How Does Science Define Truth?

[BAYLOR]

Yes I know probably might not be an unanswerable question and probably not quantifiable but, things have been getting a bit quite in this section of the forum and I figured , why not? What are are you thoughts on the definition of Truth?

Let the speculation begin, silly and serious begin .

 

[Dennis E. Taylor]

Science doesn't define Truth (with a capital T). It defines statements as falsifiable or non-falsifiable.

 

[BAYLOR]

Science doesn't define Truth (with a capital T). It defines statements as falsifiable or non-falsifiable.

Water is liquid, gravity falls causes things to fall down ect.


Nothing outside the box.

 

[Ray McCarthy]

Water is liquid, gravity falls causes things to fall down ect.

Water can be plasma, gas, liquid, several different kinds of solid (ice).

We don't know what gravity is. Negative mass (if it exists) will "fall up". Falling down and up is inaccurate. Bodies experience an attraction force proportional to sum of masses and inversely the square of distance, we call this "gravity".

 

[The Ace]

Science doesn't define Truth (with a capital T). It defines statements as falsifiable or non-falsifiable.

Yup, if you want Truth, try philosophy.

Science extrapolates from observable facts, and tests the resulting hypotheses.

 

[Stephen Palmer]

Science assumes the independent existence of the world - independent of human imagination, which was the original notion - a real world that we can test etc. Thank goodness for that.

 

[Droflet]

I suppose if science comes up with a lie detector that cannot be beat, truth would be the outcome. (I'm thinking of the book Little Fuzzy).
But I tend to go with the majority on this one. Honest, I'm not lying.

 

[Venusian Broon]

Yes, science adds very little to the deep understanding of the question 'what is truth'. Most science is in fact shallow flim-flam of the 'well it works, so we'll just accept whatever that implies for the moment till we find something that changes our mind.' Hardly fertile ground for really getting a firm handle on the truth.

So we must go to proper philosophy to people who do ponder this very question. Roger Scruton's Lectures in Modern Philosphy: An Introduction and Survey, (Chapter 9) has a nice summing up of what we might expect if we got deeper into 'Truth' with them:

"What is it, for something to be true or false? Intuitively truth is a relation - between the thing that is true, and the thing that makes it so. But both terms of the relation are in dispute, as is the relation itself. Philosophers differ as to whether the 'truth-bearer' is a sentence, a proposition, a thought, a statement, a belief, or some other entity, whether linguistic or mental. They differ too as to what truth consists in. Some speak of correspondence - but with what? (And here again there are a variety of positions, summarised in the terms 'fact', 'situation', 'reality' and 'state of affairs'.) Others replace correspondence with some other relation: coherence, for example. Others still reject the whole idea of truth as a relation, regarding it instead as an intrinsic property of whatever possesses it. There are even those who argue that truth is neither a property nor a relation, and that the concept is merely redundant."

Personally, such discussions, which can be highly interesting and fascinating, really don't seem to move me any closer to what Truth actually is...

...However many of the questions posed by Greek thinkers in the 6th Century BC remain unanswered, such is the world we live in. Perhaps, if a few decades/hundreds of years of hard thinking could 'solve and answer everything', then perhaps we'd be in a universe that was too simple for thoughtful beings to exist in?

 

[Ray McCarthy]

Mathematics is derived from Logic. In the mathematical sense an entity / relationship can be True (2x2 = 4), False (2 x 2 = 5), Ambiguous ( x=y where we don't know y) or inherently nonsense ("This statement is false"), usually called a paradox.
But this is very narrow definition of true.
Set Theory and cardinal counting can be derived from Logic.
All mathematics can thus derived from logic. A computer has a CPU made from subsystems that are defined in pure logic "Truth tables", a table of true and false outputs for all possible combinations of true and false on the inputs. This CPU (containing the ALU, Arithmetic Logic Unit made from logic gates) then executes the program in a totally deterministic way (the same program instructions with same data give same answer always). Our brains do not work like this, no matter what some people claim.
You can replace all the electronic logic gates with relays or cogwheels & pins* that have the same truth tables and the same program and data will give the same results, just a lot slower. Complexity only allows faster execution of the program, or more complex compound instructions, or programs running in parallel. Never ever will more memory capacity or circuit complexity make it more "intelligent". It has no original thought or consciousness. It's only a collection of interconnected truth tables "executing" program instructions that are 100% based on logic and deterministic. It's a machine running on Truth. If true A.I. was something we knew how to program we could have had a slow one over 50 years ago. More powerful / faster / complex computers makes it no easier a problem to solve. Essentially we have no idea how to codify any aspect of Intelligence, creativity, problem solving or originality as tables of True and False. We are not even sure what they are!

