretical physicist, but he isn't a highly regarded one.

 

[Stephen Tashi]

Something like these linguistics symbols for phonetic sounds:

No, the diagrams looked like a little like those we used in high school to "diagram sentences". However, I couldn't interpret them in that manner. They did not seem to deal with phonetics.

 

[mfb]

What you describe is a purely experimental result. It's not a theory, the author being a theoretical physicist by occupation doesn't matter. It is an experimental result and can be published as such. "Perform these steps to time travel".

This sounds like a phenomenon. It's observed, can't be denied, but is yet to be explained.
Another term might be The <Discoverer's Name> Effect.

Stigler's law of eponymy - nothing is named after its original discoverer. And as you have guessed, Stigler was not the first to discover this rule.

 

[member 656954]

I was wondering that if there was a theory that didn't have a mathematical formula yet but had experimental data how would a scientist go about publishing it?

There are a number of publishing sites that will accept content of this kind, usually for a fee, that are not as strict about aspects such as peer review. A scientist should be able to write their article in a format suitable for these sites, even if they had no theoretical explanation for the data, because content creation for publication is very likely a necessary part of their job.

Wikipedia has a list of such, https://en.wikipedia.org/wiki/List_of_academic_journals_by_preprint_policy, or you could make your own preprint platform name up for the purpose of your story.

The platforms usually describe their process of publishing, and you can often read articles to get an idea of their format, but unless you're getting into the weeds, a little handwavium regarding the process should be sufficient for a sci fi novel.

Alternatively, if the discovery was sufficiently amazing, the scientist could go straight to the media. They run the risk of being lambasted if it turns out to be nothing - think the cold fusion situation - but if they're sure of their results, and can reliably repeat them, that's another avenue for getting the news out there.

 

[compcomplexity]

Some years ago, there was a new AI system developed at Cornell U now named Eureqa that could discern the equations of motion just from data gleaned from measurements on a system liked a compound pendulum. It was quite impressive.

Later other scientists in Biology used the system to get equations from their cell measurement data. It worked it and they got the equations they wanted. But there was a problem. They couldn't publish anything because they couldn't explain the equations they had.

So in a sense this is what you are looking for where we have the data, we have the equations but we have no theory of how it works from first principles.

https://en.wikipedia.org/wiki/Eureqa

Just an FYI: Eureqa has been removed from the market in 2017 after being purchased by a consulting company.

Recently an alternative called TuringBot has been developed, and it has been shown in arXiv:2010.11328 to be more efficient than Eureqa at finding formulas. The authors used Physics-inspired problems extracted from The Feynman Lectures on Physics to do this comparison, which was nice.

 

[jedishrfu]

Thanks for the info on TuringBot. They have a nice page that even points to DataRobot where Eureqa now resides. It appears though that Eureqa is still available on the DataRobot site as is some tutorials on how to use Eureqa.

I'll need to check out TuringBot more. They use simulated annealing over genetic algorithms. I did a partial implementation using GA and Python. In my case, I was interested in generating the equations to test.

I saw another symbolic regression tool that used units of measure to further hone their equation filtering. I thought that could reduce the equation search space even more.

 

[jedishrfu]

Here's a link to Eureqa at DataRobot. Apparently, they hide it unless you search for it.

https://www.datarobot.com/nutonian/

and more specifically:

https://community.datarobot.com/t5/eureqa-academic/tkb-p/EureqaID