[* a totally mechanical copy of an Intel Pentium, or rather easier, a 6502 CPU is possible. It would be very very slow. Programs would be stored and run from punched cards or paper tape, like in fact computers used to use, derived from Hollerith census machine cards (Hollerith became IBM), developed from Jacquard Loom 1801 programming cards. Player piano is a similar idea. The motive power is anything to turn the drums and cogs, such as water, wind, "clockwork"(wound spring), steam engine or even electric engine. This is what Babbage was trying to do. He ran out of time and money to build it. Conrad Zuse in Germany, about 1938 made an electro-mechanical computer out of relays]

 

[Dennis E. Taylor]

I suppose if science comes up with a lie detector that cannot be beat, truth would be the outcome. (I'm thinking of the book Little Fuzzy).
But I tend to go with the majority on this one. Honest, I'm not lying.

But here's the problem with that kind of 'truth'. If you put a creationist under such a lie detector and asked them if God exists and if the world is 6000 years old, they'd say yes and the machine would register it as the truth. A lot of the time, capital-t-Truth is conflated with strong belief.

 

[Venusian Broon]

All mathematics can thus derived from logic.

Although mathematics does have a problem with logic deep down - Godel's incompleteness theorem did actually proved that.

"there are true statements expressible in its [the formal system of mathematics] language that are unprovable within the system. Thus no formal system (satisfying the hypotheses of the theorem) that aims to characterize the natural numbers can actually do so, as there will be true number-theoretical statements that that system cannot prove."

So I'd disagree with you on the broadest definition. It does utilise logic all the time, so in the practical sense, yes (which is what I think you're saying )

 

[Ray McCarthy]

It does utilise logic all the time, so in the practical sense, yes (which is what I think you're saying )

It is.
There is a definitional problem that all of logic and mathematics can't be fully defined within the system.

For any such system, there will always be statements about the natural numbers that are true, but that are unprovable within the system. The second incompleteness theorem, an extension of the first, shows that such a system cannot demonstrate its own consistency.

Godel's incompleteness theorems
So it's back to Truth. In the most basic sense we can't use a Truth Table, simple clear true/false statements to prove every axiom of mathematics. Nevertheless, on a practical level EVERY mathematical operation (and indeed EVERY operation) of a computer is 100% definable in terms of true / false tables, basic logic. Yet proving a computer program is 100% correct ("true") is almost impossible, but not for Godel's reasons. It's a more subtle problem that C.A.R. (Tony) Hoare solved using the Occam Programming language, but only for the special case of pure single threaded sequential programs in the language (Occam is a language for Parallel Programming, originally to support the Transputer. Hoare also invented/discovered "Quick Sort" and worked on early development of Pascal (and also taught my wife to program).

Hilbert was a genius and seriously hurt other mathematician's heads with his musings on Infinity.
https://en.wikipedia.org/wiki/Hilbert's_paradox_of_the_Grand_Hotel
He did many other wonderful things in mathematics, which, is as close to "truth" as normally we can get. (Formal Logic is a part of mathematics).

 

[Ray McCarthy]

lie detector

They have been proved time and again to be tech fake snake oil. But as you say, even if they did "work" they can't as they'd only be belief detectors, not truth or lie detectors. It's dubious that they can ever "work" with someone trained to defeat them. Even with drugs and "brain washing" (or any form of torture) there is the likelihood the investigator is told what they want to hear, or nonsense. This is what makes the CIA/USA/UK behaviour over "rendition", "terrorists", "waterboarding" etc so bizarre. This is well known.

 

[psikeyhackr]

That which can be verified repeatedly by experiment.

psik

 

[hardsciencefanagain]

Science just deals with the plausible and the probable

 

[Ray McCarthy]

Science just deals with the plausible and the probable

Really?
A lot of it isn't how you'd think stuff ought to work or be arranged.

 

[BAYLOR]

Science just deals with the plausible and the probable

But would that limit science a bit ?

 

[psikeyhackr]

But would that limit science a bit ?

For many years a lot of physicists did not think Black Holes were plausible. Reality does not care what people think.

Some science gets more and more intangible and based on conclusions of observations and mathematics like Dark Matter and Dark energy. We don't know what they are we just know their effects and make guesses. Science is also a process of figuring things out, so imagination and guesses and mistakes enter into it.

And intellectual ego games.

psik

 

[BAYLOR]

For many years a lot of physicists did not think Black Holes were plausible. Reality does not care what people think.

Some science gets more and more intangible and based on conclusions of observations and mathematics like Dark Matter and Dark energy. We don't know what they are we just know their effects and make guesses. Science is also a process of figuring things out, so imagination and guesses and mistakes enter into it.

And intellectual ego games.

psik

The 2 dimensional characters in Edwin Abbott's book Flatland seem to have a difficult with the concept of a third dimension. That's what we're like .

 

[tinkerdan]

The biggest problem and danger with lie detectors today is not the inability to detect lies; but it is more the ability of the operator to manipulate the person being examined. Someone who is led to believe that the lie detector is accurate can be manipulated just by someone responding to the last answer as though it were a lie and creating an uncertainty in the believer until they begin to question themselves and if that particular subject may be one they know they were not wholly truthful about it can lead to a cascade of confession that is based mostly on a gut feeling the operator had despite the lie detector showing nothing but truth.

I have no scientific knowledge to back this up.

Only the information from a conference for a job I worked, where a 'security specialist' (and lie-detector operator) explained how he obtained confessions using a lie-detectors.